- * Users of this class are encouraged to use a {@link TreeSet} with the {@link String#CASE_INSENSITIVE_ORDER} - * comparator if case-insensitive comparison is needed (like when dealing with HTML tags). - */ - public static class SetEmptyElementHandler - implements EmptyElementHandler - { - final protected Set mEmptyElements; - - public SetEmptyElementHandler(Set emptyElements) - { - mEmptyElements = emptyElements; - } - - public boolean allowEmptyElement(String prefix, String localName, String nsURI, boolean allowEmpty) - { - return mEmptyElements.contains(localName); - } - } - - /** - * HTML specific empty element handler. - * Extends the {@link SetEmptyElementHandler} and configures - * the HTML elements that must be self-closing according to the W3C: - * http://www.w3.org/TR/html4/index/elements.html - *
- * Note that element name comparison is case-insensitive as required - * by HTML specification. - */ - public static class HtmlEmptyElementHandler - extends SetEmptyElementHandler - { - private final static HtmlEmptyElementHandler sInstance = new HtmlEmptyElementHandler(); - - public static HtmlEmptyElementHandler getInstance() { return sInstance; } - - protected HtmlEmptyElementHandler() - { - super(new TreeSet(String.CASE_INSENSITIVE_ORDER)); - mEmptyElements.add("area"); - mEmptyElements.add("base"); - mEmptyElements.add("basefont"); - mEmptyElements.add("br"); - mEmptyElements.add("col"); - mEmptyElements.add("frame"); - mEmptyElements.add("hr"); - mEmptyElements.add("input"); - mEmptyElements.add("img"); - mEmptyElements.add("isindex"); - mEmptyElements.add("link"); - mEmptyElements.add("meta"); - mEmptyElements.add("param"); - } - } -} - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/InvalidCharHandler.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/InvalidCharHandler.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/InvalidCharHandler.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/InvalidCharHandler.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,91 +0,0 @@ -package com.ctc.wstx.api; - -import java.io.IOException; - -/** - * Simple converter interface designed to be used with stream writer property - * {@link WstxOutputProperties#P_OUTPUT_INVALID_CHAR_HANDLER}. - * The idea is that it should be easy to have a way to convert invalid - * characters such as Ascii control characters into something that - * is legal to include in XML content. This only allows for simple - * char-by-char replacements, instead of something more advanced such - * as escaping. If escaping is needed, check out - * {@link org.codehaus.stax2.XMLOutputFactory2#P_TEXT_ESCAPER} instead. - *
- * Note about exceptions: choice of only allowing throwing of
- * {@link IOException}s is due to the way Woodstox stream writer
- * backend works; XmlWriter can only throw IOExceptions.
- */
-public interface InvalidCharHandler
-{
- public char convertInvalidChar(int invalidChar) throws IOException;
-
- /**
- * This handler implementation just throws an exception for
- * all invalid characters encountered. It is the default handler
- * used if nothing else has been specified.
- */
- public static class FailingHandler
- implements InvalidCharHandler
- {
- public final static int SURR1_FIRST = 0xD800;
- public final static int SURR1_LAST = 0xDBFF;
- public final static int SURR2_FIRST = 0xDC00;
- public final static int SURR2_LAST = 0xDFFF;
-
- private final static FailingHandler sInstance = new FailingHandler();
-
- protected FailingHandler() { }
-
- public static FailingHandler getInstance() { return sInstance; }
-
- public char convertInvalidChar(int c) throws IOException
- {
- /* 17-May-2006, TSa: Would really be useful if we could throw
- * XMLStreamExceptions; esp. to indicate actual output location.
- * However, this causes problem with methods that call us and
- * can only throw IOExceptions (when invoked via Writer proxy).
- * Need to figure out how to resolve this.
- */
- if (c == 0) {
- throw new IOException("Invalid null character in text to output");
- }
- if (c < ' ' || (c >= 0x7F && c <= 0x9F)) {
- String msg = "Invalid white space character (0x"+Integer.toHexString(c)+") in text to output (in xml 1.1, could output as a character entity)";
- throw new IOException(msg);
- }
- if (c > 0x10FFFF) {
- throw new IOException("Illegal unicode character point (0x"+Integer.toHexString(c)+") to output; max is 0x10FFFF as per RFC 3629");
- }
- /* Surrogate pair in non-quotable (not text or attribute value)
- * content, and non-unicode encoding (ISO-8859-x, Ascii)?
- */
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- throw new IOException("Illegal surrogate pair -- can only be output via character entities, which are not allowed in this content");
- }
- throw new IOException("Invalid XML character (0x"+Integer.toHexString(c)+") in text to output");
- }
- }
-
- /**
- * Alternative to the default handler, this handler converts all invalid
- * characters to the specified output character. That character will
- * not be further verified or modified by the stream writer.
- */
- public static class ReplacingHandler
- implements InvalidCharHandler
- {
- final char mReplacementChar;
-
- public ReplacingHandler(char c)
- {
- mReplacementChar = c;
- }
-
- public char convertInvalidChar(int c) throws IOException
- {
- return mReplacementChar;
- }
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,7 +0,0 @@
-
- * In addition to its main task as a configuration container, this class
- * also acts as a wrapper around simple buffer recycling functionality.
- * The reason is that while conceptually this is a separate concern,
- * there are enough commonalities with the life-cycle of this object to
- * make this a very convenience place to add that functionality...
- * (that is: conceptually this is not right, but from pragmatic viewpoint
- * it just makes sense)
- */
-public final class ReaderConfig
- extends CommonConfig
- implements InputConfigFlags
-{
- /*
- ///////////////////////////////////////////////////////////////////////
- // Constants for reader properties:
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // First, standard StAX properties:
-
- // Simple flags:
- final static int PROP_COALESCE_TEXT = 1;
- final static int PROP_NAMESPACE_AWARE = 2;
- final static int PROP_REPLACE_ENTITY_REFS = 3;
- final static int PROP_SUPPORT_EXTERNAL_ENTITIES = 4;
- final static int PROP_VALIDATE_AGAINST_DTD = 5;
- final static int PROP_SUPPORT_DTD = 6;
-
- // Object type properties
- public final static int PROP_EVENT_ALLOCATOR = 7;
- final static int PROP_WARNING_REPORTER = 8;
- final static int PROP_XML_RESOLVER = 9;
-
- // // Then StAX2 standard properties:
-
- // Simple flags:
- final static int PROP_INTERN_NS_URIS = 20;
- final static int PROP_INTERN_NAMES = 21;
- final static int PROP_REPORT_CDATA = 22;
- final static int PROP_REPORT_PROLOG_WS = 23;
- final static int PROP_PRESERVE_LOCATION = 24;
- final static int PROP_AUTO_CLOSE_INPUT = 25;
-
- // Enum / Object type properties:
- final static int PROP_SUPPORT_XMLID = 26; // shared with WriterConfig
- final static int PROP_DTD_OVERRIDE = 27;
-
- // // // Constants for additional Wstx properties:
-
- // Simple flags:
-
- /**
- * Note: this entry was deprecated for 4.0 versions up until
- * and including 4.0.7; was brought back for 4.0.8 (and will
- * be retained for 4.1)
- */
- final static int PROP_NORMALIZE_LFS = 40;
-
- /* This entry was deprecated for 3.2 and removed in 4.0
- * version. There are no plans to bring it back.
- */
- //final static int PROP_NORMALIZE_ATTR_VALUES = 41;
-
- final static int PROP_CACHE_DTDS = 42;
- final static int PROP_CACHE_DTDS_BY_PUBLIC_ID = 43;
- final static int PROP_LAZY_PARSING = 44;
- final static int PROP_SUPPORT_DTDPP = 45;
- final static int PROP_TREAT_CHAR_REFS_AS_ENTS = 46;
-
- // Object type properties:
-
- final static int PROP_INPUT_BUFFER_LENGTH = 50;
- //final static int PROP_TEXT_BUFFER_LENGTH = 51;
- final static int PROP_MIN_TEXT_SEGMENT = 52;
- final static int PROP_CUSTOM_INTERNAL_ENTITIES = 53;
- final static int PROP_DTD_RESOLVER = 54;
- final static int PROP_ENTITY_RESOLVER = 55;
- final static int PROP_UNDECLARED_ENTITY_RESOLVER = 56;
- final static int PROP_BASE_URL = 57;
- final static int PROP_INPUT_PARSING_MODE = 58;
-
- /*
- ////////////////////////////////////////////////
- // Limits for numeric properties
- ////////////////////////////////////////////////
- */
-
- /**
- * Need to set a minimum size, since there are some limitations to
- * smallest consequtive block that can be used.
- */
- final static int MIN_INPUT_BUFFER_LENGTH = 8; // 16 bytes
-
- /**
- * Let's allow caching of just a dozen DTDs... shouldn't really
- * matter, how many DTDs does one really use?
- */
- final static int DTD_CACHE_SIZE_J2SE = 12;
-
- final static int DTD_CACHE_SIZE_J2ME = 5;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Default values for custom properties:
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * By default, let's require minimum of 64 chars to be delivered
- * as shortest partial (piece of) text (CDATA, text) segment;
- * same for both J2ME subset and full readers. Prevents tiniest
- * runts from getting passed
- */
- final static int DEFAULT_SHORTEST_TEXT_SEGMENT = 64;
-
- /**
- * Default config flags are converted from individual settings,
- * to conform to StAX 1.0 specifications.
- */
- final static int DEFAULT_FLAGS_FULL =
- 0
- // First, default settings StAX specs dictate:
-
- | CFG_NAMESPACE_AWARE
- // Coalescing to be disabled
- //| CFG_COALESCE_TEXT
- | CFG_REPLACE_ENTITY_REFS
- | CFG_SUPPORT_EXTERNAL_ENTITIES
- | CFG_SUPPORT_DTD
-
- // and then custom setting defaults:
-
- // and namespace URI interning
- | CFG_INTERN_NAMES
- | CFG_INTERN_NS_URIS
-
- // we will also accurately report CDATA, by default
- | CFG_REPORT_CDATA
-
- /* 20-Jan-2006, TSa: As per discussions on stax-builders list
- * (and input from xml experts), 4.0 will revert to "do not
- * report SPACE events outside root element by default"
- * settings. Conceptually this is what xml specification
- * implies should be done: there is no content outside of
- * the element tree, including any ignorable content, just
- * processing instructions and comments.
- */
- //| CFG_REPORT_PROLOG_WS
-
- /* but enable DTD caching (if they are handled):
- * (... maybe J2ME subset shouldn't do it?)
- */
- | CFG_CACHE_DTDS
- /* 29-Mar-2006, TSa: But note, no caching by public-id, due
- * to problems with cases where public-id/system-id were
- * inconsistently used, leading to problems.
- */
-
- /* by default, let's also allow lazy parsing, since it tends
- * to improve performance
- */
- | CFG_LAZY_PARSING
-
- /* and also make Event objects preserve location info...
- * can be turned off for maximum performance
- */
- | CFG_PRESERVE_LOCATION
-
- // As per Stax 1.0 specs, we can not enable this by default:
- //| CFG_AUTO_CLOSE_INPUT);
-
- /* Also, let's enable dtd++ support (shouldn't hurt with non-dtd++
- * dtds)
- */
-
- | CFG_SUPPORT_DTDPP
-
- /*
- * Set this as a default, as this is required in xml;
- */
- | CFG_NORMALIZE_LFS
-
- /* Regarding Xml:id, let's enabled typing by default, but not
- * uniqueness validity checks: latter will be taken care of
- * by DTD validation if enabled, otherwise needs to be explicitly
- * enabled
- */
- | CFG_XMLID_TYPING
- // | CFG_XMLID_UNIQ_CHECKS
- ;
-
- /**
- * For now defaults for J2ME flags can be identical to 'full' set;
- * differences are in buffer sizes.
- */
- final static int DEFAULT_FLAGS_J2ME = DEFAULT_FLAGS_FULL;
-
- // // //
-
- /**
- * Map to use for converting from String property ids to ints
- * described above; useful to allow use of switch later on.
- */
- final static HashMap sProperties = new HashMap(64); // we have about 40 entries
- static {
- // Standard ones; support for features
- sProperties.put(XMLInputFactory.IS_COALESCING,
- DataUtil.Integer(PROP_COALESCE_TEXT));
- sProperties.put(XMLInputFactory.IS_NAMESPACE_AWARE,
- DataUtil.Integer(PROP_NAMESPACE_AWARE));
- sProperties.put(XMLInputFactory.IS_REPLACING_ENTITY_REFERENCES,
- DataUtil.Integer(PROP_REPLACE_ENTITY_REFS));
- sProperties.put(XMLInputFactory.IS_SUPPORTING_EXTERNAL_ENTITIES,
- DataUtil.Integer(PROP_SUPPORT_EXTERNAL_ENTITIES));
- sProperties.put(XMLInputFactory.IS_VALIDATING,
- DataUtil.Integer(PROP_VALIDATE_AGAINST_DTD));
- sProperties.put(XMLInputFactory.SUPPORT_DTD,
- DataUtil.Integer(PROP_SUPPORT_DTD));
-
- // Standard ones; pluggable components
- sProperties.put(XMLInputFactory.ALLOCATOR,
- DataUtil.Integer(PROP_EVENT_ALLOCATOR));
- sProperties.put(XMLInputFactory.REPORTER,
- DataUtil.Integer(PROP_WARNING_REPORTER));
- sProperties.put(XMLInputFactory.RESOLVER,
- DataUtil.Integer(PROP_XML_RESOLVER));
-
- // StAX2-introduced flags:
- sProperties.put(XMLInputFactory2.P_INTERN_NAMES,
- DataUtil.Integer(PROP_INTERN_NAMES));
- sProperties.put(XMLInputFactory2.P_INTERN_NS_URIS,
- DataUtil.Integer(PROP_INTERN_NS_URIS));
- sProperties.put(XMLInputFactory2.P_REPORT_CDATA,
- DataUtil.Integer(PROP_REPORT_CDATA));
- sProperties.put(XMLInputFactory2.P_REPORT_PROLOG_WHITESPACE,
- DataUtil.Integer(PROP_REPORT_PROLOG_WS));
- sProperties.put(XMLInputFactory2.P_PRESERVE_LOCATION,
- DataUtil.Integer(PROP_PRESERVE_LOCATION));
- sProperties.put(XMLInputFactory2.P_AUTO_CLOSE_INPUT,
- DataUtil.Integer(PROP_AUTO_CLOSE_INPUT));
- sProperties.put(XMLInputFactory2.XSP_SUPPORT_XMLID,
- DataUtil.Integer(PROP_SUPPORT_XMLID));
- sProperties.put(XMLInputFactory2.P_DTD_OVERRIDE,
- DataUtil.Integer(PROP_DTD_OVERRIDE));
-
- // Non-standard ones, flags:
-
- sProperties.put(WstxInputProperties.P_CACHE_DTDS,
- DataUtil.Integer(PROP_CACHE_DTDS));
- sProperties.put(WstxInputProperties.P_CACHE_DTDS_BY_PUBLIC_ID,
- DataUtil.Integer(PROP_CACHE_DTDS_BY_PUBLIC_ID));
- sProperties.put(XMLInputFactory2.P_LAZY_PARSING,
- DataUtil.Integer(PROP_LAZY_PARSING));
- sProperties.put(WstxInputProperties.P_SUPPORT_DTDPP,
- DataUtil.Integer(PROP_SUPPORT_DTDPP));
- sProperties.put(WstxInputProperties.P_TREAT_CHAR_REFS_AS_ENTS,
- DataUtil.Integer(PROP_TREAT_CHAR_REFS_AS_ENTS));
- sProperties.put(WstxInputProperties.P_NORMALIZE_LFS,
- DataUtil.Integer(PROP_NORMALIZE_LFS));
-
-
- // Non-standard ones, non-flags:
-
- sProperties.put(WstxInputProperties.P_INPUT_BUFFER_LENGTH,
- DataUtil.Integer(PROP_INPUT_BUFFER_LENGTH));
- sProperties.put(WstxInputProperties.P_MIN_TEXT_SEGMENT,
- DataUtil.Integer(PROP_MIN_TEXT_SEGMENT));
- sProperties.put(WstxInputProperties.P_CUSTOM_INTERNAL_ENTITIES,
- DataUtil.Integer(PROP_CUSTOM_INTERNAL_ENTITIES));
- sProperties.put(WstxInputProperties.P_DTD_RESOLVER,
- DataUtil.Integer(PROP_DTD_RESOLVER));
- sProperties.put(WstxInputProperties.P_ENTITY_RESOLVER,
- DataUtil.Integer(PROP_ENTITY_RESOLVER));
- sProperties.put(WstxInputProperties.P_UNDECLARED_ENTITY_RESOLVER,
- DataUtil.Integer(PROP_UNDECLARED_ENTITY_RESOLVER));
- sProperties.put(WstxInputProperties.P_BASE_URL,
- DataUtil.Integer(PROP_BASE_URL));
- sProperties.put(WstxInputProperties.P_INPUT_PARSING_MODE,
- DataUtil.Integer(PROP_INPUT_PARSING_MODE));
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Current config state:
- ///////////////////////////////////////////////////////////////////////
- */
-
- final boolean mIsJ2MESubset;
-
- final SymbolTable mSymbols;
-
- /**
- * Bitset that contains state of on/off properties; initialized
- * to defaults, but can be set/cleared.
- */
- int mConfigFlags;
-
- /**
- * Bitset that indicates explicit changes to {@link #mConfigFlags}
- * through calls; empty bit means that the corresponding property
- * has its default value, set bit that an explicit call has been
- * made.
- */
- int mConfigFlagMods;
-
- /**
- * 13-Nov-2008, tatus: Need to be able to keep track of whether
- * name-interning has been explicitly enabled/disable or not
- * (not if it's whatever defaults we have)
- */
- final static int PROP_INTERN_NAMES_EXPLICIT = 26;
- final static int PROP_INTERN_NS_URIS_EXPLICIT = 27;
-
-
- int mInputBufferLen;
- int mMinTextSegmentLen;
-
- /**
- * Base URL to use as the resolution context for relative entity
- * references
- */
- URL mBaseURL = null;
-
- /**
- * Parsing mode can be changed from the default xml compliant
- * behavior to one of alternate modes (fragment processing,
- * multiple document processing).
- */
- WstxInputProperties.ParsingMode mParsingMode =
- WstxInputProperties.PARSING_MODE_DOCUMENT;
-
- /**
- * This boolean flag is set if the input document requires
- * xml 1.1 (or above) compliant processing: default is xml 1.0
- * compliant. Note that unlike most other properties, this
- * does not come from configuration settings, but from processed
- * document itself.
- */
- boolean mXml11 = false;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Common configuration objects
- ///////////////////////////////////////////////////////////////////////
- */
-
- XMLReporter mReporter;
-
- XMLResolver mDtdResolver = null;
- XMLResolver mEntityResolver = null;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // More special(ized) configuration objects
- ///////////////////////////////////////////////////////////////////////
- */
-
- //Map mCustomEntities;
- //XMLResolver mUndeclaredEntityResolver;
- //DTDEventListener mDTDEventListener;
-
- Object[] mSpecialProperties = null;
-
- private final static int SPEC_PROC_COUNT = 4;
-
- private final static int SP_IX_CUSTOM_ENTITIES = 0;
- private final static int SP_IX_UNDECL_ENT_RESOLVER = 1;
- private final static int SP_IX_DTD_EVENT_LISTENER = 2;
- private final static int SP_IX_DTD_OVERRIDE = 3;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Buffer recycling:
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * This ThreadLocal contains a {@link SoftRerefence}
- * to a {@link BufferRecycler} used to provide a low-cost
- * buffer recycling between Reader instances.
- */
- final static ThreadLocal mRecyclerRef = new ThreadLocal();
-
- /**
- * This is the actually container of the recyclable buffers. It
- * is obtained via ThreadLocal/SoftReference combination, if one
- * exists, when Config instance is created. If one does not
- * exist, it will created first time a buffer is returned.
- */
- BufferRecycler mCurrRecycler = null;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////////////////////
- */
-
- private ReaderConfig(boolean j2meSubset, SymbolTable symbols,
- int configFlags, int configFlagMods,
- int inputBufLen,
- int minTextSegmentLen)
- {
- mIsJ2MESubset = j2meSubset;
- mSymbols = symbols;
-
- mConfigFlags = configFlags;
- mConfigFlagMods = configFlagMods;
-
- mInputBufferLen = inputBufLen;
- mMinTextSegmentLen = minTextSegmentLen;
-
- /* Ok, let's then see if we can find a buffer recycler. Since they
- * are lazily constructed, and since GC may just flush them out
- * on its whims, it's possible we might not find one. That's ok;
- * we can reconstruct one if and when we are to return one or more
- * buffers.
- */
- SoftReference ref = (SoftReference) mRecyclerRef.get();
- if (ref != null) {
- mCurrRecycler = (BufferRecycler) ref.get();
- }
- }
-
- public static ReaderConfig createJ2MEDefaults()
- {
- /* For J2ME we'll use slightly smaller buffer sizes by
- * default, on assumption lower memory usage is desireable:
- */
- ReaderConfig rc = new ReaderConfig
- (true, null, DEFAULT_FLAGS_J2ME, 0,
- // 4k input buffer (2000 chars):
- 2000,
- DEFAULT_SHORTEST_TEXT_SEGMENT);
- return rc;
- }
-
- public static ReaderConfig createFullDefaults()
- {
- /* For full version, can use bit larger buffers to achieve better
- * overall performance.
- */
- ReaderConfig rc = new ReaderConfig
- (false, null, DEFAULT_FLAGS_FULL, 0,
- // 8k input buffer (4000 chars):
- 4000,
- DEFAULT_SHORTEST_TEXT_SEGMENT);
- return rc;
- }
-
- public ReaderConfig createNonShared(SymbolTable sym)
- {
- // should we throw an exception?
- //if (sym == null) { }
- ReaderConfig rc = new ReaderConfig(mIsJ2MESubset, sym,
- mConfigFlags, mConfigFlagMods,
- mInputBufferLen,
- mMinTextSegmentLen);
- rc.mReporter = mReporter;
- rc.mDtdResolver = mDtdResolver;
- rc.mEntityResolver = mEntityResolver;
- rc.mBaseURL = mBaseURL;
- rc.mParsingMode = mParsingMode;
- if (mSpecialProperties != null) {
- int len = mSpecialProperties.length;
- Object[] specProps = new Object[len];
- System.arraycopy(mSpecialProperties, 0, specProps, 0, len);
- rc.mSpecialProperties = specProps;
- }
-
- return rc;
- }
-
- /**
- * Unlike name suggests there is also some limited state information
- * associated with the config object. If these objects are reused,
- * that state needs to be reset between reuses, to avoid carrying
- * over incorrect state.
- */
- public void resetState()
- {
- // Current, only xml 1.0 vs 1.1 state is stored here:
- mXml11 = false;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Implementation of abstract methods
- ///////////////////////////////////////////////////////////////////////
- */
-
- protected int findPropertyId(String propName)
- {
- Integer I = (Integer) sProperties.get(propName);
- return (I == null) ? -1 : I.intValue();
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Public API, accessors
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // Accessors for immutable configuration:
-
- public SymbolTable getSymbols() { return mSymbols; }
-
- /**
- * In future this property could/should be made configurable?
- */
-
- public int getDtdCacheSize() {
- return mIsJ2MESubset ? DTD_CACHE_SIZE_J2ME : DTD_CACHE_SIZE_J2SE;
- }
-
- // // // "Raw" accessors for on/off properties:
-
- public int getConfigFlags() { return mConfigFlags; }
-
- // // // Standard StAX on/off property accessors
-
- public boolean willCoalesceText() {
- return _hasConfigFlag(CFG_COALESCE_TEXT);
- }
-
- public boolean willSupportNamespaces() {
- return _hasConfigFlag(CFG_NAMESPACE_AWARE);
- }
-
- public boolean willReplaceEntityRefs() {
- return _hasConfigFlag(CFG_REPLACE_ENTITY_REFS);
- }
-
- public boolean willSupportExternalEntities() {
- return _hasConfigFlag(CFG_SUPPORT_EXTERNAL_ENTITIES);
- }
-
- public boolean willSupportDTDs() {
- return _hasConfigFlag(CFG_SUPPORT_DTD);
- }
-
- public boolean willValidateWithDTD() {
- return _hasConfigFlag(CFG_VALIDATE_AGAINST_DTD);
- }
-
- // // // Stax2 on/off property accessors
-
- public boolean willReportCData() {
- return _hasConfigFlag(CFG_REPORT_CDATA);
- }
-
- public boolean willParseLazily() {
- return _hasConfigFlag(CFG_LAZY_PARSING);
- }
-
- public boolean willInternNames() {
- return _hasConfigFlag(CFG_INTERN_NAMES);
- }
-
- public boolean willInternNsURIs() {
- return _hasConfigFlag(CFG_INTERN_NS_URIS);
- }
-
- public boolean willPreserveLocation() {
- return _hasConfigFlag(CFG_PRESERVE_LOCATION);
- }
-
- public boolean willAutoCloseInput() {
- return _hasConfigFlag(CFG_AUTO_CLOSE_INPUT);
- }
-
- // // // Woodstox on/off property accessors
-
- public boolean willReportPrologWhitespace() {
- return _hasConfigFlag(CFG_REPORT_PROLOG_WS);
- }
-
- public boolean willCacheDTDs() {
- return _hasConfigFlag(CFG_CACHE_DTDS);
- }
-
- public boolean willCacheDTDsByPublicId() {
- return _hasConfigFlag(CFG_CACHE_DTDS_BY_PUBLIC_ID);
- }
-
- public boolean willDoXmlIdTyping() {
- return _hasConfigFlag(CFG_XMLID_TYPING);
- }
-
- public boolean willDoXmlIdUniqChecks() {
- return _hasConfigFlag(CFG_XMLID_UNIQ_CHECKS);
- }
-
- public boolean willSupportDTDPP() {
- return _hasConfigFlag(CFG_SUPPORT_DTDPP);
- }
-
- public boolean willNormalizeLFs() {
- return _hasConfigFlag(CFG_NORMALIZE_LFS);
- }
-
- public boolean willTreatCharRefsAsEnts() {
- return _hasConfigFlag(CFG_TREAT_CHAR_REFS_AS_ENTS);
- }
-
- public int getInputBufferLength() { return mInputBufferLen; }
-
- public int getShortestReportedTextSegment() { return mMinTextSegmentLen; }
-
- public Map getCustomInternalEntities()
- {
- Map custEnt = (Map) _getSpecialProperty(SP_IX_CUSTOM_ENTITIES);
- if (custEnt == null) {
- return Collections.EMPTY_MAP;
- }
- // Better be defensive and just return a copy...
- int len = custEnt.size();
- HashMap m = new HashMap(len + (len >> 2), 0.81f);
- Iterator it = custEnt.entrySet().iterator();
- while (it.hasNext()) {
- Map.Entry me = (Map.Entry) it.next();
- /* Cast is there just as a safe-guard (assertion), and to
- * document the type...
- */
- m.put(me.getKey(), (EntityDecl) me.getValue());
- }
- return m;
- }
-
- public EntityDecl findCustomInternalEntity(String id)
- {
- Map custEnt = (Map) _getSpecialProperty(SP_IX_CUSTOM_ENTITIES);
- if (custEnt == null) {
- return null;
- }
- return (EntityDecl) custEnt.get(id);
- }
-
- public XMLReporter getXMLReporter() { return mReporter; }
-
- public XMLResolver getXMLResolver() { return mEntityResolver; }
-
- public XMLResolver getDtdResolver() { return mDtdResolver; }
- public XMLResolver getEntityResolver() { return mEntityResolver; }
- public XMLResolver getUndeclaredEntityResolver() {
- return (XMLResolver) _getSpecialProperty(SP_IX_UNDECL_ENT_RESOLVER);
- }
-
- public URL getBaseURL() { return mBaseURL; }
-
- public WstxInputProperties.ParsingMode getInputParsingMode() {
- return mParsingMode;
- }
-
- public boolean inputParsingModeDocuments() {
- return mParsingMode == WstxInputProperties.PARSING_MODE_DOCUMENTS;
- }
-
- public boolean inputParsingModeFragment() {
- return mParsingMode == WstxInputProperties.PARSING_MODE_FRAGMENT;
- }
-
- /**
- * @return True if the input well-formedness and validation checks
- * should be done according to xml 1.1 specification; false if
- * xml 1.0 specification.
- */
- public boolean isXml11() {
- return mXml11;
- }
-
- public DTDEventListener getDTDEventListener() {
- return (DTDEventListener) _getSpecialProperty(SP_IX_DTD_EVENT_LISTENER);
- }
-
- public DTDValidationSchema getDTDOverride() {
- return (DTDValidationSchema) _getSpecialProperty(SP_IX_DTD_OVERRIDE);
- }
-
- /**
- * Special accessor to use to verify whether name interning has
- * explicitly been enabled; true if call was been made to set
- * it to true; false otherwise (default, or set to false)
- */
- public boolean hasInternNamesBeenEnabled() {
- return _hasExplicitConfigFlag(CFG_INTERN_NAMES);
- }
-
- public boolean hasInternNsURIsBeenEnabled() {
- return _hasExplicitConfigFlag(CFG_INTERN_NS_URIS);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Simple mutators
- ///////////////////////////////////////////////////////////////////////
- */
-
- public void setConfigFlag(int flag) {
- mConfigFlags |= flag;
- mConfigFlagMods |= flag;
- }
-
- public void clearConfigFlag(int flag) {
- mConfigFlags &= ~flag;
- mConfigFlagMods |= flag;
- }
-
- // // // Mutators for standard StAX properties
-
- public void doCoalesceText(boolean state) {
- setConfigFlag(CFG_COALESCE_TEXT, state);
- }
-
- public void doSupportNamespaces(boolean state) {
- setConfigFlag(CFG_NAMESPACE_AWARE, state);
- }
-
- public void doReplaceEntityRefs(boolean state) {
- setConfigFlag(CFG_REPLACE_ENTITY_REFS, state);
- }
-
- public void doSupportExternalEntities(boolean state) {
- setConfigFlag(CFG_SUPPORT_EXTERNAL_ENTITIES, state);
- }
-
- public void doSupportDTDs(boolean state) {
- setConfigFlag(CFG_SUPPORT_DTD, state);
- }
-
- public void doValidateWithDTD(boolean state) {
- setConfigFlag(CFG_VALIDATE_AGAINST_DTD, state);
- }
-
- // // // Mutators for Woodstox-specific properties
-
- public void doInternNames(boolean state) {
- setConfigFlag(CFG_INTERN_NAMES, state);
- }
-
- public void doInternNsURIs(boolean state) {
- setConfigFlag(CFG_INTERN_NS_URIS, state);
- }
-
- public void doReportPrologWhitespace(boolean state) {
- setConfigFlag(CFG_REPORT_PROLOG_WS, state);
- }
-
- public void doReportCData(boolean state) {
- setConfigFlag(CFG_REPORT_CDATA, state);
- }
-
- public void doCacheDTDs(boolean state) {
- setConfigFlag(CFG_CACHE_DTDS, state);
- }
-
- public void doCacheDTDsByPublicId(boolean state) {
- setConfigFlag(CFG_CACHE_DTDS_BY_PUBLIC_ID, state);
- }
-
- public void doParseLazily(boolean state) {
- setConfigFlag(CFG_LAZY_PARSING, state);
- }
-
- public void doXmlIdTyping(boolean state) {
- setConfigFlag(CFG_XMLID_TYPING, state);
- }
-
- public void doXmlIdUniqChecks(boolean state) {
- setConfigFlag(CFG_XMLID_UNIQ_CHECKS, state);
- }
-
- public void doPreserveLocation(boolean state) {
- setConfigFlag(CFG_PRESERVE_LOCATION, state);
- }
-
- public void doAutoCloseInput(boolean state) {
- setConfigFlag(CFG_AUTO_CLOSE_INPUT, state);
- }
-
- public void doSupportDTDPP(boolean state) {
- setConfigFlag(CFG_SUPPORT_DTDPP, state);
- }
-
- public void doTreatCharRefsAsEnts(final boolean state) {
- setConfigFlag(CFG_TREAT_CHAR_REFS_AS_ENTS, state);
- }
-
- public void doNormalizeLFs(final boolean state) {
- setConfigFlag(CFG_NORMALIZE_LFS, state);
- }
-
- public void setInputBufferLength(int value)
- {
- /* Let's enforce minimum here; necessary to allow longest
- * consequtive text span to be available (xml decl, etc)
- */
- if (value < MIN_INPUT_BUFFER_LENGTH) {
- value = MIN_INPUT_BUFFER_LENGTH;
- }
- mInputBufferLen = value;
- }
-
- public void setShortestReportedTextSegment(int value) {
- mMinTextSegmentLen = value;
- }
-
- public void setCustomInternalEntities(Map m)
- {
- Map entMap;
- if (m == null || m.size() < 1) {
- entMap = Collections.EMPTY_MAP;
- } else {
- int len = m.size();
- entMap = new HashMap(len + (len >> 1), 0.75f);
- Iterator it = m.entrySet().iterator();
- while (it.hasNext()) {
- Map.Entry me = (Map.Entry) it.next();
- Object val = me.getValue();
- char[] ch;
- if (val == null) {
- ch = DataUtil.getEmptyCharArray();
- } else if (val instanceof char[]) {
- ch = (char[]) val;
- } else {
- // Probably String, but let's just ensure that
- String str = val.toString();
- ch = str.toCharArray();
- }
- String name = (String) me.getKey();
- entMap.put(name, IntEntity.create(name, ch));
- }
- }
- _setSpecialProperty(SP_IX_CUSTOM_ENTITIES, entMap);
- }
-
- public void setXMLReporter(XMLReporter r) {
- mReporter = r;
- }
-
- /**
- * Note: for better granularity, you should call {@link #setEntityResolver}
- * and {@link #setDtdResolver} instead.
- */
- public void setXMLResolver(XMLResolver r) {
- mEntityResolver = r;
- mDtdResolver = r;
- }
-
- public void setDtdResolver(XMLResolver r) {
- mDtdResolver = r;
- }
-
- public void setEntityResolver(XMLResolver r) {
- mEntityResolver = r;
- }
-
- public void setUndeclaredEntityResolver(XMLResolver r) {
- _setSpecialProperty(SP_IX_UNDECL_ENT_RESOLVER, r);
- }
-
- public void setBaseURL(URL baseURL) { mBaseURL = baseURL; }
-
- public void setInputParsingMode(WstxInputProperties.ParsingMode mode) {
- mParsingMode = mode;
- }
-
- /**
- * Method called to enable or disable 1.1 compliant processing; if
- * disabled, defaults to xml 1.0 compliant processing.
- */
- public void enableXml11(boolean state) {
- mXml11 = state;
- }
-
- public void setDTDEventListener(DTDEventListener l) {
- _setSpecialProperty(SP_IX_DTD_EVENT_LISTENER, l);
- }
-
- public void setDTDOverride(DTDValidationSchema schema) {
- _setSpecialProperty(SP_IX_DTD_OVERRIDE, schema);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Profile mutators:
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Method to call to make Reader created conform as closely to XML
- * standard as possible, doing all checks and transformations mandated
- * (linefeed conversions, attr value normalizations).
- * See {@link XMLInputFactory2#configureForXmlConformance} for
- * required settings for standard StAX/StAX properties.
- *
- * In addition to the standard settings, following Woodstox-specific - * settings are also done: - *
- * Notes: Does NOT change 'performance' settings (buffer sizes, - * DTD caching, coalescing, interning, accurate location info). - */ - public void configureForXmlConformance() - { - // // StAX 1.0 settings - doSupportNamespaces(true); - doSupportDTDs(true); - doSupportExternalEntities(true); - doReplaceEntityRefs(true); - - // // Stax2 additional settings - - // Better enable full xml:id checks: - doXmlIdTyping(true); - doXmlIdUniqChecks(true); - - // Woodstox-specific ones: - } - - /** - * Method to call to make Reader created be as "convenient" to use - * as possible; ie try to avoid having to deal with some of things - * like segmented text chunks. This may incur some slight performance - * penalties, but should not affect XML conformance. - * See {@link XMLInputFactory2#configureForConvenience} for - * required settings for standard StAX/StAX properties. - *
- * In addition to the standard settings, following Woodstox-specific - * settings are also done: - *
XMLStreamFactory2.P_LAZY_PARSING (to allow for synchronous
- * error notification by forcing full XML events to be completely
- * parsed when reader's next() is called)
- * - * In addition to the standard settings, following Woodstox-specific - * settings are also done: - *
P_CACHE_DTDS.
- * XMLStremaFactory2.P_LAZY_PARSING (can improve performance
- * especially when skipping text segments)
- * P_MIN_TEXT_SEGMENT, to allow
- * reader to optimize segment length it uses (and possibly avoids
- * one copy operation in the process)
- * P_INPUT_BUFFER_LENGTH a bit from default,
- * to allow for longer consequtive read operations; also reduces cases
- * where partial text segments are on input buffer boundaries.
- * - * See {@link XMLInputFactory2#configureForLowMemUsage} for - * required settings for standard StAX/StAX properties. - *
- * In addition to the standard settings, following Woodstox-specific - * settings are also done: - *
P_CACHE_DTDS
- * P_PARSE_LAZILY
- * P_MIN_TEXT_SEGMENT to the (somewhat low)
- * default value.
- * P_INPUT_BUFFER_LENGTH a bit from the default
- * - * See {@link XMLInputFactory2#configureForLowMemUsage} for - * required settings for standard StAX/StAX properties. - *
- * In addition to the standard settings, following Woodstox-specific - * settings are also done: - *
P_MIN_TEXT_SEGMENT to the maximum value so
- * that all original text segment chunks are reported without
- * segmentation (but without coalescing with adjacent CDATA segments)
- * P_TREAT_CHAR_REFS_AS_ENTS to true, so the all the
- * original character references are reported with their position,
- * original text, and the replacement text.
- * ThreadLocal contains a {@link SoftRerefence}
- * to a {@link BufferRecycler} used to provide a low-cost
- * buffer recycling between Reader instances.
- */
- final static ThreadLocal mRecyclerRef = new ThreadLocal();
-
- /**
- * This is the actually container of the recyclable buffers. It
- * is obtained via ThreadLocal/SoftReference combination, if one
- * exists, when Config instance is created. If one does not
- * exists, it will created first time a buffer is returned.
- */
- BufferRecycler mCurrRecycler = null;
-
- /*
- //////////////////////////////////////////////////////////
- // Life-cycle:
- //////////////////////////////////////////////////////////
- */
-
- private WriterConfig(boolean j2meSubset, int flags, Object[] specProps)
- {
- mIsJ2MESubset = j2meSubset;
- mConfigFlags = flags;
- mSpecialProperties = specProps;
-
- /* Ok, let's then see if we can find a buffer recycler. Since they
- * are lazily constructed, and since GC may just flush them out
- * on its whims, it's possible we might not find one. That's ok;
- * we can reconstruct one if and when we are to return one or more
- * buffers.
- */
- SoftReference ref = (SoftReference) mRecyclerRef.get();
- if (ref != null) {
- mCurrRecycler = (BufferRecycler) ref.get();
- }
- }
-
- public static WriterConfig createJ2MEDefaults()
- {
- return new WriterConfig(true, DEFAULT_FLAGS_J2ME, null);
- }
-
- public static WriterConfig createFullDefaults()
- {
- return new WriterConfig(true, DEFAULT_FLAGS_FULL, null);
- }
-
- public WriterConfig createNonShared()
- {
- Object[] specProps;
-
- if (mSpecialProperties != null) {
- int len = mSpecialProperties.length;
- specProps = new Object[len];
- System.arraycopy(mSpecialProperties, 0, specProps, 0, len);
- } else {
- specProps = null;
- }
- return new WriterConfig(mIsJ2MESubset, mConfigFlags, specProps);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Implementation of abstract methods
- //////////////////////////////////////////////////////////
- */
-
- protected int findPropertyId(String propName)
- {
- Integer I = (Integer) sProperties.get(propName);
- return (I == null) ? -1 : I.intValue();
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Public API
- //////////////////////////////////////////////////////////
- */
-
- public Object getProperty(int id)
- {
- switch (id) {
-
- // First, Stax 1.0 properties:
-
- case PROP_AUTOMATIC_NS:
- return automaticNamespacesEnabled() ? Boolean.TRUE : Boolean.FALSE;
-
- // Then Stax2 properties:
-
- // First, properties common to input/output factories:
-
- case PROP_ENABLE_NS:
- return willSupportNamespaces() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_PROBLEM_REPORTER:
- return getProblemReporter();
-
- // Then output-specific properties:
- case PROP_AUTOMATIC_EMPTY_ELEMENTS:
- return automaticEmptyElementsEnabled() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_AUTO_CLOSE_OUTPUT:
- return willAutoCloseOutput() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_AUTOMATIC_NS_PREFIX:
- return getAutomaticNsPrefix();
- case PROP_TEXT_ESCAPER:
- return getTextEscaperFactory();
- case PROP_ATTR_VALUE_ESCAPER:
- return getAttrValueEscaperFactory();
-
- // // // Then Woodstox-specific properties:
-
- case PROP_OUTPUT_CDATA_AS_TEXT:
- return willOutputCDataAsText() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_COPY_DEFAULT_ATTRS:
- return willCopyDefaultAttrs() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_ESCAPE_CR:
- return willEscapeCr() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_ADD_SPACE_AFTER_EMPTY_ELEM:
- return willAddSpaceAfterEmptyElem() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_AUTOMATIC_END_ELEMENTS:
- return automaticEndElementsEnabled() ? Boolean.TRUE : Boolean.FALSE;
-
- case PROP_VALIDATE_STRUCTURE:
- return willValidateStructure() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_VALIDATE_CONTENT:
- return willValidateContent() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_VALIDATE_ATTR:
- return willValidateAttributes() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_VALIDATE_NAMES:
- return willValidateNames() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_FIX_CONTENT:
- return willFixContent() ? Boolean.TRUE : Boolean.FALSE;
- case PROP_OUTPUT_INVALID_CHAR_HANDLER:
- return getInvalidCharHandler();
- case PROP_OUTPUT_EMPTY_ELEMENT_HANDLER:
- return getEmptyElementHandler();
-
- // And then per-instance properties: not valid via config object
- case PROP_UNDERLYING_STREAM:
- case PROP_UNDERLYING_WRITER:
- throw new IllegalStateException("Can not access per-stream-writer properties via factory");
- }
-
- throw new IllegalStateException("Internal error: no handler for property with internal id "+id+".");
- }
-
- /**
- * @return True, if the specified property was succesfully
- * set to specified value; false if its value was not changed
- */
- public boolean setProperty(String name, int id, Object value)
- {
- switch (id) {
- // First, Stax 1.0 properties:
-
- case PROP_AUTOMATIC_NS:
- enableAutomaticNamespaces(ArgUtil.convertToBoolean(name, value));
- break;
-
- // // // Then Stax2 ones:
-
- case PROP_ENABLE_NS:
- doSupportNamespaces(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_PROBLEM_REPORTER:
- setProblemReporter((XMLReporter) value);
- break;
-
- case PROP_AUTOMATIC_EMPTY_ELEMENTS:
- enableAutomaticEmptyElements(ArgUtil.convertToBoolean(name, value));
- break;
-
- case PROP_AUTO_CLOSE_OUTPUT:
- doAutoCloseOutput(ArgUtil.convertToBoolean(name, value));
- break;
-
- case PROP_AUTOMATIC_NS_PREFIX:
- // value should be a String, but let's verify that:
- setAutomaticNsPrefix(value.toString());
- break;
-
- case PROP_TEXT_ESCAPER:
- setTextEscaperFactory((EscapingWriterFactory) value);
- break;
-
- case PROP_ATTR_VALUE_ESCAPER:
- setAttrValueEscaperFactory((EscapingWriterFactory) value);
- break;
-
- // // // Then Woodstox-specific ones:
-
- case PROP_OUTPUT_CDATA_AS_TEXT:
- doOutputCDataAsText(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_COPY_DEFAULT_ATTRS:
- doCopyDefaultAttrs(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_ESCAPE_CR:
- doEscapeCr(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_ADD_SPACE_AFTER_EMPTY_ELEM:
- doAddSpaceAfterEmptyElem(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_AUTOMATIC_END_ELEMENTS:
- enableAutomaticEndElements(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_VALIDATE_STRUCTURE:
- doValidateStructure(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_VALIDATE_CONTENT:
- doValidateContent(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_VALIDATE_ATTR:
- doValidateAttributes(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_VALIDATE_NAMES:
- doValidateNames(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_FIX_CONTENT:
- doFixContent(ArgUtil.convertToBoolean(name, value));
- break;
- case PROP_OUTPUT_INVALID_CHAR_HANDLER:
- setInvalidCharHandler((InvalidCharHandler) value);
- break;
- case PROP_OUTPUT_EMPTY_ELEMENT_HANDLER:
- setEmptyElementHandler((EmptyElementHandler) value);
- break;
-
- case PROP_UNDERLYING_STREAM:
- case PROP_UNDERLYING_WRITER:
- throw new IllegalStateException("Can not modify per-stream-writer properties via factory");
-
- default:
- throw new IllegalStateException("Internal error: no handler for property with internal id "+id+".");
- }
-
- return true;
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Extended Woodstox API, accessors/modifiers
- //////////////////////////////////////////////////////////
- */
-
- // // // "Raw" accessors for on/off properties:
-
- public int getConfigFlags() { return mConfigFlags; }
-
-
- // // // Accessors, standard properties:
-
- public boolean automaticNamespacesEnabled() {
- return hasConfigFlag(CFG_AUTOMATIC_NS);
- }
-
- // // // Accessors, Woodstox properties:
-
- public boolean automaticEmptyElementsEnabled() {
- return hasConfigFlag(CFG_AUTOMATIC_EMPTY_ELEMENTS);
- }
-
-public boolean willAutoCloseOutput() {
- return hasConfigFlag(CFG_AUTO_CLOSE_OUTPUT);
- }
-
- public boolean willSupportNamespaces() {
- return hasConfigFlag(CFG_ENABLE_NS);
- }
-
- public boolean willOutputCDataAsText() {
- return hasConfigFlag(CFG_OUTPUT_CDATA_AS_TEXT);
- }
-
- public boolean willCopyDefaultAttrs() {
- return hasConfigFlag(CFG_COPY_DEFAULT_ATTRS);
- }
-
- public boolean willEscapeCr() {
- return hasConfigFlag(CFG_ESCAPE_CR);
- }
-
- public boolean willAddSpaceAfterEmptyElem() {
- return hasConfigFlag(CFG_ADD_SPACE_AFTER_EMPTY_ELEM);
- }
-
- public boolean automaticEndElementsEnabled() {
- return hasConfigFlag(CFG_AUTOMATIC_END_ELEMENTS);
- }
-
- public boolean willValidateStructure() {
- return hasConfigFlag(CFG_VALIDATE_STRUCTURE);
- }
-
- public boolean willValidateContent() {
- return hasConfigFlag(CFG_VALIDATE_CONTENT);
- }
-
- public boolean willValidateAttributes() {
- return hasConfigFlag(CFG_VALIDATE_ATTR);
- }
-
- public boolean willValidateNames() {
- return hasConfigFlag(CFG_VALIDATE_NAMES);
- }
-
- public boolean willFixContent() {
- return hasConfigFlag(CFG_FIX_CONTENT);
- }
-
- /**
- * @return Prefix to use as the base for automatically generated
- * namespace prefixes ("namespace prefix prefix", so to speak).
- * Defaults to "wstxns".
- */
- public String getAutomaticNsPrefix() {
- String prefix = (String) getSpecialProperty(SP_IX_AUTO_NS_PREFIX);
- if (prefix == null) {
- prefix = DEFAULT_AUTOMATIC_NS_PREFIX;
- }
- return prefix;
- }
-
- public EscapingWriterFactory getTextEscaperFactory() {
- return (EscapingWriterFactory) getSpecialProperty(SP_IX_TEXT_ESCAPER_FACTORY);
- }
-
- public EscapingWriterFactory getAttrValueEscaperFactory() {
- return (EscapingWriterFactory) getSpecialProperty(SP_IX_ATTR_VALUE_ESCAPER_FACTORY);
- }
-
- public XMLReporter getProblemReporter() {
- return (XMLReporter) getSpecialProperty(SP_IX_PROBLEM_REPORTER);
- }
-
- public InvalidCharHandler getInvalidCharHandler() {
- return (InvalidCharHandler) getSpecialProperty(SP_IX_INVALID_CHAR_HANDLER);
- }
-
- public EmptyElementHandler getEmptyElementHandler() {
- return (EmptyElementHandler) getSpecialProperty(SP_IX_EMPTY_ELEMENT_HANDLER);
- }
-
- // // // Mutators:
-
- // Standard properies:
-
- public void enableAutomaticNamespaces(boolean state) {
- setConfigFlag(CFG_AUTOMATIC_NS, state);
- }
-
- // Wstx properies:
-
- public void enableAutomaticEmptyElements(boolean state) {
- setConfigFlag(CFG_AUTOMATIC_EMPTY_ELEMENTS, state);
- }
-
- public void doAutoCloseOutput(boolean state) {
- setConfigFlag(CFG_AUTO_CLOSE_OUTPUT, state);
- }
-
- public void doSupportNamespaces(boolean state) {
- setConfigFlag(CFG_ENABLE_NS, state);
- }
-
- public void doOutputCDataAsText(boolean state) {
- setConfigFlag(CFG_OUTPUT_CDATA_AS_TEXT, state);
- }
-
- public void doCopyDefaultAttrs(boolean state) {
- setConfigFlag(CFG_COPY_DEFAULT_ATTRS, state);
- }
-
- public void doEscapeCr(boolean state) {
- setConfigFlag(CFG_ESCAPE_CR, state);
- }
-
- public void doAddSpaceAfterEmptyElem(boolean state) {
- setConfigFlag(CFG_ADD_SPACE_AFTER_EMPTY_ELEM, state);
- }
-
- public void enableAutomaticEndElements(boolean state) {
- setConfigFlag(CFG_AUTOMATIC_END_ELEMENTS, state);
- }
-
- public void doValidateStructure(boolean state) {
- setConfigFlag(CFG_VALIDATE_STRUCTURE, state);
- }
-
- public void doValidateContent(boolean state) {
- setConfigFlag(CFG_VALIDATE_CONTENT, state);
- }
-
- public void doValidateAttributes(boolean state) {
- setConfigFlag(CFG_VALIDATE_ATTR, state);
- }
-
- public void doValidateNames(boolean state) {
- setConfigFlag(CFG_VALIDATE_NAMES, state);
- }
-
- public void doFixContent(boolean state) {
- setConfigFlag(CFG_FIX_CONTENT, state);
- }
-
- /**
- * @param prefix Prefix to use as the base for automatically generated
- * namespace prefixes ("namespace prefix prefix", so to speak).
- */
- public void setAutomaticNsPrefix(String prefix) {
- setSpecialProperty(SP_IX_AUTO_NS_PREFIX, prefix);
- }
-
- public void setTextEscaperFactory(EscapingWriterFactory f) {
- setSpecialProperty(SP_IX_TEXT_ESCAPER_FACTORY, f);
- }
-
- public void setAttrValueEscaperFactory(EscapingWriterFactory f) {
- setSpecialProperty(SP_IX_ATTR_VALUE_ESCAPER_FACTORY, f);
- }
-
- public void setProblemReporter(XMLReporter rep) {
- setSpecialProperty(SP_IX_PROBLEM_REPORTER, rep);
- }
-
- public void setInvalidCharHandler(InvalidCharHandler h) {
- setSpecialProperty(SP_IX_INVALID_CHAR_HANDLER, h);
- }
-
- public void setEmptyElementHandler(EmptyElementHandler h) {
- setSpecialProperty(SP_IX_EMPTY_ELEMENT_HANDLER, h);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Extended Woodstox API, profiles
- //////////////////////////////////////////////////////////
- */
-
- /**
- * For Woodstox, this profile enables all basic well-formedness checks,
- * including checking for name validity.
- */
- public void configureForXmlConformance()
- {
- doValidateAttributes(true);
- doValidateContent(true);
- doValidateStructure(true);
- doValidateNames(true);
- }
-
- /**
- * For Woodstox, this profile enables all basic well-formedness checks,
- * including checking for name validity, and also enables all matching
- * "fix-me" properties (currently only content-fixing property exists).
- */
- public void configureForRobustness()
- {
- doValidateAttributes(true);
- doValidateStructure(true);
- doValidateNames(true);
-
- /* This the actual "meat": we do want to not only check if the
- * content is ok, but also "fix" it if not, and if there's a way
- * to fix it:
- */
- doValidateContent(true);
- doFixContent(true);
- }
-
- /**
- * For Woodstox, setting this profile disables most checks for validity;
- * specifically anything that can have measurable performance impact.
- *
- */
- public void configureForSpeed()
- {
- doValidateAttributes(false);
- doValidateContent(false);
- doValidateNames(false);
-
- // Structural validation is cheap: can be left enabled (if already so)
- //doValidateStructure(false);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Buffer recycling:
- /////////////////////////////////////////////////////
- */
-
- /**
- * Method called to allocate intermediate recyclable copy buffers
- */
- public char[] allocMediumCBuffer(int minSize)
- {
- if (mCurrRecycler != null) {
- char[] result = mCurrRecycler.getMediumCBuffer(minSize);
- if (result != null) {
- return result;
- }
- }
- return new char[minSize];
- }
-
- public void freeMediumCBuffer(char[] buffer)
- {
- // Need to create (and assign) the buffer?
- if (mCurrRecycler == null) {
- mCurrRecycler = createRecycler();
- }
- mCurrRecycler.returnMediumCBuffer(buffer);
- }
-
- public char[] allocFullCBuffer(int minSize)
- {
- if (mCurrRecycler != null) {
- char[] result = mCurrRecycler.getFullCBuffer(minSize);
- if (result != null) {
- return result;
- }
- }
- return new char[minSize];
- }
-
- public void freeFullCBuffer(char[] buffer)
- {
- // Need to create (and assign) the buffer?
- if (mCurrRecycler == null) {
- mCurrRecycler = createRecycler();
- }
- mCurrRecycler.returnFullCBuffer(buffer);
- }
-
- public byte[] allocFullBBuffer(int minSize)
- {
- if (mCurrRecycler != null) {
- byte[] result = mCurrRecycler.getFullBBuffer(minSize);
- if (result != null) {
- return result;
- }
- }
- return new byte[minSize];
- }
-
- public void freeFullBBuffer(byte[] buffer)
- {
- // Need to create (and assign) the buffer?
- if (mCurrRecycler == null) {
- mCurrRecycler = createRecycler();
- }
- mCurrRecycler.returnFullBBuffer(buffer);
- }
-
- static int Counter = 0;
-
- private BufferRecycler createRecycler()
- {
- BufferRecycler recycler = new BufferRecycler();
- // No way to reuse/reset SoftReference, have to create new always:
- mRecyclerRef.set(new SoftReference(recycler));
- return recycler;
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Internal methods
- //////////////////////////////////////////////////////////
- */
-
- private void setConfigFlag(int flag, boolean state) {
- if (state) {
- mConfigFlags |= flag;
- } else {
- mConfigFlags &= ~flag;
- }
- }
-
- private final boolean hasConfigFlag(int flag) {
- return ((mConfigFlags & flag) == flag);
- }
-
- private final Object getSpecialProperty(int ix)
- {
- if (mSpecialProperties == null) {
- return null;
- }
- return mSpecialProperties[ix];
- }
-
- private final void setSpecialProperty(int ix, Object value)
- {
- if (mSpecialProperties == null) {
- mSpecialProperties = new Object[SPEC_PROC_COUNT];
- }
- mSpecialProperties[ix] = value;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/WstxInputProperties.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/WstxInputProperties.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/WstxInputProperties.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/WstxInputProperties.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,263 +0,0 @@
-package com.ctc.wstx.api;
-
-import javax.xml.stream.XMLResolver;
-
-import org.codehaus.stax2.XMLInputFactory2;
-
-/**
- * Class that contains constant for property names used to configure
- * cursor and event readers produced by Wstx implementation of
- * {@link javax.xml.stream.XMLInputFactory}.
- *- * TODO: - * - * - CHECK_CHAR_VALIDITY (separate for white spaces?) - * - CATALOG_RESOLVER? (or at least, ENABLE_CATALOGS) - */ -public final class WstxInputProperties -{ - /** - * Constants used when no DTD handling is done, and we do not know the - * 'real' type of an attribute. Seems like CDATA is the safe choice. - */ - public final static String UNKNOWN_ATTR_TYPE = "CDATA"; - - /* - /////////////////////////////////////////////////////// - // Simple on/off settings: - /////////////////////////////////////////////////////// - */ - - // // // Normalization: - - /** - * Feature that controls whether linefeeds are normalized into - * canonical linefeed as mandated by xml specification. - *
- * Note that disabling this property (from its default enabled - * state) will result in non-conforming XML processing. It may - * be useful for use cases where changes to input content should - * be minimized. - *
- * Note: this property was initially removed from Woodstox 4.0, - * but was reintroduced in 4.0.8 due to user request. - */ - public final static String P_NORMALIZE_LFS = "com.ctc.wstx.normalizeLFs"; - - //public final static String P_NORMALIZE_ATTR_VALUES = "com.ctc.wstx.normalizeAttrValues"; - - // // // XML character validation: - - /** - * Whether readers will verify that characters in text content are fully - * valid XML characters (not just Unicode). If true, will check - * that they are valid (including white space); if false, will not - * check. - *
- * Turning this option off may improve parsing performance; leaving - * it on guarantees compatibility with XML 1.0 specs regarding character - * validity rules. - */ - public final static String P_VALIDATE_TEXT_CHARS = "com.ctc.wstx.validateTextChars"; - - - // // // Caching: - - /** - * Whether readers will try to cache parsed external DTD subsets or not. - */ - - public final static String P_CACHE_DTDS = "com.ctc.wstx.cacheDTDs"; - - /** - * Whether reader is to cache DTDs (when caching enabled) based on public id - * or not: if not, system id will be primarily used. Although theoretically - * public IDs should be unique, and should be good caching keys, sometimes - * broken documents use 'wrong' public IDs, and such by default caching keys - * are based on system id only. - */ - public final static String P_CACHE_DTDS_BY_PUBLIC_ID = "com.ctc.wstx.cacheDTDsByPublicId"; - - - // // // Enabling/disabling lazy/incomplete parsing - - /** - * Whether stream readers are allowed to do lazy parsing, meaning - * to parse minimal part of the event when - * {@link javax.xml.stream.XMLStreamReader#next} is called, and only parse the rest - * as needed (or skip remainder of no extra information is needed). - * Alternative to lazy parsing is called "eager parsing", and is - * what most xml parsers use by default. - *
- * Enabling lazy parsing can improve performance for tasks where - * number of textual events are skipped. The downside is that - * not all well-formedness problems are reported when - * {@link javax.xml.stream.XMLStreamReader#next} is called, but only when the - * rest of event are read or skipped. - *
- * Default value for Woodstox is such that lazy parsing is - * enabled. - * - * @deprecated As of Woodstox 4.0 use - * {@link XMLInputFactory2#P_LAZY_PARSING} instead (from - * Stax2 extension API, v3.0) - */ - public final static String P_LAZY_PARSING = XMLInputFactory2.P_LAZY_PARSING; - - // // // API behavior (for backwards compatibility) - - /** - * This read-only property indicates whether null is returned for default name space prefix; - * Boolean.TRUE indicates it does, Boolean.FALSE that it does not. - *
- * Default value for 4.1 is 'false'; this will most likely change for 5.0 since - * Stax API actually specifies null to be used. - * - * @since 4.1.2 - */ - public final static String P_RETURN_NULL_FOR_DEFAULT_NAMESPACE = "com.ctc.wstx.returnNullForDefaultNamespace"; - - // // // Enabling/disabling support for dtd++ - - /** - * Whether the Reader will recognized DTD++ extensions when parsing - * DTD subsets. - *
- * Note: not implemented as of 2.0.x - */ - public final static String P_SUPPORT_DTDPP = "com.ctc.wstx.supportDTDPP"; - - - /** - * Whether the Reader will treat character references as entities while parsing - * XML documents. - */ - public static final String P_TREAT_CHAR_REFS_AS_ENTS = "com.ctc.wstx.treatCharRefsAsEnts"; - - // // // Enabling alternate mode for parsing XML fragments instead - // // // of full documents - - // Automatic W3C Schema support? - /* - * Whether W3C Schema hint attributes are recognized within document, - * and used to locate Schema to use for validation. - */ - //public final static String P_AUTOMATIC_W3C_SCHEMA = 0x00100000; - - /* - /////////////////////////////////////////////////////// - // More complex settings: - /////////////////////////////////////////////////////// - */ - - // // // Buffer sizes; - - /** - * Size of input buffer (in chars), to use for reading XML content - * from input stream/reader. - */ - public final static String P_INPUT_BUFFER_LENGTH = "com.ctc.wstx.inputBufferLength"; - - // // // Constraints on sizes of text segments parsed: - - - /** - * Property to specify shortest non-complete text segment (part of - * CDATA section or text content) that parser is allowed to return, - * if not required to coalesce text. - */ - public final static String P_MIN_TEXT_SEGMENT = "com.ctc.wstx.minTextSegment"; - - // // // Entity handling - - /** - * Property of type {@link java.util.Map}, that defines explicit set of - * internal (generic) entities that will define of override any entities - * defined in internal or external subsets; except for the 5 pre-defined - * entities (lt, gt, amp, apos, quot). Can be used to explicitly define - * entites that would normally come from a DTD. - *
- * @deprecated This feature may be removed from future versions of
- * Woodstox, since the same functionality can be achieved by using
- * custom entity resolvers.
- */
- public final static String P_CUSTOM_INTERNAL_ENTITIES = "com.ctc.wstx.customInternalEntities";
-
- /**
- * Property of type {@link XMLResolver}, that
- * will allow overriding of default DTD and external parameter entity
- * resolution.
- */
- public final static String P_DTD_RESOLVER = "com.ctc.wstx.dtdResolver";
-
- /**
- * Property of type {@link XMLResolver}, that
- * will allow overriding of default external general entity
- * resolution. Note that using this property overrides settings done
- * using {@link javax.xml.stream.XMLInputFactory#RESOLVER} (and vice versa).
- */
- public final static String P_ENTITY_RESOLVER = "com.ctc.wstx.entityResolver";
-
- /**
- * Property of type {@link XMLResolver}, that
- * will allow graceful handling of references to undeclared (general)
- * entities.
- */
- public final static String P_UNDECLARED_ENTITY_RESOLVER = "com.ctc.wstx.undeclaredEntityResolver";
-
- /**
- * Property of type {@link java.net.URL}, that will allow specifying
- * context URL to use when resolving relative references, for the
- * main-level entities (external DTD subset, references from the internal
- * DTD subset).
- */
- public final static String P_BASE_URL = "com.ctc.wstx.baseURL";
-
- // // // Alternate parsing modes
-
- /**
- * Three-valued property (one of
- * {@link #PARSING_MODE_DOCUMENT},
- * {@link #PARSING_MODE_FRAGMENT} or
- * {@link #PARSING_MODE_DOCUMENTS}; default being the document mode)
- * that can be used to handle "non-standard" XML content. The default
- * mode (PARSING_MODE_DOCUMENT) allows parsing of only
- * well-formed XML documents, but the other two modes allow more lenient
- * parsing. Fragment mode allows parsing of XML content that does not
- * have a single root element (can have zero or more), nor can have
- * XML or DOCTYPE declarations: this may be useful if parsing a subset
- * of a full XML document. Multi-document
- * (PARSING_MODE_DOCUMENTS) mode on the other hand allows
- * parsing of a stream that contains multiple consequtive well-formed
- * documents, with possibly multiple XML and DOCTYPE declarations.
- *
- * The main difference from the API perspective is that in first two - * modes, START_DOCUMENT and END_DOCUMENT are used as usual (as the first - * and last events returned), whereas the multi-document mode can return - * multiple pairs of these events: although it is still true that the - * first event (one cursor points to when reader is instantiated or - * returned by the event reader), there may be intervening pairs that - * signal boundary between two adjacent enclosed documents. - */ - public final static String P_INPUT_PARSING_MODE = "com.ctc.wstx.fragmentMode"; - - // // // DTD defaulting, overriding - - /* - //////////////////////////////////////////////////////////////////// - // Helper classes, values enumerations - //////////////////////////////////////////////////////////////////// - */ - - public final static ParsingMode PARSING_MODE_DOCUMENT = new ParsingMode(); - public final static ParsingMode PARSING_MODE_FRAGMENT = new ParsingMode(); - public final static ParsingMode PARSING_MODE_DOCUMENTS = new ParsingMode(); - - /** - * Inner class used for creating type-safe enumerations (prior to JDK 1.5). - */ - public final static class ParsingMode - { - ParsingMode() { } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/WstxOutputProperties.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/WstxOutputProperties.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/api/WstxOutputProperties.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/api/WstxOutputProperties.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,178 +0,0 @@ -package com.ctc.wstx.api; - -/** - * Class that contains constant for property names used to configure - * cursor and event writers produced by Wstx implementation of - * {@link javax.xml.stream.XMLOutputFactory}. - *
- */ -public final class WstxOutputProperties -{ - /** - * Default xml version number output, if none was specified by - * application. Version 1.0 is used - * to try to maximize compatibility (some older parsers - * may barf on 1.1 and later...) - */ - public final static String DEFAULT_XML_VERSION = "1.0"; - - /** - * If no encoding is passed, we should just default to what xml - * in general expects (and can determine), UTF-8. - *
- * Note: you can check out bug entry [WSTX-18] for more details - */ - public final static String DEFAULT_OUTPUT_ENCODING = "UTF-8"; - - // // // Output options, simple on/off settings: - - /** - * Whether writer should just automatically convert all calls that - * would normally produce CDATA to produce (quoted) text. - */ - public final static String P_OUTPUT_CDATA_AS_TEXT = "com.ctc.wstx.outputCDataAsText"; - - /** - * Whether writer should copy attributes that were initially expanded - * using default settings ("implicit" attributes) or not. - */ - public final static String P_COPY_DEFAULT_ATTRS = "com.ctc.wstx.copyDefaultAttrs"; - - /** - * Whether writer is to add a single white space before closing "/>" - * of the empty element or not. It is sometimes useful to add to - * increase compatibility with HTML browsers, or to increase - * readability. - *
- * The default value is 'false', up to Woodstox 4.x. - *
- * NOTE: JavaDocs for versions 4.0.0 - 4.0.7 incorrectly state that - * default is 'true': this is NOT the case. - *
- * Note: added to resolve Jira entry
- * WSTX-125.
- */
- public final static String P_ADD_SPACE_AFTER_EMPTY_ELEM = "com.ctc.wstx.addSpaceAfterEmptyElem";
-
- /**
- * Whether stream writer is to automatically add end elements that are
- * needed to properly close the output tree, when the stream is closed
- * (either explicitly by a call to close or
- * closeCompletely, or implicitly by a call
- * to writeEndDocument.
- *
- * The default value is 'true' as of Woodstox 4.x. - * Prior to 4.0, this feature was always enabled and there was no - * way to disable it) - * - * @since 3.2.8 - */ - public final static String P_AUTOMATIC_END_ELEMENTS = "com.ctc.wstx.automaticEndElements"; - - // // // Validation options: - - /** - * Whether output classes should do basic verification that the output - * structure is well-formed (start and end elements match); that - * there is one and only one root, and that there is no textual content - * in prolog/epilog. If false, won't do any checking regarding structure. - */ - public final static String P_OUTPUT_VALIDATE_STRUCTURE = "com.ctc.wstx.outputValidateStructure"; - - /** - * Whether output classes should do basic verification that the textual - * content output as part of nodes should be checked for validity, - * if there's a possibility of invalid content. Nodes that include - * such constraints are: comment/'--', cdata/']]>', - * proc. instr/'?>'. - */ - public final static String P_OUTPUT_VALIDATE_CONTENT = "com.ctc.wstx.outputValidateContent"; - - /** - * Whether output classes should check uniqueness of attribute names, - * to prevent accidental output of duplicate attributes. - */ - public final static String P_OUTPUT_VALIDATE_ATTR = "com.ctc.wstx.outputValidateAttr"; - - /** - * Whether output classes should check validity of names, ie that they - * only contain legal XML identifier characters. - */ - public final static String P_OUTPUT_VALIDATE_NAMES = "com.ctc.wstx.outputValidateNames"; - - /** - * Property that further modifies handling of invalid content so - * that if {@link #P_OUTPUT_VALIDATE_CONTENT} is enabled, instead of - * reporting an error, writer will try to fix the problem. - * Invalid content in this context refers to comment - * content with "--", CDATA with "]]>" and proc. instr data with "?>". - * This can - * be done for some content (CDATA, possibly comment), by splitting - * content into separate - * segments; but not for others (proc. instr, since that might - * change the semantics in unintended ways). - */ - public final static String P_OUTPUT_FIX_CONTENT = "com.ctc.wstx.outputFixContent"; - - /** - * Property that determines whether Carriage Return (\r) characters are - * to be escaped when output or not. If enabled, all instances of - * of character \r are escaped using a character entity (where possible, - * that is, within CHARACTERS events, and attribute values). Otherwise - * they are output as is. The main reason to enable this property is - * to ensure that carriage returns are preserved as is through parsing, - * since otherwise they will be converted to canonical xml linefeeds - * (\n), when occuring along or as part of \r\n pair. - */ - public final static String P_OUTPUT_ESCAPE_CR = "com.ctc.wstx.outputEscapeCr"; - - /** - * Property that defines a {@link InvalidCharHandler} used to determine - * what to do with a Java character that app tries to output but which - * is not a valid xml character. Alternatives are converting it to - * another character or throw an exception: default implementations - * exist for both behaviors. - */ - public final static String P_OUTPUT_INVALID_CHAR_HANDLER = "com.ctc.wstx.outputInvalidCharHandler"; - - /** - * Property that defines an {@link EmptyElementHandler} used to determine - * if the end tag for an empty element should be written or not. - * - * If specified {@link org.codehaus.stax2.XMLOutputFactory2#P_AUTOMATIC_EMPTY_ELEMENTS} is ignored. - */ - public final static String P_OUTPUT_EMPTY_ELEMENT_HANDLER = "com.ctc.wstx.outputEmptyElementHandler"; - - // // // Per-instance access to underlying output objects - - /** - * Property that can be used to find out the underlying - * {@link java.io.OutputStream} that an - * {@link javax.xml.stream.XMLStreamWriter} instance is using, - * if known (not known if constructed with a {@link java.io.Writer}, - * or other non-stream destination). Null is returned, if not - * known. - *
- * Note: in general it is dangerous to operate on returned stream - * (if any), due to buffering stream writer can do. As such, caller - * has to take care to know what he is doing, including properly - * flushing output. - */ - public final static String P_OUTPUT_UNDERLYING_STREAM = "com.ctc.wstx.outputUnderlyingStream"; - - /** - * Property that can be used to find out the underlying - * {@link java.io.Writer} that an - * {@link javax.xml.stream.XMLStreamWriter} instance is using, - * if known (may not be known if constructed with a {@link java.io.OutputStream}, - * or other non-Writer destination). Null is returned, if not - * known. Note that the Writer may be an internal wrapper over - * an output stream. - *
- * Note: in general it is dangerous to operate on returned Writer - * (if any), due to buffering stream writer can do. As such, caller - * has to take care to know what he is doing, including properly - * flushing output. - */ - public final static String P_OUTPUT_UNDERLYING_WRITER = "com.ctc.wstx.outputUnderlyingWriter"; -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/ErrorConsts.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/ErrorConsts.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/ErrorConsts.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/ErrorConsts.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,193 +0,0 @@ -package com.ctc.wstx.cfg; - -import javax.xml.XMLConstants; -import javax.xml.stream.XMLStreamConstants; - -/** - * "Static" class that contains error message constants. Note that the - * error message constants are NOT made final; reason is that doing so - * would make compiler inline them in other classes. Doing so would increase - * class size (although not mem usage -- Strings do get interned), with - * minimal performance impact. - */ -public class ErrorConsts - implements XMLStreamConstants -{ - // // // Types of warnings we issue via XMLReporter - - public static String WT_ENT_DECL = "entity declaration"; - public static String WT_ELEM_DECL = "element declaration"; - public static String WT_ATTR_DECL = "attribute declaration"; - public static String WT_XML_DECL = "xml declaration"; - public static String WT_DT_DECL = "doctype declaration"; - public static String WT_NS_DECL = "namespace declaration"; - - /** - * This is the generic type for warnings based on XMLValidationProblem - * objects. - */ - public static String WT_VALIDATION = "schema validation"; - - // // And then warning strings - - public static String W_UNDEFINED_ELEM = "Undefined element \"{0}\"; referred to by attribute(s)"; - public static String W_MIXED_ENCODINGS = "Inconsistent text encoding; declared as \"{0}\" in xml declaration, application had passed \"{1}\""; - public static String W_MISSING_DTD = "Missing DOCTYPE declaration in validating mode; can not validate elements or attributes"; - public static String W_DTD_DUP_ATTR = "Attribute \"{0}\" (for element <{1}>) declared multiple times"; - public static String W_DTD_ATTR_REDECL = "Attribute \"{0}\" already declared for element <{1}>; ignoring re-declaration"; - - // // // Generic errors: - - public static String ERR_INTERNAL = "Internal error"; - public static String ERR_NULL_ARG = "Illegal to pass null as argument"; - public static String ERR_UNKNOWN_FEATURE = "Unrecognized feature \"{0}\""; - - // // // Wrong reader state: - - public static String ERR_STATE_NOT_STELEM = "Current event not START_ELEMENT"; - public static String ERR_STATE_NOT_ELEM = "Current event not START_ELEMENT or END_ELEMENT"; - public static String ERR_STATE_NOT_PI = "Current event not PROCESSING_INSTRUCTION"; - public static String ERR_STATE_NOT_ELEM_OR_TEXT = "Current event ({0}) not START_ELEMENT, END_ELEMENT, CHARACTERS or CDATA"; - - // // // XML declaration related problems - - public static String ERR_XML_10_VS_11 = "XML 1.0 document can not refer to XML 1.1 parsed external entities"; - - // // // Structural problems, prolog/epilog: - - public static String ERR_DTD_IN_EPILOG = "Can not have DOCTYPE declaration in epilog"; - public static String ERR_DTD_DUP = "Duplicate DOCTYPE declaration"; - public static String ERR_CDATA_IN_EPILOG = " (CDATA not allowed in prolog/epilog)"; - - // // // Illegal input: - - public static String ERR_HYPHENS_IN_COMMENT = "String '--' not allowed in comment (missing '>'?)"; - public static String ERR_BRACKET_IN_TEXT = "String ']]>' not allowed in textual content, except as the end marker of CDATA section"; - - // // // Generic parsing errors: - - public static String ERR_UNEXP_KEYWORD = "Unexpected keyword \"{0}\"; expected \"{1}\""; - - public static String ERR_WF_PI_MISSING_TARGET = "Missing processing instruction target"; - public static String ERR_WF_PI_XML_TARGET = "Illegal processing instruction target (\"{0}\"); 'xml' (case insensitive) is reserved by the specs."; - public static String ERR_WF_PI_XML_MISSING_SPACE = "excepted either space or \"?>\" after PI target"; - - // // // Entity problems: - - public static String ERR_WF_ENTITY_EXT_DECLARED = "Entity \"{0}\" declared externally, but referenced from a document declared 'standalone=\"yes\"'"; - - public static String ERR_WF_GE_UNDECLARED = "Undeclared general entity \"{0}\""; - - public static String ERR_WF_GE_UNDECLARED_SA = "Undeclared general entity \"{0}\" (document in stand-alone mode; perhaps declared externally?)"; - - // // // Namespace problems: - - public static String ERR_NS_UNDECLARED = "Undeclared namespace prefix \"{0}\""; - public static String ERR_NS_UNDECLARED_FOR_ATTR = "Undeclared namespace prefix \"{0}\" (for attribute \"{1}\")"; - - public static String ERR_NS_REDECL_XML = "Trying to redeclare prefix 'xml' from its default URI '" - +XMLConstants.XML_NS_URI - +"' to \"{0}\""; - - public static String ERR_NS_REDECL_XMLNS = "Trying to declare prefix 'xmlns' (illegal as per NS 1.1 #4)"; - - public static String ERR_NS_REDECL_XML_URI = "Trying to bind URI '" - +XMLConstants.XML_NS_URI+" to prefix \"{0}\" (can only bind to 'xml')"; - - public static String ERR_NS_REDECL_XMLNS_URI = "Trying to bind URI '" - +XMLConstants.XMLNS_ATTRIBUTE_NS_URI+" to prefix \"{0}\" (can not be explicitly bound)"; - - public static String ERR_NS_EMPTY = -"Non-default namespace can not map to empty URI (as per Namespace 1.0 # 2) in XML 1.0 documents"; - - - // // // DTD-specific: - - public static String ERR_DTD_MAINLEVEL_KEYWORD = "; expected a keyword (ATTLIST, ELEMENT, ENTITY, NOTATION), comment, or conditional section"; - - public static String ERR_DTD_ATTR_TYPE = "; expected one of type (CDATA, ID, IDREF, IDREFS, ENTITY, ENTITIES NOTATION, NMTOKEN or NMTOKENS)"; - - public static String ERR_DTD_DEFAULT_TYPE = "; expected #REQUIRED, #IMPLIED or #FIXED"; - - public static String ERR_DTD_ELEM_REDEFD = - "Trying to redefine element \"{0}\" (originally defined at {1})"; - public static String ERR_DTD_NOTATION_REDEFD = - "Trying to redefine notation \"{0}\" (originally defined at {1})"; - - public static String ERR_DTD_UNDECLARED_ENTITY = - "Undeclared {0} entity \"{1}\""; - - public static String ERR_DTD_XML_SPACE = "Attribute xml:space has to be defined of type enumerated, and have 1 or 2 values, 'default' and/or 'preserve'"; - - public static String ERR_DTD_XML_ID = "Attribute xml:id has to have attribute type of ID, as per Xml:id specification"; - - - // // // DTD-validation: - - public static String ERR_VLD_UNKNOWN_ELEM = "Undefined element <{0}> encountered"; - - public static String ERR_VLD_EMPTY = "Element <{0}> has EMPTY content specification; can not contain {1}"; - public static String ERR_VLD_NON_MIXED = "Element <{0}> has non-mixed content specification; can not contain non-white space text, or any CDATA sections"; - public static String ERR_VLD_ANY = "Element <{0}> has ANY content specification; can not contain {1}"; - public static String ERR_VLD_UNKNOWN_ATTR = "Element <{0}> has no attribute \"{1}\""; - public static String ERR_VLD_WRONG_ROOT = "Unexpected root element <{0}>; expected <{0}> as per DOCTYPE declaration"; - - // // // Output problems: - - public static String WERR_PROLOG_CDATA = - "Trying to output a CDATA block outside main element tree (in prolog or epilog)"; - public static String WERR_PROLOG_NONWS_TEXT = - "Trying to output non-whitespace characters outside main element tree (in prolog or epilog)"; - public static String WERR_PROLOG_SECOND_ROOT = - "Trying to output second root, <{0}>"; - - public static String WERR_CDATA_CONTENT = - "Illegal input: CDATA block has embedded ']]>' in it (index {0})"; - public static String WERR_COMMENT_CONTENT = - "Illegal input: comment content has embedded '--' in it (index {0})"; - - public static String WERR_ATTR_NO_ELEM = - "Trying to write an attribute when there is no open start element."; - - public static String WERR_NAME_EMPTY = "Illegal to pass empty name"; - - public static String WERR_NAME_ILLEGAL_FIRST_CHAR = "Illegal first name character {0}"; - public static String WERR_NAME_ILLEGAL_CHAR = "Illegal name character {0}"; - - /* - //////////////////////////////////////////////////// - // Utility methods - //////////////////////////////////////////////////// - */ - - public static String tokenTypeDesc(int type) - { - switch (type) { - case START_ELEMENT: - return "START_ELEMENT"; - case END_ELEMENT: - return "END_ELEMENT"; - case START_DOCUMENT: - return "START_DOCUMENT"; - case END_DOCUMENT: - return "END_DOCUMENT"; - - case CHARACTERS: - return "CHARACTERS"; - case CDATA: - return "CDATA"; - case SPACE: - return "SPACE"; - - case COMMENT: - return "COMMENT"; - case PROCESSING_INSTRUCTION: - return "PROCESSING_INSTRUCTION"; - case DTD: - return "DTD"; - case ENTITY_REFERENCE: - return "ENTITY_REFERENCE"; - } - return "["+type+"]"; - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/InputConfigFlags.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/InputConfigFlags.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/InputConfigFlags.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/InputConfigFlags.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,204 +0,0 @@ -package com.ctc.wstx.cfg; - -/** - * Constant interface that contains configuration flag used by parser - * and parser factory, as well as some other input constants. - */ -public interface InputConfigFlags -{ - /* - ////////////////////////////////////////////////////// - // Flags for standard StAX features: - ////////////////////////////////////////////////////// - */ - - // // // Namespace handling: - - /** - * If true, parser will handle namespaces according to XML specs; if - * false, will only pass them as part of element/attribute name value - * information. - */ - final static int CFG_NAMESPACE_AWARE = 0x0001; - - - // // // Text normalization - - - /// Flag that indicates iterator should coalesce all text segments. - final static int CFG_COALESCE_TEXT = 0x0002; - - - // // // Entity handling - - /** - * Flag that enables automatic replacement of internal entities - */ - final static int CFG_REPLACE_ENTITY_REFS = 0x0004; - - /** - * Flag that enables support for expanding external entities. Woodstox - * pretty much ignores the setting, since effectively it is irrelevant, - * as {@link #CFG_REPLACE_ENTITY_REFS} and {@link #CFG_SUPPORT_DTD} - * both need to be enabled for external entities to be supported. - */ - final static int CFG_SUPPORT_EXTERNAL_ENTITIES = 0x0008; - - // // // DTD handling - - /** - * Whether DTD handling is enabled or disabled; disabling means both - * internal and external subsets will just be skipped unprocessed. - */ - final static int CFG_SUPPORT_DTD = 0x0010; - - /** - * Not yet (fully) supported; added as the placeholder - */ - final static int CFG_VALIDATE_AGAINST_DTD = 0x0020; - - // // Note: can add 2 more 'standard' flags here... - - /* - ////////////////////////////////////////////////////// - // Flags for StAX2 features - ////////////////////////////////////////////////////// - */ - - /** - * If true, parser will report (ignorable) white space events in prolog - * and epilog; if false, it will silently ignore them. - */ - final static int CFG_REPORT_PROLOG_WS = 0x0100; - - // // // Type conversions: - - - /** - * If true, parser will accurately report CDATA event as such (unless - * coalescing); otherwise will always report them as CHARACTERS - * independent of coalescing settings. - */ - final static int CFG_REPORT_CDATA = 0x0200; - - // // // String interning: - - /** - * If true, will guarantee that all names (attribute/element local names - * have been intern()ed. If false, this is not guaranteed although - * implementation may still choose to do it. - */ - final static int CFG_INTERN_NAMES = 0x0400; - - /** - * It true, will call intern() on all namespace URIs parsed; otherwise - * will just use 'regular' Strings created from parsed contents. Interning - * makes namespace-based access faster, but has initial overhead of - * intern() call. - */ - final static int CFG_INTERN_NS_URIS = 0x0800; - - // // // Lazy/incomplete parsing - - /** - * Property that determines whether Event objects created will - * contain (accurate) {@link javax.xml.stream.Location} information or not. If not, - * Location may be null or a fixed location (beginning of main - * XML file). - *
- * Note, however, that the underlying parser will still keep track - * of location information for error reporting purposes; it's only - * Event objects that are affected. - */ - final static int CFG_PRESERVE_LOCATION = 0x1000; - - // // // Input source handling - - /** - * Property that enables/disables stream reader to close the underlying - * input source, either when it is asked to (.close() is called), or - * when it doesn't need it any more (reaching EOF, hitting an - * unrecoverable exception). - * As per Stax 1.0 specification, automatic closing is NOT enabled by - * default; except if the caller has no access to the target (i.e. - * when factory created it) - */ - final static int CFG_AUTO_CLOSE_INPUT = 0x2000; - - /* - ////////////////////////////////////////////////////// - // Flags for Woodstox-specific features - ////////////////////////////////////////////////////// - */ - - // // // Content normalization - - // 20-Jan-2007, TSa: These properties removed from 4.0, deprecated: - final static int CFG_NORMALIZE_LFS = 0x4000; - //final static int CFG_NORMALIZE_ATTR_VALUES = 0x8000; - - // // // Caching - - /** - * If true, input factory is allowed cache parsed external DTD subsets, - * potentially speeding up things for which DTDs are needed for: entity - * substitution, attribute defaulting, and of course DTD-based validation. - */ - final static int CFG_CACHE_DTDS = 0x00010000; - - /** - * If true, key used for matching DTD subsets can be the public id, - * if false, only system id can be used. - */ - final static int CFG_CACHE_DTDS_BY_PUBLIC_ID = 0x00020000; - - // // // Lazy/incomplete parsing - - /** - * If true, input factory can defer parsing of nodes until data is - * actually needed; if false, it has to read all the data in right - * away when next type is requested. Setting it to true is good for - * performance, in the cases where some of the nodes (like comments, - * processing instructions, or whole subtrees) are ignored. Otherwise - * setting will not make much of a difference. Downside is that error - * reporting is also done 'lazily'; not right away when getting the next - * even type but when either accessing data, or skipping it. - */ - final static int CFG_LAZY_PARSING = 0x00040000; - - // // // Validation support - - // DTD++ support - - /** - * If true, DTD-parser will recognize DTD++ features, and the validator - * will also use any such information found from DTD when DTD validation - * is enabled. - */ - final static int CFG_SUPPORT_DTDPP = 0x00080000; - - // Automatic W3C Schema support? - //final static int CFG_AUTOMATIC_W3C_SCHEMA = 0x00100000; - - // // // Xml:id support - - /** - * If true, xml:id attribute type assignment and matching checks will - * be done as per Xml:id specification. Needs to be enabled for xml:id - * uniqueness checks to function properly - */ - final static int CFG_XMLID_TYPING = 0x00200000; - - /** - * If true, xml:id attribute uniqueness constraints are enforced, even - * if not validating against DTD otherwise. - */ - final static int CFG_XMLID_UNIQ_CHECKS = 0x00400000; - - /** - * If true, the XML parser will treat character references as entities. - * - */ - final static int CFG_TREAT_CHAR_REFS_AS_ENTS = 0x00800000; -} - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/OutputConfigFlags.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/OutputConfigFlags.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/OutputConfigFlags.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/OutputConfigFlags.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,105 +0,0 @@ -package com.ctc.wstx.cfg; - -/** - * Constant interface that contains configuration flag used by output - * classes internally, for presenting on/off configuration options. - */ -public interface OutputConfigFlags -{ - /** - * Flag that indicates whether writer is namespace-aware or not; if not, - * only local part is ever used. - */ - final static int CFG_ENABLE_NS = 0x0001; - - /// Flag that indicates that output class should auto-generate namespace prefixes as necessary. - final static int CFG_AUTOMATIC_NS = 0x0002; - - /// Flag that indicates we can output 'automatic' empty elements. - final static int CFG_AUTOMATIC_EMPTY_ELEMENTS = 0x0004; - - /** - * Whether writer should just automatically convert all calls that - * would normally produce CDATA to produce (quoted) text. - */ - final static int CFG_OUTPUT_CDATA_AS_TEXT = 0x0008; - - /** - * Flag that indicates whether attributes expanded from default attribute - * values should be copied to output, when using copy methods. - */ - final static int CFG_COPY_DEFAULT_ATTRS = 0x0010; - - /** - * Flag that indicates whether CR (\r, ascii 13) characters occuring - * in text (CHARACTERS) and attribute values should be escaped using - * character entities or not. Escaping is needed to enable seamless - * round-tripping (preserving CR characters). - */ - final static int CFG_ESCAPE_CR = 0x0020; - - /** - * Flag that indicates - * whether writer is to add a single white space before closing "/>" - * of the empty element or not. It is sometimes useful to add to - * increase compatibility with HTML browsers, or to increase - * readability. - */ - final static int CFG_ADD_SPACE_AFTER_EMPTY_ELEM = 0x0040; - - /** - * Flag that indicates we can output 'automatic' empty elements; - * end elements needed to close the logical output tree when - * stream writer is closed (by closing it explicitly, or by writing - * end-document event) - * - * @since 3.2.8 - */ - final static int CFG_AUTOMATIC_END_ELEMENTS = 0x0080; - - /// Flag that indicates we should check validity of output XML structure. - final static int CFG_VALIDATE_STRUCTURE = 0x0100; - - /** - * Flag that indicates we should check validity of textual content of - * nodes that have constraints. - *
- * Specifically: comments can not have '--', CDATA sections can not - * have ']]>' and processing instruction can not have '?<' character - * combinations in content passed in. - */ - final static int CFG_VALIDATE_CONTENT = 0x0200; - - /** - * Flag that indicates we should check validity of names (element and - * attribute names and prefixes; processing instruction names), that they - * contain only legal identifier characters. - */ - final static int CFG_VALIDATE_NAMES = 0x0400; - - /** - * Flag that indicates we should check uniqueness of attribute names, - * to prevent accidental output of duplicate attributes. - */ - final static int CFG_VALIDATE_ATTR = 0x0800; - - /** - * Flag that will enable writer that checks for validity of content - * to try to fix the problem, by splitting output segments as - * necessary. If disabled, validation will throw an exception; and - * without validation no problem is noticed by writer (but instead - * invalid output is created). - */ - final static int CFG_FIX_CONTENT = 0x1000; - - /** - * Property that enables/disables stream write to close the underlying - * output target, either when it is asked to (.close() is called), or - * when it doesn't need it any more (reaching EOF, hitting an - * unrecoverable exception). - * As per Stax 1.0 specification, automatic closing is NOT enabled by - * default; except if the caller has no access to the target (i.e. - * when factory created it) - */ - final static int CFG_AUTO_CLOSE_OUTPUT = 0x2000; -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/package.html --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/package.html 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/package.html 1970-01-01 00:00:00.000000000 +0000 @@ -1,3 +0,0 @@ -
-Package that contains internal configuration settings for Woodstox. - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/ParsingErrorMsgs.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/ParsingErrorMsgs.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/ParsingErrorMsgs.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/ParsingErrorMsgs.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,26 +0,0 @@ -package com.ctc.wstx.cfg; - -public interface ParsingErrorMsgs -{ - // // // EOF problems: - - final static String SUFFIX_IN_ATTR_VALUE = " in attribute value"; - final static String SUFFIX_IN_DEF_ATTR_VALUE = " in attribute default value"; - final static String SUFFIX_IN_CDATA = " in CDATA section"; - final static String SUFFIX_IN_CLOSE_ELEMENT = " in end tag"; - final static String SUFFIX_IN_COMMENT = " in comment"; - final static String SUFFIX_IN_DTD = " in DOCTYPE declaration"; - final static String SUFFIX_IN_DTD_EXTERNAL = " in external DTD subset"; - final static String SUFFIX_IN_DTD_INTERNAL = " in internal DTD subset"; - final static String SUFFIX_IN_DOC = " in main document content"; - final static String SUFFIX_IN_ELEMENT = " in start tag"; - final static String SUFFIX_IN_ENTITY_REF = " in entity reference"; - final static String SUFFIX_IN_EPILOG = " in epilog"; - final static String SUFFIX_IN_NAME = " in name token"; - final static String SUFFIX_IN_PROC_INSTR = " in processing instruction"; - final static String SUFFIX_IN_PROLOG = " in prolog"; - final static String SUFFIX_IN_TEXT = " in document text content"; - final static String SUFFIX_IN_XML_DECL = " in xml declaration"; - - final static String SUFFIX_EOF_EXP_NAME = "; expected an identifier"; -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/XmlConsts.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/XmlConsts.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/cfg/XmlConsts.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/cfg/XmlConsts.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,66 +0,0 @@ -package com.ctc.wstx.cfg; - -/** - * Simple constant container interface, shared by input and output - * sides. - */ -public interface XmlConsts -{ - // // // Constants for XML declaration - - public final static String XML_DECL_KW_ENCODING = "encoding"; - public final static String XML_DECL_KW_VERSION = "version"; - public final static String XML_DECL_KW_STANDALONE = "standalone"; - - public final static String XML_V_10_STR = "1.0"; - public final static String XML_V_11_STR = "1.1"; - - /** - * This constants refers to cases where the version has not been - * declared explicitly; and needs to be considered to be 1.0. - */ - public final static int XML_V_UNKNOWN = 0x0000; - - public final static int XML_V_10 = 0x0100; - public final static int XML_V_11 = 0x0110; - - public final static String XML_SA_YES = "yes"; - public final static String XML_SA_NO = "no"; - - // // // Stax specs mandates some settings: but since exact - // // // definitions have been re-interpreted a few times, - // // // let's isolate them in a single place - - /* 13-Mar-2008, TSa: As per latest reading of Stax specs, - * all of these are expected to be "", not null. - */ - - public final static String ELEM_NO_NS_URI = ""; - - public final static String ATTR_NO_NS_URI = ""; - - public final static String ELEM_NO_PREFIX = ""; - - public final static String ATTR_NO_PREFIX = ""; - - /** - * Top-most namespace URI assigned for root element, if not specifically - * defined (default namespace unbound). - *- * As per Stax specs, related clarifying discussion on - * the mailing list, and especially JDK 1.6 definitions - * in {@link javax.xml.XMLConstants} constants, empty String - * should be used instead of null. - */ - public final static String DEFAULT_NAMESPACE_URI = ELEM_NO_NS_URI; - - // // // Well, these are not strictly xml constants, but for - // // // now can live here - - /** - * This constant defines the highest Unicode character allowed - * in XML content. - */ - final static int MAX_UNICODE_CHAR = 0x10FFFF; - -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/compat/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/compat/package.html --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/compat/package.html 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/compat/package.html 1970-01-01 00:00:00.000000000 +0000 @@ -1,7 +0,0 @@ -
-Package that contains classes that allow abstracting out the details of -JDK platform version being used. Currently classes are used to implement -"graceful degradation" of functionality, when certain JDK classes are only -fully implemented as part of JDKs later than the minimum version -Woodstox needs for running (for Woodstox versions up to 3.x it was JDK 1.2, for 4.x JDK 1.4). - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/compat/QNameCreator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/compat/QNameCreator.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/compat/QNameCreator.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/compat/QNameCreator.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,70 +0,0 @@ -package com.ctc.wstx.compat; - -import java.util.logging.Logger; - -import javax.xml.namespace.QName; - -/** - * Helper class used to solve [WSTX-174]: some older AppServers were - * shipped with incompatible version of QName class, which is missing - * the 3 argument constructor. To address this, we'll use bit of - * ClassLoader hacker to gracefully (?) downgrade to using 2 arg - * alternatives if necessary. - *- * Note: choice of java.util.logging logging is only based on the - * fact that it is guaranteed to be present (we have JDK 1.4 baseline - * requirement) so that we do not add external dependencies. - * It is not a recommendation for using JUL per se; most users would - * do well to just use slf4j or log4j directly instead. - * - * @author Tatu Saloranta - * - * @since 3.2.8 - */ -public final class QNameCreator -{ - /** - * Creator object that creates QNames using proper 3-arg constructor. - * If dynamic class loading fails - */ - private final static Helper _helper; - static { - Helper h = null; - try { - // Not sure where it'll fail, constructor or create... - Helper h0 = new Helper(); - /*QName n =*/ h0.create("elem", "http://dummy", "ns"); - h = h0; - } catch (Throwable t) { - String msg = "Could not construct QNameCreator.Helper; assume 3-arg QName constructor not available and use 2-arg method instead. Problem: "+t.getMessage(); - try { - Logger.getLogger("com.ctc.wstx.compat.QNameCreator").warning(msg); - } catch (Throwable t2) { // just in case JUL craps out... - System.err.println("ERROR: failed to log error using Logger (problem "+t.getMessage()+"), original problem: "+msg); - } - } - _helper = h; - } - - public static QName create(String uri, String localName, String prefix) - { - if (_helper == null) { // can't use 3-arg constructor; but 2-arg will be there - return new QName(uri, localName); - } - return _helper.create(uri, localName, prefix); - } - - /** - * Helper class used to encapsulate calls to the missing method. - */ - private final static class Helper - { - public Helper() { } - - public QName create(String localName, String nsURI, String prefix) - { - return new QName(localName, nsURI, prefix); - } - } -} - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/DOMOutputElement.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/DOMOutputElement.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/DOMOutputElement.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/DOMOutputElement.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,212 +0,0 @@ -package com.ctc.wstx.dom; - -import javax.xml.namespace.NamespaceContext; - -import org.w3c.dom.Element; -import org.w3c.dom.Node; - -import com.ctc.wstx.sw.OutputElementBase; -import com.ctc.wstx.util.BijectiveNsMap; - -/** - * Context object that holds information about an open element - * (one for which START_ELEMENT has been sent, but no END_ELEMENT) - */ -public final class DOMOutputElement - extends OutputElementBase -{ - /** - * Reference to the parent element, element enclosing this element. - * Null for root element. - * Non-final to allow temporary pooling - * (on per-writer basis, to keep these short-lived). - */ - private DOMOutputElement mParent; - - /** - * DOM node that is the root under which content is written, in case - * where there is no parent (mParent == null). If mParent is not null, - * this will be null. - * Value is of type - * {@link Document}, {@link DocumentFragment} or {@link Element} - */ - private final Node mRootNode; - - /** - * Actual DOM element for which this element object acts as a proxy. - */ - private Element mElement; - - private boolean mDefaultNsSet; - - /** - * Constructor for the virtual root element - */ - private DOMOutputElement(Node rootNode) - { - super(); - mRootNode = rootNode; - mParent = null; - mElement = null; - mNsMapping = null; - mNsMapShared = false; - mDefaultNsURI = ""; - mRootNsContext = null; - mDefaultNsSet = false; - } - - private DOMOutputElement(DOMOutputElement parent, Element element, BijectiveNsMap ns) - { - super(parent, ns); - mRootNode = null; - mParent = parent; - mElement = element; - mNsMapping = ns; - mNsMapShared = (ns != null); - mDefaultNsURI = parent.mDefaultNsURI; - mRootNsContext = parent.mRootNsContext; - mDefaultNsSet = false; - } - - /** - * Method called to reuse a pooled instance. - * - * @return Chained pooled instance that should now be head of the - * reuse chain - */ - private void relink(DOMOutputElement parent, Element element) - { - super.relink(parent); - mParent = parent; - mElement = element; - parent.appendNode(element); - mDefaultNsSet = false; - } - - public static DOMOutputElement createRoot(Node rootNode) - { - return new DOMOutputElement(rootNode); - } - - /** - * Simplest factory method, which gets called when a 1-argument - * element output method is called. It is, then, assumed to - * use the default namespace. - * Will both create the child element and attach it to parent element, - * or lacking own owner document. - */ - protected DOMOutputElement createAndAttachChild(Element element) - { - if (mRootNode != null) { - mRootNode.appendChild(element); - } else { - mElement.appendChild(element); - } - return createChild(element); - } - - protected DOMOutputElement createChild(Element element) - { - return new DOMOutputElement(this, element, mNsMapping); - } - - /** - * @return New head of the recycle pool - */ - protected DOMOutputElement reuseAsChild(DOMOutputElement parent, Element element) - { - DOMOutputElement poolHead = mParent; - relink(parent, element); - return poolHead; - } - - /** - * Method called to temporarily link this instance to a pool, to - * allow reusing of instances with the same reader. - */ - protected void addToPool(DOMOutputElement poolHead) - { - mParent = poolHead; - } - - /* - /////////////////////////////////////////////////////////// - // Public API, accessors - /////////////////////////////////////////////////////////// - */ - - public DOMOutputElement getParent() { - return mParent; - } - - public boolean isRoot() { - // (Virtual) Root element has no parent... - return (mParent == null); - } - - /** - * @return String presentation of the fully-qualified name, in - * "prefix:localName" format (no URI). Useful for error and - * debugging messages. - */ - public String getNameDesc() { - if(mElement != null) { - return mElement.getLocalName(); - } - return "#error"; // unexpected case - } - - /* - /////////////////////////////////////////////////////////// - // Public API, mutators - /////////////////////////////////////////////////////////// - */ - - public void setDefaultNsUri(String uri) { - mDefaultNsURI = uri; - mDefaultNsSet = true; - } - - protected void setRootNsContext(NamespaceContext ctxt) - { - mRootNsContext = ctxt; - /* Let's also see if we have an active default ns mapping: - * (provided it hasn't yet explicitly been set for this element) - */ - if (!mDefaultNsSet) { - String defURI = ctxt.getNamespaceURI(""); - if (defURI != null && defURI.length() > 0) { - mDefaultNsURI = defURI; - } - } - } - - /* - /////////////////////////////////////////////////////////// - // Public API, DOM manipulation - /////////////////////////////////////////////////////////// - */ - - protected void appendNode(Node n) - { - if (mRootNode != null) { - mRootNode.appendChild(n); - } else { - mElement.appendChild(n); - } - } - - protected void addAttribute(String pname, String value) - { - mElement.setAttribute(pname, value); - } - - protected void addAttribute(String uri, String qname, String value) - { - mElement.setAttributeNS(uri, qname, value); - } - - public void appendChild(Node n) { - mElement.appendChild(n); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,91 +0,0 @@ -package com.ctc.wstx.dom; - -import java.util.Collections; - -import javax.xml.stream.*; -import javax.xml.transform.dom.DOMSource; - -//import org.codehaus.stax2.XMLInputFactory2; -import org.codehaus.stax2.ri.dom.DOMWrappingReader; - -import com.ctc.wstx.api.ReaderConfig; -import com.ctc.wstx.exc.WstxParsingException; - -public class WstxDOMWrappingReader - extends DOMWrappingReader -{ - protected final ReaderConfig mConfig; - - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - protected WstxDOMWrappingReader(DOMSource src, ReaderConfig cfg) - throws XMLStreamException - { - super(src, cfg.willSupportNamespaces(), cfg.willCoalesceText()); - mConfig = cfg; - // [WSTX-162]: allow enabling/disabling name/ns intern()ing - if (cfg.hasInternNamesBeenEnabled()) { - setInternNames(true); - } - if (cfg.hasInternNsURIsBeenEnabled()) { - setInternNsURIs(true); - } - } - - public static WstxDOMWrappingReader createFrom(DOMSource src, ReaderConfig cfg) - throws XMLStreamException - { - return new WstxDOMWrappingReader(src, cfg); - } - - /* - /////////////////////////////////////////////////// - // Defined/Overridden config methods - /////////////////////////////////////////////////// - */ - - public boolean isPropertySupported(String name) - { - // !!! TBI: not all these properties are really supported - return mConfig.isPropertySupported(name); - } - - public Object getProperty(String name) - { - if (name.equals("javax.xml.stream.entities")) { - // !!! TBI - return Collections.EMPTY_LIST; - } - if (name.equals("javax.xml.stream.notations")) { - // !!! TBI - return Collections.EMPTY_LIST; - } - return mConfig.getProperty(name); - } - - public boolean setProperty(String name, Object value) - { - // Basic config accessor works just fine... - return mConfig.setProperty(name, value); - } - - /* - /////////////////////////////////////////////////// - // Defined/Overridden error reporting - /////////////////////////////////////////////////// - */ - - // @Override - protected void throwStreamException(String msg, Location loc) - throws XMLStreamException - { - if (loc == null) { - throw new WstxParsingException(msg); - } - throw new WstxParsingException(msg, loc); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,744 +0,0 @@ -package com.ctc.wstx.dom; - -import java.util.*; - -import javax.xml.XMLConstants; -import javax.xml.namespace.*; -import javax.xml.stream.*; -import javax.xml.transform.dom.DOMResult; - -import org.w3c.dom.*; - -import org.codehaus.stax2.ri.EmptyNamespaceContext; -import org.codehaus.stax2.ri.dom.DOMWrappingWriter; - -import com.ctc.wstx.api.WriterConfig; -import com.ctc.wstx.cfg.ErrorConsts; -import com.ctc.wstx.sw.OutputElementBase; - -/* TODO: - * - * - validator interface implementation - */ - -/** - * This is an adapter class that allows building a DOM tree using - * {@link XMLStreamWriter} interface. - *
- * Note that the implementation is only to be used for use with
- * javax.xml.transform.dom.DOMResult.
- *
- * Some notes regarding missing/incomplete functionality: - *
- * Note: explicit empty element (written using
- * writeEmptyElement) will never become
- * current element.
- */
- protected DOMOutputElement mCurrElem;
-
- /**
- * This element is non-null right after a call to
- * either writeStartElement and
- * writeEmptyElement, and can be used to
- * add attributes and namespace declarations.
- *
- * Note: while this is often the same as {@link #mCurrElem},
- * it's not always. Specifically, an empty element (written
- * explicitly using writeEmptyElement) will
- * become open element but NOT current element. Conversely,
- * regular elements will remain current element when
- * non elements are written (text, comments, PI), but
- * not the open element.
- */
- protected DOMOutputElement mOpenElement;
-
- /**
- * for NsRepairing mode
- */
- protected int[] mAutoNsSeq;
- protected String mSuggestedDefNs = null;
- protected String mAutomaticNsPrefix;
-
- /**
- * Map that contains URI-to-prefix entries that point out suggested
- * prefixes for URIs. These are populated by calls to
- * {@link #setPrefix}, and they are only used as hints for binding;
- * if there are conflicts, repairing writer can just use some other
- * prefix.
- */
- HashMap mSuggestedPrefixes = null;
-
- /*
- ///////////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////////
- */
-
- private WstxDOMWrappingWriter(WriterConfig cfg, Node treeRoot)
- throws XMLStreamException
- {
- super(treeRoot, cfg.willSupportNamespaces(), cfg.automaticNamespacesEnabled());
- mConfig = cfg;
- mAutoNsSeq = null;
- mAutomaticNsPrefix = mNsRepairing ? mConfig.getAutomaticNsPrefix() : null;
-
- /* Ok; we need a document node; or an element node; or a document
- * fragment node.
- */
- switch (treeRoot.getNodeType()) {
- case Node.DOCUMENT_NODE:
- case Node.DOCUMENT_FRAGMENT_NODE:
- // both are ok, but no current element
- mCurrElem = DOMOutputElement.createRoot(treeRoot);
- mOpenElement = null;
- break;
-
- case Node.ELEMENT_NODE: // can make sub-tree... ok
- {
- // still need a virtual root node as parent
- DOMOutputElement root = DOMOutputElement.createRoot(treeRoot);
- Element elem = (Element) treeRoot;
- mOpenElement = mCurrElem = root.createChild(elem);
- }
- break;
-
- default: // other Nodes not usable
- throw new XMLStreamException("Can not create an XMLStreamWriter for a DOM node of type "+treeRoot.getClass());
- }
- }
-
- public static WstxDOMWrappingWriter createFrom(WriterConfig cfg, DOMResult dst)
- throws XMLStreamException
- {
- Node rootNode = dst.getNode();
- return new WstxDOMWrappingWriter(cfg, rootNode);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter API (Stax 1.0)
- ///////////////////////////////////////////////////////////
- */
-
- //public void close() { }
- //public void flush() { }
-
-
- public NamespaceContext getNamespaceContext()
- {
- if (!mNsAware) {
- return EmptyNamespaceContext.getInstance();
- }
- return mCurrElem;
- }
-
- public String getPrefix(String uri)
- {
- if (!mNsAware) {
- return null;
- }
- if (mNsContext != null) {
- String prefix = mNsContext.getPrefix(uri);
- if (prefix != null) {
- return prefix;
- }
- }
- return mCurrElem.getPrefix(uri);
- }
-
- public Object getProperty(String name) {
- return mConfig.getProperty(name);
- }
-
- public void setDefaultNamespace(String uri) {
- mSuggestedDefNs = (uri == null || uri.length() == 0) ? null : uri;
- }
-
- //public void setNamespaceContext(NamespaceContext context)
-
- public void setPrefix(String prefix, String uri)
- throws XMLStreamException
- {
- if (prefix == null) {
- throw new NullPointerException("Can not pass null 'prefix' value");
- }
- // Are we actually trying to set the default namespace?
- if (prefix.length() == 0) {
- setDefaultNamespace(uri);
- return;
- }
- if (uri == null) {
- throw new NullPointerException("Can not pass null 'uri' value");
- }
-
- /* Let's verify that xml/xmlns are never (mis)declared; as
- * mandated by XML NS specification
- */
- {
- if (prefix.equals("xml")) {
- if (!uri.equals(XMLConstants.XML_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XML, uri);
- }
- } else if (prefix.equals("xmlns")) { // prefix "xmlns"
- if (!uri.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XMLNS, uri);
- }
- // At any rate; we are NOT to output it
- return;
- } else {
- // Neither of prefixes.. but how about URIs?
- if (uri.equals(XMLConstants.XML_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XML_URI, prefix);
- } else if (uri.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XMLNS_URI, prefix);
- }
- }
- }
-
- if (mSuggestedPrefixes == null) {
- mSuggestedPrefixes = new HashMap(16);
- }
- mSuggestedPrefixes.put(uri, prefix);
-
- }
-
- public void writeAttribute(String localName, String value)
- throws XMLStreamException
- {
- outputAttribute(null, null, localName, value);
- }
-
- public void writeAttribute(String nsURI, String localName, String value)
- throws XMLStreamException
- {
- outputAttribute(nsURI, null, localName, value);
- }
-
- public void writeAttribute(String prefix, String nsURI, String localName, String value)
- throws XMLStreamException
- {
- outputAttribute(nsURI, prefix, localName, value);
- }
-
- //public void writeCData(String data)
- //public void writeCharacters(char[] text, int start, int len)
- //public void writeCharacters(String text)
- //public void writeComment(String data)
-
- public void writeDefaultNamespace(String nsURI)
- {
- if (mOpenElement == null) {
- throw new IllegalStateException("No currently open START_ELEMENT, cannot write attribute");
- }
- setDefaultNamespace(nsURI);
- mOpenElement.addAttribute(XMLConstants.XMLNS_ATTRIBUTE_NS_URI, "xmlns", nsURI);
- }
-
- //public void writeDTD(String dtd)
-
- public void writeEmptyElement(String localName)
- throws XMLStreamException
- {
- writeEmptyElement(null, localName);
- }
-
- public void writeEmptyElement(String nsURI, String localName)
- throws XMLStreamException
- {
- // First things first: must
-
- /* Note: can not just call writeStartElement(), since this
- * element will only become the open elem, but not a parent elem
- */
- createStartElem(nsURI, null, localName, true);
- }
-
- public void writeEmptyElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- if (prefix == null) { // passing null would mean "dont care", if repairing
- prefix = "";
- }
- createStartElem(nsURI, prefix, localName, true);
- }
-
- public void writeEndDocument()
- {
- mCurrElem = mOpenElement = null;
- }
-
- public void writeEndElement()
- {
- // Simple, just need to traverse up... if we can
- if (mCurrElem == null || mCurrElem.isRoot()) {
- throw new IllegalStateException("No open start element to close");
- }
- mOpenElement = null; // just in case it was open
- mCurrElem = mCurrElem.getParent();
- }
-
- //public void writeEntityRef(String name)
-
- public void writeNamespace(String prefix, String nsURI) throws XMLStreamException
- {
- if (prefix == null || prefix.length() == 0) {
- writeDefaultNamespace(nsURI);
- return;
- }
- if (!mNsAware) {
- throwOutputError("Can not write namespaces with non-namespace writer.");
- }
- outputAttribute(XMLConstants.XMLNS_ATTRIBUTE_NS_URI, "xmlns", prefix, nsURI);
- mCurrElem.addPrefix(prefix, nsURI);
- }
-
- //public void writeProcessingInstruction(String target)
- //public void writeProcessingInstruction(String target, String data)
-
- //public void writeStartDocument()
- //public void writeStartDocument(String version)
- //public void writeStartDocument(String encoding, String version)
-
- public void writeStartElement(String localName)
- throws XMLStreamException
- {
- writeStartElement(null, localName);
- }
-
- public void writeStartElement(String nsURI, String localName)
- throws XMLStreamException
- {
- createStartElem(nsURI, null, localName, false);
- }
-
- public void writeStartElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- createStartElem(nsURI, prefix, localName, false);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter2 API (Stax2 v3.0):
- // additional accessors
- ///////////////////////////////////////////////////////////
- */
-
- //public XMLStreamLocation2 getLocation()
- //public String getEncoding()
-
- public boolean isPropertySupported(String name)
- {
- // !!! TBI: not all these properties are really supported
- return mConfig.isPropertySupported(name);
- }
-
- public boolean setProperty(String name, Object value)
- {
- /* Note: can not call local method, since it'll return false for
- * recognized but non-mutable properties
- */
- return mConfig.setProperty(name, value);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter2 API (Stax2 v2.0):
- // extended write methods
- ///////////////////////////////////////////////////////////
- */
-
- //public void writeCData(char[] text, int start, int len)
-
- public void writeDTD(String rootName, String systemId, String publicId,
- String internalSubset)
- throws XMLStreamException
- {
- /* Alas: although we can create a DocumentType object, there
- * doesn't seem to be a way to attach it in DOM-2!
- */
- if (mCurrElem != null) {
- throw new IllegalStateException("Operation only allowed to the document before adding root element");
- }
- reportUnsupported("writeDTD()");
- }
-
- //public void writeFullEndElement() throws XMLStreamException
-
- //public void writeSpace(char[] text, int start, int len)
- //public void writeSpace(String text)
-
-
- //public void writeStartDocument(String version, String encoding, boolean standAlone)
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter2 API (Stax2 v2.0): validation
- ///////////////////////////////////////////////////////////
- */
-
- //public XMLValidator validateAgainst(XMLValidationSchema schema)
- //public XMLValidator stopValidatingAgainst(XMLValidationSchema schema)
- //public XMLValidator stopValidatingAgainst(XMLValidator validator)
- //public ValidationProblemHandler setValidationProblemHandler(ValidationProblemHandler h)
-
- /*
- ///////////////////////////////////////////////////////////
- // Impls of abstract methods from base class
- ///////////////////////////////////////////////////////////
- */
-
- protected void appendLeaf(Node n)
- throws IllegalStateException
- {
- mCurrElem.appendNode(n);
- mOpenElement = null;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////
- */
-
- /* Note: copied from regular RepairingNsStreamWriter#writeStartOrEmpty
- * (and its non-repairing counterpart).
- */
-
- /**
- * Method called by all start element write methods.
- *
- * @param nsURI Namespace URI to use: null and empty String denote 'no namespace'
- */
- protected void createStartElem(String nsURI, String prefix, String localName, boolean isEmpty)
- throws XMLStreamException
- {
- DOMOutputElement elem;
-
- if (!mNsAware) {
- if(nsURI != null && nsURI.length() > 0) {
- throwOutputError("Can not specify non-empty uri/prefix in non-namespace mode");
- }
- elem = mCurrElem.createAndAttachChild(mDocument.createElement(localName));
- } else {
- if (mNsRepairing) {
- String actPrefix = validateElemPrefix(prefix, nsURI, mCurrElem);
- if (actPrefix != null) { // fine, an existing binding we can use:
- if (actPrefix.length() != 0) {
- elem = mCurrElem.createAndAttachChild(mDocument.createElementNS(nsURI, actPrefix+":"+localName));
- } else {
- elem = mCurrElem.createAndAttachChild(mDocument.createElementNS(nsURI, localName));
- }
- } else { // nah, need to create a new binding...
- /* Need to ensure that we'll pass "" as prefix, not null,
- * so it is understood as "I want to use the default NS",
- * not as "whatever prefix, I don't care"
- */
- if (prefix == null) {
- prefix = "";
- }
- actPrefix = generateElemPrefix(prefix, nsURI, mCurrElem);
- boolean hasPrefix = (actPrefix.length() != 0);
- if (hasPrefix) {
- localName = actPrefix + ":" + localName;
- }
- elem = mCurrElem.createAndAttachChild(mDocument.createElementNS(nsURI, localName));
- /* Hmmh. writeNamespace method requires open element
- * to be defined. So we'll need to set it first
- * (will be set again at a later point -- would be
- * good to refactor this method into separate
- * sub-classes or so)
- */
- mOpenElement = elem;
- // Need to add new ns declaration as well
- if (hasPrefix) {
- writeNamespace(actPrefix, nsURI);
- elem.addPrefix(actPrefix, nsURI);
- } else {
- writeDefaultNamespace(nsURI);
- elem.setDefaultNsUri(nsURI);
- }
- }
- } else {
- /* Non-repairing; if non-null prefix (including "" to
- * indicate "no prefix") passed, use as is, otherwise
- * try to locate the prefix if got namespace.
- */
- if (prefix == null && nsURI != null && nsURI.length() > 0) {
- if (nsURI == null) {
- nsURI = "";
- }
- prefix = (mSuggestedPrefixes == null) ? null : (String) mSuggestedPrefixes.get(nsURI);
- if (prefix == null) {
- throwOutputError("Can not find prefix for namespace \""+nsURI+"\"");
- }
- }
- if (prefix != null && prefix.length() != 0) {
- localName = prefix + ":" +localName;
- }
- elem = mCurrElem.createAndAttachChild(mDocument.createElementNS(nsURI, localName));
- }
- }
- /* Got the element; need to make it the open element, and
- * if it's not an (explicit) empty element, current element as well
- */
- mOpenElement = elem;
- if (!isEmpty) {
- mCurrElem = elem;
- }
- }
-
- protected void outputAttribute(String nsURI, String prefix, String localName, String value)
- throws XMLStreamException
- {
- if (mOpenElement == null) {
- throw new IllegalStateException("No currently open START_ELEMENT, cannot write attribute");
- }
-
- if (mNsAware) {
- if (mNsRepairing) {
- prefix = findOrCreateAttrPrefix(prefix, nsURI, mOpenElement);
- }
- if (prefix != null && prefix.length() > 0) {
- localName = prefix + ":" + localName;
- }
- mOpenElement.addAttribute(nsURI, localName, value);
- } else { // non-ns, simple
- if (prefix != null && prefix.length() > 0) {
- localName = prefix + ":" + localName;
- }
- mOpenElement.addAttribute(localName, value);
- }
- }
-
- private final String validateElemPrefix(String prefix, String nsURI,
- DOMOutputElement elem)
- throws XMLStreamException
- {
- /* 06-Feb-2005, TSa: Special care needs to be taken for the
- * "empty" (or missing) namespace:
- * (see comments from findOrCreatePrefix())
- */
- if (nsURI == null || nsURI.length() == 0) {
- String currURL = elem.getDefaultNsUri();
- if (currURL == null || currURL.length() == 0) {
- // Ok, good:
- return "";
- }
- // Nope, needs to be re-bound:
- return null;
- }
-
- int status = elem.isPrefixValid(prefix, nsURI, true);
- if (status == DOMOutputElement.PREFIX_OK) {
- return prefix;
- }
- return null;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method called to find an existing prefix for the given namespace,
- * if any exists in the scope. If one is found, it's returned (including
- * "" for the current default namespace); if not, null is returned.
- *
- * @param nsURI URI of namespace for which we need a prefix
- */
- protected final String findElemPrefix(String nsURI, DOMOutputElement elem)
- throws XMLStreamException
- {
- /* Special case: empty NS URI can only be bound to the empty
- * prefix...
- */
- if (nsURI == null || nsURI.length() == 0) {
- String currDefNsURI = elem.getDefaultNsUri();
- if (currDefNsURI != null && currDefNsURI.length() > 0) {
- // Nope; won't do... has to be re-bound, but not here:
- return null;
- }
- return "";
- }
- return mCurrElem.getPrefix(nsURI);
- }
-
-
- /**
- * Method called after {@link #findElemPrefix} has returned null,
- * to create and bind a namespace mapping for specified namespace.
- */
- protected final String generateElemPrefix(String suggPrefix, String nsURI,
- DOMOutputElement elem)
- throws XMLStreamException
- {
- /* Ok... now, since we do not have an existing mapping, let's
- * see if we have a preferred prefix to use.
- */
- /* Except if we need the empty namespace... that can only be
- * bound to the empty prefix:
- */
- if (nsURI == null || nsURI.length() == 0) {
- return "";
- }
-
- /* Ok; with elements this is easy: the preferred prefix can
- * ALWAYS be used, since it can mask preceding bindings:
- */
- if (suggPrefix == null) {
- // caller wants this URI to map as the default namespace?
- if (mSuggestedDefNs != null && mSuggestedDefNs.equals(nsURI)) {
- suggPrefix = "";
- } else {
- suggPrefix = (mSuggestedPrefixes == null) ? null:
- (String) mSuggestedPrefixes.get(nsURI);
- if (suggPrefix == null) {
- /* 16-Oct-2005, TSa: We have 2 choices here, essentially;
- * could make elements always try to override the def
- * ns... or can just generate new one. Let's do latter
- * for now.
- */
- if (mAutoNsSeq == null) {
- mAutoNsSeq = new int[1];
- mAutoNsSeq[0] = 1;
- }
- suggPrefix = elem.generateMapping(mAutomaticNsPrefix, nsURI,
- mAutoNsSeq);
- }
- }
- }
-
- // Ok; let's let the caller deal with bindings
- return suggPrefix;
- }
-
-
- /**
- * Method called to somehow find a prefix for given namespace, to be
- * used for a new start element; either use an existing one, or
- * generate a new one. If a new mapping needs to be generated,
- * it will also be automatically bound, and necessary namespace
- * declaration output.
- *
- * @param suggPrefix Suggested prefix to bind, if any; may be null
- * to indicate "no preference"
- * @param nsURI URI of namespace for which we need a prefix
- * @param elem Currently open start element, on which the attribute
- * will be added.
- */
- protected final String findOrCreateAttrPrefix(String suggPrefix, String nsURI,
- DOMOutputElement elem)
- throws XMLStreamException
- {
- if (nsURI == null || nsURI.length() == 0) {
- /* Attributes never use the default namespace; missing
- * prefix always leads to the empty ns... so nothing
- * special is needed here.
- */
- return null;
- }
- // Maybe the suggested prefix is properly bound?
- if (suggPrefix != null) {
- int status = elem.isPrefixValid(suggPrefix, nsURI, false);
- if (status == OutputElementBase.PREFIX_OK) {
- return suggPrefix;
- }
- /* Otherwise, if the prefix is unbound, let's just bind
- * it -- if caller specified a prefix, it probably prefers
- * binding that prefix even if another prefix already existed?
- * The remaining case (already bound to another URI) we don't
- * want to touch, at least not yet: it may or not be safe
- * to change binding, so let's just not try it.
- */
- if (status == OutputElementBase.PREFIX_UNBOUND) {
- elem.addPrefix(suggPrefix, nsURI);
- writeNamespace(suggPrefix, nsURI);
- return suggPrefix;
- }
- }
-
- // If not, perhaps there's another existing binding available?
- String prefix = elem.getExplicitPrefix(nsURI);
- if (prefix != null) { // already had a mapping for the URI... cool.
- return prefix;
- }
-
- /* Nope, need to create one. First, let's see if there's a
- * preference...
- */
- if (suggPrefix != null) {
- prefix = suggPrefix;
- } else if (mSuggestedPrefixes != null) {
- prefix = (String) mSuggestedPrefixes.get(nsURI);
- // note: def ns is never added to suggested prefix map
- }
-
- if (prefix != null) {
- /* Can not use default namespace for attributes.
- * Also, re-binding is tricky for attributes; can't
- * re-bind anything that's bound on this scope... or
- * used in this scope. So, to simplify life, let's not
- * re-bind anything for attributes.
- */
- if (prefix.length() == 0
- || (elem.getNamespaceURI(prefix) != null)) {
- prefix = null;
- }
- }
-
- if (prefix == null) {
- if (mAutoNsSeq == null) {
- mAutoNsSeq = new int[1];
- mAutoNsSeq[0] = 1;
- }
- prefix = mCurrElem.generateMapping(mAutomaticNsPrefix, nsURI,
- mAutoNsSeq);
- }
-
- // Ok; so far so good: let's now bind and output the namespace:
- elem.addPrefix(prefix, nsURI);
- writeNamespace(prefix, nsURI);
- return prefix;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ChoiceContentSpec.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ChoiceContentSpec.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ChoiceContentSpec.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ChoiceContentSpec.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,240 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-import com.ctc.wstx.util.ExceptionUtil;
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Content specification that defines content model that has
- * multiple alternative elements; including mixed content model.
- */
-public class ChoiceContentSpec
- extends ContentSpec
-{
- final boolean mNsAware;
-
- /**
- * Whether this is a mixed content model; mostly affects String
- * representation
- */
- final boolean mHasMixed;
-
- final ContentSpec[] mContentSpecs;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- private ChoiceContentSpec(boolean nsAware, char arity, boolean mixed,
- ContentSpec[] specs)
- {
- super(arity);
- mNsAware = nsAware;
- mHasMixed = mixed;
- mContentSpecs = specs;
- }
-
- private ChoiceContentSpec(boolean nsAware, char arity, boolean mixed, Collection specs)
- {
- super(arity);
- mNsAware = nsAware;
- mHasMixed = mixed;
- mContentSpecs = new ContentSpec[specs.size()];
- specs.toArray(mContentSpecs);
- }
-
- public static ChoiceContentSpec constructChoice(boolean nsAware, char arity,
- Collection specs)
- {
- return new ChoiceContentSpec(nsAware, arity, false, specs);
- }
-
- public static ChoiceContentSpec constructMixed(boolean nsAware, Collection specs)
- {
- return new ChoiceContentSpec(nsAware, '*', true, specs);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- public StructValidator getSimpleValidator()
- {
- /* Can we create a simple validator? Yes, if the sub-specs are
- * all simple (leaves == element tokens with no arity modifier);
- * this is always true for mixed.
- */
- ContentSpec[] specs = mContentSpecs;
- int len = specs.length;
- int i;
-
- if (mHasMixed) {
- i = len;
- } else {
- i = 0;
- for (; i < len; ++i) {
- if (!specs[i].isLeaf()) {
- break;
- }
- }
- }
-
- if (i == len) { // all leaves, kewl
- PrefixedNameSet keyset = namesetFromSpecs(mNsAware, specs);
- return new Validator(mArity, keyset);
- }
-
- // Nah, need a DFA...
- return null;
- }
-
- public ModelNode rewrite()
- {
- // First, need to convert sub-specs:
- ContentSpec[] specs = mContentSpecs;
- int len = specs.length;
- ModelNode[] models = new ModelNode[len];
- for (int i = 0; i < len; ++i) {
- models[i] = specs[i].rewrite();
- }
- ChoiceModel model = new ChoiceModel(models);
-
- // and then resolve arity modifiers, if necessary:
- if (mArity == '*') {
- return new StarModel(model);
- }
- if (mArity == '?') {
- return new OptionalModel(model);
- }
- if (mArity == '+') {
- return new ConcatModel(model,
- new StarModel(model.cloneModel()));
- }
- return model;
- }
-
- public String toString()
- {
- StringBuffer sb = new StringBuffer();
-
- if (mHasMixed) {
- sb.append("(#PCDATA | ");
- } else {
- sb.append('(');
- }
- for (int i = 0; i < mContentSpecs.length; ++i) {
- if (i > 0) {
- sb.append(" | ");
- }
- sb.append(mContentSpecs[i].toString());
- }
- sb.append(')');
-
- if (mArity != ' ') {
- sb.append(mArity);
- }
- return sb.toString();
- }
-
- /*
- ///////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////
- */
-
- /*
- ///////////////////////////////////////////////////
- // Package methods
- ///////////////////////////////////////////////////
- */
-
- protected static PrefixedNameSet namesetFromSpecs(boolean nsAware, ContentSpec[] specs)
- {
- int len = specs.length;
- PrefixedName[] nameArray = new PrefixedName[len];
- for (int i = 0; i < len; ++i) {
- nameArray[i] = ((TokenContentSpec)specs[i]).getName();
- }
-
- if (len < 5) { // 4 or fewer elements -> small
- return new SmallPrefixedNameSet(nsAware, nameArray);
- }
- return new LargePrefixedNameSet(nsAware, nameArray);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Validator class that can be used for simple
- // choices (including mixed content)
- ///////////////////////////////////////////////////
- */
-
- final static class Validator
- extends StructValidator
- {
- final char mArity;
- final PrefixedNameSet mNames;
-
- int mCount = 0;
-
- public Validator(char arity, PrefixedNameSet names)
- {
- mArity = arity;
- mNames = names;
- }
-
- /**
- * Rules for reuse are simple: if we can have any number of
- * repetitions, we can just use a shared root instance. Although
- * its count variable will get updated this doesn't really
- * matter as it won't be used. Otherwise a new instance has to
- * be created always, to keep track of instance counts.
- */
- public StructValidator newInstance() {
- return (mArity == '*') ? this : new Validator(mArity, mNames);
- }
-
- public String tryToValidate(PrefixedName elemName)
- {
- if (!mNames.contains(elemName)) {
- if (mNames.hasMultiple()) {
- return "Expected one of ("+mNames.toString(" | ")+")";
- }
- return "Expected <"+mNames.toString("")+">";
- }
- if (++mCount > 1 && (mArity == '?' || mArity == ' ')) {
- if (mNames.hasMultiple()) {
- return "Expected $END (already had one of ["
- +mNames.toString(" | ")+"]";
- }
- return "Expected $END (already had one <"
- +mNames.toString("")+">]";
- }
- return null;
- }
-
- public String fullyValid()
- {
- switch (mArity) {
- case '*':
- case '?':
- return null;
- case '+': // need at least one (and multiples checked earlier)
- case ' ':
- if (mCount > 0) {
- return null;
- }
- return "Expected "+(mArity == '+' ? "at least" : "")
- +" one of elements ("+mNames+")";
- }
- // should never happen:
- ExceptionUtil.throwGenericInternal();
- return null;
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ChoiceModel.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ChoiceModel.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ChoiceModel.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ChoiceModel.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,110 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-/**
- * Model class that encapsulates set of sub-models, of which one (and only
- * one) needs to be matched.
- */
-public class ChoiceModel
- extends ModelNode
-{
- final ModelNode[] mSubModels;
-
- boolean mNullable = false;
-
- BitSet mFirstPos, mLastPos;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- protected ChoiceModel(ModelNode[] subModels)
- {
- super();
- mSubModels = subModels;
- boolean nullable = false;
- for (int i = 0, len = subModels.length; i < len; ++i) {
- if (subModels[i].isNullable()) {
- nullable = true;
- break;
- }
- }
- mNullable = nullable;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- public String toString()
- {
- StringBuffer sb = new StringBuffer();
- for (int i = 0; i < mSubModels.length; ++i) {
- if (i > 0) {
- sb.append(" | ");
- }
- sb.append(mSubModels[i].toString());
- }
- sb.append(')');
- return sb.toString();
- }
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public ModelNode cloneModel()
- {
- int len = mSubModels.length;
- ModelNode[] newModels = new ModelNode[len];
- for (int i = 0; i < len; ++i) {
- newModels[i] = mSubModels[i].cloneModel();
- }
- return new ChoiceModel(newModels);
- }
-
- public boolean isNullable() {
- return mNullable;
- }
-
- public void indexTokens(List tokens)
- {
- // First, let's ask sub-models to calc their settings
- for (int i = 0, len = mSubModels.length; i < len; ++i) {
- mSubModels[i].indexTokens(tokens);
- }
- }
-
- public void addFirstPos(BitSet firstPos) {
- if (mFirstPos == null) {
- mFirstPos = new BitSet();
- for (int i = 0, len = mSubModels.length; i < len; ++i) {
- mSubModels[i].addFirstPos(mFirstPos);
- }
- }
- firstPos.or(mFirstPos);
- }
-
- public void addLastPos(BitSet lastPos) {
- if (mLastPos == null) {
- mLastPos = new BitSet();
- for (int i = 0, len = mSubModels.length; i < len; ++i) {
- mSubModels[i].addLastPos(mLastPos);
- }
- }
- lastPos.or(mLastPos);
- }
-
- public void calcFollowPos(BitSet[] followPosSets)
- {
- // need to let child models do their stuff:
- for (int i = 0, len = mSubModels.length; i < len; ++i) {
- mSubModels[i].calcFollowPos(followPosSets);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ConcatModel.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ConcatModel.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ConcatModel.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ConcatModel.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,114 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-/**
- * Model class that represents sequence of 2 sub-models, needed to be
- * matched in the order.
- */
-public class ConcatModel
- extends ModelNode
-{
- ModelNode mLeftModel;
- ModelNode mRightModel;
-
- final boolean mNullable;
-
- BitSet mFirstPos, mLastPos;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public ConcatModel(ModelNode left, ModelNode right)
- {
- super();
- mLeftModel = left;
- mRightModel = right;
- mNullable = mLeftModel.isNullable() && mRightModel.isNullable();
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public ModelNode cloneModel() {
- return new ConcatModel(mLeftModel.cloneModel(), mRightModel.cloneModel());
- }
-
- public boolean isNullable() {
- return mNullable;
- }
-
- public void indexTokens(List tokens)
- {
- mLeftModel.indexTokens(tokens);
- mRightModel.indexTokens(tokens);
- }
-
- public void addFirstPos(BitSet pos) {
- if (mFirstPos == null) {
- mFirstPos = new BitSet();
- mLeftModel.addFirstPos(mFirstPos);
- if (mLeftModel.isNullable()) {
- mRightModel.addFirstPos(mFirstPos);
- }
- }
- pos.or(mFirstPos);
- }
-
- public void addLastPos(BitSet pos) {
- if (mLastPos == null) {
- mLastPos = new BitSet();
- mRightModel.addLastPos(mLastPos);
- if (mRightModel.isNullable()) {
- mLeftModel.addLastPos(mLastPos);
- }
- }
- pos.or(mLastPos);
- }
-
- public void calcFollowPos(BitSet[] followPosSets)
- {
- // Let's let sub-models do what they need to do
- mLeftModel.calcFollowPos(followPosSets);
- mRightModel.calcFollowPos(followPosSets);
-
- /* And then we can calculate follower sets between left and
- * right sub models; so that left model's last position entries
- * have right model's first position entries included
- */
- BitSet foll = new BitSet();
- mRightModel.addFirstPos(foll);
-
- BitSet toAddTo = new BitSet();
- mLeftModel.addLastPos(toAddTo);
-
- int ix = 0; // need to/can skip the null entry (index 0)
- while ((ix = toAddTo.nextSetBit(ix+1)) >= 0) {
- /* Ok; so token at this index needs to have follow positions
- * added...
- */
- followPosSets[ix].or(foll);
- }
- }
-
- public String toString()
- {
- StringBuffer sb = new StringBuffer();
- sb.append('(');
- sb.append(mLeftModel.toString());
- sb.append(", ");
- sb.append(mRightModel.toString());
- sb.append(')');
- return sb.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ContentSpec.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ContentSpec.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ContentSpec.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ContentSpec.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,68 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-/**
- * Abstract base class for classes that contain parts of a content
- * specification of an element defined in DTD. They are created
- * by {@link FullDTDReader} when parsing an DTD subset, and they
- * will be used for constructing actual validators for the element
- * content.
- */
-public abstract class ContentSpec
-{
- protected char mArity;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public ContentSpec(char arity) {
- mArity = arity;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- public final char getArity() { return mArity; }
-
- public final void setArity(char c) { mArity = c; }
-
- public boolean isLeaf() { return false; }
-
- /**
- * Method called by input element stack to get validator for
- * this content specification, if this specification is simple
- * enough not to need full DFA-based validator.
- *
- * @return Simple content model validator, if one can be directly
- * constructed, or null to indicate that a DFA needs to be
- * created.
- */
- public abstract StructValidator getSimpleValidator();
-
- /**
- * Method called as the first part of DFA construction, if necessary;
- * will usually create simpler {@link ModelNode} instances that will
- * match definition this instance contains.
- */
- public abstract ModelNode rewrite();
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DefaultAttrValue.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DefaultAttrValue.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DefaultAttrValue.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DefaultAttrValue.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,218 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.text.MessageFormat;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.ValidationContext;
-import org.codehaus.stax2.validation.XMLValidationProblem;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-
-/**
- * Simple container class used to contain information about the default
- * value for an attribute. Although for most use cases a simple String
- * would suffice, there are cases where additional information is needed
- * (especially status of 'broken' default values, which only need to be
- * reported should the default value be needed).
- */
-public final class DefaultAttrValue
-{
- /*
- ////////////////////////////////////////////////////
- // Constants
- ////////////////////////////////////////////////////
- */
-
- // // // Default value types
-
- public final static int DEF_DEFAULT = 1;
- public final static int DEF_IMPLIED = 2;
- public final static int DEF_REQUIRED = 3;
- public final static int DEF_FIXED = 4;
-
- /*
- ////////////////////////////////////////////////////
- // Singleton instances
- ////////////////////////////////////////////////////
- */
-
- final static DefaultAttrValue sImplied = new DefaultAttrValue(DEF_IMPLIED);
-
- final static DefaultAttrValue sRequired = new DefaultAttrValue(DEF_REQUIRED);
-
- /*
- ////////////////////////////////////////////////////
- // State
- ////////////////////////////////////////////////////
- */
-
- final int mDefValueType;
-
- /**
- * Actual expanded textual content of the default attribute value;
- * normalized if appropriate in this mode.
- * Note that all entities have been expanded: if a GE/PE was undefined,
- * and no fatal errors were reported (non-validating mode), the
- * references were just silently removed, and matching entries added
- * to mUndeclaredEntity
- */
- private String mValue = null;
-
- /**
- * For now, let's only keep track of the first undeclared entity:
- * can be extended if necessary.
- */
- private UndeclaredEntity mUndeclaredEntity = null;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle (creation, configuration)
- ////////////////////////////////////////////////////
- */
-
- private DefaultAttrValue(int defValueType)
- {
- mDefValueType = defValueType;
- }
-
- public static DefaultAttrValue constructImplied() { return sImplied; }
- public static DefaultAttrValue constructRequired() { return sRequired; }
-
- public static DefaultAttrValue constructFixed() {
- return new DefaultAttrValue(DEF_FIXED);
- }
-
- public static DefaultAttrValue constructOptional() {
- return new DefaultAttrValue(DEF_DEFAULT);
- }
-
- public void setValue(String v) {
- mValue = v;
- }
-
- public void addUndeclaredPE(String name, Location loc)
- {
- addUndeclaredEntity(name, loc, true);
- }
-
- public void addUndeclaredGE(String name, Location loc)
- {
- addUndeclaredEntity(name, loc, false);
- }
-
- public void reportUndeclared(ValidationContext ctxt, XMLValidator dtd)
- throws XMLStreamException
- {
- mUndeclaredEntity.reportUndeclared(ctxt, dtd);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Accessors:
- ////////////////////////////////////////////////////
- */
-
- public boolean hasUndeclaredEntities() {
- return (mUndeclaredEntity != null);
- }
-
- public String getValue() {
- return mValue;
- }
-
- /**
- * @return Expanded default value String, if there were no problems
- * (no undeclared entities), or null to indicate there were problems.
- * In latter case, caller is to figure out exact type of the problem
- * and report this appropriately to the application.
- */
- public String getValueIfOk()
- {
- return (mUndeclaredEntity == null) ? mValue : null;
- }
-
- public boolean isRequired() {
- return (this == sRequired);
- }
-
- public boolean isFixed() {
- return (mDefValueType == DEF_FIXED);
- }
-
- public boolean hasDefaultValue() {
- return (mDefValueType == DEF_DEFAULT)
- || (mDefValueType == DEF_FIXED);
- }
-
- /**
- * Method used by the element to figure out if attribute needs "special"
- * checking; basically if it's required, and/or has a default value.
- * In both cases missing the attribute has specific consequences, either
- * exception or addition of a default value.
- */
- public boolean isSpecial() {
- // Only non-special if #IMPLIED
- return (this != sImplied);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////
- */
-
- private void addUndeclaredEntity(String name, Location loc, boolean isPe)
- {
- if (mUndeclaredEntity == null) {
- mUndeclaredEntity = new UndeclaredEntity(name, loc, isPe);
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Helper class(es):
- ////////////////////////////////////////////////////
- */
-
- final static class UndeclaredEntity
- {
- final String mName;
- final boolean mIsPe;
- final Location mLocation;
-
- UndeclaredEntity(String name, Location loc, boolean isPe)
- {
- mName = name;
- mIsPe = isPe;
- mLocation = loc;
- }
-
- public void reportUndeclared(ValidationContext ctxt, XMLValidator dtd)
- throws XMLStreamException
- {
- String msg = MessageFormat.format(ErrorConsts.ERR_DTD_UNDECLARED_ENTITY, new Object[] { (mIsPe ? "parsed" : "general"), mName });
- XMLValidationProblem prob = new XMLValidationProblem
- (mLocation, msg, XMLValidationProblem.SEVERITY_FATAL);
- prob.setReporter(dtd);
- ctxt.reportProblem(prob);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DFAState.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DFAState.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DFAState.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DFAState.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,207 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Class that represents a state in DFA used for validating complex
- * DTD content models.
- */
-public final class DFAState
-{
- final int mIndex;
- final boolean mAccepting;
-
- BitSet mTokenSet;
-
- HashMap mNext = new HashMap();
-
- /*
- ///////////////////////////////////////////////
- // Life-cycle:
- ///////////////////////////////////////////////
- */
-
- public DFAState(int index, BitSet tokenSet)
- {
- mIndex = index;
- // If we have a transition to state 0, it is an accepting state...
- mAccepting = tokenSet.get(0);
- mTokenSet = tokenSet;
- }
-
- public static DFAState constructDFA(ContentSpec rootSpec)
- {
- // Let's first create the real model tree:
- ModelNode modelRoot = rootSpec.rewrite();
-
- /* Then we need to add the dummy end token, and concat node
- * to contain it:
- */
- TokenModel eofToken = TokenModel.getNullToken();
- ConcatModel dummyRoot = new ConcatModel(modelRoot, eofToken);
-
- /* then need to allocate index numbers for tokens
- * (which will also calculate nullability)
- */
- ArrayList tokens = new ArrayList();
- tokens.add(eofToken); // has to be added first, explicitly
- dummyRoot.indexTokens(tokens);
-
- /* And then we can request calculation of follow pos; this will
- * also recursively calculate first/last pos as needed:
- */
- int flen = tokens.size();
- BitSet[] followPos = new BitSet[flen];
- PrefixedName[] tokenNames = new PrefixedName[flen];
- for (int i = 0; i < flen; ++i) {
- followPos[i] = new BitSet(flen);
- tokenNames[i] = ((TokenModel) tokens.get(i)).getName();
- }
- dummyRoot.calcFollowPos(followPos);
-
- /* And then we can calculate DFA stuff. First step is to get
- * firstpos set for the root node, for creating the first
- * state:
- */
- BitSet initial = new BitSet(flen);
- dummyRoot.addFirstPos(initial);
- DFAState firstState = new DFAState(0, initial);
- ArrayList stateList = new ArrayList();
- stateList.add(firstState);
- HashMap stateMap = new HashMap();
- stateMap.put(initial, firstState);
-
- int i = 0;
- while (i < stateList.size()) {
- DFAState curr = (DFAState) stateList.get(i++);
- curr.calcNext(tokenNames, followPos, stateList, stateMap);
- }
-
- // DEBUG:
- /*
- for (i = 0; i < stateList.size(); ++i) {
- //System.out.println(stateList.get(i));
- }
- */
-
- // And there we have it!
- return firstState;
- }
-
- /*
- ///////////////////////////////////////////////
- // Public API, accessors:
- ///////////////////////////////////////////////
- */
-
- public boolean isAcceptingState() {
- return mAccepting;
- }
-
- public int getIndex() {
- return mIndex;
- }
-
- public DFAState findNext(PrefixedName elemName) {
- return (DFAState) mNext.get(elemName);
- }
-
- public TreeSet getNextNames() {
- // Let's order them alphabetically
- TreeSet names = new TreeSet();
- Iterator it = mNext.keySet().iterator();
- while (it.hasNext()) {
- Object o = it.next();
- names.add(o);
- }
- return names;
- }
-
- public void calcNext(PrefixedName[] tokenNames, BitSet[] tokenFPs,
- List stateList, Map stateMap)
- {
- /* Need to loop over all included tokens, and find groups
- * of said tokens
- */
- int first = -1;
-
- /* Need to clone; can not modify in place, since the BitSet
- * is also used as the key...
- */
- BitSet tokenSet = (BitSet) mTokenSet.clone();
- // No need to keep the reference to it, though:
- mTokenSet = null;
-
- while ((first = tokenSet.nextSetBit(first+1)) >= 0) {
- PrefixedName tokenName = tokenNames[first];
-
- /* Special case; the dummy end token has null as name;
- * we can skip that one:
- */
- if (tokenName == null) {
- continue;
- }
-
- BitSet nextGroup = (BitSet) tokenFPs[first].clone();
- int second = first;
-
- while ((second = tokenSet.nextSetBit(second+1)) > 0) {
- if (tokenNames[second] == tokenName) {
- // Let's clear it, too, so we won't match it again:
- tokenSet.clear(second);
- nextGroup.or(tokenFPs[second]);
- }
- }
-
- // Ok; is it a new group?
- DFAState next = (DFAState) stateMap.get(nextGroup);
- if (next == null) { // yup!
- next = new DFAState(stateList.size(), nextGroup);
- stateList.add(next);
- stateMap.put(nextGroup, next);
- }
- mNext.put(tokenName, next);
- }
- }
-
- /*
- ///////////////////////////////////////////////
- // Other methods
- ///////////////////////////////////////////////
- */
-
- public String toString()
- {
- StringBuffer sb = new StringBuffer();
- sb.append("State #"+mIndex+":\n");
- sb.append(" Accepting: "+mAccepting);
- sb.append("\n Next states:\n");
- Iterator it = mNext.entrySet().iterator();
- while (it.hasNext()) {
- Map.Entry en = (Map.Entry) it.next();
- sb.append(en.getKey());
- sb.append(" -> ");
- DFAState next = (DFAState) en.getValue();
- sb.append(next.getIndex());
- sb.append("\n");
- }
- return sb.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DFAValidator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DFAValidator.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DFAValidator.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DFAValidator.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,77 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-import com.ctc.wstx.util.PrefixedName;
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Validator class that is based on a DFA constructed from DTD content
- * specification.
- */
-public final class DFAValidator
- extends StructValidator
-{
- /**
- * For root validator instance, the start state of DFA; for other
- * instances, current state.
- */
- DFAState mState;
-
- public DFAValidator(DFAState initialState) {
- mState = initialState;
- }
-
- public StructValidator newInstance() {
- return new DFAValidator(mState);
- }
-
- public String tryToValidate(PrefixedName elemName)
- {
- // Do we have a follow state with that key?
- DFAState next = mState.findNext(elemName);
-
- if (next == null) {
- // Nope; let's show what we'd have expected instead...
- TreeSet names = mState.getNextNames();
- if (names.size() == 0) { // expected end tag?
- return "Expected $END";
- }
-
- // Either end tag, or another tag?
- if (mState.isAcceptingState()) {
- return "Expected <"+StringUtil.concatEntries(names, ">, <", null)+"> or $END";
- }
- return "Expected <"+StringUtil.concatEntries(names,
- ">, <", "> or <")+">";
- }
-
- mState = next;
- return null;
- }
-
- public String fullyValid()
- {
- if (mState.isAcceptingState()) {
- return null;
- }
- TreeSet names = mState.getNextNames();
- return "Expected <"+StringUtil.concatEntries(names,
- ">, <", "> or <")+">";
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDAttribute.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDAttribute.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDAttribute.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDAttribute.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,508 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.util.Map;
-
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.ValidationContext;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.ent.EntityDecl;
-import com.ctc.wstx.io.WstxInputData;
-import com.ctc.wstx.sr.InputProblemReporter;
-import com.ctc.wstx.util.PrefixedName;
-import com.ctc.wstx.util.StringUtil;
-import com.ctc.wstx.util.WordResolver;
-
-/**
- * Base class for objects that contain attribute definitions from DTD.
- * Sub-classes exists for specific typed attributes (enumeration-valued,
- * non-CDATA ones); base class itself is used for attributes of type
- * CDATA.
- */
-public abstract class DTDAttribute
-{
- final static char CHAR_SPACE = (char) 0x0020;
-
- /*
- ///////////////////////////////////////////////////
- // Type constants
- ///////////////////////////////////////////////////
- */
-
- // // // Value types
-
- public final static int TYPE_CDATA = 0; // default...
- public final static int TYPE_ENUMERATED = 1;
-
- public final static int TYPE_ID = 2;
- public final static int TYPE_IDREF = 3;
- public final static int TYPE_IDREFS = 4;
-
- public final static int TYPE_ENTITY = 5;
- public final static int TYPE_ENTITIES = 6;
-
- public final static int TYPE_NOTATION = 7;
- public final static int TYPE_NMTOKEN = 8;
- public final static int TYPE_NMTOKENS = 9;
-
- /**
- * Array that has String constants matching above mentioned
- * value types
- */
- final static String[] sTypes = new String[] {
- "CDATA",
- /* 05-Feb-2006, TSa: Hmmh. Apparently SAX specs indicate that
- * enumerated type should be listed as "NMTOKEN"... but most
- * SAX parsers use ENUMERATED, plus this way application can
- * distinguish real NMTOKEN from enumerated type.
- */
- /* 26-Nov-2006, TSa: Either way, we can change type to SAX
- * compatible within SAX classes, not here.
- */
- //"NMTOKEN"
- "ENUMERATED",
- "ID",
- "IDREF",
- "IDREFS",
- "ENTITY",
- "ENTITIES",
- "NOTATION",
- "NMTOKEN",
- "NMTOKENS",
- };
-
- /*
- ///////////////////////////////////////////////////
- // Information about the attribute itself
- ///////////////////////////////////////////////////
- */
-
- protected final PrefixedName mName;
-
- /**
- * Index number amongst "special" attributes (required ones, attributes
- * that have default values), if attribute is one: -1 if not.
- */
- protected final int mSpecialIndex;
-
- protected final DefaultAttrValue mDefValue;
-
- protected final boolean mCfgNsAware;
- protected final boolean mCfgXml11;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public DTDAttribute(PrefixedName name, DefaultAttrValue defValue, int specIndex,
- boolean nsAware, boolean xml11)
- {
- mName = name;
- mDefValue = defValue;
- mSpecialIndex = specIndex;
- mCfgNsAware = nsAware;
- mCfgXml11 = xml11;
- }
-
- public abstract DTDAttribute cloneWith(int specIndex);
-
- /*
- ///////////////////////////////////////////////////
- // Public API, accessors
- ///////////////////////////////////////////////////
- */
-
- public final PrefixedName getName() { return mName; }
-
- public final String toString() {
- return mName.toString();
- }
-
- public final String getDefaultValue(ValidationContext ctxt, XMLValidator dtd)
- throws XMLStreamException
- {
- String val = mDefValue.getValueIfOk();
- if (val == null) {
- mDefValue.reportUndeclared(ctxt, dtd);
- /* should never get here, but just to be safe, let's use
- * the 'raw' value (one that does not have undeclared entities
- * included, most likely)
- */
- val = mDefValue.getValue();
- }
- return val;
- }
-
- public final int getSpecialIndex() {
- return mSpecialIndex;
- }
-
- public final boolean needsValidation() {
- return (getValueType() != TYPE_CDATA);
- }
-
- public final boolean isFixed() {
- return mDefValue.isFixed();
- }
-
- public final boolean isRequired() {
- return mDefValue.isRequired();
- }
-
- /**
- * Method used by the element to figure out if attribute needs "special"
- * checking; basically if it's required, and/or has a default value.
- * In both cases missing the attribute has specific consequences, either
- * exception or addition of a default value.
- */
- public final boolean isSpecial() {
- return mDefValue.isSpecial();
- }
-
- public final boolean hasDefaultValue() {
- return mDefValue.hasDefaultValue();
- }
-
- /**
- * Returns the value type of this attribute as an enumerated int
- * to match type (CDATA, ...)
- *
- * Note: - */ - public int getValueType() { - return TYPE_CDATA; - } - - public String getValueTypeString() - { - return sTypes[getValueType()]; - } - - public boolean typeIsId() { - return false; - } - - public boolean typeIsNotation() { - return false; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - public abstract String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException; - - /** - *
- * Note: the default implementation is not optimized, as it does - * a potentially unnecessary copy of the contents. It is expected that - * this method is seldom called (Woodstox never directly calls it; it - * only gets called for chained validators when one validator normalizes - * the value, and then following validators are passed a String, not - * char array) - */ - public String validate(DTDValidatorBase v, String value, boolean normalize) - throws XMLStreamException - { - int len = value.length(); - /* Temporary buffer has to come from the validator itself, since - * attribute objects are stateless and shared... - */ - char[] cbuf = v.getTempAttrValueBuffer(value.length()); - if (len > 0) { - value.getChars(0, len, cbuf, 0); - } - return validate(v, cbuf, 0, len, normalize); - } - - /** - * Method called by the {@link DTDValidator} - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public abstract void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException; - - /** - * Method called when no validation is to be done, but value is still - * to be normalized as much as it can. What this usually means is that - * all regular space (parser earlier on converts other white space to - * spaces, except for specific character entities; and these special - * cases are NOT to be normalized). - *
- * The only exception is that CDATA will not do any normalization. But - * for now, let's implement basic functionality that CDTA instance will - * override - * - * @return Normalized value as a String, if any changes were done; - * null if input was normalized - */ - public String normalize(DTDValidatorBase v, char[] cbuf, int start, int end) - { - return StringUtil.normalizeSpaces(cbuf, start, end); - } - - /** - * Method called to do initial normalization of the default attribute - * value, without trying to verify its validity. Thus, it's - * called independent of whether we are fully validating the document. - */ - public void normalizeDefault() - { - String val = mDefValue.getValue(); - if (val.length() > 0) { - char[] cbuf = val.toCharArray(); - String str = StringUtil.normalizeSpaces(cbuf, 0, cbuf.length); - if (str != null) { - mDefValue.setValue(str); - } - } - } - - /* - /////////////////////////////////////////////////// - // Package methods, validation helper methods - /////////////////////////////////////////////////// - */ - - protected String validateDefaultName(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String origDefValue = mDefValue.getValue(); - String defValue = origDefValue.trim(); - - if (defValue.length() == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; empty String is not a valid name"); - } - - // Ok, needs to be a valid XML name: - int illegalIx = WstxInputData.findIllegalNameChar(defValue, mCfgNsAware, mCfgXml11); - if (illegalIx >= 0) { - if (illegalIx == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue+"'; character " - +WstxInputData.getCharDesc(defValue.charAt(0)) - +") not valid first character of a name"); - } else { - reportValidationProblem(rep, "Invalid default value '"+defValue+"'; character #"+illegalIx+" (" - +WstxInputData.getCharDesc(defValue.charAt(illegalIx)) - +") not valid name character"); - } - } - - // Ok, cool it's ok... - return normalize ? defValue : origDefValue; - } - - protected String validateDefaultNames(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String defValue = mDefValue.getValue().trim(); - int len = defValue.length(); - - // Then code similar to actual value validation: - StringBuffer sb = null; - int count = 0; - int start = 0; - - main_loop: - while (start < len) { - char c = defValue.charAt(start); - - // Ok, any white space to skip? - while (true) { - if (!WstxInputData.isSpaceChar(c)) { - break; - } - if (++start >= len) { - break main_loop; - } - c = defValue.charAt(start); - } - - // Then need to find the token itself: - int i = start+1; - - for (; i < len; ++i) { - if (WstxInputData.isSpaceChar(defValue.charAt(i))) { - break; - } - } - String token = defValue.substring(start, i); - int illegalIx = WstxInputData.findIllegalNameChar(token, mCfgNsAware, mCfgXml11); - if (illegalIx >= 0) { - if (illegalIx == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; character " - +WstxInputData.getCharDesc(defValue.charAt(start)) - +") not valid first character of a name token"); - } else { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; character " - +WstxInputData.getCharDesc(c) - +") not a valid name character"); - } - } - ++count; - if (normalize) { - if (sb == null) { - sb = new StringBuffer(i - start + 32); - } else { - sb.append(' '); - } - sb.append(token); - } - start = i+1; - } - - if (count == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; empty String is not a valid name value"); - } - - return normalize ? sb.toString() : defValue; - } - - protected String validateDefaultNmToken(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String origDefValue = mDefValue.getValue(); - String defValue = origDefValue.trim(); - - if (defValue.length() == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue+"'; empty String is not a valid NMTOKEN"); - } - int illegalIx = WstxInputData.findIllegalNmtokenChar(defValue, mCfgNsAware, mCfgXml11); - if (illegalIx >= 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; character #"+illegalIx+" (" - +WstxInputData.getCharDesc(defValue.charAt(illegalIx)) - +") not valid NMTOKEN character"); - } - // Ok, cool it's ok... - return normalize ? defValue : origDefValue; - } - - /** - * Method called by validation/normalization code for enumeration-valued - * attributes, to trim - * specified attribute value (full normalization not needed -- called - * for values that CAN NOT have spaces inside; such values can not - * be legal), and then check whether it is included - * in set of words (tokens) passed in. If actual value was included, - * will return the normalized word (as well as store shared String - * locally); otherwise will return null. - */ - public String validateEnumValue(char[] cbuf, int start, int end, - boolean normalize, - WordResolver res) - { - /* Better NOT to build temporary Strings quite yet; can resolve - * matches via resolver more efficiently. - */ - // Note: at this point, should only have real spaces... - if (normalize) { - while (start < end && cbuf[start] <= CHAR_SPACE) { - ++start; - } - while (--end > start && cbuf[end] <= CHAR_SPACE) { - ; - } - ++end; // so it'll point to the first char (or beyond end of buffer) - } - - // Empty String is never legal for enums: - if (start >= end) { - return null; - } - return res.find(cbuf, start, end); - } - - protected EntityDecl findEntityDecl(DTDValidatorBase v, - char[] ch, int start, int len, int hash) - throws XMLStreamException - { - Map entMap = v.getEntityMap(); - /* !!! 13-Nov-2005, TSa: If this was to become a bottle-neck, we - * could use/share a symbol table. Or at least reuse Strings... - */ - String id = new String(ch, start, len); - EntityDecl ent = (EntityDecl) entMap.get(id); - - if (ent == null) { - reportValidationProblem(v, "Referenced entity '"+id+"' not defined"); - } else if (ent.isParsed()) { - reportValidationProblem(v, "Referenced entity '"+id+"' is not an unparsed entity"); - } - return ent; - } - - /* Too bad this method can not be combined with previous segment -- - * the reason is that DTDValidator does not implement - * InputProblemReporter... - */ - - protected void checkEntity(InputProblemReporter rep, String id, EntityDecl ent) - throws XMLStreamException - { - if (ent == null) { - rep.reportValidationProblem("Referenced entity '"+id+"' not defined"); - } else if (ent.isParsed()) { - rep.reportValidationProblem("Referenced entity '"+id+"' is not an unparsed entity"); - } - } - - /* - /////////////////////////////////////////////////// - // Package methods, error reporting - /////////////////////////////////////////////////// - */ - - protected String reportInvalidChar(DTDValidatorBase v, char c, String msg) - throws XMLStreamException - { - reportValidationProblem(v, "Invalid character "+WstxInputData.getCharDesc(c)+": "+msg); - return null; - } - - protected String reportValidationProblem(DTDValidatorBase v, String msg) - throws XMLStreamException - { - v.reportValidationProblem("Attribute '"+mName+"': "+msg); - return null; - } - - /** - * Method called during parsing of DTD schema, to report a problem. - * Note that unlike during actual validation, we have no option of - * just gracefully listing problems and ignoring them; an exception - * is always thrown. - */ - protected String reportValidationProblem(InputProblemReporter rep, String msg) - throws XMLStreamException - { - rep.reportValidationProblem("Attribute definition '"+mName+"': "+msg); - return null; - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDCdataAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDCdataAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDCdataAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDCdataAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,61 +0,0 @@ -package com.ctc.wstx.dtd; - -import org.codehaus.stax2.validation.XMLValidationException; - -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; - -/** - * Simple {@link DTDAttribute} sub-class used for plain vanilla CDATA - * valued attributes. Although base class implements most of the methods, - * it's better designwise to keep that base class abstract and have - * separate CDATA type as well. - */ -public final class DTDCdataAttr - extends DTDAttribute -{ - public DTDCdataAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDCdataAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - // @Override - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLValidationException - { - // Nothing to do for pure CDATA attributes... - return null; - } - - // @Override - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws javax.xml.stream.XMLStreamException - { - // Nothing to do for CDATA; all values are fine - } - - // @Override - public String normalize(DTDValidatorBase v, char[] cbuf, int start, int end) - { - // Nothing to do for pure CDATA attributes... - return null; - } - - // @Override - public void normalizeDefault() - { - // Nothing to do for pure CDATA attributes... - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDElement.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDElement.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDElement.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDElement.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,576 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in the file LICENSE which is - * included with the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.util.*; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLStreamException; - -import org.codehaus.stax2.validation.XMLValidator; - -import com.ctc.wstx.api.ReaderConfig; -import com.ctc.wstx.cfg.ErrorConsts; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.ExceptionUtil; -import com.ctc.wstx.util.PrefixedName; -import com.ctc.wstx.util.WordResolver; - -/** - * Class that contains element definitions from DTD. - *
- * Notes about thread-safety: this class is not thread-safe, since it does - * not have to be, in general case. That is, the only instances that can - * be shared are external subset instances, and those are used in read-only - * manner (with the exception of temporary arrays constructed on-demand). - */ -public final class DTDElement -{ - - /* - /////////////////////////////////////////////////// - // Information about the element itself - /////////////////////////////////////////////////// - */ - - final PrefixedName mName; - - /** - * Location of the (real) definition of the element; may be null for - * placeholder elements created to hold ATTLIST definitions - */ - final Location mLocation; - - /** - * Base validator object for validating content model of this element; - * may be null for some simple content models (ANY, EMPTY). - */ - StructValidator mValidator; - - int mAllowedContent; - - /** - * True if the DTD was parsed (and is to be used) in namespace-aware - * mode. - * Affects (name) validation amongst other things. - */ - final boolean mNsAware; - - /** - * True if the DTD was parsed in xml1.1 compliant mode (referenced to - * from an xml 1.1 document). - * Affects (name) validation amongst other things. - */ - final boolean mXml11; - - /* - /////////////////////////////////////////////////// - // Attribute info - /////////////////////////////////////////////////// - */ - - HashMap mAttrMap = null; - - /** - * Ordered list of attributes that have 'special' properties (attribute - * is required, has a default value [regular or fixed]); these attributes - * have to be specifically checked after actual values have been resolved. - */ - ArrayList mSpecAttrList = null; - - boolean mAnyFixed = false; - - /** - * Flag set to true if there are any attributes that have either - * basic default value, or #FIXED default value. - */ - boolean mAnyDefaults = false; - - /** - * Flag that is set to true if there is at least one attribute that - * has type that requires normalization and/or validation; that is, - * is of some other type than CDATA. - */ - boolean mValidateAttrs = false; - - /** - * Id attribute instance, if one already declared for this element; - * can only have up to one such attribute per element. - */ - DTDAttribute mIdAttr; - - /** - * Notation attribute instance, if one already declared for this element; - * can only have up to one such attribute per element. - */ - DTDAttribute mNotationAttr; - - // // // !! If you add new attributes, make sure they get copied - // // // in #define() method !! - - /* - /////////////////////////////////////////////////// - // Namespace declaration defaulting... - /////////////////////////////////////////////////// - */ - - /** - * Set of namespace declarations with default values, if any - * (regular ns pseudo-attr declarations are just ignored) - */ - HashMap mNsDefaults = null; // [String : DTDAttribute] - - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - private DTDElement(Location loc, PrefixedName name, - StructValidator val, int allowedContent, - boolean nsAware, boolean xml11) - { - mName = name; - mLocation = loc; - mValidator = val; - mAllowedContent = allowedContent; - mNsAware = nsAware; - mXml11 = xml11; - } - - /** - * Method called to create an actual element definition, matching - * an ELEMENT directive in a DTD subset. - */ - public static DTDElement createDefined(ReaderConfig cfg, Location loc, PrefixedName name, - StructValidator val, int allowedContent) - { - if (allowedContent == XMLValidator.CONTENT_ALLOW_UNDEFINED) { // sanity check - ExceptionUtil.throwInternal("trying to use XMLValidator.CONTENT_ALLOW_UNDEFINED via createDefined()"); - } - return new DTDElement(loc, name, val, allowedContent, - cfg.willSupportNamespaces(), cfg.isXml11()); - } - - /** - * Method called to create a "placeholder" element definition, needed to - * contain attribute definitions. - */ - public static DTDElement createPlaceholder(ReaderConfig cfg, Location loc, PrefixedName name) - { - return new DTDElement(loc, name, null, XMLValidator.CONTENT_ALLOW_UNDEFINED, - cfg.willSupportNamespaces(), cfg.isXml11()); - } - - /** - * Method called on placeholder element, to create a real instance that - * has all attribute definitions placeholder had (it'll always have at - * least one -- otherwise no placeholder was needed). - */ - public DTDElement define(Location loc, StructValidator val, - int allowedContent) - { - verifyUndefined(); - if (allowedContent == XMLValidator.CONTENT_ALLOW_UNDEFINED) { // sanity check - ExceptionUtil.throwInternal("trying to use CONTENT_ALLOW_UNDEFINED via define()"); - } - - DTDElement elem = new DTDElement(loc, mName, val, allowedContent, - mNsAware, mXml11); - - // Ok, need to copy state collected so far: - elem.mAttrMap = mAttrMap; - elem.mSpecAttrList = mSpecAttrList; - elem.mAnyFixed = mAnyFixed; - elem.mValidateAttrs = mValidateAttrs; - elem.mAnyDefaults = mAnyDefaults; - elem.mIdAttr = mIdAttr; - elem.mNotationAttr = mNotationAttr; - elem.mNsDefaults = mNsDefaults; - - return elem; - } - - /** - * Method called to "upgrade" a placeholder using a defined element, - * including adding attributes. - */ - public void defineFrom(InputProblemReporter rep, DTDElement definedElem, - boolean fullyValidate) - throws XMLStreamException - { - if (fullyValidate) { - verifyUndefined(); - } - mValidator = definedElem.mValidator; - mAllowedContent = definedElem.mAllowedContent; - mergeMissingAttributesFrom(rep, definedElem, fullyValidate); - } - - private void verifyUndefined() - { - if (mAllowedContent != XMLValidator.CONTENT_ALLOW_UNDEFINED) { // sanity check - ExceptionUtil.throwInternal("redefining defined element spec"); - } - } - - /** - * Method called by DTD parser when it has read information about - * an attribute that belong to this element - * - * @return Newly created attribute Object if the attribute definition was - * added (hadn't been declared yet); null if it's a duplicate, in which - * case original definition sticks. - */ - public DTDAttribute addAttribute(InputProblemReporter rep, - PrefixedName attrName, int valueType, - DefaultAttrValue defValue, WordResolver enumValues, - boolean fullyValidate) - throws XMLStreamException - { - HashMap m = mAttrMap; - if (m == null) { - mAttrMap = m = new HashMap(); - } - - List specList = defValue.isSpecial() ? getSpecialList() : null; - - DTDAttribute attr; - int specIndex = (specList == null) ? -1 : specList.size(); - - switch (valueType) { - case DTDAttribute.TYPE_CDATA: - attr = new DTDCdataAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_ENUMERATED: - attr = new DTDEnumAttr(attrName, defValue, specIndex, mNsAware, mXml11, enumValues); - break; - - case DTDAttribute.TYPE_ID: - /* note: although ID attributes are not to have default value, - * this is 'only' a validity constraint, and in dtd-aware-but- - * not-validating mode it is apparently 'legal' to add default - * values. Bleech. - */ - attr = new DTDIdAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_IDREF: - attr = new DTDIdRefAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_IDREFS: - attr = new DTDIdRefsAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_ENTITY: - attr = new DTDEntityAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_ENTITIES: - attr = new DTDEntitiesAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_NOTATION: - attr = new DTDNotationAttr(attrName, defValue, specIndex, mNsAware, mXml11, enumValues); - break; - - case DTDAttribute.TYPE_NMTOKEN: - attr = new DTDNmTokenAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - case DTDAttribute.TYPE_NMTOKENS: - attr = new DTDNmTokensAttr(attrName, defValue, specIndex, mNsAware, mXml11); - break; - - default: - // 18-Jan-2006, TSa: should never get here... - ExceptionUtil.throwGenericInternal(); - attr = null; // unreachable, but compiler wants it - } - - DTDAttribute old = doAddAttribute(m, rep, attr, specList, fullyValidate); - return (old == null) ? attr : null; - } - - /** - * Method called to add a definition of a namespace-declaration - * pseudo-attribute with a default value. - * - * @return Attribute that acts as the placeholder, if the declaration - * was added; null to indicate it - * was a dup (there was an earlier declaration) - */ - public DTDAttribute addNsDefault - (InputProblemReporter rep, PrefixedName attrName, int valueType, - DefaultAttrValue defValue, boolean fullyValidate) - throws XMLStreamException - { - /* Let's simplify handling a bit: although theoretically all - * combinations of value can be used, let's really only differentiate - * between CDATA and 'other' (for which let's use NMTOKEN) - */ - DTDAttribute nsAttr; - - switch (valueType) { - case DTDAttribute.TYPE_CDATA: - nsAttr = new DTDCdataAttr(attrName, defValue, -1, mNsAware, mXml11); - break; - default: // something else, default to NMTOKEN then - nsAttr = new DTDNmTokenAttr(attrName, defValue, -1, mNsAware, mXml11); - break; - } - - // Ok. So which prefix are we to bind? Need to access by prefix... - String prefix = attrName.getPrefix(); - if (prefix == null || prefix.length() == 0) { // defult NS -> "" - prefix = ""; - } else { // non-default, use the local name - prefix = attrName.getLocalName(); - } - - if (mNsDefaults == null) { - mNsDefaults = new HashMap(); - } else { - if (mNsDefaults.containsKey(prefix)) { - return null; - } - } - mNsDefaults.put(prefix, nsAttr); - return nsAttr; - } - - public void mergeMissingAttributesFrom(InputProblemReporter rep, DTDElement other, - boolean fullyValidate) - throws XMLStreamException - { - Map otherMap = other.getAttributes(); - HashMap m = mAttrMap; - if (m == null) { - mAttrMap = m = new HashMap(); - } - - //boolean anyAdded = false; - - if (otherMap != null && otherMap.size() > 0) { - Iterator it = otherMap.entrySet().iterator(); - while (it.hasNext()) { - Map.Entry me = (Map.Entry) it.next(); - Object key = me.getKey(); - // Should only add if no such attribute exists... - if (!m.containsKey(key)) { - // can only use as is, if it's not a special attr - DTDAttribute newAttr = (DTDAttribute) me.getValue(); - List specList; - // otherwise need to clone - if (newAttr.isSpecial()) { - specList = getSpecialList(); - newAttr = newAttr.cloneWith(specList.size()); - } else { - specList = null; - } - doAddAttribute(m, rep, newAttr, specList, fullyValidate); - } - } - } - - HashMap otherNs = other.mNsDefaults; - if (otherNs != null) { - if (mNsDefaults == null) { - mNsDefaults = new HashMap(); - } - Iterator it = otherNs.entrySet().iterator(); - while (it.hasNext()) { - Map.Entry me = (Map.Entry) it.next(); - Object key = me.getKey(); - // Should only add if no such attribute exists... - if (!mNsDefaults.containsKey(key)) { - mNsDefaults.put(key, me.getValue()); - } - } - } - } - - /** - * @return Earlier declaration of the attribute, if any; null if - * this was a new attribute - */ - private DTDAttribute doAddAttribute(Map attrMap, InputProblemReporter rep, - DTDAttribute attr, List specList, - boolean fullyValidate) - throws XMLStreamException - { - PrefixedName attrName = attr.getName(); - - // Maybe we already have it? If so, need to ignore - DTDAttribute old = (DTDAttribute) attrMap.get(attrName); - if (old != null) { - rep.reportProblem(null, ErrorConsts.WT_ATTR_DECL, ErrorConsts.W_DTD_DUP_ATTR, - attrName, mName); - return old; - } - - switch (attr.getValueType()) { - case DTDAttribute.TYPE_ID: - // Only one such attribute per element (Specs, 1.0#3.3.1) - if (fullyValidate && mIdAttr != null) { - rep.throwParseError("Invalid id attribute \"{0}\" for element <{1}>: already had id attribute \""+mIdAttr.getName()+"\"", attrName, mName); - } - mIdAttr = attr; - break; - - case DTDAttribute.TYPE_NOTATION: - // Only one such attribute per element (Specs, 1.0#3.3.1) - if (fullyValidate && mNotationAttr != null) { - rep.throwParseError("Invalid notation attribute '"+attrName+"' for element <"+mName+">: already had notation attribute '"+mNotationAttr.getName()+"'"); - } - mNotationAttr = attr; - break; - } - - attrMap.put(attrName, attr); - if (specList != null) { - specList.add(attr); - } - if (!mAnyFixed) { - mAnyFixed = attr.isFixed(); - } - if (!mValidateAttrs) { - mValidateAttrs = attr.needsValidation(); - } - if (!mAnyDefaults) { - mAnyDefaults = attr.hasDefaultValue(); - } - - return null; - } - - /* - /////////////////////////////////////////////////// - // Public API, accessors: - /////////////////////////////////////////////////// - */ - - public PrefixedName getName() { return mName; } - - public String toString() { - return mName.toString(); - } - - public String getDisplayName() { - return mName.toString(); - } - - public Location getLocation() { return mLocation; } - - public boolean isDefined() { - return (mAllowedContent != XMLValidator.CONTENT_ALLOW_UNDEFINED); - } - - /** - * @return Constant that identifies what kind of nodes are in general - * allowed inside this element. - */ - public int getAllowedContent() { - return mAllowedContent; - } - - /** - * Specialized accessor used by non-validating but typing 'validator': - * essentially, used to figure out whether #PCDATA is allowed or not; - * and based on that, return one of 2 allowable text values (only - * space, or anything). This is the relevant subset in non-validating - * modes, needed to properly type resulting character events. - */ - public int getAllowedContentIfSpace() - { - int vld = mAllowedContent; - return (vld <= XMLValidator.CONTENT_ALLOW_WS) ? - XMLValidator.CONTENT_ALLOW_WS_NONSTRICT : - XMLValidator.CONTENT_ALLOW_ANY_TEXT; - } - - public HashMap getAttributes() { - return mAttrMap; - } - - public int getSpecialCount() { - return (mSpecAttrList == null) ? 0 : mSpecAttrList.size(); - } - - public List getSpecialAttrs() { - return mSpecAttrList; - } - - /** - * @return True if at least one of the attributes has type other than - * CDATA; false if not - */ - public boolean attrsNeedValidation() { - return mValidateAttrs; - } - - public boolean hasFixedAttrs() { - return mAnyFixed; - } - - public boolean hasAttrDefaultValues() { - return mAnyDefaults; - } - - public DTDAttribute getIdAttribute() { - return mIdAttr; - } - - public DTDAttribute getNotationAttribute() { - return mNotationAttr; - } - - public boolean hasNsDefaults() { - return (mNsDefaults != null); - } - - /* - /////////////////////////////////////////////////// - // Public API, factory methods: - /////////////////////////////////////////////////// - */ - - public StructValidator getValidator() - { - return (mValidator == null) ? null : mValidator.newInstance(); - } - - protected HashMap getNsDefaults() { - return mNsDefaults; - } - - /* - /////////////////////////////////////////////////// - // Internal methods - /////////////////////////////////////////////////// - */ - - private List getSpecialList() - { - ArrayList l = mSpecAttrList; - if (l == null) { - mSpecAttrList = l = new ArrayList(); - } - return l; - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEntitiesAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEntitiesAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEntitiesAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEntitiesAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,177 +0,0 @@ -package com.ctc.wstx.dtd; - -import java.util.StringTokenizer; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.ent.EntityDecl; -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; - -/** - * Specific attribute class for attributes that contain (unique) - * identifiers. - */ -public final class DTDEntitiesAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. Note that id attributes can never have - * default values. - */ - public DTDEntitiesAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDEntitiesAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_ENTITIES; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the {@link DTDValidatorBase} - * to let the attribute do necessary normalization and/or validation - * for the value. - * - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - /* Let's skip leading/trailing white space, even if we are not - * to normalize visible attribute value. This allows for better - * round-trip handling (no changes for physical value caller - * gets), but still allows succesful validation. - */ - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - // Empty value? - if (start >= end) { - return reportValidationProblem(v, "Empty ENTITIES value"); - } - --end; // so that it now points to the last char - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok; now start points to first, last to last char (both inclusive) - String idStr = null; - StringBuffer sb = null; - - while (start <= end) { - // Ok, need to check char validity, and also calc hash code: - char c = cbuf[start]; - if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as the first ENTITIES character"); - } - int hash = (int) c; - int i = start+1; - for (; i <= end; ++i) { - c = cbuf[i]; - if (WstxInputData.isSpaceChar(c)) { - break; - } - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an ENTITIES character"); - } - hash = (hash * 31) + (int) c; - } - - EntityDecl ent = findEntityDecl(v, cbuf, start, (i - start), hash); - // only returns if entity was found... - - // Can skip the trailing space char (if there was one) - start = i+1; - - /* When normalizing, we can possibly share id String, or - * alternatively, compose normalized String if multiple - */ - if (normalize) { - if (idStr == null) { // first idref - idStr = ent.getName(); - } else { - if (sb == null) { - sb = new StringBuffer(idStr); - } - idStr = ent.getName(); - sb.append(' '); - sb.append(idStr); - } - } - - // Ok, any white space to skip? - while (start <= end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - } - - if (normalize) { - if (sb != null) { - idStr = sb.toString(); - } - return idStr; - } - - return null; - } - - /** - * Method called by the validator object - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String normStr = validateDefaultNames(rep, true); - if (normalize) { - mDefValue.setValue(normStr); - } - - // Ok, but were they declared? - - /* Performance really shouldn't be critical here (only called when - * parsing DTDs, which get cached) -- let's just - * tokenize using standard StringTokenizer - */ - StringTokenizer st = new StringTokenizer(normStr); - /* !!! 03-Dec-2004, TSa: This is rather ugly -- need to know we - * actually really get a DTD reader, and DTD reader needs - * to expose a special method... but it gets things done. - */ - MinimalDTDReader dtdr = (MinimalDTDReader) rep; - while (st.hasMoreTokens()) { - String str = st.nextToken(); - EntityDecl ent = dtdr.findEntity(str); - // Needs to exists, and be an unparsed entity... - checkEntity(rep, normStr, ent); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEntityAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEntityAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEntityAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEntityAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,118 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.ent.EntityDecl; -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; - -/** - * Specific attribute class for attributes that contain (unique) - * identifiers. - */ -public final class DTDEntityAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. Note that id attributes can never have - * default values. - */ - public DTDEntityAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDEntityAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_ENTITY; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the {@link DTDValidatorBase} - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - // Empty value? - if (start >= end) { - return reportValidationProblem(v, "Empty ENTITY value"); - } - --end; // so that it now points to the last char - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok, need to check char validity, and also calc hash code: - char c = cbuf[start]; - if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11) && c != ':') { - return reportInvalidChar(v, c, "not valid as the first ID character"); - } - int hash = (int) c; - - for (int i = start+1; i <= end; ++i) { - c = cbuf[i]; - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an ID character"); - } - hash = (hash * 31) + (int) c; - } - - EntityDecl ent = findEntityDecl(v, cbuf, start, (end - start + 1), hash); - // only returns if it succeeded... - - return normalize ? ent.getName() : null; - } - - /** - * Method called by the validator object - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String normStr = validateDefaultName(rep, normalize); - if (normalize) { - mDefValue.setValue(normStr); - } - - // Ok, but was it declared? - - /* 03-Dec-2004, TSa: This is rather ugly -- need to know we - * actually really get a DTD reader, and DTD reader needs - * to expose a special method... but it gets things done. - */ - EntityDecl ent = ((MinimalDTDReader) rep).findEntity(normStr); - checkEntity(rep, normStr, ent); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEnumAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEnumAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEnumAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEnumAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,93 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; -import com.ctc.wstx.util.WordResolver; - -/** - * Specific attribute class for attributes that have enumerated values. - */ -public final class DTDEnumAttr - extends DTDAttribute -{ - final WordResolver mEnumValues; - - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - public DTDEnumAttr(PrefixedName name, DefaultAttrValue defValue, - int specIndex, boolean nsAware, boolean xml11, - WordResolver enumValues) - { - super(name, defValue, specIndex, nsAware, xml11); - mEnumValues = enumValues; - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDEnumAttr(mName, mDefValue, specIndex, mCfgNsAware, - mCfgXml11, mEnumValues); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_ENUMERATED; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - String ok = validateEnumValue(cbuf, start, end, normalize, mEnumValues); - if (ok == null) { - String val = new String(cbuf, start, (end-start)); - return reportValidationProblem(v, "Invalid enumerated value '"+val+"': has to be one of (" - +mEnumValues+")"); - } - return ok; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String def = validateDefaultNmToken(rep, normalize); - - // And then that it's one of listed values: - String shared = mEnumValues.find(def); - if (shared == null) { - reportValidationProblem(rep, "Invalid default value '"+def+"': has to be one of (" - +mEnumValues+")"); - return; - } - - // Ok, cool it's ok... - if (normalize) { - mDefValue.setValue(shared); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEventListener.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEventListener.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDEventListener.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDEventListener.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,51 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in file LICENSE, included with - * the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.net.URL; - -import javax.xml.stream.XMLStreamException; - -public interface DTDEventListener -{ - // Configuration - - /** - * @return True, if there is a listener interested in getting comment - * events within DTD subset (since that's optional) - */ - public boolean dtdReportComments(); - - // Basic content events - - public void dtdProcessingInstruction(String target, String data); - public void dtdComment(char[] data, int offset, int len); - public void dtdSkippedEntity(String name); - - // DTD declarations that must be exposed - public void dtdNotationDecl(String name, String publicId, String systemId, URL baseURL) - throws XMLStreamException; - - public void dtdUnparsedEntityDecl(String name, String publicId, String systemId, String notationName, URL baseURL) - throws XMLStreamException; - - // DTD declarations that can be exposed - - public void attributeDecl(String eName, String aName, String type, String mode, String value); - public void dtdElementDecl(String name, String model); - public void dtdExternalEntityDecl(String name, String publicId, String systemId); - public void dtdInternalEntityDecl(String name, String value); -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,131 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.cfg.ErrorConsts; -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.ElementId; -import com.ctc.wstx.util.ElementIdMap; -import com.ctc.wstx.util.PrefixedName; - -/** - * Specific attribute class for attributes that contain (unique) - * identifiers. - */ -public final class DTDIdAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. Note that id attributes can never have - * default values. - *
- * note: although ID attributes are not to have default value, - * this is 'only' a validity constraint, and in dtd-aware-but- - * not-validating mode it is apparently 'legal' to add default - * values. - */ - public DTDIdAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDIdAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_ID; - } - - public boolean typeIsId() { - return true; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - // Let's trim leading white space first... - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - // No id? - if (start >= end) { - return reportValidationProblem(v, "Empty ID value"); - } - --end; // so that it now points to the last char - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok, need to check char validity, and also calc hash code: - char c = cbuf[start]; - if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as the first ID character"); - } - int hash = (int) c; - for (int i = start+1; i <= end; ++i) { - c = cbuf[i]; - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an ID character"); - } - hash = (hash * 31) + (int) c; - } - - // Either way, we do need to validate characters, and calculate hash - ElementIdMap m = v.getIdMap(); - PrefixedName elemName = v.getElemName(); - Location loc = v.getLocation(); - ElementId id = m.addDefined(cbuf, start, (end - start + 1), hash, - loc, elemName, mName); - - // We can detect dups by checking if Location is the one we passed: - if (id.getLocation() != loc) { - return reportValidationProblem(v, "Duplicate id '"+id.getId()+"', first declared at " - +id.getLocation()); - } - - if (normalize) { - return id.getId(); - } - return null; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - { - // Should never get called - throw new IllegalStateException(ErrorConsts.ERR_INTERNAL); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDId.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDId.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDId.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDId.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,139 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in the file LICENSE which is - * included with the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.net.URI; - -/** - * Simple key object class, used for accessing (external) DTDs when stored for - * caching. Main idea is that the primary id of a DTD (public or system id; - * latter normalized if possible) - * has to match, as well as couple of on/off settings for parsing (namespace - * support, text normalization). - * Latter restriction is needed since although DTDs do not deal - * with (or understand) namespaces, some parsing is done to be able to validate - * namespace aware/non-aware documents, and handling differs between the two. - * As to primary key part, public id is used if one was defined; if so, - * comparison is String equality. If not, then system id is compared: system - * id has to be expressed as URL if so. - */ -public final class DTDId -{ - protected final String mPublicId; - - protected final URI mSystemId; - - protected final int mConfigFlags; - - protected final boolean mXml11; - - protected int mHashCode = 0; - - /* - /////////////////////////////////////////////////////////////////////// - // Life-cycle: - /////////////////////////////////////////////////////////////////////// - */ - - private DTDId(String publicId, URI systemId, int configFlags, boolean xml11) - { - mPublicId = publicId; - mSystemId = systemId; - mConfigFlags = configFlags; - mXml11 = xml11; - } - - public static DTDId constructFromPublicId(String publicId, int configFlags, - boolean xml11) - { - if (publicId == null || publicId.length() == 0) { - throw new IllegalArgumentException("Empty/null public id."); - } - return new DTDId(publicId, null, configFlags, xml11); - } - - public static DTDId constructFromSystemId(URI systemId, int configFlags, - boolean xml11) - { - if (systemId == null) { - throw new IllegalArgumentException("Null system id."); - } - return new DTDId(null, systemId, configFlags, xml11); - } - - public static DTDId construct(String publicId, URI systemId, int configFlags, boolean xml11) - { - if (publicId != null && publicId.length() > 0) { - return new DTDId(publicId, null, configFlags, xml11); - } - if (systemId == null) { - throw new IllegalArgumentException("Illegal arguments; both public and system id null/empty."); - } - return new DTDId(null, systemId, configFlags, xml11); - } - - /* - /////////////////////////////////////////////////////////////////////// - // Overridden standard methods - /////////////////////////////////////////////////////////////////////// - */ - - public int hashCode() { - int hash = mHashCode; - if (hash == 0) { - hash = mConfigFlags; - if (mPublicId != null) { - hash ^= mPublicId.hashCode(); - } else { - hash ^= mSystemId.hashCode(); - } - if (mXml11) { - hash ^= 1; - } - mHashCode = hash; - } - return hash; - } - - public String toString() { - StringBuffer sb = new StringBuffer(60); - sb.append("Public-id: "); - sb.append(mPublicId); - sb.append(", system-id: "); - sb.append(mSystemId); - sb.append(" [config flags: 0x"); - sb.append(Integer.toHexString(mConfigFlags)); - sb.append("], xml11: "); - sb.append(mXml11); - return sb.toString(); - } - - public boolean equals(Object o) - { - if (o == this) return true; - if (o == null || o.getClass() != getClass()) return false; - DTDId other = (DTDId) o; - if (other.mConfigFlags != mConfigFlags - || other.mXml11 != mXml11) { - return false; - } - if (mPublicId != null) { - String op = other.mPublicId; - return (op != null) && op.equals(mPublicId); - } - return mSystemId.equals(other.mSystemId); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdRefAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdRefAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdRefAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdRefAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,116 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.ElementId; -import com.ctc.wstx.util.ElementIdMap; -import com.ctc.wstx.util.PrefixedName; - -/** - * Attribute class for attributes that contain references - * to elements that have matching identifier specified. - */ -public final class DTDIdRefAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. - */ - public DTDIdRefAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDIdRefAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_IDREF; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - /* Let's skip leading/trailing white space, even if we are not - * to normalize visible attribute value. This allows for better - * round-trip handling, but still allow validation. - */ - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - if (start >= end) { // empty (all white space) value? - return reportValidationProblem(v, "Empty IDREF value"); - } - - --end; // so that it now points to the last char - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok, need to check char validity, and also calc hash code: - char c = cbuf[start]; - if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as the first IDREF character"); - } - int hash = (int) c; - for (int i = start+1; i <= end; ++i) { - c = cbuf[i]; - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an IDREF character"); - } - hash = (hash * 31) + (int) c; - } - - // Ok, let's check and update id ref list... - ElementIdMap m = v.getIdMap(); - Location loc = v.getLocation(); - ElementId id = m.addReferenced(cbuf, start, (end - start + 1), hash, - loc, v.getElemName(), mName); - // and that's all; no more checks needed here - return normalize ? id.getId() : null; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String def = validateDefaultName(rep, normalize); - if (normalize) { - mDefValue.setValue(def); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,158 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.ElementId; -import com.ctc.wstx.util.ElementIdMap; -import com.ctc.wstx.util.PrefixedName; - -/** - * Attribute class for attributes that contain multiple references - * to elements that have matching identifier specified. - */ -public final class DTDIdRefsAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. - */ - public DTDIdRefsAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDIdRefsAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_IDREFS; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - /* Let's skip leading/trailing white space, even if we are not - * to normalize visible attribute value. This allows for better - * round-trip handling (no changes for physical value caller - * gets), but still allows succesful validation. - */ - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - // No id? - if (start >= end) { - return reportValidationProblem(v, "Empty IDREFS value"); - } - - --end; // so that it now points to the last char - // We now the first char is not a space by now... - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok; now start points to first, end to last char (both inclusive) - ElementIdMap m = v.getIdMap(); - Location loc = v.getLocation(); - - String idStr = null; - StringBuffer sb = null; - while (start <= end) { - // Ok, need to check char validity, and also calc hash code: - char c = cbuf[start]; - if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as the first IDREFS character"); - } - int hash = (int) c; - int i = start+1; - for (; i <= end; ++i) { - c = cbuf[i]; - if (WstxInputData.isSpaceChar(c)) { - break; - } - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an IDREFS character"); - } - hash = (hash * 31) + (int) c; - } - - // Ok, got the next id ref... - ElementId id = m.addReferenced(cbuf, start, i - start, hash, - loc, v.getElemName(), mName); - - // Can skip the trailing space char (if there was one) - start = i+1; - - /* When normalizing, we can possibly share id String, or - * alternatively, compose normalized String if multiple - */ - if (normalize) { - if (idStr == null) { // first idref - idStr = id.getId(); - } else { - if (sb == null) { - sb = new StringBuffer(idStr); - } - idStr = id.getId(); - sb.append(' '); - sb.append(idStr); - } - } - - // Ok, any white space to skip? - while (start <= end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - } - - if (normalize) { - if (sb != null) { - idStr = sb.toString(); - } - return idStr; - } - - return null; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - *
- * It's unlikely there will be default values... but just in case, - * let's implement it properly. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String def = validateDefaultNames(rep, normalize); - if (normalize) { - mDefValue.setValue(def); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,108 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; - -/** - * Specific attribute class for attributes that contain (unique) - * identifiers. - */ -public final class DTDNmTokenAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. - */ - public DTDNmTokenAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDNmTokenAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_NMTOKEN; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - int origLen = end-start; - - // Let's trim leading white space first... - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - - // Empty value? - if (start >= end) { - return reportValidationProblem(v, "Empty NMTOKEN value"); - } - - --end; // so that it now points to the last char - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - } - - // Ok, need to check char validity - for (int i = start; i <= end; ++i) { - char c = cbuf[i]; - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid NMTOKEN character"); - } - } - - if (normalize) { - // Let's only create the String if we trimmed something - int len = (end - start)+1; - if (len != origLen) { - return new String(cbuf, start, len); - } - } - return null; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String def = validateDefaultNmToken(rep, normalize); - if (normalize) { - mDefValue.setValue(def); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,205 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; - -/** - * Specific attribute class for attributes that contain (unique) - * identifiers. - */ -public final class DTDNmTokensAttr - extends DTDAttribute -{ - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - /** - * Main constructor. - */ - public DTDNmTokensAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex, - boolean nsAware, boolean xml11) - { - super(name, defValue, specIndex, nsAware, xml11); - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDNmTokensAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_NMTOKENS; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - //int origStart = start; - - /* First things first; let's ensure value is not empty (all - * white space)... - */ - while (start < end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - // Empty value? - if (start >= end) { - return reportValidationProblem(v, "Empty NMTOKENS value"); - } - - /* Then, let's have separate handling for normalizing and - * non-normalizing case, since latter is trivially easy case: - */ - if (!normalize) { - for (; start < end; ++start) { - char c = cbuf[start]; - if (!WstxInputData.isSpaceChar(c) - && !WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as NMTOKENS character"); - } - } - return null; // ok, all good - } - - //boolean trimmed = (origStart != start); - //origStart = start; - - --end; // so that it now points to the last char - // Wouldn't absolutely have to trim trailing... but is easy to do - while (end > start && WstxInputData.isSpaceChar(cbuf[end])) { - --end; - //trimmed = true; - } - - /* Ok, now, need to check we only have valid chars, and maybe - * also coalesce multiple spaces, if any. - */ - StringBuffer sb = null; - - while (start <= end) { - int i = start; - for (; i <= end; ++i) { - char c = cbuf[i]; - if (WstxInputData.isSpaceChar(c)) { - break; - } - if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) { - return reportInvalidChar(v, c, "not valid as an NMTOKENS character"); - } - } - - if (sb == null) { - sb = new StringBuffer(end - start + 1); - } else { - sb.append(' '); - } - sb.append(cbuf, start, (i - start)); - - start = i + 1; - // Ok, any white space to skip? - while (start <= end && WstxInputData.isSpaceChar(cbuf[start])) { - ++start; - } - } - - /* 27-Nov-2005, TSa: Could actually optimize trimming, and often - * avoid using StringBuffer... but let's only do it if it turns - * out dealing with NMTOKENS normalization shows up on profiling... - */ - return sb.toString(); - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - String defValue = mDefValue.getValue(); - int len = defValue.length(); - - // Then code similar to actual value validation: - StringBuffer sb = null; - int count = 0; - int start = 0; - - main_loop: - while (start < len) { - char c = defValue.charAt(start); - - // Ok, any white space to skip? - while (true) { - if (!WstxInputData.isSpaceChar(c)) { - break; - } - if (++start >= len) { - break main_loop; - } - c = defValue.charAt(start); - } - - int i = start+1; - - do { - if (++i >= len) { - break; - } - c = defValue.charAt(i); - } while (!WstxInputData.isSpaceChar(c)); - ++count; - String token = defValue.substring(start, i); - int illegalIx = WstxInputData.findIllegalNmtokenChar(token, mCfgNsAware, mCfgXml11); - if (illegalIx >= 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; character #"+illegalIx+" (" - +WstxInputData.getCharDesc(defValue.charAt(illegalIx)) - +") not a valid NMTOKENS character"); - } - - if (normalize) { - if (sb == null) { - sb = new StringBuffer(i - start + 32); - } else { - sb.append(' '); - } - sb.append(token); - } - start = i+1; - } - - if (count == 0) { - reportValidationProblem(rep, "Invalid default value '"+defValue - +"'; empty String is not a valid NMTOKENS value"); - return; - } - - if (normalize) { - mDefValue.setValue(sb.toString()); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNotationAttr.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNotationAttr.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDNotationAttr.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDNotationAttr.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,100 +0,0 @@ -package com.ctc.wstx.dtd; - -import javax.xml.stream.XMLStreamException; - -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.PrefixedName; -import com.ctc.wstx.util.WordResolver; - -/** - * Specific attribute class for attributes that are of NOTATION type, - * and also contain enumerated set of legal values. - */ -public final class DTDNotationAttr - extends DTDAttribute -{ - final WordResolver mEnumValues; - - /* - /////////////////////////////////////////////////// - // Life-cycle - /////////////////////////////////////////////////// - */ - - public DTDNotationAttr(PrefixedName name, DefaultAttrValue defValue, - int specIndex, boolean nsAware, boolean xml11, - WordResolver enumValues) - { - super(name, defValue, specIndex, nsAware, xml11); - mEnumValues = enumValues; - } - - public DTDAttribute cloneWith(int specIndex) - { - return new DTDNotationAttr(mName, mDefValue, specIndex, - mCfgNsAware, mCfgXml11, mEnumValues); - } - - /* - /////////////////////////////////////////////////// - // Public API - /////////////////////////////////////////////////// - */ - - public int getValueType() { - return TYPE_NOTATION; - } - - public boolean typeIsNotation() { - return true; - } - - /* - /////////////////////////////////////////////////// - // Public API, validation - /////////////////////////////////////////////////// - */ - - /** - * Method called by the validator - * to let the attribute do necessary normalization and/or validation - * for the value. - *
- * Note: identical to the implementation in {@link DTDEnumAttr} - */ - public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize) - throws XMLStreamException - { - String ok = validateEnumValue(cbuf, start, end, normalize, mEnumValues); - if (ok == null) { - String val = new String(cbuf, start, (end-start)); - return reportValidationProblem(v, "Invalid notation value '"+val+"': has to be one of (" - +mEnumValues+")"); - } - return ok; - } - - /** - * Method called by the validator - * to ask attribute to verify that the default it has (if any) is - * valid for such type. - */ - public void validateDefault(InputProblemReporter rep, boolean normalize) - throws XMLStreamException - { - // First, basic checks that it's a valid non-empty name: - String def = validateDefaultName(rep, normalize); - - // And then that it's one of listed values: - String shared = mEnumValues.find(def); - if (shared == null) { - reportValidationProblem(rep, "Invalid default value '"+def+"': has to be one of (" - +mEnumValues+")"); - } - - // Ok, cool it's ok... - if (normalize) { - mDefValue.setValue(shared); - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSchemaFactory.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSchemaFactory.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSchemaFactory.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSchemaFactory.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,202 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in the file LICENSE which is - * included with the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.io.*; -import java.net.URL; - -import javax.xml.stream.*; - -import org.codehaus.stax2.validation.*; - -import com.ctc.wstx.api.ReaderConfig; -import com.ctc.wstx.api.ValidatorConfig; -import com.ctc.wstx.cfg.XmlConsts; -import com.ctc.wstx.exc.WstxIOException; -import com.ctc.wstx.io.*; -import com.ctc.wstx.util.DefaultXmlSymbolTable; -import com.ctc.wstx.util.SymbolTable; -import com.ctc.wstx.util.URLUtil; - -/** - * Factory for creating DTD validator schema objects (shareable stateless - * "blueprints" for creating actual validators). - *
- * Due to close coupling of XML and DTD, some of the functionality - * implemented (like that of reading internal subsets embedded in XML - * documents) is only accessible by core Woodstox. The externally - * accessible - */ -public class DTDSchemaFactory - extends XMLValidationSchemaFactory -{ - /* - ///////////////////////////////////////////////////// - // Objects shared by actual parsers - ///////////////////////////////////////////////////// - */ - - /** - * 'Root' symbol table, used for creating actual symbol table instances, - * but never as is. - */ - final static SymbolTable mRootSymbols = DefaultXmlSymbolTable.getInstance(); - static { - mRootSymbols.setInternStrings(true); - } - - /** - * Current configurations for this factory - */ - protected final ValidatorConfig mSchemaConfig; - - /** - * This configuration object is used (instead of a more specific one) - * since the actual DTD reader uses such configuration object. - */ - protected final ReaderConfig mReaderConfig; - - public DTDSchemaFactory() - { - super(XMLValidationSchema.SCHEMA_ID_DTD); - mReaderConfig = ReaderConfig.createFullDefaults(); - mSchemaConfig = ValidatorConfig.createDefaults(); - } - - /* - //////////////////////////////////////////////////////////// - // Stax2, Configuration methods - //////////////////////////////////////////////////////////// - */ - - public boolean isPropertySupported(String propName) - { - return mSchemaConfig.isPropertySupported(propName); - } - - public boolean setProperty(String propName, Object value) - { - return mSchemaConfig.setProperty(propName, value); - } - - public Object getProperty(String propName) - { - return mSchemaConfig.getProperty(propName); - } - - /* - //////////////////////////////////////////////////////////// - // Stax2, Factory methods - //////////////////////////////////////////////////////////// - */ - - public XMLValidationSchema createSchema(InputStream in, String encoding, - String publicId, String systemId) - throws XMLStreamException - { - ReaderConfig rcfg = createPrivateReaderConfig(); - return doCreateSchema(rcfg, StreamBootstrapper.getInstance - (publicId, systemId, in), publicId, systemId, null); - } - - public XMLValidationSchema createSchema(Reader r, String publicId, - String systemId) - throws XMLStreamException - { - ReaderConfig rcfg = createPrivateReaderConfig(); - return doCreateSchema(rcfg, ReaderBootstrapper.getInstance - (publicId, systemId, r, null), publicId, systemId, null); - } - - public XMLValidationSchema createSchema(URL url) - throws XMLStreamException - { - ReaderConfig rcfg = createPrivateReaderConfig(); - try { - InputStream in = URLUtil.inputStreamFromURL(url); - return doCreateSchema(rcfg, StreamBootstrapper.getInstance - (null, null, in), - null, url.toExternalForm(), url); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - - public XMLValidationSchema createSchema(File f) - throws XMLStreamException - { - ReaderConfig rcfg = createPrivateReaderConfig(); - try { - URL url = f.toURL(); - return doCreateSchema(rcfg, StreamBootstrapper.getInstance - (null, null, new FileInputStream(f)), - null, url.toExternalForm(), url); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - - /* - //////////////////////////////////////////////////////////// - // Woodstox-specific API - //////////////////////////////////////////////////////////// - */ - - /* - //////////////////////////////////////////////////////////// - // Internal methods - //////////////////////////////////////////////////////////// - */ - - /** - * The main validator construction method, called by all externally - * visible methods. - */ - protected XMLValidationSchema doCreateSchema - (ReaderConfig rcfg, InputBootstrapper bs, String publicId, String systemId, URL ctxt) - throws XMLStreamException - { - try { - Reader r = bs.bootstrapInput(rcfg, false, XmlConsts.XML_V_UNKNOWN); - if (bs.declaredXml11()) { - rcfg.enableXml11(true); - } - if (ctxt == null) { // this is just needed as context for param entity expansion - ctxt = URLUtil.urlFromCurrentDir(); - } - /* Note: need to pass unknown for 'xmlVersion' here (as well as - * above for bootstrapping), since this is assumed to be the main - * level parsed document and no xml version compatibility checks - * should be done. - */ - WstxInputSource src = InputSourceFactory.constructEntitySource - (rcfg, null, null, bs, publicId, systemId, XmlConsts.XML_V_UNKNOWN, ctxt, r); - - /* true -> yes, fully construct for validation - * (does not mean it has to be used for validation, but required - * if it is to be used for that purpose) - */ - return FullDTDReader.readExternalSubset(src, rcfg, /*int.subset*/null, true, bs.getDeclaredVersion()); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - - private ReaderConfig createPrivateReaderConfig() - { - return mReaderConfig.createNonShared(mRootSymbols.makeChild()); - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSubsetImpl.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSubsetImpl.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSubsetImpl.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSubsetImpl.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,527 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in file LICENSE, included with - * the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.text.MessageFormat; -import java.util.*; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLStreamException; -import javax.xml.stream.events.NotationDeclaration; - -import org.codehaus.stax2.validation.*; - -import com.ctc.wstx.cfg.ErrorConsts; -import com.ctc.wstx.exc.WstxParsingException; -import com.ctc.wstx.sr.InputProblemReporter; -import com.ctc.wstx.util.DataUtil; - -/** - * The default implementation of {@link DTDSubset} - */ -public final class DTDSubsetImpl - extends DTDSubset -{ - /** - * Whether this subset is cachable. Only those external - * subsets that do not refer to PEs defined by internal subsets (or - * GEs via default attribute value expansion) are cachable. - */ - final boolean mIsCachable; - - /** - * Whether this subset has full validation information; and - * consequently whether it will do actual validation, or just allow - * access to type information, notations, entities, and add default - * attribute values. - */ - final boolean mFullyValidating; - - /** - * Flag that indicates whether any of the elements declarared - * has any attribute default values for namespace pseudo-attributes. - */ - final boolean mHasNsDefaults; - - /* - ////////////////////////////////////////////////////// - // Entity information - ////////////////////////////////////////////////////// - */ - - /** - * Map (name-to-EntityDecl) of general entity declarations (internal, - * external) for this DTD subset. - */ - final HashMap mGeneralEntities; - - /** - * Lazily instantiated List that contains all notations from - * {@link #mGeneralEntities} (preferably in their declaration order; depends - * on whether platform, ie. JDK version, has insertion-ordered - * Maps available), used by DTD event Objects. - */ - volatile transient List mGeneralEntityList = null; - - /** - * Set of names of general entities references by this subset. Note that - * only those GEs that are referenced by default attribute value - * definitions count, since GEs in text content are only expanded - * when reading documents, but attribute default values are expanded - * when reading DTD subset itself. - *
- * Needed - * for determinining if external subset materially depends on definitions - * from internal subset; if so, such subset is not cachable. - * This also - * means that information is not stored for non-cachable instance. - */ - final Set mRefdGEs; - - // // // Parameter entity info: - - /** - * Map (name-to-WEntityDeclaration) that contains all parameter entities - * defined by this subset. May be empty if such information will not be - * needed for use; for example, external subset's definitions are needed, - * nor are combined DTD set's. - */ - final HashMap mDefinedPEs; - - /** - * Set of names of parameter entities references by this subset. Needed - * when determinining if external subset materially depends on definitions - * from internal subset, which is needed to know when caching external - * subsets. - *
- * Needed - * for determinining if external subset materially depends on definitions - * from internal subset; if so, such subset is not cachable. - * This also - * means that information is not stored for non-cachable instance. - */ - final Set mRefdPEs; - - /* - ////////////////////////////////////////////////////// - // Notation definitions: - ////////////////////////////////////////////////////// - */ - - /** - * Map (name-to-NotationDecl) that this subset has defined. - */ - final HashMap mNotations; - - /** - * Lazily instantiated List that contains all notations from - * {@link #mNotations} (preferably in their declaration order; depends - * on whether platform, ie. JDK version, has insertion-ordered - * Maps available), used by DTD event Objects. - */ - transient List mNotationList = null; - - - /* - ////////////////////////////////////////////////////// - // Element definitions: - ////////////////////////////////////////////////////// - */ - - final HashMap mElements; - - /* - ////////////////////////////////////////////////////// - // Life-cycle - ////////////////////////////////////////////////////// - */ - - private DTDSubsetImpl(boolean cachable, - HashMap genEnt, Set refdGEs, - HashMap paramEnt, Set peRefs, - HashMap notations, HashMap elements, - boolean fullyValidating) - { - mIsCachable = cachable; - mGeneralEntities = genEnt; - mRefdGEs = refdGEs; - mDefinedPEs = paramEnt; - mRefdPEs = peRefs; - mNotations = notations; - mElements = elements; - mFullyValidating = fullyValidating; - - boolean anyNsDefs = false; - if (elements != null) { - Iterator it = elements.values().iterator(); - while (it.hasNext()) { - DTDElement elem = (DTDElement) it.next(); - if (elem.hasNsDefaults()) { - anyNsDefs = true; - break; - } - } - } - mHasNsDefaults = anyNsDefs; - } - - public static DTDSubsetImpl constructInstance(boolean cachable, - HashMap genEnt, Set refdGEs, - HashMap paramEnt, Set refdPEs, - HashMap notations, HashMap elements, - boolean fullyValidating) - { - return new DTDSubsetImpl(cachable, genEnt, refdGEs, - paramEnt, refdPEs, - notations, elements, - fullyValidating); - } - - /** - * Method that will combine definitions from internal and external subsets, - * producing a single DTD set. - */ - public DTDSubset combineWithExternalSubset(InputProblemReporter rep, DTDSubset extSubset) - throws XMLStreamException - { - /* First let's see if we can just reuse GE Map used by int or ext - * subset; (if only one has contents), or if not, combine them. - */ - HashMap ge1 = getGeneralEntityMap(); - HashMap ge2 = extSubset.getGeneralEntityMap(); - if (ge1 == null || ge1.isEmpty()) { - ge1 = ge2; - } else { - if (ge2 != null && !ge2.isEmpty()) { - /* Internal subset Objects are never shared or reused (and by - * extension, neither are objects they contain), so we can just - * modify GE map if necessary - */ - combineMaps(ge1, ge2); - } - } - - // Ok, then, let's combine notations similarly - HashMap n1 = getNotationMap(); - HashMap n2 = extSubset.getNotationMap(); - if (n1 == null || n1.isEmpty()) { - n1 = n2; - } else { - if (n2 != null && !n2.isEmpty()) { - /* First; let's make sure there are no colliding notation - * definitions: it's an error to try to redefine notations. - */ - checkNotations(n1, n2); - - /* Internal subset Objects are never shared or reused (and by - * extension, neither are objects they contain), so we can just - * modify notation map if necessary - */ - combineMaps(n1, n2); - } - } - - - // And finally elements, rather similarly: - HashMap e1 = getElementMap(); - HashMap e2 = extSubset.getElementMap(); - if (e1 == null || e1.isEmpty()) { - e1 = e2; - } else { - if (e2 != null && !e2.isEmpty()) { - /* Internal subset Objects are never shared or reused (and by - * extension, neither are objects they contain), so we can just - * modify element map if necessary - */ - combineElements(rep, e1, e2); - } - } - - /* Combos are not cachable, and because of that, there's no point - * in storing any PE info either. - */ - return constructInstance(false, ge1, null, null, null, n1, e1, - mFullyValidating); - } - - /* - ////////////////////////////////////////////////////// - // XMLValidationSchema implementation - ////////////////////////////////////////////////////// - */ - - public XMLValidator createValidator(ValidationContext ctxt) - throws XMLStreamException - { - if (mFullyValidating) { - return new DTDValidator(this, ctxt, mHasNsDefaults, - getElementMap(), getGeneralEntityMap()); - } - return new DTDTypingNonValidator(this, ctxt, mHasNsDefaults, - getElementMap(), getGeneralEntityMap()); - - } - - /* - ////////////////////////////////////////////////////// - // DTDValidationSchema implementation - ////////////////////////////////////////////////////// - */ - - public int getEntityCount() { - return (mGeneralEntities == null) ? 0 : mGeneralEntities.size(); - } - - public int getNotationCount() { - return (mNotations == null) ? 0 : mNotations.size(); - } - - /* - ////////////////////////////////////////////////////// - // Woodstox-specific public API - ////////////////////////////////////////////////////// - */ - - public boolean isCachable() { - return mIsCachable; - } - - public HashMap getGeneralEntityMap() { - return mGeneralEntities; - } - - public List getGeneralEntityList() - { - List l = mGeneralEntityList; - if (l == null) { - if (mGeneralEntities == null || mGeneralEntities.size() == 0) { - l = Collections.EMPTY_LIST; - } else { - l = Collections.unmodifiableList(new ArrayList(mGeneralEntities.values())); - } - mGeneralEntityList = l; - } - - return l; - } - - public HashMap getParameterEntityMap() { - return mDefinedPEs; - } - - public HashMap getNotationMap() { - return mNotations; - } - - public synchronized List getNotationList() - { - List l = mNotationList; - if (l == null) { - if (mNotations == null || mNotations.size() == 0) { - l = Collections.EMPTY_LIST; - } else { - l = Collections.unmodifiableList(new ArrayList(mNotations.values())); - } - mNotationList = l; - } - - return l; - } - - public HashMap getElementMap() { - return mElements; - } - - /** - * Method used in determining whether cached external subset instance - * can be used with specified internal subset. If ext. subset references - * any parameter/general entities int subset (re-)defines, it can not; - * otherwise it can be used. - * - * @return True if this (external) subset refers to a parameter entity - * defined in passed-in internal subset. - */ - public boolean isReusableWith(DTDSubset intSubset) - { - Set refdPEs = mRefdPEs; - - if (refdPEs != null && refdPEs.size() > 0) { - HashMap intPEs = intSubset.getParameterEntityMap(); - if (intPEs != null && intPEs.size() > 0) { - if (DataUtil.anyValuesInCommon(refdPEs, intPEs.keySet())) { - return false; - } - } - } - Set refdGEs = mRefdGEs; - - if (refdGEs != null && refdGEs.size() > 0) { - HashMap intGEs = intSubset.getGeneralEntityMap(); - if (intGEs != null && intGEs.size() > 0) { - if (DataUtil.anyValuesInCommon(refdGEs, intGEs.keySet())) { - return false; - } - } - } - return true; // yep, no dependencies overridden - } - - /* - ////////////////////////////////////////////////////// - // Overridden default methods: - ////////////////////////////////////////////////////// - */ - - public String toString() { - StringBuffer sb = new StringBuffer(); - sb.append("[DTDSubset: "); - int count = getEntityCount(); - sb.append(count); - sb.append(" general entities"); - sb.append(']'); - return sb.toString(); - } - - /* - ////////////////////////////////////////////////////// - // Convenience methods used by other classes - ////////////////////////////////////////////////////// - */ - - public static void throwNotationException(NotationDeclaration oldDecl, NotationDeclaration newDecl) - throws XMLStreamException - { - throw new WstxParsingException - (MessageFormat.format(ErrorConsts.ERR_DTD_NOTATION_REDEFD, - new Object[] { - newDecl.getName(), - oldDecl.getLocation().toString()}), - newDecl.getLocation()); - } - - public static void throwElementException(DTDElement oldElem, Location loc) - throws XMLStreamException - { - throw new WstxParsingException - (MessageFormat.format(ErrorConsts.ERR_DTD_ELEM_REDEFD, - new Object[] { - oldElem.getDisplayName(), - oldElem.getLocation().toString() }), - loc); - } - - /* - ////////////////////////////////////////////////////// - // Internal methods - ////////////////////////////////////////////////////// - */ - - /** - *
- * Note: The first Map argument WILL be modified; second one - * not. Caller needs to ensure this is acceptable. - */ - private static void combineMaps(HashMap m1, HashMap m2) - { - Iterator it = m2.entrySet().iterator(); - while (it.hasNext()) { - Map.Entry me = (Map.Entry) it.next(); - Object key = me.getKey(); - /* Int. subset has precedence, but let's guess most of - * the time there are no collisions: - */ - Object old = m1.put(key, me.getValue()); - // Oops, got value! Let's put it back - if (old != null) { - m1.put(key, old); - } - } - } - - /** - * Method that will try to merge in elements defined in the external - * subset, into internal subset; it will also check for redeclarations - * when doing this, as it's invalid to redeclare elements. Care has to - * be taken to only check actual redeclarations: placeholders should - * not cause problems. - */ - private void combineElements(InputProblemReporter rep, HashMap intElems, HashMap extElems) - throws XMLStreamException - { - Iterator it = extElems.entrySet().iterator(); - while (it.hasNext()) { - Map.Entry me = (Map.Entry) it.next(); - Object key = me.getKey(); - Object extVal = me.getValue(); - Object oldVal = intElems.get(key); - - // If there was no old value, can just merge new one in and continue - if (oldVal == null) { - intElems.put(key, extVal); - continue; - } - - DTDElement extElem = (DTDElement) extVal; - DTDElement intElem = (DTDElement) oldVal; - - // Which one is defined (if either)? - if (extElem.isDefined()) { // one from the ext subset - if (intElem.isDefined()) { // but both can't be; that's an error - throwElementException(intElem, extElem.getLocation()); - } else { - /* Note: can/should not modify the external element (by - * for example adding attributes); external element may - * be cached and shared... so, need to do the reverse, - * define the one from internal subset. - */ - intElem.defineFrom(rep, extElem, mFullyValidating); - } - } else { - if (!intElem.isDefined()) { - /* ??? Should we warn about neither of them being really - * declared? - */ - rep.reportProblem(intElem.getLocation(), - ErrorConsts.WT_ENT_DECL, - ErrorConsts.W_UNDEFINED_ELEM, - extElem.getDisplayName(), null); - - } else { - intElem.mergeMissingAttributesFrom(rep, extElem, mFullyValidating); - } - } - } - } - - private static void checkNotations(HashMap fromInt, HashMap fromExt) - throws XMLStreamException - { - /* Since it's external subset that would try to redefine things - * defined in internal subset, let's traverse definitions in - * the ext. subset first (even though that may not be the fastest - * way), so that we have a chance of catching the first problem - * (As long as Maps iterate in insertion order). - */ - Iterator it = fromExt.entrySet().iterator(); - while (it.hasNext()) { - Map.Entry en = (Map.Entry) it.next(); - if (fromInt.containsKey(en.getKey())) { - throwNotationException((NotationDeclaration) fromInt.get(en.getKey()), - (NotationDeclaration) en.getValue()); - } - } - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSubset.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSubset.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDSubset.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDSubset.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,115 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in the file LICENSE which is - * included with the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.util.*; - -import javax.xml.stream.XMLStreamException; - -import org.codehaus.stax2.validation.*; - -import com.ctc.wstx.sr.InputProblemReporter; - -/** - * This is the abstract base class that implements the standard Stax2 - * validation schema base class ({@link XMLValidationSchema}, as well - * as specifies extended Woodstox-specific interface for accessing - * DTD-specific things like entity expansions and notation properties. - *
- * API is separated from its implementation to reduce coupling; for example,
- * it is possible to have DTD subset implementations that do not implement
- * validation logics, just entity expansion.
- */
-public abstract class DTDSubset
- implements DTDValidationSchema
-{
- /*
- //////////////////////////////////////////////////////
- // Life-cycle
- //////////////////////////////////////////////////////
- */
-
- protected DTDSubset() { }
-
- /**
- * Method that will combine definitions from this internal subset with
- * definitions from passed-in external subset, producing a new combined
- * DTDSubset instance.
- */
- public abstract DTDSubset combineWithExternalSubset(InputProblemReporter rep,
- DTDSubset extSubset)
- throws XMLStreamException;
-
- /*
- //////////////////////////////////////////////////////
- // XMLValidationSchema implementation
- //////////////////////////////////////////////////////
- */
-
- public abstract XMLValidator createValidator(ValidationContext ctxt)
- throws XMLStreamException;
-
- public String getSchemaType() {
- return XMLValidationSchema.SCHEMA_ID_DTD;
- }
-
- /*
- //////////////////////////////////////////////////////
- // And extended DTDValidationSchema
- //////////////////////////////////////////////////////
- */
-
- public abstract int getEntityCount();
-
- public abstract int getNotationCount();
-
- /*
- //////////////////////////////////////////////////////
- // Woodstox-specific API, caching support
- //////////////////////////////////////////////////////
- */
-
- public abstract boolean isCachable();
-
- /**
- * Method used in determining whether cached external subset instance
- * can be used with specified internal subset. If ext. subset references
- * any parameter entities int subset (re-)defines, it can not; otherwise
- * it can be used.
- *
- * @return True if this (external) subset refers to a parameter entity
- * defined in passed-in internal subset.
- */
- public abstract boolean isReusableWith(DTDSubset intSubset);
-
- /*
- //////////////////////////////////////////////////////
- // Woodstox-specific API, entity/notation handling
- //////////////////////////////////////////////////////
- */
-
- public abstract HashMap getGeneralEntityMap();
-
- public abstract List getGeneralEntityList();
-
- public abstract HashMap getParameterEntityMap();
-
- public abstract HashMap getNotationMap();
-
- public abstract List getNotationList();
-
- public abstract HashMap getElementMap();
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDTypingNonValidator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDTypingNonValidator.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDTypingNonValidator.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDTypingNonValidator.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,315 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.util.DataUtil;
-import com.ctc.wstx.util.ElementIdMap;
-import com.ctc.wstx.util.ExceptionUtil;
-
-/**
- * This class is a "non-validating validator"; a validator-like object
- * that handles DTD-based non-validation functionality: determining type
- * information and default values. This instance does NOT implement any
- * actual DTD-validation, and is to be used in DTD-aware non-validating
- * mode.
- */
-public class DTDTypingNonValidator
- extends DTDValidatorBase
-{
- /*
- ///////////////////////////////////////////
- // Element def/spec/validator stack, state
- ///////////////////////////////////////////
- */
-
- /**
- * Flag that indicates if current element has any attributes that
- * have default values.
- */
- protected boolean mHasAttrDefaults = false;
-
- /**
- * Bitset used for keeping track of defaulted attributes for which values
- * have been found. Only non-null when current element does have such
- * attributes
- */
- protected BitSet mCurrDefaultAttrs = null;
-
- /**
- * Flag that indicates whether any of the attributes is potentially
- * normalizable, and we are in attribute-normalizing mode.
- */
- protected boolean mHasNormalizableAttrs = false;
-
- /*
- ///////////////////////////////////////
- // Temporary helper objects
- ///////////////////////////////////////
- */
-
- /**
- * Reusable lazily instantiated BitSet; needed to keep track of
- * 'missing' attributes with default values (normal default, #FIXED).
- */
- BitSet mTmpDefaultAttrs;
-
- /*
- ///////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////
- */
-
- public DTDTypingNonValidator(DTDSubset schema, ValidationContext ctxt, boolean hasNsDefaults,
- Map elemSpecs, Map genEntities)
- {
- super(schema, ctxt, hasNsDefaults, elemSpecs, genEntities);
- }
-
- /**
- * @return False, since this is not a real validator
- */
- public final boolean reallyValidating() { return false; }
-
- /*
- ///////////////////////////////////////
- // Configuration
- ///////////////////////////////////////
- */
-
- /**
- * This 'validator' will not normalize any attributes,
- * so let's implement this as no-op.
- */
- public void setAttrValueNormalization(boolean state) {
- // nop
- }
-
- /*
- ///////////////////////////////////////
- // XMLValidator implementation
- ///////////////////////////////////////
- */
-
- //public XMLValidationSchema getSchema()
-
- public void validateElementStart(String localName, String uri, String prefix)
- throws XMLStreamException
- {
- // Ok, can we find the element definition?
- mTmpKey.reset(prefix, localName);
- DTDElement elem = (DTDElement) mElemSpecs.get(mTmpKey);
- // whether it's found or not, let's add a stack frame:
- int elemCount = mElemCount++;
- if (elemCount >= mElems.length) {
- mElems = (DTDElement[]) DataUtil.growArrayBy50Pct(mElems);
- }
-
- mElems[elemCount] = mCurrElem = elem;
- mAttrCount = 0;
- mIdAttrIndex = -2; // -2 as a "don't know yet" marker
-
- /* but if not found, can not obtain any type information. Not
- * a validation problem though, since we are doing none...
- * Oh, also, unlike with real validation, not having actual element
- * information is ok; can still have attributes!
- */
- if (elem == null) { // || !elem.isDefined())
- mCurrAttrDefs = EMPTY_MAP;
- mHasAttrDefaults = false;
- mCurrDefaultAttrs = null;
- mHasNormalizableAttrs = false;
- return;
- }
-
- // If element found, does it have any attributes?
- mCurrAttrDefs = elem.getAttributes();
- if (mCurrAttrDefs == null) {
- mCurrAttrDefs = EMPTY_MAP;
- mHasAttrDefaults = false;
- mCurrDefaultAttrs = null;
- mHasNormalizableAttrs = false;
- return;
- }
-
- // Any normalization needed?
- mHasNormalizableAttrs = mNormAttrs || elem.attrsNeedValidation();
-
- // Any default values?
- mHasAttrDefaults = elem.hasAttrDefaultValues();
- if (mHasAttrDefaults) {
- /* Special count also contains ones with #REQUIRED value, but
- * that's a minor sub-optimality...
- */
- int specCount = elem.getSpecialCount();
- BitSet bs = mTmpDefaultAttrs;
- if (bs == null) {
- mTmpDefaultAttrs = bs = new BitSet(specCount);
- } else {
- bs.clear();
- }
- mCurrDefaultAttrs = bs;
- } else {
- mCurrDefaultAttrs = null;
- }
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix, String value)
- throws XMLStreamException
- {
- /* no need to do any validation; however, need to do following:
- *
- * (a) Figure out type info, if any (to get data type, id index etc);
- * if yes, do:
- * (1) If attribute has default value, note down it's not needed due
- * to explicit definition
- * (2) If attribute is normalizable, normalize it without validation
- */
- DTDAttribute attr = (DTDAttribute) mCurrAttrDefs.get(mTmpKey.reset(prefix, localName));
- int index = mAttrCount++;
- if (index >= mAttrSpecs.length) {
- mAttrSpecs = (DTDAttribute[]) DataUtil.growArrayBy50Pct(mAttrSpecs);
- }
- mAttrSpecs[index] = attr;
-
- /* Although undeclared attribute would be a validation error,
- * we don't care here... just need to skip it
- */
- if (attr != null) {
- if (mHasAttrDefaults) {
- /* Once again, let's use more generic 'special' index,
- * even though it also includes #REQUIRED values
- */
- int specIndex = attr.getSpecialIndex();
- if (specIndex >= 0) {
- mCurrDefaultAttrs.set(specIndex);
- }
- }
- if (mHasNormalizableAttrs) {
- // !!! TBI
- }
- }
- return null; // fine as is
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix,
- char[] valueChars, int valueStart,
- int valueEnd)
- throws XMLStreamException
- {
- // note: cut'n pasted from above...
- DTDAttribute attr = (DTDAttribute) mCurrAttrDefs.get(mTmpKey.reset(prefix, localName));
- int index = mAttrCount++;
- if (index >= mAttrSpecs.length) {
- mAttrSpecs = (DTDAttribute[]) DataUtil.growArrayBy50Pct(mAttrSpecs);
- }
- mAttrSpecs[index] = attr;
- if (attr != null) {
- if (mHasAttrDefaults) {
- int specIndex = attr.getSpecialIndex();
- if (specIndex >= 0) {
- mCurrDefaultAttrs.set(specIndex);
- }
- }
- if (mHasNormalizableAttrs) { // may get normalized, after all
- return attr.normalize(this, valueChars, valueStart, valueEnd);
- }
- }
- return null; // fine as is
- }
-
- public int validateElementAndAttributes()
- throws XMLStreamException
- {
- /* Ok; since we are not really validating, we just need to add possible
- * attribute default values, and return "anything goes"
- * as the allowable content:
- */
- DTDElement elem = mCurrElem;
- if (mHasAttrDefaults) {
- BitSet specBits = mCurrDefaultAttrs;
- int specCount = elem.getSpecialCount();
- int ix = specBits.nextClearBit(0);
- while (ix < specCount) { // something amiss!
- List specAttrs = elem.getSpecialAttrs();
- DTDAttribute attr = (DTDAttribute) specAttrs.get(ix);
- if (attr.hasDefaultValue()) { // no default for #REQUIRED...
- doAddDefaultValue(attr);
- }
- ix = specBits.nextClearBit(ix+1);
- }
- }
- /* However: we should indicate cases where PCDATA is not supposed
- * to occur -- although it won't be considered an error, when not
- * validating, info is needed to determine type of SPACE instead
- * of CHARACTERS. Other validation types are not to be returned,
- * however, since caller doesn't know how to deal with such
- * cases.
- */
- return (elem == null) ? XMLValidator.CONTENT_ALLOW_ANY_TEXT :
- elem.getAllowedContentIfSpace();
- }
-
- public int validateElementEnd(String localName, String uri, String prefix)
- throws XMLStreamException
- {
- /* Since we are not really validating, only need to maintain
- * the element stack, and return "anything goes" as allowable content:
- */
- int ix = --mElemCount;
- mElems[ix] = null;
- if (ix < 1) {
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- }
- DTDElement elem = mElems[ix-1];
- return (elem == null) ? XMLValidator.CONTENT_ALLOW_ANY_TEXT :
- mElems[ix-1].getAllowedContentIfSpace();
- }
-
- // base class implements these ok:
- //public void validateText(String text, boolean lastTextSegment)
- //public void validateText(char[] cbuf, int textStart, int textEnd, boolean lastTextSegment)
-
- public void validationCompleted(boolean eod)
- //throws XMLStreamException
- {
- // fine, great, nothing to do...
- }
-
- /*
- ///////////////////////////////////////
- // Package methods, accessors
- ///////////////////////////////////////
- */
-
- protected ElementIdMap getIdMap()
- {
- /* should never be called; for now let's throw an exception, if it
- * turns out it does get called can/should return an empty immutable
- * map or something
- */
- ExceptionUtil.throwGenericInternal();
- return null; // never gets here
- }
-
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDValidatorBase.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDValidatorBase.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDValidatorBase.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDValidatorBase.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,531 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.text.MessageFormat;
-import java.util.*;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.api.WstxInputProperties;
-import com.ctc.wstx.sr.NsDefaultProvider;
-import com.ctc.wstx.sr.InputElementStack;
-import com.ctc.wstx.util.DataUtil;
-import com.ctc.wstx.util.ElementIdMap;
-import com.ctc.wstx.util.ExceptionUtil;
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Shared abstract base class for Woodstox implementations
- * of {@link XMLValidator} for DTD validation.
- * Since there are 2 sub-types -- full actual DTD validator, and a dummy
- * one that only adds type information and default values, with no actual
- * validation -- common functionality was refactored into this base
- * class.
- */
-public abstract class DTDValidatorBase
- extends XMLValidator
- implements NsDefaultProvider // for namespace attr defaults
-{
- /*
- /////////////////////////////////////////////////////
- // Constants
- /////////////////////////////////////////////////////
- */
-
- /**
- * Estimated maximum depth of typical documents; used to allocate
- * the array for element stack
- */
- final static int DEFAULT_STACK_SIZE = 16;
-
- /**
- * Estimated maximum number of attributes for a single element
- */
- final static int EXP_MAX_ATTRS = 16;
-
- /**
- * Let's actually just reuse a local Map...
- */
- protected final static HashMap EMPTY_MAP = new HashMap();
-
- /*
- ///////////////////////////////////////
- // Configuration
- ///////////////////////////////////////
- */
-
- /**
- * Flag that indicates whether any of the elements declared has default
- * attribute values for namespace declaration pseudo-attributes.
- */
- final boolean mHasNsDefaults;
-
- /**
- * DTD schema ({@link DTDSubsetImpl}) object that created this validator
- * instance.
- */
- final DTDSubset mSchema;
-
- /**
- * Validation context (owner) for this validator. Needed for adding
- * default attribute values, for example.
- */
- final ValidationContext mContext;
-
- /**
- * Map that contains element specifications from DTD; null if no
- * DOCTYPE declaration found.
- */
- final Map mElemSpecs;
-
- /**
- * General entities defined in DTD subsets; needed for validating
- * ENTITY/ENTITIES attributes.
- */
- final Map mGeneralEntities;
-
- /**
- * Flag that indicates whether parser wants the attribute values
- * to be normalized (according to XML specs) or not (which may be
- * more efficient, although not compliant with the specs)
- */
- protected boolean mNormAttrs;
-
- /*
- ///////////////////////////////////////////
- // Element def/spec/validator stack, state
- ///////////////////////////////////////////
- */
-
- /**
- * This is the element that is currently being validated; valid
- * during
- * validateElementStart,
- * validateAttribute,
- * validateElementAndAttributes calls.
- */
- protected DTDElement mCurrElem = null;
-
- /**
- * Stack of element definitions matching the current active element stack.
- * Instances are elements definitions read from DTD.
- */
- protected DTDElement[] mElems = null;
-
- /**
- * Number of elements in {@link #mElems}.
- */
- protected int mElemCount = 0;
-
- /**
- * Attribute definitions for attributes the current element may have
- */
- protected HashMap mCurrAttrDefs = null;
-
- /**
- * List of attribute declarations/specifications, one for each
- * attribute of the current element, for which there is a matching
- * value (either explicitly defined, or assigned via defaulting).
- */
- protected DTDAttribute[] mAttrSpecs = new DTDAttribute[EXP_MAX_ATTRS];
-
- /**
- * Number of attribute specification Objects in
- * {@link #mAttrSpecs}; needed to store in case type information
- * is requested later on.
- */
- protected int mAttrCount = 0;
-
- /**
- * Index of the attribute of type ID, within current element's
- * attribute list. Track of this is kept separate from other
- * attribute since id attributes often need to be used for resolving
- * cross-references.
- */
- protected int mIdAttrIndex = -1;
-
- /*
- ///////////////////////////////////////
- // Temporary helper objects
- ///////////////////////////////////////
- */
-
- protected final transient PrefixedName mTmpKey = new PrefixedName(null, null);
-
- /**
- * Temporary buffer attribute instances can share for validation
- * purposes
- */
- char[] mTmpAttrValueBuffer = null;
-
- /*
- ///////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////
- */
-
- public DTDValidatorBase(DTDSubset schema, ValidationContext ctxt, boolean hasNsDefaults,
- Map elemSpecs, Map genEntities)
- {
- mSchema = schema;
- mContext = ctxt;
- mHasNsDefaults = hasNsDefaults;
- mElemSpecs = (elemSpecs == null || elemSpecs.size() == 0) ?
- Collections.EMPTY_MAP : elemSpecs;
- mGeneralEntities = genEntities;
- // By default, let's assume attrs are to be normalized (fully xml compliant)
- mNormAttrs = true;
- mElems = new DTDElement[DEFAULT_STACK_SIZE];
- }
-
- /*
- ///////////////////////////////////////
- // Configuration
- ///////////////////////////////////////
- */
-
- /**
- * Method that allows enabling/disabling attribute value normalization.
- * In general, readers by default enable normalization (to be fully xml
- * compliant),
- * whereas writers do not (since there is usually little to gain, if
- * anything -- it is even possible value may be written before validation
- * is called in some cases)
- */
- public void setAttrValueNormalization(boolean state) {
- mNormAttrs = state;
- }
-
- /**
- * @return True for validator object that actually do validate
- * content; false for objects that only use DTD type information.
- */
- public abstract boolean reallyValidating();
-
- /*
- ///////////////////////////////////////
- // XMLValidator implementation
- ///////////////////////////////////////
- */
-
- public final XMLValidationSchema getSchema() {
- return mSchema;
- }
-
- /**
- * Method called to update information about the newly encountered (start)
- * element. At this point namespace information has been resolved, but
- * no DTD validation has been done. Validator is to do these validations,
- * including checking for attribute value (and existence) compatibility.
- */
- public abstract void validateElementStart(String localName, String uri, String prefix)
- throws XMLStreamException;
-
- public abstract String validateAttribute(String localName, String uri,
- String prefix, String value)
- throws XMLStreamException;
-
- public abstract String validateAttribute(String localName, String uri,
- String prefix,
- char[] valueChars, int valueStart,
- int valueEnd)
- throws XMLStreamException;
-
- public abstract int validateElementAndAttributes()
- throws XMLStreamException;
-
- /**
- * @return Validation state that should be effective for the parent
- * element state
- */
- public abstract int validateElementEnd(String localName, String uri, String prefix)
- throws XMLStreamException;
-
- public void validateText(String text, boolean lastTextSegment)
- throws XMLStreamException
- {
- /* This method is a NOP, since basic DTD has no mechanism for
- * validating textual content.
- */
- }
-
- public void validateText(char[] cbuf, int textStart, int textEnd,
- boolean lastTextSegment)
- throws XMLStreamException
- {
- /* This method is a NOP, since basic DTD has no mechanism for
- * validating textual content.
- */
- }
-
- public abstract void validationCompleted(boolean eod)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////
- // Attribute info access
- ///////////////////////////////////////
- */
-
- // // // Access to type info
-
- public String getAttributeType(int index)
- {
- DTDAttribute attr = mAttrSpecs[index];
- return (attr == null) ? WstxInputProperties.UNKNOWN_ATTR_TYPE :
- attr.getValueTypeString();
- }
-
- /**
- * Method for finding out the index of the attribute (collected using
- * the attribute collector; having DTD-derived info in same order)
- * that is of type ID. DTD explicitly specifies that at most one
- * attribute can have this type for any element.
- *
- * @return Index of the attribute with type ID, in the current
- * element, if one exists: -1 otherwise
- */
- public int getIdAttrIndex()
- {
- // Let's figure out the index only when needed
- int ix = mIdAttrIndex;
- if (ix == -2) {
- ix = -1;
- if (mCurrElem != null) {
- DTDAttribute idAttr = mCurrElem.getIdAttribute();
- if (idAttr != null) {
- DTDAttribute[] attrs = mAttrSpecs;
- for (int i = 0, len = attrs.length; i < len; ++i) {
- if (attrs[i] == idAttr) {
- ix = i;
- break;
- }
- }
- }
- }
- mIdAttrIndex = ix;
- }
- return ix;
- }
-
- /**
- * Method for finding out the index of the attribute (collected using
- * the attribute collector; having DTD-derived info in same order)
- * that is of type NOTATION. DTD explicitly specifies that at most one
- * attribute can have this type for any element.
- *
- * @return Index of the attribute with type NOTATION, in the current
- * element, if one exists: -1 otherwise
- */
- public int getNotationAttrIndex()
- {
- /* If necessary, we could find this index when resolving the
- * element, could avoid linear search. But who knows how often
- * it's really needed...
- */
- for (int i = 0, len = mAttrCount; i < len; ++i) {
- if (mAttrSpecs[i].typeIsNotation()) {
- return i;
- }
- }
- return -1;
- }
-
- /*
- /////////////////////////////////////////////////////
- // NsDefaultProvider interface
- /////////////////////////////////////////////////////
- */
-
- /**
- * Calling this method before {@link #checkNsDefaults} is necessary
- * to pass information regarding the current element; although
- * it will become available later on (via normal XMLValidator interface),
- * that's too late (after namespace binding and resolving).
- */
- public boolean mayHaveNsDefaults(String elemPrefix, String elemLN)
- {
- mTmpKey.reset(elemPrefix, elemLN);
- DTDElement elem = (DTDElement) mElemSpecs.get(mTmpKey);
- mCurrElem = elem;
- return (elem != null) && elem.hasNsDefaults();
- }
-
- public void checkNsDefaults(InputElementStack nsStack)
- throws XMLStreamException
- {
- // We only get called if mCurrElem != null, and has defaults
- HashMap m = mCurrElem.getNsDefaults();
- if (m != null) {
- Iterator it = m.entrySet().iterator();
- while (it.hasNext()) {
- Map.Entry me = (Map.Entry) it.next();
- String prefix = (String) me.getKey();
- if (!nsStack.isPrefixLocallyDeclared(prefix)) {
- DTDAttribute attr = (DTDAttribute) me.getValue();
- String uri = attr.getDefaultValue(mContext, this);
- nsStack.addNsBinding(prefix, uri);
- }
- }
- }
- }
-
- /*
- ///////////////////////////////////////
- // Package methods, accessors
- ///////////////////////////////////////
- */
-
- /**
- * Name of current element on the top of the element stack.
- */
- PrefixedName getElemName() {
- DTDElement elem = mElems[mElemCount-1];
- return elem.getName();
- }
-
- Location getLocation() {
- return mContext.getValidationLocation();
- }
-
- protected abstract ElementIdMap getIdMap();
-
- Map getEntityMap() {
- return mGeneralEntities;
- }
-
- char[] getTempAttrValueBuffer(int neededLength)
- {
- if (mTmpAttrValueBuffer == null
- || mTmpAttrValueBuffer.length < neededLength) {
- int size = (neededLength < 100) ? 100 : neededLength;
- mTmpAttrValueBuffer = new char[size];
- }
- return mTmpAttrValueBuffer;
- }
-
- public boolean hasNsDefaults() {
- return mHasNsDefaults;
- }
-
- /*
- ///////////////////////////////////////
- // Package methods, error handling
- ///////////////////////////////////////
- */
-
- /**
- * Method called to report validity problems; depending on mode, will
- * either throw an exception, or add a problem notification to the
- * list of problems.
- */
- void reportValidationProblem(String msg)
- throws XMLStreamException
- {
- doReportValidationProblem(msg, null);
- }
-
- void reportValidationProblem(String msg, Location loc)
- throws XMLStreamException
- {
- doReportValidationProblem(msg, loc);
- }
-
- void reportValidationProblem(String format, Object arg)
- throws XMLStreamException
- {
- doReportValidationProblem(MessageFormat.format(format, new Object[] { arg }),
- null);
- }
-
- void reportValidationProblem(String format, Object arg1, Object arg2)
- throws XMLStreamException
- {
- doReportValidationProblem(MessageFormat.format(format, new Object[] { arg1, arg2 }),
- null);
- }
-
- /*
- ///////////////////////////////////////
- // Private/sub-class methods
- ///////////////////////////////////////
- */
-
- protected void doReportValidationProblem(String msg, Location loc)
- throws XMLStreamException
- {
- if (loc == null) {
- loc = getLocation();
- }
- XMLValidationProblem prob = new XMLValidationProblem(loc, msg, XMLValidationProblem.SEVERITY_ERROR);
- prob.setReporter(this);
- mContext.reportProblem(prob);
- }
-
- protected void doAddDefaultValue(DTDAttribute attr)
- throws XMLStreamException
- {
- /* If we get here, we should have a non-null (possibly empty) default
- * value:
- */
- String def = attr.getDefaultValue(mContext, this);
- if (def == null) {
- ExceptionUtil.throwInternal("null default attribute value");
- }
- PrefixedName an = attr.getName();
- // Ok, do we need to find the URI?
- String prefix = an.getPrefix();
- String uri = "";
- if (prefix != null && prefix.length() > 0) {
- uri = mContext.getNamespaceURI(prefix);
- // Can not map to empty NS!
- if (uri == null || uri.length() == 0) {
- /* Hmmh. This is a weird case where we do have to
- * throw a validity exception; even though it really
- * is more a ns-well-formedness error...
- */
- reportValidationProblem("Unbound namespace prefix \"{0}\" for default attribute \"{1}\"", prefix, attr);
- // May continue if we don't throw errors, just collect them to a list
- uri = "";
- }
- }
- int defIx = mContext.addDefaultAttribute(an.getLocalName(), uri, prefix, def);
- if (defIx < 0) {
- /* 13-Dec-2005, Tatus: Hmmh. For readers this is an error
- * condition, but writers may just indicate they are not
- * interested in defaults. So let's let context report
- * problem(s) if it has any regarding the request.
- */
- // nop
- } else {
- while (defIx >= mAttrSpecs.length) {
- mAttrSpecs = (DTDAttribute[]) DataUtil.growArrayBy50Pct(mAttrSpecs);
- }
- /* Any intervening empty slots? (can happen if other
- * validators add default attributes...)
- */
- while (mAttrCount < defIx) {
- mAttrSpecs[mAttrCount++] = null;
- }
- mAttrSpecs[defIx] = attr;
- mAttrCount = defIx+1;
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDValidator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDValidator.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDValidator.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDValidator.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,406 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.util.DataUtil;
-import com.ctc.wstx.util.ElementId;
-import com.ctc.wstx.util.ElementIdMap;
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Woodstox implementation of {@link XMLValidator}; the class that
- * handles DTD-based validation.
- */
-public class DTDValidator
- extends DTDValidatorBase
-{
- /*
- ///////////////////////////////////////
- // Configuration
- ///////////////////////////////////////
- */
-
- /**
- * Determines if identical problems (definition of the same element,
- * for example) should cause multiple error notifications or not:
- * if true, will get one error per instance, if false, only the first
- * one will get reported.
- */
- protected boolean mReportDuplicateErrors = false;
-
- /*
- ///////////////////////////////////////
- // Id/idref state
- ///////////////////////////////////////
- */
-
- /**
- * Information about declared and referenced element ids (unique
- * ids that attributes may defined, as defined by DTD)
- */
- protected ElementIdMap mIdMap = null;
-
- /*
- ///////////////////////////////////////////
- // Element def/spec/validator stack, state
- ///////////////////////////////////////////
- */
-
- /**
- * Stack of validators for open elements
- */
- protected StructValidator[] mValidators = null;
-
- /**
- * Bitset used for keeping track of required and defaulted attributes
- * for which values have been found.
- */
- protected BitSet mCurrSpecialAttrs = null;
-
- boolean mCurrHasAnyFixed = false;
-
- /*
- ///////////////////////////////////////
- // Temporary helper objects
- ///////////////////////////////////////
- */
-
- /**
- * Reusable lazily instantiated BitSet; needed to keep track of
- * missing 'special' attributes (required ones, ones with default
- * values)
- */
- BitSet mTmpSpecialAttrs;
-
- /*
- ///////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////
- */
-
- public DTDValidator(DTDSubset schema, ValidationContext ctxt, boolean hasNsDefaults,
- Map elemSpecs, Map genEntities)
- {
- super(schema, ctxt, hasNsDefaults, elemSpecs, genEntities);
- mValidators = new StructValidator[DEFAULT_STACK_SIZE];
- }
-
- public final boolean reallyValidating() { return true; }
-
- /*
- ///////////////////////////////////////
- // XMLValidator implementation
- ///////////////////////////////////////
- */
-
- //public XMLValidationSchema getSchema();
-
- /**
- * Method called to update information about the newly encountered (start)
- * element. At this point namespace information has been resolved, but
- * no DTD validation has been done. Validator is to do these validations,
- * including checking for attribute value (and existence) compatibility.
- */
- public void validateElementStart(String localName, String uri, String prefix)
- throws XMLStreamException
- {
- /* Ok, need to find the element definition; if not found (or
- * only implicitly defined), need to throw the exception.
- */
- mTmpKey.reset(prefix, localName);
-
- DTDElement elem = (DTDElement) mElemSpecs.get(mTmpKey);
-
- /* Let's add the entry in (even if it's a null); this is necessary
- * to keep things in-sync if allowing graceful handling of validity
- * errors
- */
- int elemCount = mElemCount++;
- if (elemCount >= mElems.length) {
- mElems = (DTDElement[]) DataUtil.growArrayBy50Pct(mElems);
- mValidators = (StructValidator[]) DataUtil.growArrayBy50Pct(mValidators);
- }
- mElems[elemCount] = mCurrElem = elem;
- if (elem == null || !elem.isDefined()) {
- reportValidationProblem(ErrorConsts.ERR_VLD_UNKNOWN_ELEM, mTmpKey.toString());
- }
-
- // Is this element legal under the parent element?
- StructValidator pv = (elemCount > 0) ? mValidators[elemCount-1] : null;
-
- if (pv != null && elem != null) {
- String msg = pv.tryToValidate(elem.getName());
- if (msg != null) {
- int ix = msg.indexOf("$END");
- String pname = mElems[elemCount-1].toString();
- if (ix >= 0) {
- msg = msg.substring(0, ix) + ""
- +msg.substring(ix+4);
- }
- reportValidationProblem("Validation error, encountered element <"
- +elem.getName()+"> as a child of <"
- +pname+">: "+msg);
- }
- }
-
- mAttrCount = 0;
- mIdAttrIndex = -2; // -2 as a "don't know yet" marker
-
- // Ok, need to get the child validator, then:
- if (elem == null) {
- mValidators[elemCount] = null;
- mCurrAttrDefs = EMPTY_MAP;
- mCurrHasAnyFixed = false;
- mCurrSpecialAttrs = null;
- } else {
- mValidators[elemCount] = elem.getValidator();
- mCurrAttrDefs = elem.getAttributes();
- if (mCurrAttrDefs == null) {
- mCurrAttrDefs = EMPTY_MAP;
- }
- mCurrHasAnyFixed = elem.hasFixedAttrs();
- int specCount = elem.getSpecialCount();
- if (specCount == 0) {
- mCurrSpecialAttrs = null;
- } else {
- BitSet bs = mTmpSpecialAttrs;
- if (bs == null) {
- mTmpSpecialAttrs = bs = new BitSet(specCount);
- } else {
- bs.clear();
- }
- mCurrSpecialAttrs = bs;
- }
- }
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix, String value)
- throws XMLStreamException
- {
- DTDAttribute attr = (DTDAttribute) mCurrAttrDefs.get(mTmpKey.reset(prefix, localName));
- if (attr == null) {
- // Only report error if not already covering from an error:
- if (mCurrElem != null) {
- reportValidationProblem(ErrorConsts.ERR_VLD_UNKNOWN_ATTR,
- mCurrElem.toString(), mTmpKey.toString());
- }
- /* [WSTX-190] NPE if we continued (after reported didn't
- * throw an exception); nothing more to do, let's leave
- */
- return value;
- }
- int index = mAttrCount++;
- if (index >= mAttrSpecs.length) {
- mAttrSpecs = (DTDAttribute[]) DataUtil.growArrayBy50Pct(mAttrSpecs);
- }
- mAttrSpecs[index] = attr;
- if (mCurrSpecialAttrs != null) { // Need to mark that we got it
- int specIndex = attr.getSpecialIndex();
- if (specIndex >= 0) {
- mCurrSpecialAttrs.set(specIndex);
- }
- }
- String result = attr.validate(this, value, mNormAttrs);
- if (mCurrHasAnyFixed && attr.isFixed()) {
- String act = (result == null) ? value : result;
- String exp = attr.getDefaultValue(mContext, this);
- if (!act.equals(exp)) {
- reportValidationProblem("Value of attribute \""+attr+"\" (element <"+mCurrElem+">) not \""+exp+"\" as expected, but \""+act+"\"");
- }
- }
- return result;
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix,
- char[] valueChars, int valueStart,
- int valueEnd)
- throws XMLStreamException
- {
- DTDAttribute attr = (DTDAttribute) mCurrAttrDefs.get(mTmpKey.reset(prefix, localName));
- if (attr == null) {
- // Only report error if not already covering from an error:
- if (mCurrElem != null) {
- reportValidationProblem(ErrorConsts.ERR_VLD_UNKNOWN_ATTR,
- mCurrElem.toString(), mTmpKey.toString());
- }
- /* [WSTX-190] NPE if we continued (after reported didn't
- * throw an exception); nothing more to do, let's leave
- */
- return new String(valueChars, valueStart, valueEnd);
- }
- int index = mAttrCount++;
- if (index >= mAttrSpecs.length) {
- mAttrSpecs = (DTDAttribute[]) DataUtil.growArrayBy50Pct(mAttrSpecs);
- }
- mAttrSpecs[index] = attr;
- if (mCurrSpecialAttrs != null) { // Need to mark that we got it
- int specIndex = attr.getSpecialIndex();
- if (specIndex >= 0) {
- mCurrSpecialAttrs.set(specIndex);
- }
- }
- String result = attr.validate(this, valueChars, valueStart, valueEnd, mNormAttrs);
- if (mCurrHasAnyFixed && attr.isFixed()) {
- String exp = attr.getDefaultValue(mContext, this);
- boolean match;
- if (result == null) {
- match = StringUtil.matches(exp, valueChars, valueStart, valueEnd - valueStart);
- } else {
- match = exp.equals(result);
- }
- if (!match) {
- String act = (result == null) ?
- new String(valueChars, valueStart, valueEnd) : result;
- reportValidationProblem("Value of #FIXED attribute \""+attr+"\" (element <"+mCurrElem+">) not \""+exp+"\" as expected, but \""+act+"\"");
- }
- }
- return result;
- }
-
- public int validateElementAndAttributes()
- throws XMLStreamException
- {
- // Ok: are we fine with the attributes?
- DTDElement elem = mCurrElem;
- if (elem == null) { // had an error, most likely no such element defined...
- // need to just return, nothing to do here
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- }
-
- // Any special attributes missing?
- if (mCurrSpecialAttrs != null) {
- BitSet specBits = mCurrSpecialAttrs;
- int specCount = elem.getSpecialCount();
- int ix = specBits.nextClearBit(0);
- while (ix < specCount) { // something amiss!
- List specAttrs = elem.getSpecialAttrs();
- DTDAttribute attr = (DTDAttribute) specAttrs.get(ix);
-
- /* [WSTX-155]: Problems if reportValidationProblem returns
- * ok (which happens if a reporter handles it). So what
- * to do with missing required value? First thought is
- * to just leave it as is.
- */
- if (attr.isRequired()) {
- reportValidationProblem("Required attribute \"{0}\" missing from element <{1}>", attr, elem);
- } else {
- doAddDefaultValue(attr);
- }
- ix = specBits.nextClearBit(ix+1);
- }
- }
-
- return elem.getAllowedContent();
- }
-
- /**
- * @return Validation state that should be effective for the parent
- * element state
- */
- public int validateElementEnd(String localName, String uri, String prefix)
- throws XMLStreamException
- {
- // First, let's remove the top:
- int ix = mElemCount-1;
- /* [WSTX-200]: need to avoid problems when doing sub-tree
- * validation...
- */
- if (ix < 0) {
- return XMLValidator.CONTENT_ALLOW_WS;
- }
- mElemCount = ix;
-
- DTDElement closingElem = mElems[ix];
- mElems[ix] = null;
- StructValidator v = mValidators[ix];
- mValidators[ix] = null;
-
- // Validation?
- if (v != null) {
- String msg = v.fullyValid();
- if (msg != null) {
- reportValidationProblem("Validation error, element : "+msg);
- }
- }
-
- // Then let's get info from parent, if any
- if (ix < 1) { // root element closing..
- // doesn't really matter; epilog/prolog differently handled:
- return XMLValidator.CONTENT_ALLOW_WS;
- }
- return mElems[ix-1].getAllowedContent();
- }
-
- //public void validateText(String text, boolean lastTextSegment) ;
- //public void validateText(char[] cbuf, int textStart, int textEnd, boolean lastTextSegment) ;
-
- public void validationCompleted(boolean eod)
- throws XMLStreamException
- {
- /* Need to now ensure that all IDREF/IDREFS references
- * point to defined ID attributes
- */
- checkIdRefs();
- }
-
- /*
- ///////////////////////////////////////
- // Package methods, accessors
- ///////////////////////////////////////
- */
-
- protected ElementIdMap getIdMap() {
- if (mIdMap == null) {
- mIdMap = new ElementIdMap();
- }
- return mIdMap;
- }
-
- /*
- ///////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////
- */
-
- protected void checkIdRefs()
- throws XMLStreamException
- {
- /* 02-Oct-2004, TSa: Now we can also check that all id references
- * pointed to ids that actually are defined
- */
- if (mIdMap != null) {
- ElementId ref = mIdMap.getFirstUndefined();
- if (ref != null) { // problem!
- reportValidationProblem("Undefined id '"+ref.getId()
- +"': referenced from element <"
- +ref.getElemName()+">, attribute '"
- +ref.getAttrName()+"'",
- ref.getLocation());
- }
- }
- }
-
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/DTDWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/DTDWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,164 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.io.IOException;
-import java.io.Writer;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.exc.WstxIOException;
-
-/**
- * Simple utility class used by {@link DTDReader} when writing out
- * flattened external DTD subset file. Writing functionality encapsulated
- * here since it's specific to one mode of operation (flattening).
- *
- * Note, too, that underlying {@link IOException}s are generally wrapped - * as {@link XMLStreamException}s. This is needed to reduce amount of - * work caller has to do for wrapping. It will still be possible to - * unwrap these exceptions further up the call stack if need be. - */ -final class DTDWriter -{ - /* - ////////////////////////////////////////////////// - // Configuration - ////////////////////////////////////////////////// - */ - - final Writer mWriter; - - final boolean mIncludeComments; - - final boolean mIncludeConditionals; - - final boolean mIncludePEs; - - /* - ////////////////////////////////////////////////// - // Output status - ////////////////////////////////////////////////// - */ - - /** - * Counter that indicates whether flattened output should be written to - * (non-null) mWriter; values above zero indicate output is enabled, - * zero and below that output is disabled. - * Only enabled if mWriter is not - * null; will be temporarily disabled during processing of content - * that is not to be included (PE reference; or comments / conditional - * sections if comment/cs output is suppressed) - */ - int mIsFlattening = 0; - - /** - * Pointer to first character in the current input buffer that - * has not yet been written to flatten writer. - */ - int mFlattenStart = 0; - - /* - ////////////////////////////////////////////////// - // Life-cycle - ////////////////////////////////////////////////// - */ - - public DTDWriter(Writer out, boolean inclComments, boolean inclCond, - boolean inclPEs) - { - mWriter = out; - mIncludeComments = inclComments; - mIncludeConditionals = inclCond; - mIncludePEs = inclPEs; - - mIsFlattening = 1; // starts enabled - } - - /* - ////////////////////////////////////////////////// - // Public API, accessors, state change - ////////////////////////////////////////////////// - */ - - public boolean includeComments() { - return mIncludeComments; - } - - public boolean includeConditionals() { - return mIncludeConditionals; - } - - public boolean includeParamEntities() { - return mIncludePEs; - } - - public void disableOutput() - { - --mIsFlattening; - } - - public void enableOutput(int newStart) - { - ++mIsFlattening; - mFlattenStart = newStart; - } - - public void setFlattenStart(int ptr) { - mFlattenStart = ptr; - } - - public int getFlattenStart() { - return mFlattenStart; - } - - /* - ////////////////////////////////////////////////// - // Public API, output methods: - ////////////////////////////////////////////////// - */ - - public void flush(char[] buf, int upUntil) - throws XMLStreamException - { - if (mFlattenStart < upUntil) { - if (mIsFlattening > 0) { - try { - mWriter.write(buf, mFlattenStart, upUntil - mFlattenStart); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - mFlattenStart = upUntil; - } - } - - /** - * Method called when explicit output has to be done for flatten output: - * this is usually done when there's need to do speculative checks - * before it's known if some chars are output (when suppressing comments - * or conditional sections) - */ - public void output(String output) - throws XMLStreamException - { - if (mIsFlattening > 0) { - try { - mWriter.write(output); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - } - - public void output(char c) - throws XMLStreamException - { - if (mIsFlattening > 0) { - try { - mWriter.write(c); - } catch (IOException ioe) { - throw new WstxIOException(ioe); - } - } - } -} - diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/EmptyValidator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/EmptyValidator.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/EmptyValidator.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/EmptyValidator.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,48 +0,0 @@ -package com.ctc.wstx.dtd; - -import com.ctc.wstx.util.PrefixedName; - -/** - * Simple content model validator that accepts no elements, ever; this - * is true for pure #PCDATA content model as well as EMPTY content model. - * Can be used as a singleton, since all info needed for diagnostics - * is passed via methods. - */ -public class EmptyValidator - extends StructValidator -{ - final static EmptyValidator sPcdataInstance = new EmptyValidator("No elements allowed in pure #PCDATA content model"); - - final static EmptyValidator sEmptyInstance = new EmptyValidator("No elements allowed in EMPTY content model"); - - final String mErrorMsg; - - private EmptyValidator(String errorMsg) { - mErrorMsg = errorMsg; - } - - public static EmptyValidator getPcdataInstance() { return sPcdataInstance; } - public static EmptyValidator getEmptyInstance() { return sPcdataInstance; } - - /** - * Simple; can always (re)use instance itself; no state information - * is kept. - */ - public StructValidator newInstance() { - return this; - } - - public String tryToValidate(PrefixedName elemName) - { - return mErrorMsg; - } - - /** - * If we ever get as far as element closing, things are all good; - * can just return null. - */ - public String fullyValid() - { - return null; - } -} diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/FullDTDReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/FullDTDReader.java --- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/FullDTDReader.java 2012-04-24 04:38:20.000000000 +0000 +++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/FullDTDReader.java 1970-01-01 00:00:00.000000000 +0000 @@ -1,3499 +0,0 @@ -/* Woodstox XML processor - * - * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi - * - * Licensed under the License specified in file LICENSE, included with - * the source code. - * You may not use this file except in compliance with the License. - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package com.ctc.wstx.dtd; - -import java.io.IOException; -import java.io.Writer; -import java.net.URL; -import java.text.MessageFormat; -import java.util.*; - -import javax.xml.stream.Location; -import javax.xml.stream.XMLReporter; -import javax.xml.stream.XMLStreamException; -import javax.xml.stream.events.NotationDeclaration; - -import org.codehaus.stax2.validation.XMLValidationProblem; -import org.codehaus.stax2.validation.XMLValidator; - -import com.ctc.wstx.api.ReaderConfig; -import com.ctc.wstx.cfg.ErrorConsts; -import com.ctc.wstx.cfg.XmlConsts; -import com.ctc.wstx.ent.*; -import com.ctc.wstx.evt.WNotationDeclaration; -import com.ctc.wstx.io.WstxInputData; -import com.ctc.wstx.io.WstxInputSource; -import com.ctc.wstx.util.*; - -/** - * Reader that reads in DTD information from internal or external subset. - *
- * There are 2 main modes for DTDReader, depending on whether it is parsing - * internal or external subset. Parsing of internal subset is somewhat - * simpler, since no dependency checking is needed. For external subset, - * handling of parameter entities is bit more complicated, as care has to - * be taken to distinguish between using PEs defined in int. subset, and - * ones defined in ext. subset itself. This determines cachability of - * external subsets. - *
- * Reader also implements simple stand-alone functionality for flattening - * DTD files (expanding all references to their eventual textual form); - * this is sometimes useful when optimizing modularized DTDs - * (which are more maintainable) into single monolithic DTDs (which in - * general can be more performant). - * - * @author Tatu Saloranta - */ -public class FullDTDReader - extends MinimalDTDReader -{ - /** - * Flag that can be changed to enable or disable interning of shared - * names; shared names are used for enumerated values to reduce - * memory usage. - */ - final static boolean INTERN_SHARED_NAMES = false; - - // // // Entity expansion types: - - final static Boolean ENTITY_EXP_GE = Boolean.FALSE; - - final static Boolean ENTITY_EXP_PE = Boolean.TRUE; - - /* - ////////////////////////////////////////////////// - // Configuration - ////////////////////////////////////////////////// - */ - - final int mConfigFlags; - - // Extracted wstx-specific settings: - - final boolean mCfgSupportDTDPP; - - /** - * This flag indicates whether we should build a validating 'real' - * validator (true, the usual case), - * or a simpler pseudo-validator that can do all non-validation tasks - * that are based on DTD info (entity expansion, notation references, - * default attribute values). Latter is used in non-validating mode. - *
- */ - final boolean mCfgFullyValidating; - - /* - ////////////////////////////////////////////////// - // Entity handling, parameter entities (PEs) - ////////////////////////////////////////////////// - */ - - /** - * Set of parameter entities defined so far in the currently parsed - * subset. Note: the first definition sticks, entities can not be - * redefined. - *
- * Keys are entity name Strings; values are instances of EntityDecl - */ - HashMap mParamEntities; - - /** - * Set of parameter entities already defined for the subset being - * parsed; namely, PEs defined in the internal subset passed when - * parsing matching external subset. Null when parsing internal - * subset. - */ - final HashMap mPredefdPEs; - - /** - * Set of parameter entities (ids) that have been referenced by this - * DTD; only maintained for external subsets, and only as long as - * no pre-defined PE has been referenced. - */ - Set mRefdPEs; - - /* - ////////////////////////////////////////////////// - // Entity handling, general entities (GEs) - ////////////////////////////////////////////////// - */ - - /** - * Set of generic entities defined so far in this subset. - * As with parameter entities, the first definition sticks. - *
- * Keys are entity name Strings; values are instances of EntityDecl - *
- * Note: this Map only contains entities declared and defined in the - * subset being parsed; no previously defined values are passed. - */ - HashMap mGeneralEntities; - - /** - * Set of general entities already defined for the subset being - * parsed; namely, PEs defined in the internal subset passed when - * parsing matching external subset. Null when parsing internal - * subset. Such entities are only needed directly for one purpose; - * to be expanded when reading attribute default value definitions. - */ - final HashMap mPredefdGEs; - - /** - * Set of general entities (ids) that have been referenced by this - * DTD; only maintained for external subsets, and only as long as - * no pre-defined GEs have been referenced. - */ - Set mRefdGEs; - - /* - ////////////////////////////////////////////////// - // Entity handling, both PEs and GEs - ////////////////////////////////////////////////// - */ - - /** - * Flag used to keep track of whether current (external) subset - * has referenced at least one PE that was pre-defined. - */ - boolean mUsesPredefdEntities = false; - - /* - ////////////////////////////////////////////////// - // Notation settings - ////////////////////////////////////////////////// - */ - - /** - * Set of notations defined so far. Since it's illegal to (try to) - * redefine notations, there's no specific precedence. - *
- * Keys are entity name Strings; values are instances of
- * NotationDecl objects
- */
- HashMap/*
- * Note: see base class for some additional remarks about this
- * method.
- */
- public EntityDecl findEntity(String entName)
- {
- if (mPredefdGEs != null) {
- EntityDecl decl = (EntityDecl) mPredefdGEs.get(entName);
- if (decl != null) {
- return decl;
- }
- }
- return (EntityDecl) mGeneralEntities.get(entName);
- }
-
- /*
- //////////////////////////////////////////////////
- // Main-level parsing methods
- //////////////////////////////////////////////////
- */
-
- protected DTDSubset parseDTD()
- throws XMLStreamException
- {
- while (true) {
- mCheckForbiddenPEs = false; // PEs are ok at this point
- int i = getNextAfterWS();
- if (i < 0) {
- if (mIsExternal) { // ok for external DTDs
- break;
- }
- // Error for internal subset
- throwUnexpectedEOF(SUFFIX_IN_DTD_INTERNAL);
- }
- if (i == '%') { // parameter entity
- expandPE();
- continue;
- }
-
- /* First, let's keep track of start of the directive; needed for
- * entity and notation declaration events.
- */
- mTokenInputTotal = mCurrInputProcessed + mInputPtr;
- mTokenInputRow = mCurrInputRow;
- mTokenInputCol = mInputPtr - mCurrInputRowStart;
-
- if (i == '<') {
- // PEs not allowed within declarations, in the internal subset proper
- mCheckForbiddenPEs = !mIsExternal && (mInput == mRootInput);
- if (mFlattenWriter == null) {
- parseDirective();
- } else {
- parseDirectiveFlattened();
- }
- continue;
- }
-
- if (i == ']') {
- if (mIncludeCount == 0 && !mIsExternal) { // End of internal subset
- break;
- }
- if (mIncludeCount > 0) { // active INCLUDE block(s) open?
- boolean suppress = (mFlattenWriter != null) && !mFlattenWriter.includeConditionals();
-
- if (suppress) {
- mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
- mFlattenWriter.disableOutput();
- }
-
- try {
- // ]]> needs to be a token, can not come from PE:
- char c = dtdNextFromCurr();
- if (c == ']') {
- c = dtdNextFromCurr();
- if (c == '>') {
- // Ok, fine, conditional include section ended.
- --mIncludeCount;
- continue;
- }
- }
- throwDTDUnexpectedChar(c, "; expected ']]>' to close conditional include section");
- } finally {
- if (suppress) {
- mFlattenWriter.enableOutput(mInputPtr);
- }
- }
- }
- // otherwise will fall through, and give an error
- }
-
- if (mIsExternal) {
- throwDTDUnexpectedChar(i, "; expected a '<' to start a directive");
- }
- throwDTDUnexpectedChar(i, "; expected a '<' to start a directive, or \"]>\" to end internal subset");
- }
-
- /* 05-Feb-2006, TSa: Not allowed to have unclosed INCLUDE/IGNORE
- * blocks...
- */
- if (mIncludeCount > 0) { // active INCLUDE block(s) open?
- String suffix = (mIncludeCount == 1) ? "an INCLUDE block" : (""+mIncludeCount+" INCLUDE blocks");
- throwUnexpectedEOF(getErrorMsg()+"; expected closing marker for "+suffix);
- }
-
- /* First check: have all notation references been resolved?
- * (related to [WSTX-121])
- */
- if (mNotationForwardRefs != null && mNotationForwardRefs.size() > 0) {
- _reportUndefinedNotationRefs();
- }
-
- // Ok; time to construct and return DTD data object.
- DTDSubset ss;
-
- // There are more settings for ext. subsets:
- if (mIsExternal) {
- /* External subsets are cachable if they did not refer to any
- * PEs or GEs defined in internal subset passed in (if any),
- * nor to any notations.
- * We don't care about PEs it defined itself, but need to pass
- * in Set of PEs it refers to, to check if cached copy can be
- * used with different int. subsets.
- * We need not worry about notations referred, since they are
- * not allowed to be re-defined.
- */
- boolean cachable = !mUsesPredefdEntities && !mUsesPredefdNotations;
- ss = DTDSubsetImpl.constructInstance(cachable,
- mGeneralEntities, mRefdGEs,
- null, mRefdPEs,
- mNotations, mElements,
- mCfgFullyValidating);
- } else {
- /* Internal subsets are not cachable (no unique way to refer
- * to unique internal subsets), and there can be no references
- * to pre-defined PEs, as none were passed.
- */
- ss = DTDSubsetImpl.constructInstance(false, mGeneralEntities, null,
- mParamEntities, null,
- mNotations, mElements,
- mCfgFullyValidating);
- }
-
- return ss;
- }
-
- protected void parseDirective()
- throws XMLStreamException
- {
- /* Hmmh. Don't think PEs are allowed to contain starting
- * '!' (or '?')... and it has to come from the same
- * input source too (no splits)
- */
- char c = dtdNextFromCurr();
- if (c == '?') { // xml decl?
- readPI();
- return;
- }
- if (c != '!') { // nothing valid
- throwDTDUnexpectedChar(c, "; expected '!' to start a directive");
- }
-
- /* ignore/include, comment, or directive; we are still getting
- * token from same section though
- */
- c = dtdNextFromCurr();
- if (c == '-') { // plain comment
- c = dtdNextFromCurr();
- if (c != '-') {
- throwDTDUnexpectedChar(c, "; expected '-' for a comment");
- }
- if (mEventListener != null && mEventListener.dtdReportComments()) {
- readComment(mEventListener);
- } else {
- skipComment();
- }
- } else if (c == '[') {
- checkInclusion();
- } else if (c >= 'A' && c <= 'Z') {
- handleDeclaration(c);
- } else {
- throwDTDUnexpectedChar(c, ErrorConsts.ERR_DTD_MAINLEVEL_KEYWORD);
- }
- }
-
- /**
- * Method similar to {@link #parseDirective}, but one that takes care
- * to properly output dtd contents via {@link com.ctc.wstx.dtd.DTDWriter}
- * as necessary.
- * Separated to simplify both methods; otherwise would end up with
- * 'if (... flatten...) ... else ...' spaghetti code.
- */
- protected void parseDirectiveFlattened()
- throws XMLStreamException
- {
- /* First, need to flush any flattened output there may be, at
- * this point (except for opening lt char): and then need to
- * temporarily disable more output until we know the type and
- * whether it should be output or not:
- */
- mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
- mFlattenWriter.disableOutput();
-
- /* Let's determine type here, and call appropriate skip/parse
- * methods.
- */
- char c = dtdNextFromCurr();
- if (c == '?') { // xml decl?
- mFlattenWriter.enableOutput(mInputPtr);
- mFlattenWriter.output("= 'A' && c <= 'Z') {
- mFlattenWriter.enableOutput(mInputPtr);
- mFlattenWriter.output(" 0) {
- if (mFlattenWriter != null) {
- mFlattenWriter.setFlattenStart(mInputPtr);
- }
- return true;
- }
- input.close();
- } catch (IOException ioe) {
- throw constructFromIOE(ioe);
- }
- if (input == mRootInput) {
- return false;
- }
- WstxInputSource parent = input.getParent();
- if (parent == null) { // sanity check!
- throwNullParent(input);
- }
- /* 13-Feb-2006, TSa: Ok, do we violate a proper nesting constraints
- * with this input block closure?
- */
- if (mCurrDepth != input.getScopeId()) {
- handleIncompleteEntityProblem(input);
- }
-
- mInput = input = parent;
- input.restoreContext(this);
- if (mFlattenWriter != null) {
- mFlattenWriter.setFlattenStart(mInputPtr);
- }
- mInputTopDepth = input.getScopeId();
- /* 21-Feb-2006, TSa: Since linefeed normalization needs to be
- * suppressed for internal entity expansion, we may need to
- * change the state...
- */
- if (!mNormalizeLFs) {
- mNormalizeLFs = !input.fromInternalEntity();
- }
- // Maybe there are leftovers from that input in buffer now?
- } while (mInputPtr >= mInputEnd);
-
- return true;
- }
-
- protected boolean loadMoreFromCurrent()
- throws XMLStreamException
- {
- // Any flattened not-yet-output input to flush?
- if (mFlattenWriter != null) {
- mFlattenWriter.flush(mInputBuffer, mInputEnd);
- }
-
- // Need to update offsets properly
- mCurrInputProcessed += mInputEnd;
- mCurrInputRowStart -= mInputEnd;
- try {
- int count = mInput.readInto(this);
- if (count > 0) {
- if (mFlattenWriter != null) {
- mFlattenWriter.setFlattenStart(mInputPtr);
- }
- return true;
- }
- } catch (IOException ie) {
- throwFromIOE(ie);
- }
- return false;
- }
-
- protected boolean ensureInput(int minAmount)
- throws XMLStreamException
- {
- int currAmount = mInputEnd - mInputPtr;
- if (currAmount >= minAmount) {
- return true;
- }
- // Any flattened not-yet-output input to flush?
- if (mFlattenWriter != null) {
- mFlattenWriter.flush(mInputBuffer, mInputEnd);
- }
- try {
- if (mInput.readMore(this, minAmount)) {
- if (mFlattenWriter != null) {
- //mFlattenWriter.setFlattenStart(mInputPtr);
- mFlattenWriter.setFlattenStart(currAmount);
- }
- return true;
- }
- } catch (IOException ie) {
- throwFromIOE(ie);
- }
- return false;
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, input access:
- //////////////////////////////////////////////////
- */
-
- private void loadMoreScoped(WstxInputSource currScope,
- String entityName, Location loc)
- throws XMLStreamException
- {
- boolean check = (mInput == currScope);
- loadMore(getErrorMsg());
- // Did we get out of the scope?
- if (check && (mInput != currScope)) {
- _reportWFCViolation("Unterminated entity value for entity '"
- +entityName+"' (definition started at "
- +loc+")");
- }
- }
-
- /**
- * @return Next character from the current input block, if any left;
- * NULL if end of block (entity expansion)
- */
- private char dtdNextIfAvailable()
- throws XMLStreamException
- {
- char c;
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- int i = peekNext();
- if (i < 0) {
- return CHAR_NULL;
- }
- ++mInputPtr;
- c = (char) i;
- }
- if (c == CHAR_NULL) {
- throwNullChar();
- }
- return c;
- }
-
- /**
- * Method that will get next character, and either return it as is (for
- * normal chars), or expand parameter entity that starts with next
- * character (which has to be '%').
- */
- private char getNextExpanded()
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- if (c != '%') {
- return c;
- }
- expandPE();
- }
- }
-
- private char skipDtdWs(boolean handlePEs)
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- if (c > CHAR_SPACE) {
- if (c == '%' && handlePEs) {
- expandPE();
- continue;
- }
- return c;
- }
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- }
- }
-
- /**
- * Note: Apparently a parameter entity expansion does also count
- * as white space (that is, PEs outside of quoted text are considered
- * to be separated by white spaces on both sides). Fortunately this
- * can be handled by 2 little hacks: both a start of a PE, and an
- * end of input block (== end of PE expansion) count as succesful
- * spaces.
- *
- * @return Character following the obligatory boundary (white space
- * or PE start/end)
- */
- private char skipObligatoryDtdWs()
- throws XMLStreamException
- {
- /* Ok; since we need at least one space, or a PE, or end of input
- * block, let's do this unique check first...
- */
- int i = peekNext();
- char c;
-
- if (i == -1) { // just means 'local' EOF (since peek only checks current)
- c = getNextChar(getErrorMsg());
- // Non-space, non PE is ok, due to end-of-block...
- if (c > CHAR_SPACE && c != '%') {
- return c;
- }
- } else {
- c = mInputBuffer[mInputPtr++]; // was peek, need to read
- if (c > CHAR_SPACE && c != '%') {
- throwDTDUnexpectedChar(c, "; expected a separating white space");
- }
- }
-
- // Ok, got it, now can loop...
- while (true) {
- if (c == '%') {
- expandPE();
- } else if (c > CHAR_SPACE) {
- break;
- } else {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- }
- /* Now we got one space (or end of input block) -- no need to
- * restrict get next on current block (in case PE ends); happens
- * with xmltest/valid/not-sa/003.xml, for eaxmple.
- */
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- }
- return c;
- }
-
- /**
- * Method called to handle expansion of parameter entities. When called,
- * '%' character has been encountered as a reference indicator, and
- * now we should get parameter entity name.
- */
- private void expandPE()
- throws XMLStreamException
- {
- String id;
- char c;
-
- if (mCheckForbiddenPEs) {
- /* Ok; we hit a PE where we should not have (within the internal
- * dtd subset proper, within a declaration). This is a WF error.
- */
- throwForbiddenPE();
- }
-
- // 01-Jul-2004, TSa: When flattening, need to flush previous output
- if (mFlattenWriter != null) {
- // Flush up to but not including ampersand...
- mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
- mFlattenWriter.disableOutput();
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- id = readDTDName(c);
- try {
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- } finally {
- // will ignore name and colon (or whatever was parsed)
- mFlattenWriter.enableOutput(mInputPtr);
- }
- } else {
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- id = readDTDName(c);
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- }
-
- // Should now get semicolon...
- if (c != ';') {
- throwDTDUnexpectedChar(c, "; expected ';' to end parameter entity name");
- }
- mExpandingPE = true;
- expandEntity(id, true, ENTITY_EXP_PE);
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, low-level parsing:
- //////////////////////////////////////////////////
- */
-
- /**
- * Method called to verify whether input has specified keyword; if it
- * has, returns null and points to char after the keyword; if not,
- * returns whatever constitutes a keyword matched, for error
- * reporting purposes.
- */
- protected String checkDTDKeyword(String exp)
- throws XMLStreamException
- {
- int i = 0;
- int len = exp.length();
- char c = ' ';
-
- for (; i < len; ++i) {
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- c = dtdNextIfAvailable();
- if (c == CHAR_NULL) { // end of block, fine
- return exp.substring(0, i);
- }
- }
- if (c != exp.charAt(i)) {
- break;
- }
- }
-
- if (i == len) {
- // Got a match? Cool... except if identifier still continues...
- c = dtdNextIfAvailable();
- if (c == CHAR_NULL) { // EOB, fine
- return null;
- }
- if (!isNameChar(c)) {
- --mInputPtr; // to push it back
- return null;
- }
- }
- StringBuffer sb = new StringBuffer(exp.substring(0, i));
- sb.append(c);
- while (true) {
- c = dtdNextIfAvailable();
- if (c == CHAR_NULL) { // EOB, fine
- break;
- }
- if (!isNameChar(c) && c != ':') {
- --mInputPtr; // to push it back
- break;
- }
- sb.append(c);
- }
- return sb.toString();
- }
-
- /**
- * Method called usually to indicate an error condition; will read rest
- * of specified keyword (including characters that can be part of XML
- * identifiers), append that to passed prefix (which is optional), and
- * return resulting String.
- *
- * @param prefix Part of keyword already read in.
- */
- protected String readDTDKeyword(String prefix)
- throws XMLStreamException
- {
- StringBuffer sb = new StringBuffer(prefix);
-
- while (true) {
- char c;
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- // Don't want to cross block boundary
- c = dtdNextIfAvailable();
- if (c == CHAR_NULL) {
- break; // end-of-block
- }
- }
- if (!isNameChar(c) && c != ':') {
- --mInputPtr;
- break;
- }
- sb.append(c);
- }
- return sb.toString();
- }
-
- /**
- * @return True, if input contains 'PUBLIC' keyword; false if it
- * contains 'SYSTEM'; otherwise throws an exception.
- */
- private boolean checkPublicSystemKeyword(char c)
- throws XMLStreamException
- {
- String errId;
-
- if (c == 'P') {
- errId = checkDTDKeyword("UBLIC");
- if (errId == null) {
- return true;
- }
- errId = "P" + errId;
- } else if (c == 'S') {
- errId = checkDTDKeyword("YSTEM");
- if (errId == null) {
- return false;
- }
- errId = "S" + errId;
- } else {
- if (!isNameStartChar(c)) {
- throwDTDUnexpectedChar(c, "; expected 'PUBLIC' or 'SYSTEM' keyword");
- }
- errId = readDTDKeyword(String.valueOf(c));
- }
-
- _reportWFCViolation("Unrecognized keyword '"+errId+"'; expected 'PUBLIC' or 'SYSTEM'");
- return false; // never gets here
- }
-
- private String readDTDName(char c)
- throws XMLStreamException
- {
- // Let's just check this before trying to parse the id...
- if (!isNameStartChar(c)) {
- throwDTDUnexpectedChar(c, "; expected an identifier");
- }
- return parseFullName(c);
- }
-
- private String readDTDLocalName(char c, boolean checkChar)
- throws XMLStreamException
- {
- /* Let's just check this first, to get better error msg
- * (parseLocalName() will double-check it too)
- */
- if (checkChar && !isNameStartChar(c)) {
- throwDTDUnexpectedChar(c, "; expected an identifier");
- }
- return parseLocalName(c);
- }
-
- /**
- * Similar to {@link #readDTDName}, except that the rules are bit looser,
- * ie. there are no additional restrictions for the first char
- */
- private String readDTDNmtoken(char c)
- throws XMLStreamException
- {
- char[] outBuf = getNameBuffer(64);
- int outLen = outBuf.length;
- int outPtr = 0;
-
- while (true) {
- /* Note: colon not included in name char array, since it has
- * special meaning WRT QNames, need to add into account here:
- */
- if (!isNameChar(c) && c != ':') {
- // Need to get at least one char
- if (outPtr == 0) {
- throwDTDUnexpectedChar(c, "; expected a NMTOKEN character to start a NMTOKEN");
- }
- --mInputPtr;
- break;
- }
- if (outPtr >= outLen) {
- outBuf = expandBy50Pct(outBuf);
- outLen = outBuf.length;
- }
- outBuf[outPtr++] = c;
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- c = dtdNextIfAvailable();
- if (c == CHAR_NULL) { // end-of-block
- break;
- }
- }
- }
-
- /* Nmtokens need not be canonicalized; they will be processed
- * as necessary later on:
- */
- return new String(outBuf, 0, outPtr);
- }
-
- /**
- * Method that will read an element or attribute name from DTD; depending
- * on namespace mode, it can have prefix as well.
- *
- * Note: returned {@link PrefixedName} instances are canonicalized so that
- * all instances read during parsing of a single DTD subset so that
- * identity comparison can be used instead of calling
- * Note: c is known to be a letter (from 'A' to 'Z') at this poit.
- */
- private void handleDeclaration(char c)
- throws XMLStreamException
- {
- String keyw = null;
-
- /* We need to ensure that PEs do not span declaration boundaries
- * (similar to element nesting wrt. GE expansion for xml content).
- * This VC is defined in xml 1.0, section 2.8 as
- * "VC: Proper Declaration/PE Nesting"
- */
- /* We have binary depths within DTDs, for now: since the declaration
- * just started, we should now have 1 as the depth:
- */
- mCurrDepth = 1;
-
- try {
- do { // dummy loop, for break
- if (c == 'A') { // ATTLIST?
- keyw = checkDTDKeyword("TTLIST");
- if (keyw == null) {
- mCurrDeclaration = "ATTLIST";
- handleAttlistDecl();
- break;
- }
- keyw = "A" + keyw;
- } else if (c == 'E') { // ENTITY, ELEMENT?
- c = dtdNextFromCurr();
- if (c == 'N') {
- keyw = checkDTDKeyword("TITY");
- if (keyw == null) {
- mCurrDeclaration = "ENTITY";
- handleEntityDecl(false);
- break;
- }
- keyw = "EN" + keyw;
- } else if (c == 'L') {
- keyw = checkDTDKeyword("EMENT");
- if (keyw == null) {
- mCurrDeclaration = "ELEMENT";
- handleElementDecl();
- break;
- }
- keyw = "EL" + keyw;
- } else {
- keyw = readDTDKeyword("E");
- }
- } else if (c == 'N') { // NOTATION?
- keyw = checkDTDKeyword("OTATION");
- if (keyw == null) {
- mCurrDeclaration = "NOTATION";
- handleNotationDecl();
- break;
- }
- keyw = "N" + keyw;
- } else if (c == 'T' && mCfgSupportDTDPP) { // (dtd++ only) TARGETNS?
- keyw = checkDTDKeyword("ARGETNS");
- if (keyw == null) {
- mCurrDeclaration = "TARGETNS";
- handleTargetNsDecl();
- break;
- }
- keyw = "T" + keyw;
- } else {
- keyw = readDTDKeyword(String.valueOf(c));
- }
-
- // If we got this far, we got a problem...
- _reportBadDirective(keyw);
- } while (false);
- /* Ok: now, the current input can not have been started
- * within the scope... so:
- */
- if (mInput.getScopeId() > 0) {
- handleGreedyEntityProblem(mInput);
- }
-
- } finally {
- // Either way, declaration has ended now...
- mCurrDepth = 0;
- mCurrDeclaration = null;
- }
- }
-
- /**
- * Specialized method that handles potentially suppressable entity
- * declaration. Specifically: at this point it is known that first
- * letter is 'E', that we are outputting flattened DTD info,
- * and that parameter entity declarations are to be suppressed.
- * Furthermore, flatten output is still being disabled, and needs
- * to be enabled by the method at some point.
- */
- private void handleSuppressedDeclaration()
- throws XMLStreamException
- {
- String keyw;
- char c = dtdNextFromCurr();
-
- if (c == 'N') {
- keyw = checkDTDKeyword("TITY");
- if (keyw == null) {
- handleEntityDecl(true);
- return;
- }
- keyw = "EN" + keyw;
- mFlattenWriter.enableOutput(mInputPtr); // error condition...
- } else {
- mFlattenWriter.enableOutput(mInputPtr);
- mFlattenWriter.output("' char;
- * otherwise it's obligatory.
- */
- c = getNextExpanded();
- if (isSpaceChar(c)) {
- // Let's push it back in case it's LF, to be handled properly
- --mInputPtr;
- c = skipDtdWs(true);
-
- /* 26-Jan-2006, TSa: actually there are edge cases where
- * we may get the attribute name right away (esp.
- * with PEs...); so let's defer possible error for
- * later on. Should not allow missing spaces between
- * attribute declarations... ?
- */
- /*
- } else if (c != '>') {
- throwDTDUnexpectedChar(c, "; excepted either '>' closing ATTLIST declaration, or a white space character separating individual attribute declarations");
- */
- }
- if (c == '>') {
- break;
- }
- handleAttrDecl(elem, c, index, loc);
- ++index;
- }
- }
-
- private void handleElementDecl()
- throws XMLStreamException
- {
- char c = skipObligatoryDtdWs();
- final PrefixedName elemName = readDTDQName(c);
-
- /* Ok, event needs to know its exact starting point (opening '<'
- * char), let's get that info now (note: data has been preserved
- * earlier)
- */
- Location loc = getStartLocation();
-
- // Ok; name got, need some white space next
- c = skipObligatoryDtdWs();
-
- /* Then the content spec: either a special case (ANY, EMPTY), or
- * a parenthesis group for 'real' content spec
- */
- StructValidator val = null;
- int vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT;
-
- if (c == '(') { // real content model
- c = skipDtdWs(true);
- if (c == '#') {
- val = readMixedSpec(elemName, mCfgFullyValidating);
- vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT; // checked against DTD
- } else {
- --mInputPtr; // let's push it back...
- ContentSpec spec = readContentSpec(elemName, true, mCfgFullyValidating);
- val = spec.getSimpleValidator();
- if (val == null) {
- val = new DFAValidator(DFAState.constructDFA(spec));
- }
- vldContent = XMLValidator.CONTENT_ALLOW_WS; // checked against DTD
- }
- } else if (isNameStartChar(c)) {
- do { // dummy loop to allow break:
- String keyw = null;
- if (c == 'A') {
- keyw = checkDTDKeyword("NY");
- if (keyw == null) {
- val = null;
- vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT; // no DTD checks
- break;
- }
- keyw = "A"+keyw;
- } else if (c == 'E') {
- keyw = checkDTDKeyword("MPTY");
- if (keyw == null) {
- val = EmptyValidator.getPcdataInstance();
- vldContent = XMLValidator.CONTENT_ALLOW_NONE; // needed to prevent non-elements too
- break;
- }
- keyw = "E"+keyw;
- } else {
- --mInputPtr;
- keyw = readDTDKeyword(String.valueOf(c));
- }
- _reportWFCViolation("Unrecognized DTD content spec keyword '"
- +keyw+"' (for element <"+elemName+">); expected ANY or EMPTY");
- } while (false);
- } else {
- throwDTDUnexpectedChar(c, ": excepted '(' to start content specification for element <"+elemName+">");
- }
-
- // Ok, still need the trailing gt-char to close the declaration:
- c = skipDtdWs(true);
- if (c != '>') {
- throwDTDUnexpectedChar(c, "; expected '>' to finish the element declaration for <"+elemName+">");
- }
-
- HashMap m = getElementMap();
- DTDElement oldElem = (DTDElement) m.get(elemName);
- // Ok to have it if it's not 'really' declared
-
- if (oldElem != null) {
- if (oldElem.isDefined()) { // oops, a problem!
- /* 03-Feb-2006, TSa: Hmmh. Apparently all other XML parsers
- * consider it's ok in non-validating mode. All right.
- */
- if (mCfgFullyValidating) {
- DTDSubsetImpl.throwElementException(oldElem, loc);
- } else {
- // let's just ignore re-definition if not validating
- return;
- }
- }
-
- /* 09-Sep-2004, TSa: Need to transfer existing attribute
- * definitions, however...
- */
- oldElem = oldElem.define(loc, val, vldContent);
- } else {
- // Sweet, let's then add the definition:
- oldElem = DTDElement.createDefined(mConfig, loc, elemName, val, vldContent);
- }
- m.put(elemName, oldElem);
- }
-
- /**
- * This method is tricky to implement, since it can contain parameter
- * entities in multiple combinations... and yet declare one as well.
- *
- * @param suppressPEDecl If true, will need to take of enabling/disabling
- * of flattened output.
- */
- private void handleEntityDecl(boolean suppressPEDecl)
- throws XMLStreamException
- {
- /* Hmmh. It seems that PE reference are actually accepted
- * even here... which makes distinguishing definition from
- * reference bit challenging.
- */
- char c = dtdNextFromCurr();
- boolean gotSeparator = false;
- boolean isParam = false;
-
- while (true) {
- if (c == '%') { // reference?
- // note: end-of-block acceptable, same as space
- char d = dtdNextIfAvailable();
- if (d == CHAR_NULL || isSpaceChar(d)) { // ok, PE declaration
- isParam = true;
- if (d == '\n' || c == '\r') {
- skipCRLF(d);
- }
- break;
- }
- // Reference?
- if (!isNameStartChar(d)) {
- throwDTDUnexpectedChar(d, "; expected a space (for PE declaration) or PE reference name");
- }
- --mInputPtr; // need to push the first char back, then
- gotSeparator = true;
- expandPE();
- // need the next char, from the new scope... or if it gets closed, this one
- c = dtdNextChar();
- } else if (!isSpaceChar(c)) { // non-PE entity?
- break;
- } else {
- gotSeparator = true;
- c = dtdNextFromCurr();
- }
- }
-
- if (!gotSeparator) {
- throwDTDUnexpectedChar(c, "; expected a space separating ENTITY keyword and entity name");
- }
-
- /* Ok; fair enough: now must have either '%', or a name start
- * character:
- */
- if (isParam) {
- /* PE definition: at this point we already know that there must
- * have been a space... just need to skip the rest, if any.
- * Also, can still get a PE to expand (name of a PE to define
- * from a PE reference)
- */
- c = skipDtdWs(true);
- }
-
- if (suppressPEDecl) { // only if mFlattenWriter != null
- if (!isParam) {
- mFlattenWriter.enableOutput(mInputPtr);
- mFlattenWriter.output(" 0 && (old = m.get(id)) != null) {
- // Application may want to know about the problem...
- XMLReporter rep = mConfig.getXMLReporter();
- if (rep != null) {
- EntityDecl oldED = (EntityDecl) old;
- String str = " entity '"+id+"' defined more than once: first declaration at "
- + oldED.getLocation();
- if (isParam) {
- str = "Parameter" + str;
- } else {
- str = "General" + str;
- }
- _reportWarning(rep, ErrorConsts.WT_ENT_DECL, str, evtLoc);
- }
- } else {
- m.put(id, ent);
- }
-
- // And finally, let's notify listener, if we have one...
- if (mEventListener != null) {
- if (ent.isParsed()) { // Parsed GE or PE
- } else { // unparsed GE
- mEventListener.dtdUnparsedEntityDecl(id, ent.getPublicId(), ent.getSystemId(), ent.getNotationName(), mInput.getSource());
- }
- }
- }
-
- /**
- * Method called to handle declaration.
- */
- private void handleNotationDecl()
- throws XMLStreamException
- {
- char c = skipObligatoryDtdWs();
- String id = readDTDName(c);
-
- c = skipObligatoryDtdWs();
- boolean isPublic = checkPublicSystemKeyword(c);
-
- String pubId, sysId;
-
- c = skipObligatoryDtdWs();
-
- // Ok, now we can parse the reference; first public id if needed:
- if (isPublic) {
- if (c != '"' && c != '\'') {
- throwDTDUnexpectedChar(c, "; expected a quote to start the public identifier");
- }
- pubId = parsePublicId(c, getErrorMsg());
- c = skipDtdWs(true);
- } else {
- pubId = null;
- }
-
- /* And then we may need the system id; one NOTATION oddity, if
- * there's public id, system one is optional.
- */
- if (c == '"' || c == '\'') {
- sysId = parseSystemId(c, mNormalizeLFs, getErrorMsg());
- c = skipDtdWs(true);
- } else {
- if (!isPublic) {
- throwDTDUnexpectedChar(c, "; expected a quote to start the system identifier");
- }
- sysId = null;
- }
-
- // And then we should get the closing '>'
- if (c != '>') {
- throwDTDUnexpectedChar(c, "; expected closing '>' after NOTATION declaration");
- }
- URL baseURL = mInput.getSource();
-
- // Any external listeners?
- if (mEventListener != null) {
- mEventListener.dtdNotationDecl(id, pubId, sysId, baseURL);
- }
-
- /* Ok, event needs to know its exact starting point (opening '<'
- * char), let's get that info now (note: data has been preserved
- * earlier)
- */
- Location evtLoc = getStartLocation();
- NotationDeclaration nd = new WNotationDeclaration(evtLoc, id, pubId, sysId, baseURL);
-
- // Any definitions from the internal subset?
- if (mPredefdNotations != null) {
- NotationDeclaration oldDecl = (NotationDeclaration) mPredefdNotations.get(id);
- if (oldDecl != null) { // oops, a problem!
- DTDSubsetImpl.throwNotationException(oldDecl, nd);
- }
- }
-
- HashMap m = mNotations;
- if (m == null) {
- /* Let's try to get insert-ordered Map, to be able to
- * report redefinition problems in proper order when validating
- * subset compatibility
- */
- mNotations = m = new LinkedHashMap();
- } else {
- NotationDeclaration oldDecl = (NotationDeclaration) m.get(id);
- if (oldDecl != null) { // oops, a problem!
- DTDSubsetImpl.throwNotationException(oldDecl, nd);
- }
- }
- // Does this resolve a dangling reference?
- if (mNotationForwardRefs != null) {
- mNotationForwardRefs.remove(id);
- }
- m.put(id, nd);
- }
-
- /**
- * Method called to handle declaration (the only
- * new declaration type for DTD++)
- *
- * Note: only valid for DTD++, in 'plain DTD' mode shouldn't get
- * called.
- */
- private void handleTargetNsDecl()
- throws XMLStreamException
- {
- mAnyDTDppFeatures = true;
-
- char c = skipObligatoryDtdWs();
- String name;
-
- // Explicit namespace name?
- if (isNameStartChar(c)) {
- name = readDTDLocalName(c, false);
- c = skipObligatoryDtdWs();
- } else { // no, default namespace (or error)
- name = null;
- }
-
- // Either way, should now get a quote:
- if (c != '"' && c != '\'') {
- if (c == '>') { // slightly more accurate error
- _reportWFCViolation("Missing namespace URI for TARGETNS directive");
- }
- throwDTDUnexpectedChar(c, "; expected a single or double quote to enclose the namespace URI");
- }
-
- /* !!! 07-Nov-2004, TSa: what's the exact value we should get
- * here? Ns declarations can have any attr value...
- */
- String uri = parseSystemId(c, false, "in namespace URI");
-
- // Do we need to normalize the URI?
- if ((mConfigFlags & CFG_INTERN_NS_URIS) != 0) {
- uri = InternCache.getInstance().intern(uri);
- }
-
- // Ok, and then the closing '>':
- c = skipDtdWs(true);
- if (c != '>') {
- throwDTDUnexpectedChar(c, "; expected '>' to end TARGETNS directive");
- }
-
- if (name == null) { // default NS URI
- mDefaultNsURI = uri;
- } else {
- if (mNamespaces == null) {
- mNamespaces = new HashMap();
- }
- mNamespaces.put(name, uri);
- }
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, secondary decl parsing methods
- /////////////////////////////////////////////////////
- */
-
- /**
- * @param elem Element that contains this attribute
- * @param c First character of what should be the attribute name
- * @param index Sequential index number of this attribute as children
- * of the element; used for creating bit masks later on.
- * @param loc Location of the element name in attribute list declaration
- */
- private void handleAttrDecl(DTDElement elem, char c, int index,
- Location loc)
- throws XMLStreamException
- {
- // First attribute name
- PrefixedName attrName = readDTDQName(c);
-
- // then type:
- c = skipObligatoryDtdWs();
-
- int type = 0;
- WordResolver enumValues = null;
-
- if (c == '(') { // enumerated type
- enumValues = parseEnumerated(elem, attrName, false);
- type = DTDAttribute.TYPE_ENUMERATED;
- } else {
- String typeStr = readDTDName(c);
-
- dummy:
- do { // dummy loop
- switch (typeStr.charAt(0)) {
- case 'C': // CDATA
- if (typeStr == "CDATA") {
- type = DTDAttribute.TYPE_CDATA;
- break dummy;
- }
- break;
- case 'I': // ID, IDREF, IDREFS
- if (typeStr == "ID") {
- type = DTDAttribute.TYPE_ID;
- break dummy;
- } else if (typeStr == "IDREF") {
- type = DTDAttribute.TYPE_IDREF;
- break dummy;
- } else if (typeStr == "IDREFS") {
- type = DTDAttribute.TYPE_IDREFS;
- break dummy;
- }
- break;
- case 'E': // ENTITY, ENTITIES
- if (typeStr == "ENTITY") {
- type = DTDAttribute.TYPE_ENTITY;
- break dummy;
- } else if (typeStr == "ENTITIES") {
- type = DTDAttribute.TYPE_ENTITIES;
- break dummy;
- }
- break;
- case 'N': // NOTATION, NMTOKEN, NMTOKENS
- if (typeStr == "NOTATION") {
- type = DTDAttribute.TYPE_NOTATION;
- /* Special case; is followed by a list of
- * enumerated ids...
- */
- c = skipObligatoryDtdWs();
- if (c != '(') {
- throwDTDUnexpectedChar(c, "Excepted '(' to start the list of NOTATION ids");
- }
- enumValues = parseEnumerated(elem, attrName, true);
- break dummy;
- } else if (typeStr == "NMTOKEN") {
- type = DTDAttribute.TYPE_NMTOKEN;
- break dummy;
- } else if (typeStr == "NMTOKENS") {
- type = DTDAttribute.TYPE_NMTOKENS;
- break dummy;
- }
- break;
- }
-
- // Problem:
- throwDTDAttrError("Unrecognized attribute type '"+typeStr+"'"
- +ErrorConsts.ERR_DTD_ATTR_TYPE,
- elem, attrName);
- } while (false);
- }
- DefaultAttrValue defVal;
-
- // Ok, and how about the default declaration?
- c = skipObligatoryDtdWs();
- if (c == '#') {
- String defTypeStr = readDTDName(getNextExpanded());
- if (defTypeStr == "REQUIRED") {
- defVal = DefaultAttrValue.constructRequired();
- } else if (defTypeStr == "IMPLIED") {
- defVal = DefaultAttrValue.constructImplied();
- } else if (defTypeStr == "FIXED") {
- defVal = DefaultAttrValue.constructFixed();
- c = skipObligatoryDtdWs();
- parseAttrDefaultValue(defVal, c, attrName, loc, true);
- } else {
- throwDTDAttrError("Unrecognized attribute default value directive #"+defTypeStr
- +ErrorConsts.ERR_DTD_DEFAULT_TYPE,
- elem, attrName);
- defVal = null; // never gets here...
- }
- } else {
- defVal = DefaultAttrValue.constructOptional();
- parseAttrDefaultValue(defVal, c, attrName, loc, false);
- }
-
- /* There are some checks that can/need to be done now, such as:
- *
- * - [#3.3.1/VC: ID Attribute default] def. value type can not
- * be #FIXED
- */
- if (type == DTDAttribute.TYPE_ID && defVal.hasDefaultValue()) {
- // Just a VC, not WFC... so:
- if (mCfgFullyValidating) {
- throwDTDAttrError("has type ID; can not have a default (or #FIXED) value (XML 1.0/#3.3.1)",
- elem, attrName);
- }
- } else { // not an ID... shouldn't be xml:id, then
- if (mConfig.willDoXmlIdTyping()) {
- if (attrName.isXmlReservedAttr(mCfgNsEnabled, "id")) {
- // 26-Sep-2006, TSa: For [WSTX-22], need to verify 'xml:id'
- checkXmlIdAttr(type);
- }
- }
- }
-
- /* 01-Sep-2006, TSa: To address [WSTX-23], we should verify declaration
- * of 'xml:space' attribute
- */
- if (attrName.isXmlReservedAttr(mCfgNsEnabled, "space")) {
- checkXmlSpaceAttr(type, enumValues);
- }
-
- DTDAttribute attr;
-
- /* 17-Feb-2006, TSa: Ok. So some (legacy?) DTDs do declare namespace
- * declarations too... sometimes including default values.
- */
- if (mCfgNsEnabled && attrName.isaNsDeclaration()) { // only check in ns mode
- /* Ok: just declaring them is unnecessary, and can be safely
- * ignored. It's only the default values that matter (and yes,
- * let's not worry about #REQUIRED for now)
- */
- if (!defVal.hasDefaultValue()) {
- return;
- }
- // But defaulting... Hmmh.
- attr = elem.addNsDefault(this, attrName, type,
- defVal, mCfgFullyValidating);
- } else {
- attr = elem.addAttribute(this, attrName, type,
- defVal, enumValues,
- mCfgFullyValidating);
- }
-
- // getting null means this is a dup...
- if (attr == null) {
- // anyone interested in knowing about possible problem?
- XMLReporter rep = mConfig.getXMLReporter();
- if (rep != null) {
- String msg = MessageFormat.format(ErrorConsts.W_DTD_ATTR_REDECL, new Object[] { attrName, elem });
- _reportWarning(rep, ErrorConsts.WT_ATTR_DECL, msg, loc);
- }
- } else {
- if (defVal.hasDefaultValue()) {
- // always normalize
- attr.normalizeDefault();
- // but only validate in validating mode:
- if (mCfgFullyValidating) {
- attr.validateDefault(this, true);
- }
- }
- }
- }
-
- /**
- * Parsing method that reads a list of one or more space-separated
- * tokens (nmtoken or name, depending on 'isNotation' argument)
- */
- private WordResolver parseEnumerated(DTDElement elem, PrefixedName attrName,
- boolean isNotation)
- throws XMLStreamException
- {
- /* Need to use tree set to be able to construct the data
- * structs we need later on...
- */
- TreeSet set = new TreeSet();
-
- char c = skipDtdWs(true);
- if (c == ')') { // just to give more meaningful error msgs
- throwDTDUnexpectedChar(c, " (empty list; missing identifier(s))?");
- }
-
- HashMap sharedEnums;
-
- if (isNotation) {
- sharedEnums = null;
- } else {
- sharedEnums = mSharedEnumValues;
- if (sharedEnums == null && !isNotation) {
- mSharedEnumValues = sharedEnums = new HashMap();
- }
- }
-
- String id = isNotation ? readNotationEntry(c, attrName, elem.getLocation())
- : readEnumEntry(c, sharedEnums);
- set.add(id);
-
- while (true) {
- c = skipDtdWs(true);
- if (c == ')') {
- break;
- }
- if (c != '|') {
- throwDTDUnexpectedChar(c, "; missing '|' separator?");
- }
- c = skipDtdWs(true);
- id = isNotation ? readNotationEntry(c, attrName, elem.getLocation())
- : readEnumEntry(c, sharedEnums);
- if (!set.add(id)) {
- /* 03-Feb-2006, TSa: Hmmh. Apparently all other XML parsers
- * consider it's ok in non-validating mode. All right.
- */
- if (mCfgFullyValidating) {
- throwDTDAttrError("Duplicate enumeration value '"+id+"'",
- elem, attrName);
- }
- }
- }
-
- // Ok, let's construct the minimal data struct, then:
- return WordResolver.constructInstance(set);
- }
-
- /**
- * Method called to read a notation reference entry; done both for
- * attributes of type NOTATION, and for external unparsed entities
- * that refer to a notation. In both cases, notation referenced
- * needs to have been defined earlier; but only if we are building
- * a fully validating DTD subset object (there is the alternative
- * of a minimal DTD in DTD-aware mode, which does no validation
- * but allows attribute defaulting and normalization, as well as
- * access to entity and notation declarations).
- *
- * @param refLoc Starting location of the DTD component that contains
- * the reference
- */
- private String readNotationEntry(char c, PrefixedName attrName, Location refLoc)
- throws XMLStreamException
- {
- String id = readDTDName(c);
-
- /* Need to check whether we have a reference to a "pre-defined"
- * notation: pre-defined here means that it was defined in the
- * internal subset (prior to this parsing which then would external
- * subset). This is needed to know if the subset can be cached or
- * not.
- */
- if (mPredefdNotations != null) {
- NotationDeclaration decl = (NotationDeclaration) mPredefdNotations.get(id);
- if (decl != null) {
- mUsesPredefdNotations = true;
- return decl.getName();
- }
- }
-
- NotationDeclaration decl = (mNotations == null) ? null :
- (NotationDeclaration) mNotations.get(id);
- if (decl == null) {
- // In validating mode, this may be a problem (otherwise not)
- if (mCfgFullyValidating) {
- if (mNotationForwardRefs == null) {
- mNotationForwardRefs = new LinkedHashMap();
- }
- mNotationForwardRefs.put(id, refLoc);
- }
- return id;
- }
- return decl.getName();
- }
-
- private String readEnumEntry(char c, HashMap sharedEnums)
- throws XMLStreamException
- {
- String id = readDTDNmtoken(c);
-
- /* Let's make sure it's shared for this DTD subset; saves memory
- * both for DTDs and resulting docs. Could also intern Strings?
- */
- String sid = (String) sharedEnums.get(id);
- if (sid == null) {
- sid = id;
- if (INTERN_SHARED_NAMES) {
- /* 19-Nov-2004, TSa: Let's not use intern cache here...
- * shouldn't be performance critical (DTDs themselves
- * cached), and would add more entries to cache.
- */
- sid = sid.intern();
- }
- sharedEnums.put(sid, sid);
- }
- return sid;
- }
-
-
- /**
- * Method called to parse what seems like a mixed content specification.
- *
- * @param construct If true, will build full object for validating content
- * within mixed content model; if false, will just parse and discard
- * information (done in non-validating DTD-supporting mode)
- */
- private StructValidator readMixedSpec(PrefixedName elemName, boolean construct)
- throws XMLStreamException
- {
- String keyw = checkDTDKeyword("PCDATA");
- if (keyw != null) {
- _reportWFCViolation("Unrecognized directive #"+keyw+"'; expected #PCDATA (or element name)");
- }
-
- HashMap m = new LinkedHashMap();
- while (true) {
- char c = skipDtdWs(true);
- if (c == ')') {
- break;
- }
- if (c == '|') {
- c = skipDtdWs(true);
- } else if (c == ',') {
- throwDTDUnexpectedChar(c, " (sequences not allowed within mixed content)");
- } else if (c == '(') {
- throwDTDUnexpectedChar(c, " (sub-content specs not allowed within mixed content)");
- } else {
- throwDTDUnexpectedChar(c, "; expected either '|' to separate elements, or ')' to close the list");
- }
- PrefixedName n = readDTDQName(c);
- Object old = m.put(n, TokenContentSpec.construct(' ', n));
- if (old != null) {
- /* 03-Feb-2006, TSa: Hmmh. Apparently all other XML parsers
- * consider it's ok in non-validating mode. All right.
- */
- if (mCfgFullyValidating) {
- throwDTDElemError("duplicate child element <"+n+"> in mixed content model",
- elemName);
- }
- }
- }
-
- /* One more check: can have a trailing asterisk; in fact, have
- * to have one if there were any elements.
- */
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- if (c != '*') {
- if (m.size() > 0) {
- _reportWFCViolation("Missing trailing '*' after a non-empty mixed content specification");
- }
- --mInputPtr; // need to push it back
- }
- if (!construct) { // no one cares?
- return null;
- }
-
- /* Without elements, it's considered "pure" PCDATA, which can use a
- * specific 'empty' validator:
- */
- if (m.isEmpty()) {
- return EmptyValidator.getPcdataInstance();
- }
- ContentSpec spec = ChoiceContentSpec.constructMixed(mCfgNsEnabled, m.values());
- StructValidator val = spec.getSimpleValidator();
- if (val == null) {
- DFAState dfa = DFAState.constructDFA(spec);
- val = new DFAValidator(dfa);
- }
- return val;
- }
-
- private ContentSpec readContentSpec(PrefixedName elemName, boolean mainLevel,
- boolean construct)
- throws XMLStreamException
- {
- ArrayList subSpecs = new ArrayList();
- boolean isChoice = false; // default to sequence
- boolean choiceSet = false;
-
- while (true) {
- char c = skipDtdWs(true);
- if (c == ')') {
- // Need to have had at least one entry...
- if (subSpecs.isEmpty()) {
- _reportWFCViolation("Empty content specification for '"+elemName+"' (need at least one entry)");
- }
- break;
- }
- if (c == '|' || c == ',') { // choice/seq indicator
- boolean newChoice = (c == '|');
- if (!choiceSet) {
- isChoice = newChoice;
- choiceSet = true;
- } else {
- if (isChoice != newChoice) {
- _reportWFCViolation("Can not mix content spec separators ('|' and ','); need to use parenthesis groups");
- }
- }
- c = skipDtdWs(true);
- } else {
- // Need separator between subspecs...
- if (!subSpecs.isEmpty()) {
- throwDTDUnexpectedChar(c, " (missing separator '|' or ','?)");
- }
- }
- if (c == '(') {
- ContentSpec cs = readContentSpec(elemName, false, construct);
- subSpecs.add(cs);
- continue;
- }
-
- // Just to get better error messages:
- if (c == '|' || c == ',') {
- throwDTDUnexpectedChar(c, " (missing element name?)");
- }
- PrefixedName thisName = readDTDQName(c);
-
- /* Now... it's also legal to directly tag arity marker to a
- * single element name, too...
- */
- char arity = readArity();
- ContentSpec cs = construct ?
- TokenContentSpec.construct(arity, thisName)
- : TokenContentSpec.getDummySpec();
- subSpecs.add(cs);
- }
-
- char arity = readArity();
-
- /* Not really interested in constructing anything? Let's just
- * return the dummy placeholder.
- */
- if (!construct) {
- return TokenContentSpec.getDummySpec();
- }
-
- // Just one entry? Can just return it as is, combining arities
- if (subSpecs.size() == 1) {
- ContentSpec cs = (ContentSpec) subSpecs.get(0);
- char otherArity = cs.getArity();
- if (arity != otherArity) {
- cs.setArity(combineArities(arity, otherArity));
- }
- return cs;
- }
-
- if (isChoice) {
- return ChoiceContentSpec.constructChoice(mCfgNsEnabled, arity, subSpecs);
- }
- return SeqContentSpec.construct(mCfgNsEnabled, arity, subSpecs);
- }
-
- private static char combineArities(char arity1, char arity2)
- {
- if (arity1 == arity2) {
- return arity1;
- }
-
- // no modifier doesn't matter:
- if (arity1 == ' ') {
- return arity2;
- }
- if (arity2 == ' ') {
- return arity1;
- }
- // Asterisk is most liberal, supercedes others:
- if (arity1 == '*' || arity2 == '*') {
- return '*';
- }
-
- /* Ok, can only have '+' and '?'; which combine to
- * '*'
- */
- return '*';
- }
-
- /**
- * Method that handles rest of external entity declaration, after
- * it's been figured out entity is not internal (does not continue
- * with a quote).
- *
- * @param inputSource Input source for the start of the declaration.
- * Needed for resolving relative system references, if any.
- * @param isParam True if this a parameter entity declaration; false
- * if general entity declaration
- * @param evtLoc Location where entity declaration directive started;
- * needed when construction event Objects for declarations.
- */
- private EntityDecl handleExternalEntityDecl(WstxInputSource inputSource,
- boolean isParam, String id,
- char c, Location evtLoc)
- throws XMLStreamException
- {
- boolean isPublic = checkPublicSystemKeyword(c);
-
- String pubId = null;
-
- // Ok, now we can parse the reference; first public id if needed:
- if (isPublic) {
- c = skipObligatoryDtdWs();
- if (c != '"' && c != '\'') {
- throwDTDUnexpectedChar(c, "; expected a quote to start the public identifier");
- }
- pubId = parsePublicId(c, getErrorMsg());
- /* 30-Sep-2005, TSa: SGML has public ids that miss the system
- * id. Although not legal with XML DTDs, let's give bit more
- * meaningful error in those cases...
- */
- c = getNextExpanded();
- if (c <= CHAR_SPACE) { // good
- c = skipDtdWs(true);
- } else { // not good...
- // Let's just push it back and generate normal error then:
- if (c != '>') { // this is handled below though
- --mInputPtr;
- c = skipObligatoryDtdWs();
- }
- }
-
- /* But here let's deal with one case that we are familiar with:
- * SGML does NOT require system id after public one...
- */
- if (c == '>') {
- _reportWFCViolation("Unexpected end of ENTITY declaration (expected a system id after public id): trying to use an SGML DTD instead of XML one?");
- }
- } else {
- // Just need some white space here
- c = skipObligatoryDtdWs();
- }
- if (c != '"' && c != '\'') {
- throwDTDUnexpectedChar(c, "; expected a quote to start the system identifier");
- }
- String sysId = parseSystemId(c, mNormalizeLFs, getErrorMsg());
-
- // Ok; how about notation?
- String notationId = null;
-
- /* Ok; PEs are simpler, as they always are parsed (see production
- * #72 in xml 1.0 specs)
- */
- if (isParam) {
- c = skipDtdWs(true);
- } else {
- /* GEs can be unparsed, too, so it's bit more complicated;
- * if we get '>', don't need space; otherwise need separating
- * space (or PE boundary). Thus, need bit more code.
- */
- int i = peekNext();
- if (i == '>') { // good
- c = '>';
- ++mInputPtr;
- } else if (i < 0) { // local EOF, ok
- c = skipDtdWs(true);
- } else if (i == '%') {
- c = getNextExpanded();
- } else {
- ++mInputPtr;
- c = (char) i;
- if (!isSpaceChar(c)) {
- throwDTDUnexpectedChar(c, "; expected a separating space or closing '>'");
- }
- c = skipDtdWs(true);
- }
-
- if (c != '>') {
- if (!isNameStartChar(c)) {
- throwDTDUnexpectedChar(c, "; expected either NDATA keyword, or closing '>'");
- }
- String keyw = checkDTDKeyword("DATA");
- if (keyw != null) {
- _reportWFCViolation("Unrecognized keyword '"+keyw+"'; expected NOTATION (or closing '>')");
- }
- c = skipObligatoryDtdWs();
- notationId = readNotationEntry(c, null, evtLoc);
- c = skipDtdWs(true);
- }
- }
-
- // Ok, better have '>' now:
- if (c != '>') {
- throwDTDUnexpectedChar(c, "; expected closing '>'");
- }
-
- URL ctxt = inputSource.getSource();
- if (notationId == null) { // parsed entity:
- return new ParsedExtEntity(evtLoc, id, ctxt, pubId, sysId);
- }
- return new UnparsedExtEntity(evtLoc, id, ctxt, pubId, sysId, notationId);
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Data struct access
- ///////////////////////////////////////////////////////
- */
-
- private HashMap getElementMap() {
- HashMap m = mElements;
- if (m == null) {
- /* Let's try to get insert-ordered Map, to be able to
- * report redefinition problems in proper order when validating
- * subset compatibility
- */
- mElements = m = new LinkedHashMap();
- }
- return m;
- }
-
- final PrefixedName mAccessKey = new PrefixedName(null, null);
-
- /**
- * Method used to 'intern()' qualified names; main benefit is reduced
- * memory usage as the name objects are shared. May also slightly
- * speed up Map access, as more often identity comparisons catch
- * matches.
- *
- * Note: it is assumed at this point that access is only from a single
- * thread, and non-recursive -- generally valid assumption as readers are
- * not shared. Restriction is needed since the method is not re-entrant:
- * it uses mAccessKey during the method call.
- */
- private PrefixedName findSharedName(String prefix, String localName)
- {
- HashMap m = mSharedNames;
-
- if (mSharedNames == null) {
- mSharedNames = m = new HashMap();
- } else {
- // Maybe we already have a shared instance... ?
- PrefixedName key = mAccessKey;
- key.reset(prefix, localName);
- key = (PrefixedName) m.get(key);
- if (key != null) { // gotcha
- return key;
- }
- }
-
- // Not found; let's create, cache and return it:
- PrefixedName result = new PrefixedName(prefix, localName);
- m.put(result, result);
- return result;
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Implementations of abstract methods from StreamScanner
- //////////////////////////////////////////////////////////
- */
-
- // @Override
- /**
- * @param arg If Boolean.TRUE, we are expanding a general entity
- *
- */
- protected EntityDecl findEntity(String id, Object arg)
- {
- // Expand a Parameter Entity?
- if (arg == ENTITY_EXP_PE) { // for attribute default
- EntityDecl ed = (mPredefdPEs == null) ? null : (EntityDecl) mPredefdPEs.get(id);
- if (ed != null) { // Entity from int. subset...
- mUsesPredefdEntities = true;
- /* No need to further keep track of internal references,
- * since this subset can not be cached, so let's just free
- * up Map if it has been created
- */
- mRefdPEs = null;
- } else if (mParamEntities != null) {
- ed = (EntityDecl) mParamEntities.get(id);
- if (ed != null) {
- if (!mUsesPredefdEntities) {
- // Let's also mark down the fact we referenced the entity:
- Set used = mRefdPEs;
- if (used == null) {
- mRefdPEs = used = new HashSet();
- }
- used.add(id);
- }
- }
- }
- return ed;
- }
-
- // Nope, General Entity (within attribute default value)?
- if (arg == ENTITY_EXP_GE) {
- /* This is only complicated for external subsets, since
- * they may 'inherit' entity definitions from preceding
- * internal subset...
- */
- EntityDecl ed = (mPredefdGEs == null) ? null : (EntityDecl) mPredefdGEs.get(id);
- if (ed != null) {
- mUsesPredefdEntities = true;
- /* No need to further keep track of references,
- * as this means this subset is not cachable...
- * so let's just free up Map if it has been created
- */
- mRefdGEs = null;
- } else if (mGeneralEntities != null) {
- ed = (EntityDecl) mGeneralEntities.get(id);
- if (ed != null) {
- // Ok, just need to mark reference, if we still care:
- if (!mUsesPredefdEntities) {
- // Let's also mark down the fact we referenced the entity:
- if (mRefdGEs == null) {
- mRefdGEs = new HashSet();
- }
- mRefdGEs.add(id);
- }
- }
- }
- return ed;
- }
-
- throw new IllegalStateException(ErrorConsts.ERR_INTERNAL);
- }
-
- // @Override
- /**
- * Undeclared parameter entity is a VC, not WFC...
- */
- protected void handleUndeclaredEntity(String id)
- throws XMLStreamException
- {
- _reportVCViolation("Undeclared parameter entity '"+id+"'.");
- if (mCurrAttrDefault != null) {
- Location loc = getLastCharLocation();
- if (mExpandingPE) {
- mCurrAttrDefault.addUndeclaredPE(id, loc);
- } else {
- mCurrAttrDefault.addUndeclaredGE(id, loc);
- }
- }
- if (mEventListener != null) {
- // GEs only matter when expanding...
- if (mExpandingPE) {
- mEventListener.dtdSkippedEntity("%"+id);
- }
- }
- }
-
- // @Override
- /**
- * Handling of PE matching problems is actually intricate; one type
- * will be a WFC ("PE Between Declarations", which refers to PEs that
- * start from outside declarations), and another just a VC
- * ("Proper Declaration/PE Nesting", when PE is contained within
- * declaration)
- */
- protected void handleIncompleteEntityProblem(WstxInputSource closing)
- throws XMLStreamException
- {
- // Did it start outside of declaration?
- if (closing.getScopeId() == 0) { // yup
- // and being WFC, need not be validating
- _reportWFCViolation(entityDesc(closing) + ": "
- +"Incomplete PE: has to fully contain a declaration (as per xml 1.0.3, section 2.8, WFC 'PE Between Declarations')");
- } else {
- // whereas the other one is only sent in validating mode..
- if (mCfgFullyValidating) {
- _reportVCViolation(entityDesc(closing) + ": "
- +"Incomplete PE: has to be fully contained in a declaration (as per xml 1.0.3, section 2.8, VC 'Proper Declaration/PE Nesting')");
- }
- }
- }
-
- protected void handleGreedyEntityProblem(WstxInputSource input)
- throws XMLStreamException
- {
- // Here it can only be of VC kind...
- if (mCfgFullyValidating) { // since it's a VC, not WFC
- _reportWFCViolation(entityDesc(input) + ": " +
- "Unbalanced PE: has to be fully contained in a declaration (as per xml 1.0.3, section 2.8, VC 'Proper Declaration/PE Nesting')");
- }
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Additional validity checking
- ///////////////////////////////////////////////////////
- */
-
- protected void checkXmlSpaceAttr(int type, WordResolver enumValues)
- throws XMLStreamException
- {
- boolean ok = (type == DTDAttribute.TYPE_ENUMERATED);
- if (ok) {
- switch (enumValues.size()) {
- case 1:
- ok = (enumValues.find("preserve") != null)
- || (enumValues.find("default") != null);
- break;
- case 2:
- ok = (enumValues.find("preserve") != null)
- && (enumValues.find("default") != null);
- break;
- default:
- ok = false;
- }
- }
-
- if (!ok) {
- _reportVCViolation(ErrorConsts.ERR_DTD_XML_SPACE);
- }
- }
-
- protected void checkXmlIdAttr(int type)
- throws XMLStreamException
- {
- if (type != DTDAttribute.TYPE_ID) {
- _reportVCViolation(ErrorConsts.ERR_DTD_XML_ID);
- }
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Error handling
- ///////////////////////////////////////////////////////
- */
-
- private void _reportWarning(XMLReporter rep, String probType, String msg,
- Location loc)
- throws XMLStreamException
- {
- if (rep != null) {
- /* Note: since the problem occurs at DTD (schema) parsing,
- * not during validation, can not set reporter.
- */
- XMLValidationProblem prob = new XMLValidationProblem
- (loc, msg, XMLValidationProblem.SEVERITY_WARNING, probType);
- rep.report(msg, probType, prob, loc);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/LargePrefixedNameSet.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/LargePrefixedNameSet.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/LargePrefixedNameSet.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/LargePrefixedNameSet.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,184 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.Iterator;
-import java.util.TreeSet;
-
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Implementation of {@link PrefixedNameSet} suitable for storing large number
- * of entries; basically anything above trivially small sets (4 or less).
- *
- * Notes about usage:
- *
- * 03-Dec-2004, TSa: This is not particularly elegant: should be
- * able to pass the information some other way. But for now it
- * works and is necessary.
- */
- public EntityDecl findEntity(String entName) {
- return null;
- }
-
- /*
- //////////////////////////////////////////////////
- // Main-level skipping method(s)
- //////////////////////////////////////////////////
- */
-
- /**
- * Method that will skip through internal DTD subset, without doing
- * any parsing, except for trying to match end of subset properly.
- */
- protected void skipInternalSubset()
- throws XMLStreamException
- {
- while (true) {
- int i = getNextAfterWS();
- if (i < 0) {
- // Error for internal subset
- throwUnexpectedEOF(SUFFIX_IN_DTD_INTERNAL);
- }
- if (i == '%') { // parameter entity
- skipPE();
- continue;
- }
- if (i == '<') {
- /* Let's determine type here, and call appropriate skip
- * methods.
- */
- char c = getNextSkippingPEs();
- if (c == '?') { // xml decl?
- /* Not sure if PIs are really allowed in DTDs, but let's
- * just allow them until proven otherwise. XML declaration
- * is legal in the beginning, anyhow
- */
- skipPI();
- } else if (c == '!') { // ignore/include, comment, declaration?
- c = getNextSkippingPEs();
- if (c == '[') {
- /* Shouldn't really get these, as they are not allowed
- * in the internal subset? So let's just leave it
- * as is, and see what happens. :-)
- */
- ;
- } else if (c == '-') { // plain comment
- skipComment();
- } else if (c >= 'A' && c <= 'Z') {
- skipDeclaration(c);
- } else {
- /* Hmmh, let's not care too much; but we need to try
- * to match the closing gt-char nonetheless?
- */
- skipDeclaration(c);
- }
- } else {
- /* Shouldn't fail (since we are to completely ignore
- * subset); let's just push it back and continue.
- */
- --mInputPtr;
- }
- continue;
- }
-
- if (i == ']') {
- // Int. subset has no conditional sections, has to be the end...
- /* 18-Jul-2004, TSa: Let's just make sure it happened
- * in the main input source, not at external entity...
- */
- if (mInput != mRootInput) {
- throwParseError("Encountered int. subset end marker ']]>' in an expanded entity; has to be at main level.");
- }
- // End of internal subset
- break;
- }
- throwUnexpectedChar(i, SUFFIX_IN_DTD_INTERNAL+"; expected a '<' to start a directive, or \"]>\" to end internal subset.");
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, input access:
- //////////////////////////////////////////////////
- */
-
- protected char dtdNextFromCurr()
- throws XMLStreamException
- {
- return (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(getErrorMsg());
- }
-
- protected char dtdNextChar()
- throws XMLStreamException
- {
- return (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- }
-
- protected char getNextSkippingPEs()
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- if (c != '%') {
- return c;
- }
- skipPE();
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, skipping:
- //////////////////////////////////////////////////
- */
-
- private void skipPE()
- throws XMLStreamException
- {
- skipDTDName();
- /* Should now get semicolon... let's try to find and skip it; but
- * if none found, let's not throw an exception -- we are just skipping
- * internal subset here.
- */
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c != ';') {
- --mInputPtr;
- }
- }
-
- protected void skipComment()
- throws XMLStreamException
- {
- skipCommentContent();
- // Now, we may be getting end mark; first need second marker char:.
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c != '>') {
- throwParseError("String '--' not allowed in comment (missing '>'?)");
- }
- }
-
- protected void skipCommentContent()
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '-') {
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '-') {
- return;
- }
- } else if (c == '\n' || c == '\r') {
- skipCRLF(c);
- }
- }
- }
-
- protected void skipPI()
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '?') {
- do {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- } while (c == '?');
- if (c == '>') {
- break;
- }
- }
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- }
- }
- }
-
- private void skipDeclaration(char c)
- throws XMLStreamException
- {
- while (c != '>') {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- /* No need for specific handling for PE refs; they just have
- * identifier that'll get properly skipped.
- */
- /* 17-Jul-2004, TSa: But we do need to properly handle literals;
- * it is possible to add '>' char in entity expansion values.
- */
- } else if (c == '\'' || c == '"') {
- skipLiteral(c);
- }
- }
- }
-
- private void skipLiteral(char quoteChar)
- throws XMLStreamException
- {
- while (true) {
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c == quoteChar) {
- break;
- }
- /* No need for specific handling for PE refs, should be ignored
- * just ok (plus they need to properly nested in any case)
- */
- }
- }
-
- private void skipDTDName()
- throws XMLStreamException
- {
- /*int len =*/ skipFullName(getNextChar(getErrorMsg()));
- /* Should we give an error about missing name? For now,
- * let's just exit.
- */
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, error handling:
- //////////////////////////////////////////////////
- */
-
- protected String getErrorMsg() {
- return mIsExternal ? SUFFIX_IN_DTD_EXTERNAL : SUFFIX_IN_DTD_INTERNAL;
- }
-
-
- protected void throwIllegalCall()
- throws Error
- {
- throw new IllegalStateException("Internal error: this method should never be called");
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ModelNode.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ModelNode.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/ModelNode.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/ModelNode.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,35 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.BitSet;
-import java.util.List;
-
-/**
- * Abstract base class for classes constructed from {@link ContentSpec}
- * objects, when they get rewritten (when their {@link ContentSpec#rewrite}
- * gets called). These nodes are then used for constructing complete DFA
- * states for validation.
- */
-public abstract class ModelNode
-{
- /*
- ///////////////////////////////////////////////////
- // Methods needed for DFA construction
- ///////////////////////////////////////////////////
- */
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public abstract ModelNode cloneModel();
-
- public abstract boolean isNullable();
-
- public abstract void indexTokens(List tokens);
-
- public abstract void addFirstPos(BitSet firstPos);
-
- public abstract void addLastPos(BitSet firstPos);
-
- public abstract void calcFollowPos(BitSet[] followPosSets);
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/OptionalModel.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/OptionalModel.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/OptionalModel.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/OptionalModel.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,65 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-/**
- * Content specification class that represents an optional specification.
- * Optional specifications are generally a result of '?' arity marker,
- * and are created when {@link ContentSpec#rewrite} is called
- * on a specification with '?' arity modifier.
- */
-public class OptionalModel
- extends ModelNode
-{
- ModelNode mModel;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public OptionalModel(ModelNode model) {
- mModel = model;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public ModelNode cloneModel() {
- return new OptionalModel(mModel.cloneModel());
- }
-
- public boolean isNullable() {
- return true;
- }
-
- public void indexTokens(List tokens) {
- mModel.indexTokens(tokens);
- }
-
- public void addFirstPos(BitSet pos) {
- mModel.addFirstPos(pos);
- }
-
- public void addLastPos(BitSet pos) {
- mModel.addLastPos(pos);
- }
-
- public void calcFollowPos(BitSet[] followPosSets)
- {
- // Let's let sub-model do its stuff
- mModel.calcFollowPos(followPosSets);
- }
-
- public String toString() {
- return mModel + "[?]";
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,3 +0,0 @@
-
- * Notes about usage:
- *
- * Note: this is not really safe, as caller can modify the array, but
- * since this method is thought to provide fast access, let's avoid making
- * copy here.
- */
- public char[] getReplacementChars() {
- return mRepl;
- }
-
- // // // Type information
-
- public boolean isExternal() { return false; }
-
- public boolean isParsed() { return true; }
-
- public WstxInputSource expand(WstxInputSource parent,
- XMLResolver res, ReaderConfig cfg,
- int xmlVersion)
- {
- /* 26-Dec-2006, TSa: Better leave source as null, since internal
- * entity declaration context should never be used: when expanding,
- * reference context is to be used.
- */
- return InputSourceFactory.constructCharArraySource
- //(parent, mName, mRepl, 0, mRepl.length, mContentLocation, getSource());
- (parent, mName, mRepl, 0, mRepl.length, mContentLocation, null);
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,5 +0,0 @@
-
- * One of few complications here is the way start elements are constructed.
- * The pattern used is double-indirection, needed to get a callback from
- * the stream reader, with data we need for constructing even Object...
- * but without stream reader having any understanding of event Objects
- * per se.
- *
- * 03-Dec-2004, TSa: One additional twist is that it's now possible to
- * create slightly faster event handling, by indicating that the
- * fully accurate Location information is not necessary. If so,
- * allocator will just use one shared Location object passed to
- * all event objects constructed.
- */
-public class DefaultEventAllocator
- extends ElemCallback
- implements XMLEventAllocator, XMLStreamConstants
-{
- final static DefaultEventAllocator sStdInstance = new DefaultEventAllocator(true);
-
- /*
- ////////////////////////////////////////
- // Configuration
- ////////////////////////////////////////
- */
-
- protected final boolean mAccurateLocation;
-
- /*
- ////////////////////////////////////////
- // Recycled objects
- ////////////////////////////////////////
- */
-
- /**
- * Last used location info; only relevant to non-accurate-location
- * allocators.
- */
- protected Location mLastLocation = null;
-
- /**
- * @param accurateLocation If true, allocator will construct instances
- * that have accurate location information; if false, instances
- * will only have some generic shared Location info. Latter option
- * will reduce memory usage/thrashing a bit, and may improve speed.
- */
- protected DefaultEventAllocator(boolean accurateLocation) {
- mAccurateLocation = accurateLocation;
- }
-
- public static DefaultEventAllocator getDefaultInstance() {
- /* Default (accurate location) instance can be shared as it
- * has no state
- */
- return sStdInstance;
- }
-
- public static DefaultEventAllocator getFastInstance() {
- /* Can not share instances, due to QName caching, as well as because
- * of Location object related state
- */
- return new DefaultEventAllocator(false);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // XMLEventAllocator implementation
- //////////////////////////////////////////////////////////
- */
-
- public XMLEvent allocate(XMLStreamReader r)
- throws XMLStreamException
- {
- Location loc;
-
- // Need to keep track of accurate location info?
- if (mAccurateLocation) {
- loc = r.getLocation();
- } else {
- loc = mLastLocation;
- /* And even if we can just share one instance, we need that
- * first instance...
- */
- if (loc == null) {
- loc = mLastLocation = r.getLocation();
- }
- }
-
- switch (r.getEventType()) {
- case CDATA:
- return new CharactersEventImpl(loc, r.getText(), true);
- case CHARACTERS:
- return new CharactersEventImpl(loc, r.getText(), false);
- case COMMENT:
- return new CommentEventImpl(loc, r.getText());
- case DTD:
- // Not sure if we really need this defensive coding but...
- if (r instanceof XMLStreamReader2) {
- XMLStreamReader2 sr2 = (XMLStreamReader2) r;
- DTDInfo dtd = sr2.getDTDInfo();
- return new WDTD(loc,
- dtd.getDTDRootName(),
- dtd.getDTDSystemId(), dtd.getDTDPublicId(),
- dtd.getDTDInternalSubset(),
- (DTDSubset) dtd.getProcessedDTD());
- }
- /* No way to get all information... the real big problem is
- * that of how to access root name: it's obligatory for
- * DOCTYPE construct. :-/
- */
- return new WDTD(loc, null, r.getText());
-
- case END_DOCUMENT:
- return new EndDocumentEventImpl(loc);
-
- case END_ELEMENT:
- return new EndElementEventImpl(loc, r);
-
- case PROCESSING_INSTRUCTION:
- return new ProcInstrEventImpl(loc, r.getPITarget(), r.getPIData());
- case SPACE:
- {
- CharactersEventImpl ch = new CharactersEventImpl(loc, r.getText(), false);
- ch.setWhitespaceStatus(true);
- return ch;
- }
- case START_DOCUMENT:
- return new StartDocumentEventImpl(loc, r);
-
- case START_ELEMENT:
- {
- /* Creating the event is bit complicated, as the stream
- * reader is not to know anything about event objects.
- * To do this, we do double-indirection, which means that
- * this object actually gets a callback:
- */
- /* 19-Jul-2006, TSa: WSTX-61 points out that the code was
- * assuming it's always Woodstox reader we had... not
- * necessarily so.
- */
- if (r instanceof StreamReaderImpl) {
- StreamReaderImpl sr = (StreamReaderImpl) r;
- BaseStartElement be = (BaseStartElement) sr.withStartElement(this, loc);
- if (be == null) { // incorrect state
- throw new WstxException("Trying to create START_ELEMENT when current event is "
- +ErrorConsts.tokenTypeDesc(sr.getEventType()),
- loc);
- }
- return be;
- }
- /* Ok, not woodstox impl, will be bit more work (plus less
- * efficient, and may miss some info)... but can be done.
- */
- NamespaceContext nsCtxt = null;
- if (r instanceof XMLStreamReader2) {
- nsCtxt = ((XMLStreamReader2) r).getNonTransientNamespaceContext();
- }
- Map attrs;
- {
- int attrCount = r.getAttributeCount();
- if (attrCount < 1) {
- attrs = null;
- } else {
- attrs = new LinkedHashMap();
- for (int i = 0; i < attrCount; ++i) {
- QName aname = r.getAttributeName(i);
- attrs.put(aname, new AttributeEventImpl(loc, aname, r.getAttributeValue(i), r.isAttributeSpecified(i)));
- }
- }
- }
- List ns;
- {
- int nsCount = r.getNamespaceCount();
- if (nsCount < 1) {
- ns = null;
- } else {
- ns = new ArrayList(nsCount);
- for (int i = 0; i < nsCount; ++i) {
- ns.add(NamespaceEventImpl.constructNamespace(loc, r.getNamespacePrefix(i), r.getNamespaceURI(i)));
- }
- }
- }
-
- return SimpleStartElement.construct(loc, r.getName(), attrs, ns, nsCtxt);
- }
-
- case ENTITY_REFERENCE:
- {
- /* 19-Jul-2006, TSa: Let's also allow other impls, although
- * we can't get actual declaration if so...
- */
- if (r instanceof StreamReaderImpl) {
- EntityDecl ed = ((StreamReaderImpl) r).getCurrentEntityDecl();
- if (ed == null) { // undefined?
- // We'll still know the name though...
- return new WEntityReference(loc, r.getLocalName());
- }
- return new WEntityReference(loc, ed);
- }
- return new WEntityReference(loc, r.getLocalName());
- }
-
-
- /* Following 2 types should never get in here; they are directly
- * handled by DTDReader, and can only be accessed via DTD event
- * element.
- */
- case ENTITY_DECLARATION:
- case NOTATION_DECLARATION:
- /* Following 2 types should never get in here; they are directly
- * handled by the reader, and can only be accessed via start
- * element.
- */
- case NAMESPACE:
- case ATTRIBUTE:
- throw new WstxException("Internal error: should not get "
- +ErrorConsts.tokenTypeDesc(r.getEventType()));
- default:
- throw new IllegalStateException("Unrecognized event type "+r.getEventType()+".");
- }
- }
-
- public void allocate(XMLStreamReader r, XMLEventConsumer consumer)
- throws XMLStreamException
- {
- consumer.add(allocate(r));
- }
-
- public XMLEventAllocator newInstance() {
- return new DefaultEventAllocator(mAccurateLocation);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // ElemCallback implementation
- //////////////////////////////////////////////////////////
- */
-
- public Object withStartElement(Location loc, QName name,
- BaseNsContext nsCtxt, ElemAttrs attrs,
- boolean wasEmpty)
- {
- return new CompactStartElement(loc, name, nsCtxt, attrs);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/MergedNsContext.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/MergedNsContext.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/MergedNsContext.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/MergedNsContext.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,212 +0,0 @@
-package com.ctc.wstx.evt;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.util.*;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.stream.events.Namespace;
-
-import org.codehaus.stax2.ri.EmptyIterator;
-
-import com.ctc.wstx.util.BaseNsContext;
-
-/**
- * Hierarchic {@link NamespaceContext} implementation used when constructing
- * event and namespace information explicitly via
- * {@link javax.xml.stream.XMLEventFactory},
- * not by a stream reader.
- *
- * TODO:
- *
- * The current implementation does not add much beyond basic
- * {@link XMLValidationException}, except for fixing some problems that
- * underlying {@link XMLStreamException} has.
- *
- * Note that some of the code is shared with {@link WstxException}. Unfortunately
- * it is not possible to extend it, however, since it extends basic
- * {@link XMLStreamException}, not {@link XMLValidationException}.
- *
- * One more thing to note: unlike some other exception classes, these
- * exceptions do not have chained root causes. That's why no special
- * handling is necessary for setting the root cause in backwards compatible
- * way.
- */
-public class WstxValidationException
- extends XMLValidationException
-{
- private static final long serialVersionUID = 1L;
-
- protected WstxValidationException(XMLValidationProblem cause, String msg)
- {
- super(cause, msg);
- }
-
- protected WstxValidationException(XMLValidationProblem cause, String msg,
- Location loc)
- {
- super(cause, msg, loc);
- }
-
- public static WstxValidationException create(XMLValidationProblem cause)
- {
- // Should always get a message
- Location loc = cause.getLocation();
- if (loc == null) {
- return new WstxValidationException(cause, cause.getMessage());
- }
- return new WstxValidationException(cause, cause.getMessage(), loc);
- }
-
- /*
- /////////////////////////////////////////////////////////
- // Overridden methods from XMLStreamException
- /////////////////////////////////////////////////////////
- */
-
- /**
- * Method is overridden for two main reasons: first, default method
- * does not display public/system id information, even if it exists, and
- * second, default implementation can not handle nested Location
- * information.
- */
- public String getMessage()
- {
- String locMsg = getLocationDesc();
- /* Better not use super's message if we do have location information,
- * since parent's message contains (part of) Location
- * info; something we can regenerate better...
- */
- if (locMsg == null) {
- return super.getMessage();
- }
- String msg = getValidationProblem().getMessage();
- StringBuffer sb = new StringBuffer(msg.length() + locMsg.length() + 20);
- sb.append(msg);
- StringUtil.appendLF(sb);
- sb.append(" at ");
- sb.append(locMsg);
- return sb.toString();
- }
-
- public String toString()
- {
- return getClass().getName()+": "+getMessage();
- }
-
- /*
- ////////////////////////////////////////////////////////
- // Internal methods:
- ////////////////////////////////////////////////////////
- */
-
- protected String getLocationDesc()
- {
- Location loc = getLocation();
- return (loc == null) ? null : loc.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/AsciiReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/AsciiReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/AsciiReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/AsciiReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,136 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.XmlConsts;
-
-/**
- * Optimized Reader that reads ascii content from an input stream.
- * In addition to doing (hopefully) optimal conversion, it can also take
- * array of "pre-read" (leftover) bytes; this is necessary when preliminary
- * stream/reader is trying to figure out XML encoding.
- */
-public final class AsciiReader
- extends BaseReader
-{
- boolean mXml11 = false;
-
- /**
- * Total read character count; used for error reporting purposes
- * (note: byte count is the same, due to fixed one-byte-per char mapping)
- */
- int mCharCount = 0;
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- public AsciiReader(ReaderConfig cfg, InputStream in, byte[] buf, int ptr, int len,
- boolean recycleBuffer)
- {
- super(cfg, in, buf, ptr, len, recycleBuffer);
- }
-
- public void setXmlCompliancy(int xmlVersion)
- {
- mXml11 = (xmlVersion == XmlConsts.XML_V_11);
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- public int read(char[] cbuf, int start, int len)
- throws IOException
- {
- // Let's first ensure there's enough room...
- if (start < 0 || (start+len) > cbuf.length) {
- reportBounds(cbuf, start, len);
- }
- // Already EOF?
- if (mByteBuffer == null) {
- return -1;
- }
- if (len < 1) {
- return 0;
- }
-
- // Need to load more data?
- int avail = mByteBufferEnd - mBytePtr;
- if (avail <= 0) {
- mCharCount += mByteBufferEnd;
- // Let's always try to read full buffers, actually...
- int count = readBytes();
- if (count <= 0) {
- if (count == 0) {
- reportStrangeStream();
- }
- /* Let's actually then free the buffer right away; shouldn't
- * yet close the underlying stream though?
- */
- freeBuffers(); // to help GC?
- return -1;
- }
- avail = count;
- }
-
- // K, have at least one byte == char, good enough:
-
- if (len > avail) {
- len = avail;
- }
- int i = mBytePtr;
- int last = i + len;
-
- for (; i < last; ) {
- char c = (char) mByteBuffer[i++];
- if (c >= CHAR_DEL) {
- if (c > CHAR_DEL) {
- reportInvalidAscii(c);
- } else {
- if (mXml11) {
- int pos = mCharCount + mBytePtr;
- reportInvalidXml11(c, pos, pos);
- }
- }
- }
- cbuf[start++] = c;
- }
-
- mBytePtr = last;
- return len;
- }
-
- /*
- ////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////
- */
-
- private void reportInvalidAscii(char c)
- throws IOException
- {
- throw new CharConversionException("Invalid ascii byte; value above 7-bit ascii range ("+((int) c)+"; at pos #"+(mCharCount + mBytePtr)+")");
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BaseInputSource.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BaseInputSource.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BaseInputSource.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BaseInputSource.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,173 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.IOException;
-import java.net.URL;
-
-import javax.xml.stream.XMLStreamException;
-
-/**
- * Abstract base class that implements shared functionality that all current
- * {@link WstxInputSource} implementations Woodstox includes need.
- */
-public abstract class BaseInputSource
- extends WstxInputSource
-{
- final String mPublicId;
-
- final String mSystemId;
-
- /**
- * URL that points to original source of input, if known; null if not
- * known (source constructed with just a stream or reader). Used for
- * resolving references from the input that's read from this source.
- * Can be overridden by reader; done if P_BASE_URL is changed on
- * stream reader for which this input source is the currently
- * active input source.
- */
- protected URL mSource;
-
- /**
- * Input buffer this input source uses, if any.
- */
- protected char[] mBuffer;
-
- /**
- * Length of the buffer, if buffer used
- */
- protected int mInputLast;
-
- /*
- ////////////////////////////////////////////////////////////////
- // Saved location information; active information is directly
- // handled by Reader, and then saved to input source when switching
- // to a nested input source.
- ////////////////////////////////////////////////////////////////
- */
-
- /**
- * Number of characters read from the current input source prior to
- * the current buffer
- */
- long mSavedInputProcessed = 0;
-
- int mSavedInputRow = 1;
- int mSavedInputRowStart = 0;
-
- int mSavedInputPtr = 0;
-
- /*
- ////////////////////////////////////////////////////////////////
- // Some simple lazy-loading/reusing support...
- ////////////////////////////////////////////////////////////////
- */
-
- transient WstxInputLocation mParentLocation = null;
-
- /*
- ////////////////////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////////////////////
- */
- protected BaseInputSource(WstxInputSource parent, String fromEntity,
- String publicId, String systemId, URL src)
- {
- super(parent, fromEntity);
- mSystemId = systemId;
- mPublicId = publicId;
- mSource = src;
- }
-
- public void overrideSource(URL src)
- {
- mSource = src;
- }
-
- public abstract boolean fromInternalEntity();
-
- public URL getSource() {
- return mSource;
- }
-
- public String getPublicId() {
- return mPublicId;
- }
-
- public String getSystemId() {
- return mSystemId;
- }
-
- protected abstract void doInitInputLocation(WstxInputData reader);
-
- public abstract int readInto(WstxInputData reader)
- throws IOException, XMLStreamException;
-
- public abstract boolean readMore(WstxInputData reader, int minAmount)
- throws IOException, XMLStreamException;
-
- public void saveContext(WstxInputData reader)
- {
- // First actual input data
- mSavedInputPtr = reader.mInputPtr;
-
- // then location
- mSavedInputProcessed = reader.mCurrInputProcessed;
- mSavedInputRow = reader.mCurrInputRow;
- mSavedInputRowStart = reader.mCurrInputRowStart;
- }
-
- public void restoreContext(WstxInputData reader)
- {
- reader.mInputBuffer = mBuffer;
- reader.mInputEnd = mInputLast;
- reader.mInputPtr = mSavedInputPtr;
-
- // then location
- reader.mCurrInputProcessed = mSavedInputProcessed;
- reader.mCurrInputRow = mSavedInputRow;
- reader.mCurrInputRowStart = mSavedInputRowStart;
- }
-
- public abstract void close() throws IOException;
-
- /*
- //////////////////////////////////////////////////////////
- // Methods for accessing location information
- //////////////////////////////////////////////////////////
- */
-
- /**
- * This method only gets called by the 'child' input source (for example,
- * contents of an expanded entity), to get the enclosing context location.
- */
- protected final WstxInputLocation getLocation()
- {
- // Note: columns are 1-based, need to add 1.
- return getLocation(mSavedInputProcessed + mSavedInputPtr - 1L,
- mSavedInputRow,
- mSavedInputPtr - mSavedInputRowStart + 1);
- }
-
- public final WstxInputLocation getLocation(long total, int row, int col)
- {
- WstxInputLocation pl;
-
- if (mParent == null) {
- pl = null;
- } else {
- /* 13-Apr-2005, TSa: We can actually reuse parent location, since
- * it can not change during lifetime of this child context...
- */
- pl = mParentLocation;
- if (pl == null) {
- mParentLocation = pl = mParent.getLocation();
- }
- pl = mParent.getLocation();
- }
- /* !!! 15-Apr-2005, TSa: This will cause overflow for total count,
- * but since StAX 1.0 API doesn't have any way to deal with that,
- * let's just let that be...
- */
- return new WstxInputLocation(pl, getPublicId(), getSystemId(),
- (int) total, row, col);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BaseReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BaseReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BaseReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BaseReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,227 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-
-/**
- * Simple basic class for optimized Readers Woodstox has; implements
- * "cookie-cutter" methods that are used by all actual implementations.
- */
-abstract class BaseReader
- extends Reader
-{
- protected final static char NULL_CHAR = (char) 0;
- protected final static char NULL_BYTE = (byte) 0;
-
- /**
- * In xml 1.1, NEL (0x85) behaves much the way \n does (can
- * be follow \r as part of the linefeed
- */
- protected final static char CONVERT_NEL_TO = '\n';
-
- /**
- * In xml 1.1, LSEP bit like \n, or \r. Need to choose one as the
- * result. Let's use \n, for simplicity
- */
- protected final static char CONVERT_LSEP_TO = '\n';
-
- /**
- * DEL character is both the last ascii char, and illegal in xml 1.1.
- */
- final static char CHAR_DEL = (char) 127;
-
- protected final ReaderConfig mConfig;
-
- private InputStream mIn;
-
- protected byte[] mByteBuffer;
-
- /**
- * Pointer to the next available byte (if any), iff less than
- *
- * Currently this Reader is only used as the main-level Reader, to allow for
- * branching of internal DTD subset to a text buffer if necessary.
- */
-public final class BranchingReaderSource
- extends ReaderSource
-{
-
- // // // Branching information
-
- TextBuffer mBranchBuffer = null;
-
- int mBranchStartOffset = 0;
-
- boolean mConvertLFs = false;
-
- /**
- * Flag that indicates that last char from previous buffer was
- * '\r', and that following '\n' (if there is one) needs to be
- * ignored.
- */
- boolean mGotCR = false;
-
- public BranchingReaderSource(ReaderConfig cfg, String pubId, String sysId, URL src,
- Reader r, boolean realClose)
- {
- /* null -> no parent,
- * null -> not from explicit entity (no id/name)
- */
- super(cfg, null, null, pubId, sysId, src, r, realClose);
- }
-
- public int readInto(WstxInputData reader)
- throws IOException, XMLStreamException
- {
- // Need to flush out branched content?
- if (mBranchBuffer != null) {
- if (mInputLast > mBranchStartOffset) {
- appendBranched(mBranchStartOffset, mInputLast);
- }
- mBranchStartOffset = 0;
- }
- return super.readInto(reader);
- }
-
- public boolean readMore(WstxInputData reader, int minAmount)
- throws IOException, XMLStreamException
- {
- // Existing data to output to branch?
- if (mBranchBuffer != null) {
- int ptr = reader.mInputPtr;
- int currAmount = mInputLast - ptr;
- if (currAmount > 0) {
- if (ptr > mBranchStartOffset) {
- appendBranched(mBranchStartOffset, ptr);
- }
- mBranchStartOffset = 0;
- }
- }
- return super.readMore(reader, minAmount);
- }
-
- /*
- //////////////////////////////////////////////////
- // Branching methods; used mostly to make a copy
- // of parsed internal subsets.
- //////////////////////////////////////////////////
- */
-
- public void startBranch(TextBuffer tb, int startOffset,
- boolean convertLFs)
- {
- mBranchBuffer = tb;
- mBranchStartOffset = startOffset;
- mConvertLFs = convertLFs;
- mGotCR = false;
- }
-
- /**
- * Currently this input source does not implement branching
- */
- public void endBranch(int endOffset)
- {
- if (mBranchBuffer != null) {
- if (endOffset > mBranchStartOffset) {
- appendBranched(mBranchStartOffset, endOffset);
- }
- // Let's also make sure no branching is done from this point on:
- mBranchBuffer = null;
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods
- //////////////////////////////////////////////////
- */
-
- private void appendBranched(int startOffset, int pastEnd) {
- // Main tricky thing here is just replacing of linefeeds...
- if (mConvertLFs) {
- char[] inBuf = mBuffer;
- /* this will also unshare() and ensure there's room for at
- * least one more char
- */
- char[] outBuf = mBranchBuffer.getCurrentSegment();
- int outPtr = mBranchBuffer.getCurrentSegmentSize();
-
- // Pending \n to skip?
- if (mGotCR) {
- if (inBuf[startOffset] == '\n') {
- ++startOffset;
- }
- }
-
- while (startOffset < pastEnd) {
- char c = inBuf[startOffset++];
- if (c == '\r') {
- if (startOffset < pastEnd) {
- if (inBuf[startOffset] == '\n') {
- ++startOffset;
- }
- } else {
- mGotCR = true;
- }
- c = '\n';
- }
-
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = mBranchBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
-
- mBranchBuffer.setCurrentLength(outPtr);
- } else {
- mBranchBuffer.append(mBuffer, startOffset, pastEnd-startOffset);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BufferRecycler.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BufferRecycler.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BufferRecycler.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BufferRecycler.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,108 +0,0 @@
-package com.ctc.wstx.io;
-
-/**
- * This is a small utility class, whose main functionality is to allow
- * simple reuse of raw byte/char buffers. It is usually used through
- *
- * Regarding implementation: the key design goal is simplicity; and to
- * that end, different types of buffers are handled separately. While
- * code may look inelegant as a result (wouldn't it be neat to just
- * have generic char[]/byte[] buffer accessors?), benefit is that
- * no data structures are needed, just simple references. As long
- * as usage pattern is well known (which it is, for stream readers)
- * this should be highly optimal and robust implementation.
- */
-public final class BufferRecycler
-{
- private char[] mSmallCBuffer = null; // temp buffers
- private char[] mMediumCBuffer = null; // text collector
- private char[] mFullCBuffer = null; // for actual parsing buffer
-
- private byte[] mFullBBuffer = null;
-
- public BufferRecycler() { }
-
- // // // Char buffers:
-
- // // Small buffers, for temporary parsing
-
- public char[] getSmallCBuffer(int minSize)
- {
- char[] result = null;
- if (mSmallCBuffer != null && mSmallCBuffer.length >= minSize) {
- result = mSmallCBuffer;
- mSmallCBuffer = null;
- }
-//System.err.println("DEBUG: Alloc CSmall: "+result);
- return result;
- }
-
- public void returnSmallCBuffer(char[] buffer)
- {
-//System.err.println("DEBUG: Return CSmall ("+buffer.length+"): "+buffer);
- mSmallCBuffer = buffer;
- }
-
- // // Medium buffers, for text output collection
-
- public char[] getMediumCBuffer(int minSize)
- {
- char[] result = null;
- if (mMediumCBuffer != null && mMediumCBuffer.length >= minSize) {
- result = mMediumCBuffer;
- mMediumCBuffer = null;
- }
-//System.err.println("DEBUG: Alloc CMed: "+result);
- return result;
- }
-
- public void returnMediumCBuffer(char[] buffer)
- {
- mMediumCBuffer = buffer;
-//System.err.println("DEBUG: Return CMed ("+buffer.length+"): "+buffer);
- }
-
- // // Full buffers, for parser buffering
-
- public char[] getFullCBuffer(int minSize)
- {
- char[] result = null;
- if (mFullCBuffer != null && mFullCBuffer.length >= minSize) {
- result = mFullCBuffer;
- mFullCBuffer = null;
- }
-//System.err.println("DEBUG: Alloc CFull: "+result);
- return result;
- }
-
- public void returnFullCBuffer(char[] buffer)
- {
- mFullCBuffer = buffer;
-//System.err.println("DEBUG: Return CFull ("+buffer.length+"): "+buffer);
- }
-
- // // // Byte buffers:
-
- // // Full byte buffers, for byte->char conversion (Readers)
-
- public byte[] getFullBBuffer(int minSize)
- {
- byte[] result = null;
- if (mFullBBuffer != null && mFullBBuffer.length >= minSize) {
- result = mFullBBuffer;
- mFullBBuffer = null;
- }
-//System.err.println("DEBUG: Alloc BFull: "+result);
- return result;
- }
-
- public void returnFullBBuffer(byte[] buffer)
- {
- mFullBBuffer = buffer;
-//System.err.println("DEBUG: Return BFull ("+buffer.length+"): "+buffer);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CharArraySource.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CharArraySource.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CharArraySource.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CharArraySource.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,107 +0,0 @@
-package com.ctc.wstx.io;
-
-import javax.xml.stream.Location;
-
-import java.net.URL;
-
-/**
- * Input source that reads input from a static char array, usually used
- * when expanding internal entities. It can also be used if the input
- * given is a raw character array.
- */
-public final class CharArraySource
- extends BaseInputSource
-{
- int mOffset;
-
- // // // Plus, location offset info:
-
- final Location mContentStart;
-
- CharArraySource(WstxInputSource parent, String fromEntity,
- char[] chars, int offset, int len,
- Location loc, URL src)
- {
- super(parent, fromEntity, loc.getPublicId(), loc.getSystemId(), src);
- mBuffer = chars;
- mOffset = offset;
- mInputLast = offset + len;
- mContentStart = loc;
- }
-
- /**
- * This is a hard-coded assumption, but yes, for now this source is
- * only created from internal entities.
- */
- public boolean fromInternalEntity() {
- return true;
- }
-
- /**
- * Unlike with reader source, we won't start from beginning of a file,
- * but usually from somewhere in the middle...
- */
- protected void doInitInputLocation(WstxInputData reader)
- {
- reader.mCurrInputProcessed = mContentStart.getCharacterOffset();
- reader.mCurrInputRow = mContentStart.getLineNumber();
- /* 13-Apr-2005, TSa: Since column offsets reported by Location
- * objects are 1-based, but internally we use 0-based counts,
- * need to offset this start by 1 to begin with.
- */
- reader.mCurrInputRowStart = -mContentStart.getColumnNumber() + 1;
- }
-
- public int readInto(WstxInputData reader)
- {
- /* Shouldn't really try to read after closing, but it may be easier
- * for caller not to have to keep track of open/close state...
- */
- if (mBuffer == null) {
- return -1;
- }
-
- /* In general, there are only 2 states; either this has been
- * read or not. Offset is used as the marker; plus, in case
- * somehow we get a dummy char source (length of 0), it'll
- * also prevent any reading.
- */
- int len = mInputLast - mOffset;
- if (len < 1) {
- return -1;
- }
- reader.mInputBuffer = mBuffer;
- reader.mInputPtr = mOffset;
- reader.mInputEnd = mInputLast;
- // Also, need to note the fact we're done
- mOffset = mInputLast;
- return len;
- }
-
- public boolean readMore(WstxInputData reader, int minAmount)
- {
- /* Only case where this may work is if we haven't yet been
- * read from at all. And that should mean caller also has no
- * existing input...
- */
- if (reader.mInputPtr >= reader.mInputEnd) {
- int len = mInputLast - mOffset;
- if (len >= minAmount) {
- return (readInto(reader) > 0);
- }
- }
- return false;
- }
-
- public void close()
- {
- /* Let's help GC a bit, in case there might be back references
- * to this Object from somewhere...
- */
- mBuffer = null;
- }
-
- public void closeCompletely() {
- close();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CharsetNames.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CharsetNames.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CharsetNames.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CharsetNames.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,294 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.OutputStreamWriter;
-import java.io.Writer;
-
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Simple utility class that normalizes given character input character
- * set names into canonical (within Woodstox, anyways) names.
- */
-public final class CharsetNames
-{
- /*
- //////////////////////////////////////////////////
- // Canonical names used internally
- //////////////////////////////////////////////////
- */
-
- // // // Unicode variants:
-
- public final static String CS_US_ASCII = "US-ASCII";
- public final static String CS_UTF8 = "UTF-8";
-
- /**
- * This constants is intentionally vague, so that some other information
- * will be needed to determine the endianness.
- */
- public final static String CS_UTF16 = "UTF-16";
-
- public final static String CS_UTF16BE = "UTF-16BE";
- public final static String CS_UTF16LE = "UTF-16LE";
- public final static String CS_UTF32 = "UTF-32";
- public final static String CS_UTF32BE = "UTF-32BE";
- public final static String CS_UTF32LE = "UTF-32LE";
-
- // // // 8-bit ISO encodings:
-
- public final static String CS_ISO_LATIN1 = "ISO-8859-1";
-
- // // // Japanese non-unicode encodings:
-
- public final static String CS_SHIFT_JIS = "Shift_JIS";
-
- // // // Other oddities:
-
- /* There are tons of EBCDIC varieties, with similar but
- * non-identical names. As a result, we can not give or use
- * just a single canonical name for general use.
- * However, we can choose a single one to use for bootstrapping;
- * that is, for parsing xml declaration to know the "real"
- * EBCDIC variant.
- */
- public final static String CS_EBCDIC_SUBSET = "IBM037";
-
- /*
- //////////////////////////////////////////////////
- // Utility methods
- //////////////////////////////////////////////////
- */
-
- public static String normalize(String csName)
- {
- if (csName == null || csName.length() < 3) {
- return csName;
- }
-
- /* Canonical charset names here are from IANA recommendation:
- * http://www.iana.org/assignments/character-sets
- * but comparison is done loosely (case-insensitive, ignoring
- * spacing, underscore vs. hyphen etc) to try to make detection
- * as extensive as possible.
- */
-
- /* But first bit of pre-filtering: it seems like 'cs' prefix
- * is applicable to pretty much all actual encodings (as per
- * IANA recommendations; csASCII, csUcs4 etc). So, let's just
- * strip out the prefix if so
- */
- boolean gotCsPrefix = false;
- char c = csName.charAt(0);
- if (c == 'c' || c == 'C'){
- char d = csName.charAt(1);
- if (d == 's' || d == 'S') {
- csName = csName.substring(2);
- c = csName.charAt(0);
- gotCsPrefix = true;
- }
- }
-
- switch (c) {
- case 'a':
- case 'A':
- if (StringUtil.equalEncodings(csName, "ASCII")) {
- return CS_US_ASCII;
- }
- break;
-
- case 'c':
- case 'C':
- /* Tons of variants: let's assume 'cpXXX' is an EBCDIC
- * variant, and should read 'IBMXXX'
- */
- if (StringUtil.encodingStartsWith(csName, "cp")) {
- return "IBM" + StringUtil.trimEncoding(csName, true).substring(2);
- }
- // Hmmh. There are boatloads of these... but what to do with them?
- if (StringUtil.encodingStartsWith(csName, "cs")) {
- // Well, "csIBMxx" means EBCDIC of "IBMxx"
- if (StringUtil.encodingStartsWith(csName, "csIBM")) {
- // So let's just peel off "cs" prefix:
- return StringUtil.trimEncoding(csName, true).substring(2);
- }
- // !!! TBI
- }
- break;
-
- case 'e':
- case 'E':
- if (csName.startsWith("EBCDIC-CP-") ||
- csName.startsWith("ebcdic-cp-")) {
- // EBCDIC, but which variety?
- // Let's trim out prefix to make comparison easier:
- String type = StringUtil.trimEncoding(csName, true).substring(8);
- // Note: these are suggested encodings of Xerces
- if (type.equals("US") || type.equals("CA")
- || type.equals("WT") || type.equals("NL")) {
- return "IBM037";
- }
- if (type.equals("DK") || type.equals("NO")) { // Denmark, Norway
- return "IBM277";
- }
- if (type.equals("FI") || type.equals("SE")) { // Finland, Sweden
- return "IBM278";
- }
- if (type.equals("ROECE") || type.equals("YU")) {
- return "IBM870";
- }
- if (type.equals("IT")) return "IBM280";
- if (type.equals("ES")) return "IBM284";
- if (type.equals("GB")) return "IBM285";
- if (type.equals("FR")) return "IBM297";
- if (type.equals("AR1")) return "IBM420";
- if (type.equals("AR2")) return "IBM918";
- if (type.equals("HE")) return "IBM424";
- if (type.equals("CH")) return "IBM500";
- if (type.equals("IS")) return "IBM871";
-
- // Dunno... let's just default to 037?
- return CS_EBCDIC_SUBSET;
- }
- break;
- case 'i':
- case 'I':
- if (StringUtil.equalEncodings(csName, "ISO-8859-1")
- || StringUtil.equalEncodings(csName, "ISO-Latin1")) {
- return CS_ISO_LATIN1;
- }
- if (StringUtil.encodingStartsWith(csName, "ISO-10646")) {
- /* Hmmh. There are boatloads of alternatives here, it
- * seems (see http://www.iana.org/assignments/character-sets
- * for details)
- */
- int ix = csName.indexOf("10646");
- String suffix = csName.substring(ix+5);
- if (StringUtil.equalEncodings(suffix, "UCS-Basic")) {
- return CS_US_ASCII;
- }
- if (StringUtil.equalEncodings(suffix, "Unicode-Latin1")) {
- return CS_ISO_LATIN1;
- }
- if (StringUtil.equalEncodings(suffix, "UCS-2")) {
- return CS_UTF16; // endianness?
- }
- if (StringUtil.equalEncodings(suffix, "UCS-4")) {
- return CS_UTF32; // endianness?
- }
- if (StringUtil.equalEncodings(suffix, "UTF-1")) {
- // "Universal Transfer Format (1), this is the multibyte encoding, that subsets ASCII-7"???
- return CS_US_ASCII;
- }
- if (StringUtil.equalEncodings(suffix, "J-1")) {
- // Name: ISO-10646-J-1, Source: ISO 10646 Japanese, see RFC 1815.
- // ... so what does that really mean? let's consider it ascii
- return CS_US_ASCII;
- }
- if (StringUtil.equalEncodings(suffix, "US-ASCII")) {
- return CS_US_ASCII;
- }
- } else if (StringUtil.encodingStartsWith(csName, "IBM")) {
- // EBCDIC of some kind... what (if anything) to do?
- // ... for now, return as is
- return csName;
- }
- break;
- case 'j':
- case 'J':
- if (StringUtil.equalEncodings(csName, "JIS_Encoding")) {
- return CS_SHIFT_JIS;
- }
- break;
- case 's':
- case 'S':
- if (StringUtil.equalEncodings(csName, "Shift_JIS")) {
- return CS_SHIFT_JIS;
- }
- break;
- case 'u':
- case 'U':
- if (csName.length() < 2) { // sanity check
- break;
- }
- switch (csName.charAt(1)) {
- case 'c':
- case 'C':
- if (StringUtil.equalEncodings(csName, "UCS-2")) {
- return CS_UTF16;
- }
- if (StringUtil.equalEncodings(csName, "UCS-4")) {
- return CS_UTF32;
- }
- break;
- case 'n': // csUnicodeXxx,
- case 'N':
- if (gotCsPrefix) {
- if (StringUtil.equalEncodings(csName, "Unicode")) {
- return CS_UTF16; // need BOM
- }
- if (StringUtil.equalEncodings(csName, "UnicodeAscii")) {
- return CS_ISO_LATIN1;
- }
- if (StringUtil.equalEncodings(csName, "UnicodeAscii")) {
- return CS_US_ASCII;
- }
- }
- break;
- case 's':
- case 'S':
- if (StringUtil.equalEncodings(csName, "US-ASCII")) {
- return CS_US_ASCII;
- }
- break;
- case 't':
- case 'T':
- if (StringUtil.equalEncodings(csName, "UTF-8")) {
- return CS_UTF8;
- }
- if (StringUtil.equalEncodings(csName, "UTF-16BE")) {
- return CS_UTF16BE;
- }
- if (StringUtil.equalEncodings(csName, "UTF-16LE")) {
- return CS_UTF16LE;
- }
- if (StringUtil.equalEncodings(csName, "UTF-16")) {
- return CS_UTF16;
- }
- if (StringUtil.equalEncodings(csName, "UTF-32BE")) {
- return CS_UTF32BE;
- }
- if (StringUtil.equalEncodings(csName, "UTF-32LE")) {
- return CS_UTF32LE;
- }
- if (StringUtil.equalEncodings(csName, "UTF-32")) {
- return CS_UTF32;
- }
- if (StringUtil.equalEncodings(csName, "UTF")) {
- // 21-Jan-2006, TSa: ??? What is this to do... ?
- return CS_UTF16;
- }
- }
- break;
- }
-
- return csName;
- }
-
- /**
- * Because of legacy encodings used by earlier JDK versions, we
- * need to be careful when accessing encoding names via JDK
- * classes.
- */
- public static String findEncodingFor(Writer w)
- {
- if (w instanceof OutputStreamWriter) {
- String enc = ((OutputStreamWriter) w).getEncoding();
- /* [WSTX-146]: It is important that we normalize this, since
- * older JDKs return legacy encoding names ("UTF8" instead of
- * canonical "UTF-8")
- */
- return normalize(enc);
- }
- return null;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CompletelyCloseable.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CompletelyCloseable.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/CompletelyCloseable.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/CompletelyCloseable.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,8 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.IOException;
-
-public interface CompletelyCloseable
-{
- public void closeCompletely() throws IOException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/DefaultInputResolver.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/DefaultInputResolver.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/DefaultInputResolver.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/DefaultInputResolver.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,323 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-import java.net.URL;
-
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.transform.Source;
-import javax.xml.transform.stream.StreamSource;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.util.URLUtil;
-
-/**
- * Static utility class that implements the entity (external DTD subset,
- * external parsed entities) resolution logics.
- */
-public final class DefaultInputResolver
-{
- /*
- ////////////////////////////
- // Life-cycle
- ////////////////////////////
- */
-
- private DefaultInputResolver() { }
-
- /*
- ////////////////////////////
- // Public API
- ////////////////////////////
- */
-
- /**
- * Basic external resource resolver implementation; usable both with
- * DTD and entity resolution.
- *
- * @param parent Input source that contains reference to be expanded.
- * @param pathCtxt Reference context to use for resolving path, if
- * known. If null, reference context of the parent will
- * be used; and if that is null (which is possible), the
- * current working directory will be assumed.
- * @param entityName Name/id of the entity being expanded, if this is an
- * entity expansion; null otherwise (for example, when resolving external
- * subset).
- * @param publicId Public identifier of the resource, if known; null/empty
- * otherwise. Default implementation just ignores the identifier.
- * @param systemId System identifier of the resource. Although interface
- * allows null/empty, default implementation considers this an error.
- * @param xmlVersion Xml version as declared by the main parsed
- * document. Currently only relevant for checking that XML 1.0 document
- * does not include XML 1.1 external parsed entities.
- * If XML_V_UNKNOWN, no checks will be done.
- * @param customResolver Custom resolver to use first for resolution,
- * if any (may be null).
- * @param cfg Reader configuration object used by the parser that is
- * resolving the entity
- *
- * @return Input source, if entity could be resolved; null if it could
- * not be resolved. In latter case processor may use its own default
- * resolution mechanism.
- */
- public static WstxInputSource resolveEntity
- (WstxInputSource parent, URL pathCtxt, String entityName,
- String publicId, String systemId,
- XMLResolver customResolver, ReaderConfig cfg, int xmlVersion)
- throws IOException, XMLStreamException
- {
- if (pathCtxt == null) {
- pathCtxt = parent.getSource();
- if (pathCtxt == null) {
- pathCtxt = URLUtil.urlFromCurrentDir();
- }
- }
-
- // Do we have a custom resolver that may be able to resolve it?
- if (customResolver != null) {
- Object source = customResolver.resolveEntity(publicId, systemId, pathCtxt.toExternalForm(), entityName);
- if (source != null) {
- return sourceFrom(parent, cfg, entityName, xmlVersion, source);
- }
- }
-
- // Have to have a system id, then...
- if (systemId == null) {
- throw new XMLStreamException("Can not resolve "
- +((entityName == null) ? "[External DTD subset]" : ("entity '"+entityName+"'"))+" without a system id (public id '"
- +publicId+"')");
- }
- URL url = URLUtil.urlFromSystemId(systemId, pathCtxt);
- return sourceFromURL(parent, cfg, entityName, xmlVersion, url, publicId);
- }
-
- /**
- * A very simple utility expansion method used generally when the
- * only way to resolve an entity is via passed resolver; and where
- * failing to resolve it is not fatal.
- */
- public static WstxInputSource resolveEntityUsing
- (WstxInputSource refCtxt, String entityName,
- String publicId, String systemId,
- XMLResolver resolver, ReaderConfig cfg, int xmlVersion)
- throws IOException, XMLStreamException
- {
- URL ctxt = (refCtxt == null) ? null : refCtxt.getSource();
- if (ctxt == null) {
- ctxt = URLUtil.urlFromCurrentDir();
- }
- Object source = resolver.resolveEntity(publicId, systemId, ctxt.toExternalForm(), entityName);
- return (source == null) ? null : sourceFrom(refCtxt, cfg, entityName, xmlVersion, source);
- }
-
- /**
- * Factory method that accepts various types of Objects, and tries to
- * create a {@link WstxInputSource} from it. Currently it's only called
- * to locate external DTD subsets, when overriding default DOCTYPE
- * declarations; not for entity expansion or for locating the main
- * document entity.
- *
- * @param parent Input source context active when resolving a new
- * "sub-source"; usually the main document source.
- * @param refName Name of the entity to be expanded, if any; may be
- * null (and currently always is)
- * @param o Object that should provide the new input source; non-type safe
- */
- protected static WstxInputSource sourceFrom(WstxInputSource parent,
- ReaderConfig cfg, String refName,
- int xmlVersion,
- Object o)
- throws IllegalArgumentException, IOException, XMLStreamException
- {
- if (o instanceof Source) {
- if (o instanceof StreamSource) {
- return sourceFromSS(parent, cfg, refName, xmlVersion, (StreamSource) o);
- }
- /* !!! 05-Feb-2006, TSa: Could check if SAXSource actually has
- * stream/reader available... ?
- */
- throw new IllegalArgumentException("Can not use other Source objects than StreamSource: got "+o.getClass());
- }
- if (o instanceof URL) {
- return sourceFromURL(parent, cfg, refName, xmlVersion, (URL) o, null);
- }
- if (o instanceof InputStream) {
- return sourceFromIS(parent, cfg, refName, xmlVersion, (InputStream) o, null, null);
- }
- if (o instanceof Reader) {
- return sourceFromR(parent, cfg, refName, xmlVersion, (Reader) o, null, null);
- }
- if (o instanceof String) {
- return sourceFromString(parent, cfg, refName, xmlVersion, (String) o);
- }
- if (o instanceof File) {
- URL u = ((File) o).toURL();
- return sourceFromURL(parent, cfg, refName, xmlVersion, u, null);
- }
-
- throw new IllegalArgumentException("Unrecognized input argument type for sourceFrom(): "+o.getClass());
- }
-
- public static Reader constructOptimizedReader(ReaderConfig cfg, InputStream in, boolean isXml11, String encoding)
- throws XMLStreamException
- {
- /* 03-Jul-2005, TSa: Since Woodstox' implementations of specialized
- * readers are faster than default JDK ones (at least for 1.4, UTF-8
- * reader is especially fast...), let's use them if possible
- */
- /* 17-Feb-2006, TSa: These should actually go via InputBootstrapper,
- * since BOM may need to be skipped; xml 1.0 vs. 1.1 should be
- * checked, and so on. Given encoding could be just verified
- * against suggested one.
- */
- int inputBufLen = cfg.getInputBufferLength();
- String normEnc = CharsetNames.normalize(encoding);
- BaseReader r;
-
- // All of these use real InputStream, and use recyclable buffer
- boolean recycleBuffer = true;
- if (normEnc == CharsetNames.CS_UTF8) {
- r = new UTF8Reader(cfg, in, cfg.allocFullBBuffer(inputBufLen), 0, 0, recycleBuffer);
- } else if (normEnc == CharsetNames.CS_ISO_LATIN1) {
- r = new ISOLatinReader(cfg, in, cfg.allocFullBBuffer(inputBufLen), 0, 0, recycleBuffer);
- } else if (normEnc == CharsetNames.CS_US_ASCII) {
- r = new AsciiReader(cfg, in, cfg.allocFullBBuffer(inputBufLen), 0, 0, recycleBuffer);
- } else if (normEnc.startsWith(CharsetNames.CS_UTF32)) {
- boolean isBE = (normEnc == CharsetNames.CS_UTF32BE);
- r = new UTF32Reader(cfg, in, cfg.allocFullBBuffer(inputBufLen), 0, 0, recycleBuffer, isBE);
- } else {
- try {
- return new InputStreamReader(in, encoding);
- } catch (UnsupportedEncodingException ex) {
- throw new XMLStreamException("[unsupported encoding]: "+ex);
- }
- }
-
- if (isXml11) { // only need to set if we are xml 1.1 compliant (1.0 is the default)
- r.setXmlCompliancy(XmlConsts.XML_V_11);
- }
-
- return r;
- }
-
- /*
- ////////////////////////////
- // Internal methods
- ////////////////////////////
- */
-
- private static WstxInputSource sourceFromSS(WstxInputSource parent, ReaderConfig cfg,
- String refName, int xmlVersion,
- StreamSource ssrc)
- throws IOException, XMLStreamException
- {
- InputBootstrapper bs;
- Reader r = ssrc.getReader();
- String pubId = ssrc.getPublicId();
- String sysId = ssrc.getSystemId();
- URL ctxt = (parent == null) ? null : parent.getSource();
- URL url = (sysId == null || sysId.length() == 0) ? null
- : URLUtil.urlFromSystemId(sysId, ctxt);
-
- if (r == null) {
- InputStream in = ssrc.getInputStream();
- if (in == null) { // Need to try just resolving the system id then
- if (url == null) {
- throw new IllegalArgumentException("Can not create Stax reader for a StreamSource -- neither reader, input stream nor system id was set.");
- }
- in = URLUtil.inputStreamFromURL(url);
- }
- bs = StreamBootstrapper.getInstance(pubId, sysId, in);
- } else {
- bs = ReaderBootstrapper.getInstance(pubId, sysId, r, null);
- }
-
- Reader r2 = bs.bootstrapInput(cfg, false, xmlVersion);
- return InputSourceFactory.constructEntitySource
- (cfg, parent, refName, bs, pubId, sysId, xmlVersion,
- ((url == null) ? ctxt : url), r2);
- }
-
- private static WstxInputSource sourceFromURL(WstxInputSource parent, ReaderConfig cfg,
- String refName, int xmlVersion,
- URL url,
- String pubId)
- throws IOException, XMLStreamException
- {
- /* And then create the input source. Note that by default URL's
- * own input stream creation creates buffered reader -- for us
- * that's useless and wasteful (adds one unnecessary level of
- * caching, halving the speed due to copy operations needed), so
- * let's avoid it.
- */
- InputStream in = URLUtil.inputStreamFromURL(url);
- String sysId = url.toExternalForm();
- StreamBootstrapper bs = StreamBootstrapper.getInstance(pubId, sysId, in);
- Reader r = bs.bootstrapInput(cfg, false, xmlVersion);
- return InputSourceFactory.constructEntitySource
- (cfg, parent, refName, bs, pubId, sysId, xmlVersion, url, r);
- }
-
- /**
- * We have multiple ways to look at what would it mean to get a String
- * as the resolved result. The most straight-forward is to consider
- * it literal replacement (with possible embedded entities), so let's
- * use that (alternative would be to consider it to be a reference
- * like URL -- those need to be returned as appropriate objects
- * instead).
- *
- * Note: public to give access for unit tests that need it...
- */
- public static WstxInputSource sourceFromString(WstxInputSource parent, ReaderConfig cfg,
- String refName, int xmlVersion,
- String refContent)
- throws IOException, XMLStreamException
- {
- /* Last null -> no app-provided encoding (doesn't matter for non-
- * main-level handling)
- */
- return sourceFromR(parent, cfg, refName, xmlVersion,
- new StringReader(refContent),
- null, refName);
- }
-
- private static WstxInputSource sourceFromIS(WstxInputSource parent,
- ReaderConfig cfg,
- String refName, int xmlVersion,
- InputStream is,
- String pubId, String sysId)
- throws IOException, XMLStreamException
- {
- StreamBootstrapper bs = StreamBootstrapper.getInstance(pubId, sysId, is);
- Reader r = bs.bootstrapInput(cfg, false, xmlVersion);
- URL ctxt = parent.getSource();
-
- // If we got a real sys id, we do know the source...
- if (sysId != null && sysId.length() > 0) {
- ctxt = URLUtil.urlFromSystemId(sysId, ctxt);
- }
- return InputSourceFactory.constructEntitySource
- (cfg, parent, refName, bs, pubId, sysId, xmlVersion, ctxt, r);
- }
-
- private static WstxInputSource sourceFromR(WstxInputSource parent, ReaderConfig cfg,
- String refName, int xmlVersion,
- Reader r,
- String pubId, String sysId)
- throws IOException, XMLStreamException
- {
- /* Last null -> no app-provided encoding (doesn't matter for non-
- * main-level handling)
- */
- ReaderBootstrapper rbs = ReaderBootstrapper.getInstance(pubId, sysId, r, null);
- // null -> no xml reporter... should have one?
- Reader r2 = rbs.bootstrapInput(cfg, false, xmlVersion);
- URL ctxt = (parent == null) ? null : parent.getSource();
- if (sysId != null && sysId.length() > 0) {
- ctxt = URLUtil.urlFromSystemId(sysId, ctxt);
- }
- return InputSourceFactory.constructEntitySource
- (cfg, parent, refName, rbs, pubId, sysId, xmlVersion, ctxt, r2);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/EBCDICCodec.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/EBCDICCodec.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/EBCDICCodec.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/EBCDICCodec.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,322 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-/**
- * This is a container class for EBCDIC code page(s) that we need
- * to properly bootstrap EBCDIC encoded xml documents.
- */
-public final class EBCDICCodec
-{
- final static int[] sCodePage037;
- static {
- /* First, let's fill these in as per:
- *
- * http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/EBCDIC/CP037.TXT
- */
- int[] ch = new int[256];
- ch[0] = 0x0000;
- ch[0x01] = 0x0001;
- ch[0x02] = 0x0002;
- ch[0x03] = 0x0003;
- ch[0x04] = 0x009C;
- ch[0x05] = 0x0009;
- ch[0x06] = 0x0086;
- ch[0x07] = 0x007F;
- ch[0x08] = 0x0097;
- ch[0x09] = 0x008D;
- ch[0x0A] = 0x008E;
- ch[0x0B] = 0x000B;
- ch[0x0C] = 0x000C;
- ch[0x0D] = 0x000D;
- ch[0x0E] = 0x000E;
- ch[0x0F] = 0x000F;
- ch[0x10] = 0x0010;
- ch[0x11] = 0x0011;
- ch[0x12] = 0x0012;
- ch[0x13] = 0x0013;
- ch[0x14] = 0x009D;
- ch[0x15] = 0x0085;
- ch[0x16] = 0x0008;
- ch[0x17] = 0x0087;
- ch[0x18] = 0x0018;
- ch[0x19] = 0x0019;
- ch[0x1A] = 0x0092;
- ch[0x1B] = 0x008F;
- ch[0x1C] = 0x001C;
- ch[0x1D] = 0x001D;
- ch[0x1E] = 0x001E;
- ch[0x1F] = 0x001F;
- ch[0x20] = 0x0080;
- ch[0x21] = 0x0081;
- ch[0x22] = 0x0082;
- ch[0x23] = 0x0083;
- ch[0x24] = 0x0084;
- ch[0x25] = 0x000A;
- ch[0x26] = 0x0017;
- ch[0x27] = 0x001B;
- ch[0x28] = 0x0088;
- ch[0x29] = 0x0089;
- ch[0x2A] = 0x008A;
- ch[0x2B] = 0x008B;
- ch[0x2C] = 0x008C;
- ch[0x2D] = 0x0005;
- ch[0x2E] = 0x0006;
- ch[0x2F] = 0x0007;
- ch[0x30] = 0x0090;
- ch[0x31] = 0x0091;
- ch[0x32] = 0x0016;
- ch[0x33] = 0x0093;
- ch[0x34] = 0x0094;
- ch[0x35] = 0x0095;
- ch[0x36] = 0x0096;
- ch[0x37] = 0x0004;
- ch[0x38] = 0x0098;
- ch[0x39] = 0x0099;
- ch[0x3A] = 0x009A;
- ch[0x3B] = 0x009B;
- ch[0x3C] = 0x0014;
- ch[0x3D] = 0x0015;
- ch[0x3E] = 0x009E;
- ch[0x3F] = 0x001A;
- ch[0x40] = 0x0020;
- ch[0x41] = 0x00A0;
- ch[0x42] = 0x00E2;
- ch[0x43] = 0x00E4;
- ch[0x44] = 0x00E0;
- ch[0x45] = 0x00E1;
- ch[0x46] = 0x00E3;
- ch[0x47] = 0x00E5;
- ch[0x48] = 0x00E7;
- ch[0x49] = 0x00F1;
- ch[0x4A] = 0x00A2;
- ch[0x4B] = 0x002E;
- ch[0x4C] = 0x003C;
- ch[0x4D] = 0x0028;
- ch[0x4E] = 0x002B;
- ch[0x4F] = 0x007C;
- ch[0x50] = 0x0026;
- ch[0x51] = 0x00E9;
- ch[0x52] = 0x00EA;
- ch[0x53] = 0x00EB;
- ch[0x54] = 0x00E8;
- ch[0x55] = 0x00ED;
- ch[0x56] = 0x00EE;
- ch[0x57] = 0x00EF;
- ch[0x58] = 0x00EC;
- ch[0x59] = 0x00DF;
- ch[0x5A] = 0x0021;
- ch[0x5B] = 0x0024;
- ch[0x5C] = 0x002A;
- ch[0x5D] = 0x0029;
- ch[0x5E] = 0x003B;
- ch[0x5F] = 0x00AC;
- ch[0x60] = 0x002D;
- ch[0x61] = 0x002F;
- ch[0x62] = 0x00C2;
- ch[0x63] = 0x00C4;
- ch[0x64] = 0x00C0;
- ch[0x65] = 0x00C1;
- ch[0x66] = 0x00C3;
- ch[0x67] = 0x00C5;
- ch[0x68] = 0x00C7;
- ch[0x69] = 0x00D1;
- ch[0x6A] = 0x00A6;
- ch[0x6B] = 0x002C;
- ch[0x6C] = 0x0025;
- ch[0x6D] = 0x005F;
- ch[0x6E] = 0x003E;
- ch[0x6F] = 0x003F;
- ch[0x70] = 0x00F8;
- ch[0x71] = 0x00C9;
- ch[0x72] = 0x00CA;
- ch[0x73] = 0x00CB;
- ch[0x74] = 0x00C8;
- ch[0x75] = 0x00CD;
- ch[0x76] = 0x00CE;
- ch[0x77] = 0x00CF;
- ch[0x78] = 0x00CC;
- ch[0x79] = 0x0060;
- ch[0x7A] = 0x003A;
- ch[0x7B] = 0x0023;
- ch[0x7C] = 0x0040;
- ch[0x7D] = 0x0027;
- ch[0x7E] = 0x003D;
- ch[0x7F] = 0x0022;
-
- ch[0x80] = 0x00D8;
- ch[0x81] = 0x0061;
- ch[0x82] = 0x0062;
- ch[0x83] = 0x0063;
- ch[0x84] = 0x0064;
- ch[0x85] = 0x0065;
- ch[0x86] = 0x0066;
- ch[0x87] = 0x0067;
- ch[0x88] = 0x0068;
- ch[0x89] = 0x0069;
- ch[0x8A] = 0x00AB;
- ch[0x8B] = 0x00BB;
- ch[0x8C] = 0x00F0;
- ch[0x8D] = 0x00FD;
- ch[0x8E] = 0x00FE;
- ch[0x8F] = 0x00B1;
- ch[0x90] = 0x00B0;
- ch[0x91] = 0x006A;
- ch[0x92] = 0x006B;
- ch[0x93] = 0x006C;
- ch[0x94] = 0x006D;
- ch[0x95] = 0x006E;
- ch[0x96] = 0x006F;
- ch[0x97] = 0x0070;
- ch[0x98] = 0x0071;
- ch[0x99] = 0x0072;
- ch[0x9A] = 0x00AA;
- ch[0x9B] = 0x00BA;
- ch[0x9C] = 0x00E6;
- ch[0x9D] = 0x00B8;
- ch[0x9E] = 0x00C6;
- ch[0x9F] = 0x00A4;
- ch[0xA0] = 0x00B5;
- ch[0xA1] = 0x007E;
- ch[0xA2] = 0x0073;
- ch[0xA3] = 0x0074;
- ch[0xA4] = 0x0075;
- ch[0xA5] = 0x0076;
- ch[0xA6] = 0x0077;
- ch[0xA7] = 0x0078;
- ch[0xA8] = 0x0079;
- ch[0xA9] = 0x007A;
- ch[0xAA] = 0x00A1;
- ch[0xAB] = 0x00BF;
- ch[0xAC] = 0x00D0;
- ch[0xAD] = 0x00DD;
- ch[0xAE] = 0x00DE;
- ch[0xAF] = 0x00AE;
- ch[0xB0] = 0x005E;
- ch[0xB1] = 0x00A3;
- ch[0xB2] = 0x00A5;
- ch[0xB3] = 0x00B7;
- ch[0xB4] = 0x00A9;
- ch[0xB5] = 0x00A7;
- ch[0xB6] = 0x00B6;
- ch[0xB7] = 0x00BC;
- ch[0xB8] = 0x00BD;
- ch[0xB9] = 0x00BE;
- ch[0xBA] = 0x005B;
- ch[0xBB] = 0x005D;
- ch[0xBC] = 0x00AF;
- ch[0xBD] = 0x00A8;
- ch[0xBE] = 0x00B4;
- ch[0xBF] = 0x00D7;
- ch[0xC0] = 0x007B;
- ch[0xC1] = 0x0041;
- ch[0xC2] = 0x0042;
- ch[0xC3] = 0x0043;
- ch[0xC4] = 0x0044;
- ch[0xC5] = 0x0045;
- ch[0xC6] = 0x0046;
- ch[0xC7] = 0x0047;
- ch[0xC8] = 0x0048;
- ch[0xC9] = 0x0049;
- ch[0xCA] = 0x00AD;
- ch[0xCB] = 0x00F4;
- ch[0xCC] = 0x00F6;
- ch[0xCD] = 0x00F2;
- ch[0xCE] = 0x00F3;
- ch[0xCF] = 0x00F5;
- ch[0xD0] = 0x007D;
- ch[0xD1] = 0x004A;
- ch[0xD2] = 0x004B;
- ch[0xD3] = 0x004C;
- ch[0xD4] = 0x004D;
- ch[0xD5] = 0x004E;
- ch[0xD6] = 0x004F;
- ch[0xD7] = 0x0050;
- ch[0xD8] = 0x0051;
- ch[0xD9] = 0x0052;
- ch[0xDA] = 0x00B9;
- ch[0xDB] = 0x00FB;
- ch[0xDC] = 0x00FC;
- ch[0xDD] = 0x00F9;
- ch[0xDE] = 0x00FA;
- ch[0xDF] = 0x00FF;
- ch[0xE0] = 0x005C;
- ch[0xE1] = 0x00F7;
- ch[0xE2] = 0x0053;
- ch[0xE3] = 0x0054;
- ch[0xE4] = 0x0055;
- ch[0xE5] = 0x0056;
- ch[0xE6] = 0x0057;
- ch[0xE7] = 0x0058;
- ch[0xE8] = 0x0059;
- ch[0xE9] = 0x005A;
- ch[0xEA] = 0x00B2;
- ch[0xEB] = 0x00D4;
- ch[0xEC] = 0x00D6;
- ch[0xED] = 0x00D2;
- ch[0xEE] = 0x00D3;
- ch[0xEF] = 0x00D5;
- ch[0xF0] = 0x0030;
- ch[0xF1] = 0x0031;
- ch[0xF2] = 0x0032;
- ch[0xF3] = 0x0033;
- ch[0xF4] = 0x0034;
- ch[0xF5] = 0x0035;
- ch[0xF6] = 0x0036;
- ch[0xF7] = 0x0037;
- ch[0xF8] = 0x0038;
- ch[0xF9] = 0x0039;
- ch[0xFA] = 0x00B3;
- ch[0xFB] = 0x00DB;
- ch[0xFC] = 0x00DC;
- ch[0xFD] = 0x00D9;
- ch[0xFE] = 0x00DA;
- ch[0xFF] = 0x009F;
-
- /* And then massage it so we can catch problems with control
- * chars same way as we'd do with, say, Latin1 input.
- */
- for (int i = 0, len = ch.length; i < len; ++i) {
- int c = ch[i];
- if (c >= 0x7F && c <= 0x9F) {
- /* 21-Sep-2007, TSa: Hmmh. In a way, it should be dealt
- * as per xml rules, and not allowed in xml declaration
- * before encoding declaration. But that won't work well
- * with real docs. Converting it earlier works better...
- * so for now that seems the way it needs to be done. ;-/
- */
- if (c == 0x85) {
- ch[i] = BaseReader.CONVERT_NEL_TO;
- } else {
- ch[i] = -c;
- }
- }
- }
-
- sCodePage037 = ch;
- }
-
- private EBCDICCodec() { }
-
- /**
- * @return Code table for EBCDIC code page 037 (US, Canada etc)
- */
- public static int[] getCp037Mapping()
- {
- return sCodePage037;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/InputBootstrapper.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/InputBootstrapper.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/InputBootstrapper.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/InputBootstrapper.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,529 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.exc.*;
-
-/**
- * Abstract base class that defines common API used with both stream and
- * reader-based input sources. Class is responsible for opening the physical
- * input source, figure out encoding (if necessary; only for streams), and
- * then handle (optional) XML declaration.
- */
-public abstract class InputBootstrapper
-{
- /*
- ////////////////////////////////////////////////////////////
- // Shared string consts
- ////////////////////////////////////////////////////////////
- */
-
- protected final static String ERR_XMLDECL_KW_VERSION = "; expected keyword '"+XmlConsts.XML_DECL_KW_VERSION+"'";
- protected final static String ERR_XMLDECL_KW_ENCODING = "; expected keyword '"+XmlConsts.XML_DECL_KW_ENCODING+"'";
- protected final static String ERR_XMLDECL_KW_STANDALONE = "; expected keyword '"+XmlConsts.XML_DECL_KW_STANDALONE+"'";
-
- protected final static String ERR_XMLDECL_END_MARKER = "; expected \"?>\" end marker";
-
- protected final static String ERR_XMLDECL_EXP_SPACE = "; expected a white space";
- protected final static String ERR_XMLDECL_EXP_EQ = "; expected '=' after ";
- protected final static String ERR_XMLDECL_EXP_ATTRVAL = "; expected a quote character enclosing value for ";
-
- /*
- ////////////////////////////////////////////////////////////
- // Other consts
- ////////////////////////////////////////////////////////////
- */
-
- public final static char CHAR_NULL = (char) 0;
- public final static char CHAR_SPACE = (char) 0x0020;
-
- public final static char CHAR_NEL = (char) 0x0085;
-
- public final static byte CHAR_CR = '\r';
- public final static byte CHAR_LF = '\n';
-
- public final static byte BYTE_NULL = (byte) 0;
-
- public final static byte BYTE_CR = (byte) '\r';
- public final static byte BYTE_LF = (byte) '\n';
-
- /*
- ////////////////////////////////////////////////////////////
- // Input source info
- ////////////////////////////////////////////////////////////
- */
-
- protected final String mPublicId;
-
- protected final String mSystemId;
-
- /*
- ////////////////////////////////////////////////////////////
- // Input location data (similar to one in WstxInputData)
- ////////////////////////////////////////////////////////////
- */
-
- /**
- * Current number of characters that were processed in previous blocks,
- * before contents of current input buffer.
- */
- protected int mInputProcessed = 0;
-
- /**
- * Current row location of current point in input buffer, starting
- * from 1
- */
- protected int mInputRow = 1;
-
- /**
- * Current index of the first character of the current row in input
- * buffer. Needed to calculate column position, if necessary; benefit
- * of not having column itself is that this only has to be updated
- * once per line.
- */
- protected int mInputRowStart = 0;
-
- /*
- ////////////////////////////////////////
- // Info from XML declaration
- ////////////////////////////////////////
- */
-
- //boolean mHadDeclaration = false;
-
- /**
- * XML declaration from the input (1.0, 1.1 or 'unknown')
- */
- int mDeclaredXmlVersion = XmlConsts.XML_V_UNKNOWN;
-
- /**
- * Value of encoding pseudo-attribute from xml declaration, if
- * one was found; null otherwise.
- */
- String mFoundEncoding;
-
- String mStandalone;
-
- /**
- * Flag that indicates whether input read from this input source
- * needs to be xml 1.1 compliant or not; if not, xml 1.0 is assumed.
- * State of this flag depends on parent context (if one existed), or if
- * not, on xml declaration of this input source.
- */
- boolean mXml11Handling = false;
-
- /*
- ////////////////////////////////////////
- // Temporary data
- ////////////////////////////////////////
- */
-
- /**
- * Need a short buffer to read in values of pseudo-attributes (version,
- * encoding, standalone). Don't really need tons of space; just enough
- * for the longest anticipated encoding id... and maybe few chars just
- * in case (for additional white space that we ignore)
- */
- final char[] mKeyword = new char[60];
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- protected InputBootstrapper(String pubId, String sysId)
- {
- mPublicId = pubId;
- mSystemId = sysId;
- }
-
- protected void initFrom(InputBootstrapper src)
- {
- mInputProcessed = src.mInputProcessed;
- mInputRow = src.mInputRow;
- mInputRowStart = src.mInputRowStart;
- mDeclaredXmlVersion = src.mDeclaredXmlVersion;
- mFoundEncoding = src.mFoundEncoding;
- mStandalone = src.mStandalone;
- mXml11Handling = src.mXml11Handling;
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- /**
- * @param xmlVersion Optional xml version identifier of the main parsed
- * document (if not bootstrapping the main document).
- * Currently only relevant for checking that XML 1.0 document does not
- * include XML 1.1 external parsed entities.
- * If null, no checks will be done; when bootstrapping parsing of the
- * main document, null should be passed for this argument.
- */
- public abstract Reader bootstrapInput(ReaderConfig cfg, boolean mainDoc,
- int xmlVersion)
- throws IOException, XMLStreamException;
-
- // // // Source information:
-
- public String getPublicId() { return mPublicId; }
- public String getSystemId() { return mSystemId; }
-
- // // // XML declaration data:
-
- public int getDeclaredVersion()
- {
- return mDeclaredXmlVersion;
- }
-
- /**
- * @return True, if the input bootstrapped declared that it conforms
- * to xml 1.1 (independent of where it was included from)
- */
- public boolean declaredXml11() {
- return (mDeclaredXmlVersion == XmlConsts.XML_V_11);
- }
-
- public String getStandalone() {
- return mStandalone;
- }
-
- /**
- * @return Encoding declaration found from the xml declaration, if any;
- * null if none.
- */
- public String getDeclaredEncoding() {
- return mFoundEncoding;
- }
-
- // // // Location/position info:
-
- /**
- * @return Total number of characters read from bootstrapped input
- * (stream, reader)
- */
- public abstract int getInputTotal();
-
- public int getInputRow() {
- return mInputRow;
- }
-
- public abstract int getInputColumn();
-
-
- // // // Misc other info
-
- /**
- * Actual character encoding used by the underlying input source;
- * may have been passed by the application, or auto-detected
- * by byte stream boot strapper (can not be determined from a
- * Reader source).
- *
- * @return Input encoding in use, if it could be determined or was
- * passed by the calling application
- */
- public abstract String getInputEncoding();
-
- /*
- ////////////////////////////////////////
- // Package methods, parsing
- ////////////////////////////////////////
- */
-
- /**
- * @param xmlVersion Optional xml version identifier of the main parsed
- * document (if not bootstrapping the main document).
- * Currently only relevant for checking that XML 1.0 document does not
- * include XML 1.1 external parsed entities.
- * If null, no checks will be done; when bootstrapping parsing of the
- * main document, null should be passed for this argument.
- */
- protected void readXmlDecl(boolean isMainDoc, int xmlVersion)
- throws IOException, WstxException
- {
- int c = getNextAfterWs(false);
-
- // First, version pseudo-attribute:
-
- if (c != 'v') { // version info obligatory for main docs
- if (isMainDoc) {
- reportUnexpectedChar(c, ERR_XMLDECL_KW_VERSION);
- }
- } else { // ok, should be version
- mDeclaredXmlVersion = readXmlVersion();
- c = getWsOrChar('?');
- }
-
- /* 17-Feb-2006, TSa: Whether we are to be xml 1.1 compliant or not
- * depends on parent context, if any; and if not, on actual
- * xml declaration. But in former case, it is illegal to include
- * xml 1.1 declared entities from xml 1.0 context.
- */
- boolean thisIs11 = (mDeclaredXmlVersion == XmlConsts.XML_V_11);
- if (xmlVersion != XmlConsts.XML_V_UNKNOWN) { // happens when reading main doc
- mXml11Handling = (XmlConsts.XML_V_11 == xmlVersion);
- // Can not refer to xml 1.1 entities from 1.0 doc:
- if (thisIs11 && !mXml11Handling) {
- reportXmlProblem(ErrorConsts.ERR_XML_10_VS_11);
- }
- } else {
- mXml11Handling = thisIs11;
- }
-
- // Then, 'encoding'
- if (c != 'e') { // obligatory for external entities
- if (!isMainDoc) {
- reportUnexpectedChar(c, ERR_XMLDECL_KW_ENCODING);
- }
- } else {
- mFoundEncoding = readXmlEncoding();
- c = getWsOrChar('?');
- }
-
- // Then, 'standalone' (for main doc)
- if (isMainDoc && c == 's') {
- mStandalone = readXmlStandalone();
- c = getWsOrChar('?');
- }
-
- // And finally, need to have closing markers
-
- if (c != '?') {
- reportUnexpectedChar(c, ERR_XMLDECL_END_MARKER);
- }
- c = getNext();
- if (c != '>') {
- reportUnexpectedChar(c, ERR_XMLDECL_END_MARKER);
- }
- }
-
- /**
- * @return Xml version declaration read
- */
- private final int readXmlVersion()
- throws IOException, WstxException
- {
- int c = checkKeyword(XmlConsts.XML_DECL_KW_VERSION);
- if (c != CHAR_NULL) {
- reportUnexpectedChar(c, XmlConsts.XML_DECL_KW_VERSION);
- }
- c = handleEq(XmlConsts.XML_DECL_KW_VERSION);
- int len = readQuotedValue(mKeyword, c);
-
- if (len == 3) {
- if (mKeyword[0] == '1' && mKeyword[1] == '.') {
- c = mKeyword[2];
- if (c == '0') {
- return XmlConsts.XML_V_10;
- }
- if (c == '1') {
- return XmlConsts.XML_V_11;
- }
- }
- }
-
- // Nope; error. -1 indicates run off...
- String got;
-
- if (len < 0) {
- got = "'"+new String(mKeyword)+"[..]'";
- } else if (len == 0) {
- got = "
- * Note: since the actual Reader to use after bootstrapping is pre-constructed,
- * the local input buffer can (and should) be quite small.
- */
-public final class ReaderBootstrapper
- extends InputBootstrapper
-{
- final static char CHAR_BOM_MARKER = (char) 0xFEFF;
-
- /*
- ////////////////////////////////////////
- // Configuration
- ////////////////////////////////////////
- */
-
- /**
- * Underlying Reader to use for reading content.
- */
- final Reader mIn;
-
- /**
- * Encoding identifier processing application passed in; if not null,
- * will be compared to actual xml declaration based encoding (if
- * declaration found)
- */
- final String mInputEncoding;
-
- /*
- ///////////////////////////////////////////////////////////////
- // Input buffering
- ///////////////////////////////////////////////////////////////
- */
-
- private char[] mCharBuffer;
-
- private int mInputPtr;
-
- private int mInputEnd;
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- private ReaderBootstrapper(String pubId, String sysId, Reader r, String appEncoding)
- {
- super(pubId, sysId);
- mIn = r;
- if (appEncoding == null) { // may still be able to figure it out
- if (r instanceof InputStreamReader) {
- appEncoding = ((InputStreamReader) r).getEncoding();
- }
- }
- mInputEncoding = appEncoding;
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- /**
- * @param r Eventual reader that will be reading actual content, after
- * bootstrapping finishes
- * @param appEncoding Encoding that application declared; may be null.
- * If not null, will be compared to actual declaration found; and
- * incompatibility reported as a potential (but not necessarily fatal)
- * problem.
- */
- public static ReaderBootstrapper getInstance(String pubId, String sysId,
- Reader r, String appEncoding)
- {
- return new ReaderBootstrapper(pubId, sysId, r, appEncoding);
- }
-
- /**
- * Method called to do actual bootstrapping.
- *
- * @return Actual reader to use for reading xml content
- */
- public Reader bootstrapInput(ReaderConfig cfg, boolean mainDoc, int xmlVersion)
- throws IOException, XMLStreamException
- {
- /* First order of business: allocate input buffer. Not done during
- * construction for simplicity; that way config object need not be
- * passed before actual bootstrap method is called
- */
- /* Let's make sure buffer is at least 6 chars (to know '")
- */
- if (mInputEnd >= 7) {
- char c = mCharBuffer[mInputPtr];
-
- // BOM to skip?
- if (c == CHAR_BOM_MARKER) {
- c = mCharBuffer[++mInputPtr];
- }
- if (c == '<') {
- if (mCharBuffer[mInputPtr+1] == '?'
- && mCharBuffer[mInputPtr+2] == 'x'
- && mCharBuffer[mInputPtr+3] == 'm'
- && mCharBuffer[mInputPtr+4] == 'l'
- && mCharBuffer[mInputPtr+5] <= CHAR_SPACE) {
- // Yup, got the declaration ok!
- mInputPtr += 6; // skip declaration
- readXmlDecl(mainDoc, xmlVersion);
-
- if (mFoundEncoding != null && mInputEncoding != null) {
- verifyXmlEncoding(cfg);
- }
- }
- } else {
- /* We may also get something that would be invalid xml
- * ("garbage" char; neither '<' nor space). If so, and
- * it's one of "well-known" cases, we can not only throw
- * an exception but also indicate a clue as to what is likely
- * to be wrong.
- */
- /* Specifically, UTF-8 read via, say, ISO-8859-1 reader, can
- * "leak" marker (0xEF, 0xBB, 0xBF). While we could just eat
- * it, there's bound to be other problems cropping up, so let's
- * inform about the problem right away.
- */
- if (c == 0xEF) {
- throw new WstxIOException("Unexpected first character (char code 0xEF), not valid in xml document: could be mangled UTF-8 BOM marker. Make sure that the Reader uses correct encoding or pass an InputStream instead");
- }
- }
- }
-
- /* Ok, now; do we have unused chars we have read that need to
- * be merged in?
- */
- if (mInputPtr < mInputEnd) {
- return new MergedReader(cfg, mIn, mCharBuffer, mInputPtr, mInputEnd);
- }
-
- return mIn;
- }
-
- public String getInputEncoding() {
- return mInputEncoding;
- }
-
- public int getInputTotal() {
- return mInputProcessed + mInputPtr;
- }
-
- public int getInputColumn() {
- return (mInputPtr - mInputRowStart);
- }
-
- /*
- ////////////////////////////////////////
- // Internal methods, parsing
- ////////////////////////////////////////
- */
-
- protected void verifyXmlEncoding(ReaderConfig cfg)
- throws XMLStreamException
- {
- String inputEnc = mInputEncoding;
-
- // Close enough?
- if (StringUtil.equalEncodings(inputEnc, mFoundEncoding)) {
- return;
- }
-
- /* Ok, maybe the difference is just with endianness indicator?
- * (UTF-16BE vs. UTF-16)?
- */
- // !!! TBI
-
- XMLReporter rep = cfg.getXMLReporter();
- if (rep != null) {
- Location loc = getLocation();
- String msg = MessageFormat.format(ErrorConsts.W_MIXED_ENCODINGS,
- new Object[] { mFoundEncoding,
- inputEnc });
- String type = ErrorConsts.WT_XML_DECL;
- /* 30-May-2008, tatus: Should wrap all the info as XMValidationProblem
- * since that's Woodstox' contract wrt. relatedInformation field.
- */
- XMLValidationProblem prob = new XMLValidationProblem(loc, msg, XMLValidationProblem.SEVERITY_WARNING, type);
- rep.report(msg, type, prob, loc);
- }
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, loading input data
- /////////////////////////////////////////////////////
- */
-
- protected boolean initialLoad(int minimum)
- throws IOException
- {
- mInputPtr = 0;
- mInputEnd = 0;
-
- while (mInputEnd < minimum) {
- int count = mIn.read(mCharBuffer, mInputEnd,
- mCharBuffer.length - mInputEnd);
- if (count < 1) {
- return false;
- }
- mInputEnd += count;
- }
- return true;
- }
-
- protected void loadMore()
- throws IOException, WstxException
- {
- /* Need to make sure offsets are properly updated for error
- * reporting purposes, and do this now while previous amounts
- * are still known.
- */
- mInputProcessed += mInputEnd;
- mInputRowStart -= mInputEnd;
-
- mInputPtr = 0;
- mInputEnd = mIn.read(mCharBuffer, 0, mCharBuffer.length);
- if (mInputEnd < 1) {
- throw new WstxEOFException(ParsingErrorMsgs.SUFFIX_IN_XML_DECL,
- getLocation());
- }
- }
-
- /*
- /////////////////////////////////////////////////////
- // Implementations of abstract parsing methods
- /////////////////////////////////////////////////////
- */
-
- protected void pushback() {
- --mInputPtr;
- }
-
- protected int getNext()
- throws IOException, WstxException
- {
- return (mInputPtr < mInputEnd) ?
- mCharBuffer[mInputPtr++] : nextChar();
- }
-
-
- protected int getNextAfterWs(boolean reqWs)
- throws IOException, WstxException
- {
- int count = 0;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mCharBuffer[mInputPtr++] : nextChar();
-
- if (c > CHAR_SPACE) {
- if (reqWs && count == 0) {
- reportUnexpectedChar(c, ERR_XMLDECL_EXP_SPACE);
- }
- return c;
- }
- if (c == CHAR_CR || c == CHAR_LF) {
- skipCRLF(c);
- } else if (c == CHAR_NULL) {
- reportNull();
- }
- ++count;
- }
- }
-
- /**
- * @return First character that does not match expected, if any;
- * CHAR_NULL if match succeeded
- */
- protected int checkKeyword(String exp)
- throws IOException, WstxException
- {
- int len = exp.length();
-
- for (int ptr = 1; ptr < len; ++ptr) {
- char c = (mInputPtr < mInputEnd) ?
- mCharBuffer[mInputPtr++] : nextChar();
-
- if (c != exp.charAt(ptr)) {
- return c;
- }
- if (c == CHAR_NULL) {
- reportNull();
- }
- }
-
- return CHAR_NULL;
- }
-
- protected int readQuotedValue(char[] kw, int quoteChar)
- throws IOException, WstxException
- {
- int i = 0;
- int len = kw.length;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mCharBuffer[mInputPtr++] : nextChar();
- if (c == CHAR_CR || c == CHAR_LF) {
- skipCRLF(c);
- } else if (c == CHAR_NULL) {
- reportNull();
- }
- if (c == quoteChar) {
- return (i < len) ? i : -1;
- }
- // Let's just truncate longer values, but match quote
- if (i < len) {
- kw[i++] = c;
- }
- }
- }
-
- protected Location getLocation()
- {
- return new WstxInputLocation(null, mPublicId, mSystemId,
- mInputProcessed + mInputPtr - 1,
- mInputRow, mInputPtr - mInputRowStart);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, single-byte access methods
- /////////////////////////////////////////////////////
- */
-
- protected char nextChar()
- throws IOException, WstxException
- {
- if (mInputPtr >= mInputEnd) {
- loadMore();
- }
- return mCharBuffer[mInputPtr++];
- }
-
- protected void skipCRLF(char lf)
- throws IOException, WstxException
- {
- if (lf == CHAR_CR) {
- char c = (mInputPtr < mInputEnd) ?
- mCharBuffer[mInputPtr++] : nextChar();
- if (c != BYTE_LF) {
- --mInputPtr; // pushback if not 2-char/byte lf
- }
- }
- ++mInputRow;
- mInputRowStart = mInputPtr;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/ReaderSource.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/ReaderSource.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/ReaderSource.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/ReaderSource.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,205 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.IOException;
-import java.io.Reader;
-import java.net.URL;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.exc.WstxException;
-
-/**
- * Input source that reads input via a Reader.
- */
-public class ReaderSource
- extends BaseInputSource
-{
- final ReaderConfig mConfig;
-
- /**
- * Underlying Reader to read character data from
- */
- Reader mReader;
-
- /**
- * If true, will close the underlying Reader when this source is closed;
- * if false will leave it open.
- */
- final boolean mDoRealClose;
-
- int mInputProcessed = 0;
- int mInputRow = 1;
- int mInputRowStart = 0;
-
- public ReaderSource(ReaderConfig cfg, WstxInputSource parent, String fromEntity,
- String pubId, String sysId, URL src,
- Reader r, boolean realClose)
- {
- super(parent, fromEntity, pubId, sysId, src);
- mConfig = cfg;
- mReader = r;
- mDoRealClose = realClose;
- int bufSize = cfg.getInputBufferLength();
- mBuffer = cfg.allocFullCBuffer(bufSize);
- }
-
- /**
- * Method called to change the default offsets this source has. Generally
- * done when the underlying Reader had been partially read earlier (like
- * reading the xml declaration before starting real parsing).
- */
- public void setInputOffsets(int proc, int row, int rowStart)
- {
- mInputProcessed = proc;
- mInputRow = row;
- mInputRowStart = rowStart;
- }
-
- /**
- * Input location is easy to set, as we'll start from the beginning
- * of a File.
- */
- protected void doInitInputLocation(WstxInputData reader)
- {
- reader.mCurrInputProcessed = mInputProcessed;
- reader.mCurrInputRow = mInputRow;
- reader.mCurrInputRowStart = mInputRowStart;
- }
-
- /**
- * This is a hard-coded assumption, for now this source is
- * only created from external entities
- */
- public boolean fromInternalEntity() {
- return false;
- }
-
- public int readInto(WstxInputData reader)
- throws IOException, XMLStreamException
- {
- /* Shouldn't really try to read after closing, but it may be easier
- * for caller not to have to keep track of closure...
- */
- if (mBuffer == null) {
- return -1;
- }
- int count = mReader.read(mBuffer, 0, mBuffer.length);
- if (count < 1) {
- /* Let's prevent caller from accidentally being able to access
- * data, first.
- */
- mInputLast = 0;
- reader.mInputPtr = 0;
- reader.mInputEnd = 0;
- if (count == 0) {
- /* Sanity check; should never happen with correctly written
- * Readers:
- */
- throw new WstxException("Reader (of type "+mReader.getClass().getName()+") returned 0 characters, even when asked to read up to "+mBuffer.length, getLocation());
- }
- return -1;
- }
- reader.mInputBuffer = mBuffer;
- reader.mInputPtr = 0;
- mInputLast = count;
- reader.mInputEnd = count;
-
- return count;
- }
-
- public boolean readMore(WstxInputData reader, int minAmount)
- throws IOException, XMLStreamException
- {
- /* Shouldn't really try to read after closing, but it may be easier
- * for caller not to have to keep track of closure...
- */
- if (mBuffer == null) {
- return false;
- }
-
- int ptr = reader.mInputPtr;
- int currAmount = mInputLast - ptr;
-
- // Let's first adjust caller's data appropriately:
- /* Since we are essentially removing 'ptr' chars that we
- * have used already, they count as past chars. Also, since
- * offsets are reduced by 'ptr', need to adjust linefeed offset
- * marker as well.
- */
- reader.mCurrInputProcessed += ptr;
- reader.mCurrInputRowStart -= ptr;
-
- // Existing data to move?
- if (currAmount > 0) {
- System.arraycopy(mBuffer, ptr, mBuffer, 0, currAmount);
- minAmount -= currAmount;
- }
- reader.mInputBuffer = mBuffer;
- reader.mInputPtr = 0;
- mInputLast = currAmount;
-
- while (minAmount > 0) {
- int amount = mBuffer.length - currAmount;
- int actual = mReader.read(mBuffer, currAmount, amount);
- if (actual < 1) {
- if (actual == 0) { // sanity check:
- throw new WstxException("Reader (of type "+mReader.getClass().getName()+") returned 0 characters, even when asked to read up to "+amount, getLocation());
- }
- reader.mInputEnd = mInputLast = currAmount;
- return false;
- }
- currAmount += actual;
- minAmount -= actual;
- }
- reader.mInputEnd = mInputLast = currAmount;
- return true;
- }
-
- public void close()
- throws IOException
- {
- /* Buffer gets nullified by call to close() or closeCompletely(),
- * no need to call second time
- */
- if (mBuffer != null) { // so that it's ok to call multiple times
- closeAndRecycle(mDoRealClose);
- }
- }
-
- public void closeCompletely()
- throws IOException
- {
- /* Only need to call if the Reader is not yet null... since
- * buffer may have been cleaned by a call to close()
- */
- if (mReader != null) { // so that it's ok to call multiple times
- closeAndRecycle(true);
- }
- }
-
- private void closeAndRecycle(boolean fullClose)
- throws IOException
- {
- char[] buf = mBuffer;
-
- // Can we recycle buffers?
- if (buf != null) {
- mBuffer = null;
- mConfig.freeFullCBuffer(buf);
- }
-
- // How about Reader; close and/or recycle its buffers?
- if (mReader != null) {
- if (mReader instanceof BaseReader) {
- ((BaseReader) mReader).freeBuffers();
- }
- if (fullClose) {
- Reader r = mReader;
- mReader = null;
- r.close();
- }
- }
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/StreamBootstrapper.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/StreamBootstrapper.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/StreamBootstrapper.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/StreamBootstrapper.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1011 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ParsingErrorMsgs;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.exc.*;
-
-/**
- * Input bootstrap class used with streams, when encoding is not known
- * (when encoding is specified by application, a reader is constructed,
- * and then reader-based bootstrapper is used).
- * after last valid byte in the buffer
- */
- private StreamBootstrapper(String pubId, String sysId, byte[] data, int start, int end)
- {
- super(pubId, sysId);
- mIn = null;
- mRecycleBuffer = false;
- mByteBuffer = data;
- mInputPtr = start;
- mInputEnd = end;
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- /**
- * Factory method used when the underlying data provider is an
- * actual stream.
- */
- public static StreamBootstrapper getInstance(String pubId, String sysId, InputStream in)
- {
- return new StreamBootstrapper(pubId, sysId, in);
- }
-
- /**
- * Factory method used when the underlying data provider is a pre-allocated
- * block source, and no stream is used.
- * Additionally the buffer passed is not owned by the bootstrapper
- * or Reader that is created, so it is not to be recycled.
- */
- public static StreamBootstrapper getInstance(String pubId, String sysId, byte[] data, int start, int end)
- {
- return new StreamBootstrapper(pubId, sysId, data, start, end);
- }
-
- public Reader bootstrapInput(ReaderConfig cfg, boolean mainDoc, int xmlVersion)
- throws IOException, XMLStreamException
- {
- String normEnc = null;
-
- // First, let's get the buffers...
- int bufSize = cfg.getInputBufferLength();
- if (bufSize < MIN_BUF_SIZE) {
- bufSize = MIN_BUF_SIZE;
- }
- if (mByteBuffer == null) { // non-null if we were passed a buffer
- mByteBuffer = cfg.allocFullBBuffer(bufSize);
- }
-
- resolveStreamEncoding();
-
- if (hasXmlDecl()) {
- // note: readXmlDecl will set mXml11Handling too
- readXmlDecl(mainDoc, xmlVersion);
- if (mFoundEncoding != null) {
- normEnc = verifyXmlEncoding(mFoundEncoding);
- }
- } else {
- /* We'll actually then just inherit whatever main doc had...
- * (or in case there was no parent, just copy the 'unknown')
- */
- mXml11Handling = (XmlConsts.XML_V_11 == xmlVersion);
- }
-
- // Now, have we figured out the encoding?
-
- if (normEnc == null) { // not via xml declaration
- /* 21-Sep-2007, TSa: As with any non-UTF-8 encoding, declaration
- * isn't optional any more. Besides, we need that information
- * anyway to know which variant it is.
- */
- if (mEBCDIC) {
- if (mFoundEncoding == null || mFoundEncoding.length() == 0) {
- reportXmlProblem("Missing encoding declaration: underlying encoding looks like an EBCDIC variant, but no xml encoding declaration found");
- }
- // Hmmh. What should be the canonical name? Let's just use found encoding?
- normEnc = mFoundEncoding;
- } else if (mBytesPerChar == 2) { // UTF-16, BE/LE
- normEnc = mBigEndian ? CharsetNames.CS_UTF16BE : CharsetNames.CS_UTF16LE;
- } else if (mBytesPerChar == 4) { // UCS-4... ?
- /* 22-Mar-2005, TSa: JDK apparently has no way of dealing
- * with these encodings... not sure if and how it should
- * be dealt with, really. Name could be UCS-4xx... or
- * perhaps UTF-32xx
- */
- normEnc = mBigEndian ? CharsetNames.CS_UTF32BE : CharsetNames.CS_UTF32LE;
- } else {
- // Ok, default has to be UTF-8, as per XML specs
- normEnc = CharsetNames.CS_UTF8;
- }
- }
-
- mInputEncoding = normEnc;
-
- /* And then the reader. Let's figure out if we can use our own fast
- * implementations first:
- */
- BaseReader r;
-
- // Normalized, can thus use straight equality checks now
- if (normEnc == CharsetNames.CS_UTF8) {
- r = new UTF8Reader(cfg, mIn, mByteBuffer, mInputPtr, mInputEnd, mRecycleBuffer);
- } else if (normEnc == CharsetNames.CS_ISO_LATIN1) {
- r = new ISOLatinReader(cfg, mIn, mByteBuffer, mInputPtr, mInputEnd, mRecycleBuffer);
- } else if (normEnc == CharsetNames.CS_US_ASCII) {
- r = new AsciiReader(cfg, mIn, mByteBuffer, mInputPtr, mInputEnd, mRecycleBuffer);
- } else if (normEnc.startsWith(CharsetNames.CS_UTF32)) {
- // let's augment with actual endianness info
- if (normEnc == CharsetNames.CS_UTF32) {
- mInputEncoding = mBigEndian ? CharsetNames.CS_UTF32BE : CharsetNames.CS_UTF32LE;
- }
- r = new UTF32Reader(cfg, mIn, mByteBuffer, mInputPtr, mInputEnd,
- mRecycleBuffer, mBigEndian);
- } else {
- // Nah, JDK needs to try it
- // Ok; first, do we need to merge stuff back?
- InputStream in = mIn;
- if (mInputPtr < mInputEnd) {
- in = new MergedStream(cfg, in, mByteBuffer, mInputPtr, mInputEnd);
- }
- /* 20-Jan-2006, TSa: Ok; although it is possible to declare
- * stream as 'UTF-16', JDK may need help in figuring out
- * the right order, so let's be explicit:
- */
- if (normEnc == CharsetNames.CS_UTF16) {
- mInputEncoding = normEnc = mBigEndian ? CharsetNames.CS_UTF16BE : CharsetNames.CS_UTF16LE;
- }
- try {
- return new InputStreamReader(in, normEnc);
- } catch (UnsupportedEncodingException usex) {
- throw new WstxIOException("Unsupported encoding: "+usex.getMessage());
- }
- }
-
- if (mXml11Handling) {
- r.setXmlCompliancy(XmlConsts.XML_V_11);
- }
-
- return r;
- }
-
- /**
- * Since this class only gets used when encoding is not explicitly
- * passed, need use the encoding that was auto-detected...
- */
- public String getInputEncoding() {
- return mInputEncoding;
- }
-
- public int getInputTotal() {
- int total = mInputProcessed + mInputPtr;
- if (mBytesPerChar > 1) {
- total /= mBytesPerChar;
- }
- return total;
- }
-
- public int getInputColumn() {
- int col = mInputPtr - mInputRowStart;
- if (mBytesPerChar > 1) {
- col /= mBytesPerChar;
- }
- return col;
- }
-
- /*
- ////////////////////////////////////////
- // Internal methods, parsing
- ////////////////////////////////////////
- */
-
- /**
- * Method called to try to figure out physical encoding the underlying
- * input stream uses.
- */
- protected void resolveStreamEncoding()
- throws IOException, WstxException
- {
- // Let's first set defaults:
- mBytesPerChar = 0;
- mBigEndian = true;
-
- /* Ok; first just need 4 bytes for determining bytes-per-char from
- * BOM or first char(s) of likely xml declaration:
- */
- if (ensureLoaded(4)) {
- bomblock:
- do { // BOM/auto-detection block
- int quartet = (mByteBuffer[0] << 24)
- | ((mByteBuffer[1] & 0xFF) << 16)
- | ((mByteBuffer[2] & 0xFF) << 8)
- | (mByteBuffer[3] & 0xFF);
-
- /* Handling of (usually) optional BOM (required for
- * multi-byte formats); first 32-bit charsets:
- */
- switch (quartet) {
- case 0x0000FEFF:
- mBigEndian = true;
- mInputPtr = mBytesPerChar = 4;
- break bomblock;
- case 0xFFFE0000: // UCS-4, LE?
- mInputPtr = mBytesPerChar = 4;
- mBigEndian = false;
- break bomblock;
- case 0x0000FFFE: // UCS-4, in-order...
- reportWeirdUCS4("2143");
- break bomblock;
- case 0x0FEFF0000: // UCS-4, in-order...
- reportWeirdUCS4("3412");
- break bomblock;
- }
-
- // Ok, if not, how about 16-bit encoding BOMs?
- int msw = quartet >>> 16;
- if (msw == 0xFEFF) { // UTF-16, BE
- mInputPtr = mBytesPerChar = 2;
- mBigEndian = true;
- break;
- }
- if (msw == 0xFFFE) { // UTF-16, LE
- mInputPtr = mBytesPerChar = 2;
- mBigEndian = false;
- break;
- }
-
- // And if not, then UTF-8 BOM?
- if ((quartet >>> 8) == 0xEFBBBF) { // UTF-8
- mInputPtr = 3;
- mBytesPerChar = 1;
- mBigEndian = true; // doesn't really matter
- break;
- }
-
- /* And if that wasn't succesful, how about auto-detection
- * for ' 0);
-
- // Let's update location markers to ignore BOM.
- mInputProcessed = -mInputPtr;
- mInputRowStart = mInputPtr;
- }
-
- /* Hmmh. If we haven't figured it out, let's just assume
- * UTF-8 as per XML specs:
- */
- mByteSizeFound = (mBytesPerChar != 0);
- if (!mByteSizeFound) {
- mBytesPerChar = 1;
- mBigEndian = true; // doesn't matter
- }
- }
-
- /**
- * @return Normalized encoding name
- */
- protected String verifyXmlEncoding(String enc)
- throws WstxException
- {
- enc = CharsetNames.normalize(enc);
-
- // Let's actually verify we got matching information:
- if (enc == CharsetNames.CS_UTF8) {
- verifyEncoding(enc, 1);
- } else if (enc == CharsetNames.CS_ISO_LATIN1) {
- verifyEncoding(enc, 1);
- } else if (enc == CharsetNames.CS_US_ASCII) {
- verifyEncoding(enc, 1);
- } else if (enc == CharsetNames.CS_UTF16) {
- // BOM is obligatory, to know the ordering
- /* 22-Mar-2005, TSa: Actually, since we don't have a
- * custom decoder, so the underlying JDK Reader may
- * have dealt with it transparently... so we can not
- * really throw an exception here.
- */
- //if (!mHadBOM) {
- //reportMissingBOM(enc);
- //}
- verifyEncoding(enc, 2);
- } else if (enc == CharsetNames.CS_UTF16LE) {
- verifyEncoding(enc, 2, false);
- } else if (enc == CharsetNames.CS_UTF16BE) {
- verifyEncoding(enc, 2, true);
-
- } else if (enc == CharsetNames.CS_UTF32) {
- // Do we require a BOM here? we can live without it...
- //if (!mHadBOM) {
- // reportMissingBOM(enc);
- //}
- verifyEncoding(enc, 4);
- } else if (enc == CharsetNames.CS_UTF32LE) {
- verifyEncoding(enc, 4, false);
- } else if (enc == CharsetNames.CS_UTF32BE) {
- verifyEncoding(enc, 4, true);
- }
- return enc;
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, loading input data
- /////////////////////////////////////////////////////
- */
-
- protected boolean ensureLoaded(int minimum)
- throws IOException
- {
- /* Let's assume here buffer has enough room -- this will always
- * be true for the limited used this method gets
- */
- int gotten = (mInputEnd - mInputPtr);
- while (gotten < minimum) {
- int count = (mIn == null) ? -1 : mIn.read(mByteBuffer, mInputEnd, mByteBuffer.length - mInputEnd);
- if (count < 1) {
- return false;
- }
- mInputEnd += count;
- gotten += count;
- }
- return true;
- }
-
- protected void loadMore()
- throws IOException, WstxException
- {
- /* Need to make sure offsets are properly updated for error
- * reporting purposes, and do this now while previous amounts
- * are still known.
- */
- /* Note: at this point these are all in bytes, not chars (for multibyte
- * encodings)
- */
- mInputProcessed += mInputEnd;
- mInputRowStart -= mInputEnd;
-
- mInputPtr = 0;
- mInputEnd = (mIn == null) ? -1 : mIn.read(mByteBuffer, 0, mByteBuffer.length);
- if (mInputEnd < 1) {
- throw new WstxEOFException(ParsingErrorMsgs.SUFFIX_IN_XML_DECL,
- getLocation());
- }
- }
-
- /*
- /////////////////////////////////////////////////////
- // Implementations of abstract parsing methods
- /////////////////////////////////////////////////////
- */
-
- protected void pushback() {
- if (mBytesPerChar < 0) {
- mInputPtr += mBytesPerChar;
- } else {
- mInputPtr -= mBytesPerChar;
- }
- }
-
- protected int getNext()
- throws IOException, WstxException
- {
- if (mBytesPerChar != 1) {
- if (mBytesPerChar == -1) { // need to translate
- return nextTranslated();
- }
- return nextMultiByte();
- }
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- return (b & 0xFF);
- }
-
-
- protected int getNextAfterWs(boolean reqWs)
- throws IOException, WstxException
- {
- int count;
-
- if (mBytesPerChar == 1) { // single byte
- count = skipSbWs();
- } else {
- if (mBytesPerChar == -1) { // translated
- count = skipTranslatedWs();
- } else { // multi byte
- count = skipMbWs();
- }
- }
-
- if (reqWs && count == 0) {
- reportUnexpectedChar(getNext(), ERR_XMLDECL_EXP_SPACE);
- }
-
- // inlined getNext()
- if (mBytesPerChar != 1) {
- if (mBytesPerChar == -1) { // translated
- return nextTranslated();
- }
- return nextMultiByte();
- }
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- return (b & 0xFF);
- }
-
- /**
- * @return First character that does not match expected, if any;
- * CHAR_NULL if match succeeded
- */
- protected int checkKeyword(String exp)
- throws IOException, WstxException
- {
- if (mBytesPerChar != 1) {
- if (mBytesPerChar == -1) {
- return checkTranslatedKeyword(exp);
- }
- return checkMbKeyword(exp);
- }
- return checkSbKeyword(exp);
- }
-
- protected int readQuotedValue(char[] kw, int quoteChar)
- throws IOException, WstxException
- {
- int i = 0;
- int len = kw.length;
- boolean simple = (mBytesPerChar == 1);
- boolean mb = !simple && (mBytesPerChar > 1);
-
- while (i < len) {
- int c;
-
- if (simple) {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- if (b == BYTE_NULL) {
- reportNull();
- }
- if (b == BYTE_CR || b == BYTE_LF) {
- skipSbLF(b);
- b = BYTE_LF;
- }
- c = (b & 0xFF);
- } else {
- if (mb) {
- c = nextMultiByte();
- if (c == CHAR_CR || c == CHAR_LF) {
- skipMbLF(c);
- c = CHAR_LF;
- }
- } else {
- c = nextTranslated();
- if (c == CHAR_CR || c == CHAR_LF) {
- skipTranslatedLF(c);
- c = CHAR_LF;
- }
- }
- }
-
- if (c == quoteChar) {
- return (i < len) ? i : -1;
- }
-
- if (i < len) {
- kw[i++] = (char) c;
- }
- }
-
- /* If we end up this far, we ran out of buffer space... let's let
- * caller figure that out, though
- */
- return -1;
- }
-
- protected boolean hasXmlDecl()
- throws IOException, WstxException
- {
- /* Separate handling for common and fast case; 1/variable byte
- * encodings that have ASCII subset:
- */
- if (mBytesPerChar == 1) {
- /* However... there has to be at least 6 bytes available; and if
- * so, can check the 'signature' easily:
- */
- if (ensureLoaded(6)) {
- if (mByteBuffer[mInputPtr] == '<'
- && mByteBuffer[mInputPtr+1] == '?'
- && mByteBuffer[mInputPtr+2] == 'x'
- && mByteBuffer[mInputPtr+3] == 'm'
- && mByteBuffer[mInputPtr+4] == 'l'
- && ((mByteBuffer[mInputPtr+5] & 0xFF) <= CHAR_SPACE)) {
-
- // Let's skip stuff so far:
- mInputPtr += 6;
- return true;
- }
- }
- } else if (mBytesPerChar == -1) { // translated (EBCDIC)
- if (ensureLoaded(6)) {
- int start = mInputPtr; // if we have to 'unread' chars
- if (nextTranslated() == '<'
- && nextTranslated() == '?'
- && nextTranslated() == 'x'
- && nextTranslated() == 'm'
- && nextTranslated() == 'l'
- && nextTranslated() <= CHAR_SPACE) {
- return true;
- }
- mInputPtr = start; // push data back
- }
- } else {
- // ... and then for slower fixed-multibyte encodings:
-
- // Is there enough data for checks?
- if (ensureLoaded (6 * mBytesPerChar)) {
- int start = mInputPtr; // if we have to 'unread' chars
- if (nextMultiByte() == '<'
- && nextMultiByte() == '?'
- && nextMultiByte() == 'x'
- && nextMultiByte() == 'm'
- && nextMultiByte() == 'l'
- && nextMultiByte() <= CHAR_SPACE) {
- return true;
- }
- mInputPtr = start; // push data back
- }
- }
-
- return false;
- }
-
- protected Location getLocation()
- {
- /* Ok; for fixed-size multi-byte encodings, need to divide numbers
- * to get character locations. For variable-length encodings the
- * good thing is that xml declaration only uses shortest codepoints,
- * ie. char count == byte count.
- */
- int total = mInputProcessed + mInputPtr;
- int col = mInputPtr - mInputRowStart;
-
- if (mBytesPerChar > 1) {
- total /= mBytesPerChar;
- col /= mBytesPerChar;
- }
-
- return new WstxInputLocation(null, mPublicId, mSystemId,
- total - 1, // 0-based
- mInputRow, col);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, single-byte access methods
- /////////////////////////////////////////////////////
- */
-
- protected byte nextByte()
- throws IOException, WstxException
- {
- if (mInputPtr >= mInputEnd) {
- loadMore();
- }
- return mByteBuffer[mInputPtr++];
- }
-
- protected int skipSbWs()
- throws IOException, WstxException
- {
- int count = 0;
-
- while (true) {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
-
- if ((b & 0xFF) > CHAR_SPACE) {
- --mInputPtr;
- break;
- }
- if (b == BYTE_CR || b == BYTE_LF) {
- skipSbLF(b);
- } else if (b == BYTE_NULL) {
- reportNull();
- }
- ++count;
- }
- return count;
- }
-
- protected void skipSbLF(byte lfByte)
- throws IOException, WstxException
- {
- if (lfByte == BYTE_CR) {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- if (b != BYTE_LF) {
- --mInputPtr; // pushback if not 2-char/byte lf
- }
- }
- ++mInputRow;
- mInputRowStart = mInputPtr;
- }
-
- /**
- * @return First character that does not match expected, if any;
- * CHAR_NULL if match succeeded
- */
- protected int checkSbKeyword(String expected)
- throws IOException, WstxException
- {
- int len = expected.length();
-
- for (int ptr = 1; ptr < len; ++ptr) {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
-
- if (b == BYTE_NULL) {
- reportNull();
- }
- if ((b & 0xFF) != expected.charAt(ptr)) {
- return (b & 0xFF);
- }
- }
-
- return CHAR_NULL;
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, multi-byte/translated access/checks
- /////////////////////////////////////////////////////
- */
-
- protected int nextMultiByte()
- throws IOException, WstxException
- {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- byte b2 = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- int c;
-
- if (mBytesPerChar == 2) {
- if (mBigEndian) {
- c = ((b & 0xFF) << 8) | (b2 & 0xFF);
- } else {
- c = (b & 0xFF) | ((b2 & 0xFF) << 8);
- }
- } else {
- // Has to be 4 bytes
- byte b3 = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- byte b4 = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
-
- if (mBigEndian) {
- c = (b << 24) | ((b2 & 0xFF) << 16)
- | ((b3 & 0xFF) << 8) | (b4 & 0xFF);
- } else {
- c = (b4 << 24) | ((b3 & 0xFF) << 16)
- | ((b2 & 0xFF) << 8) | (b & 0xFF);
- }
- }
-
- // Let's catch null chars early
- if (c == 0) {
- reportNull();
- }
- return c;
- }
-
- protected int nextTranslated()
- throws IOException, WstxException
- {
- byte b = (mInputPtr < mInputEnd) ?
- mByteBuffer[mInputPtr++] : nextByte();
- int ch = mSingleByteTranslation[b & 0xFF];
- if (ch < 0) { // special char... won't care for now
- ch = -ch;
- }
- return ch;
- }
-
- protected int skipMbWs()
- throws IOException, WstxException
- {
- int count = 0;
-
- while (true) {
- int c = nextMultiByte();
-
- if (c > CHAR_SPACE) {
- mInputPtr -= mBytesPerChar;
- break;
- }
- if (c == CHAR_CR || c == CHAR_LF) {
- skipMbLF(c);
- } else if (c == CHAR_NULL) {
- reportNull();
- }
- ++count;
- }
- return count;
- }
-
- protected int skipTranslatedWs()
- throws IOException, WstxException
- {
- int count = 0;
-
- while (true) {
- int c = nextTranslated();
-
- // Hmmh. Are we to accept NEL (0x85)?
- if (c > CHAR_SPACE && c != CHAR_NEL) {
- --mInputPtr;
- break;
- }
- if (c == CHAR_CR || c == CHAR_LF) {
- skipTranslatedLF(c);
- } else if (c == CHAR_NULL) {
- reportNull();
- }
- ++count;
- }
- return count;
- }
-
- protected void skipMbLF(int lf)
- throws IOException, WstxException
- {
- if (lf == CHAR_CR) {
- int c = nextMultiByte();
- if (c != CHAR_LF) {
- mInputPtr -= mBytesPerChar;
- }
- }
- ++mInputRow;
- mInputRowStart = mInputPtr;
- }
-
- protected void skipTranslatedLF(int lf)
- throws IOException, WstxException
- {
- if (lf == CHAR_CR) {
- int c = nextTranslated();
- if (c != CHAR_LF) {
- mInputPtr -= 1;
- }
- }
- ++mInputRow;
- mInputRowStart = mInputPtr;
- }
-
- /**
- * @return First character that does not match expected, if any;
- * CHAR_NULL if match succeeded
- */
- protected int checkMbKeyword(String expected)
- throws IOException, WstxException
- {
- int len = expected.length();
-
- for (int ptr = 1; ptr < len; ++ptr) {
- int c = nextMultiByte();
- if (c == BYTE_NULL) {
- reportNull();
- }
- if (c != expected.charAt(ptr)) {
- return c;
- }
- }
-
- return CHAR_NULL;
- }
-
- protected int checkTranslatedKeyword(String expected)
- throws IOException, WstxException
- {
- int len = expected.length();
-
- for (int ptr = 1; ptr < len; ++ptr) {
- int c = nextTranslated();
- if (c == BYTE_NULL) {
- reportNull();
- }
- if (c != expected.charAt(ptr)) {
- return c;
- }
- }
-
- return CHAR_NULL;
- }
-
- /*
- ////////////////////////////////////////
- // Other private methods:
- ////////////////////////////////////////
- */
-
- private void verifyEncoding(String id, int bpc)
- throws WstxException
- {
- if (mByteSizeFound) {
- /* Let's verify that if we matched an encoding, it's the same
- * as what was declared...
- */
- if (bpc != mBytesPerChar) {
- // [WSTX-138]: Needs to detect EBCDIC discrepancy
- if (mEBCDIC) {
- reportXmlProblem("Declared encoding '"+id+"' incompatible with auto-detected physical encoding (EBCDIC variant), can not decode input since actual code page not known");
- }
- reportXmlProblem("Declared encoding '"+id+"' uses "+bpc
- +" bytes per character; but physical encoding appeared to use "+mBytesPerChar+"; cannot decode");
- }
- }
- }
-
- private void verifyEncoding(String id, int bpc, boolean bigEndian)
- throws WstxException
- {
- if (mByteSizeFound) {
- verifyEncoding(id, bpc);
-
- if (bigEndian != mBigEndian) {
- String bigStr = bigEndian ? "big" : "little";
- reportXmlProblem
- ("Declared encoding '"+id+"' has different endianness ("
- +bigStr+" endian) than what physical ordering appeared to be; cannot decode");
- }
- }
- }
-
- private void reportWeirdUCS4(String type)
- throws IOException
- {
- throw new CharConversionException("Unsupported UCS-4 endianness ("+type+") detected");
- }
-
- /*
- private void reportMissingBOM(String enc)
- throws WstxException
- {
- throw new WstxException("Missing BOM for encoding '"+enc+"'; can not be omitted",
- getLocation());
- }
- */
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/TextEscaper.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/TextEscaper.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/TextEscaper.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/TextEscaper.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,88 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-public final class TextEscaper
-{
- private TextEscaper() { }
-
- /*
- /////////////////////////////////////////////////////////////
- // Static utility methods, for non-state-aware escaping
- /////////////////////////////////////////////////////////////
- */
-
- public static void writeEscapedAttrValue(Writer w, String value)
- throws IOException
- {
- int i = 0;
- int len = value.length();
- do {
- int start = i;
- char c = '\u0000';
-
- for (; i < len; ++i) {
- c = value.charAt(i);
- if (c == '<' || c == '&' || c == '"') {
- break;
- }
- }
- int outLen = i - start;
- if (outLen > 0) {
- w.write(value, start, outLen);
- }
- if (i < len) {
- if (c == '<') {
- w.write("<");
- } else if (c == '&') {
- w.write("&");
- } else if (c == '"') {
- w.write(""");
-
- }
- }
- } while (++i < len);
- }
-
- /**
- * Quoting method used when outputting content that will be part of
- * DTD (internal/external subset). Additional quoting is needed for
- * percentage char, which signals parameter entities.
- */
- public static void outputDTDText(Writer w, char[] ch, int offset, int len)
- throws IOException
- {
- int i = offset;
- len += offset;
- do {
- int start = i;
- char c = '\u0000';
-
- for (; i < len; ++i) {
- c = ch[i];
- if (c == '&' || c == '%' || c == '"') {
- break;
- }
- }
- int outLen = i - start;
- if (outLen > 0) {
- w.write(ch, start, outLen);
- }
- if (i < len) {
- if (c == '&') {
- /* Only need to quote to prevent it from being accidentally
- * taken as part of char entity...
- */
- w.write("&");
- } else if (c == '%') {
- // Need to quote, to prevent use as Param Entity marker
- w.write("%");
- } else if (c == '"') {
- // Need to quote assuming it encloses entity value
- w.write(""");
- }
- }
- } while (++i < len);
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF32Reader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF32Reader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF32Reader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF32Reader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,258 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.XmlConsts;
-
-/**
- * Since JDK does not come with UTF-32/UCS-4, let's implement a simple
- * decoder to use.
- */
-public final class UTF32Reader
- extends BaseReader
-{
- final boolean mBigEndian;
-
- boolean mXml11;
-
- /**
- * Although input is fine with full Unicode set, Java still uses
- * 16-bit chars, so we may have to split high-order chars into
- * surrogate pairs.
- */
- char mSurrogate = NULL_CHAR;
-
- /**
- * Total read character count; used for error reporting purposes
- */
- int mCharCount = 0;
-
- /**
- * Total read byte count; used for error reporting purposes
- */
- int mByteCount = 0;
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- public UTF32Reader(ReaderConfig cfg, InputStream in, byte[] buf, int ptr, int len,
- boolean recycleBuffer,
- boolean isBigEndian)
- {
- super(cfg, in, buf, ptr, len, recycleBuffer);
- mBigEndian = isBigEndian;
- }
-
- public void setXmlCompliancy(int xmlVersion)
- {
- mXml11 = (xmlVersion == XmlConsts.XML_V_11);
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- public int read(char[] cbuf, int start, int len)
- throws IOException
- {
- // Let's first ensure there's enough room...
- if (start < 0 || (start+len) > cbuf.length) {
- reportBounds(cbuf, start, len);
- }
- // Already EOF?
- if (mByteBuffer == null) {
- return -1;
- }
- if (len < 1) {
- return 0;
- }
-
- len += start;
- int outPtr = start;
-
- // Ok, first; do we have a surrogate from last round?
- if (mSurrogate != NULL_CHAR) {
- cbuf[outPtr++] = mSurrogate;
- mSurrogate = NULL_CHAR;
- // No need to load more, already got one char
- } else {
- /* Note: we'll try to avoid blocking as much as possible. As a
- * result, we only need to get 4 bytes for a full char.
- */
- int left = (mByteBufferEnd - mBytePtr);
- if (left < 4) {
- if (!loadMore(left)) { // (legal) EOF?
- return -1;
- }
- }
- }
-
- byte[] buf = mByteBuffer;
-
- main_loop:
- while (outPtr < len) {
- int ptr = mBytePtr;
- int ch;
-
- if (mBigEndian) {
- ch = (buf[ptr] << 24) | ((buf[ptr+1] & 0xFF) << 16)
- | ((buf[ptr+2] & 0xFF) << 8) | (buf[ptr+3] & 0xFF);
- } else {
- ch = (buf[ptr] & 0xFF) | ((buf[ptr+1] & 0xFF) << 8)
- | ((buf[ptr+2] & 0xFF) << 16) | (buf[ptr+3] << 24);
- }
- mBytePtr += 4;
-
- // Does it need to be split to surrogates?
- // (also, we can and need to verify illegal chars)
- if (ch >= 0x7F) {
- if (ch <= 0x9F) {
- if (mXml11) { // high-order ctrl char detection...
- if (ch != 0x85) {
- reportInvalid(ch, outPtr-start, "(can only be included via entity in xml 1.1)");
- }
- ch = CONVERT_NEL_TO;
- }
- } else if (ch >= 0xD800) {
- // Illegal?
- if (ch > XmlConsts.MAX_UNICODE_CHAR) {
- reportInvalid(ch, outPtr-start,
- "(above "+Integer.toHexString(XmlConsts.MAX_UNICODE_CHAR)+") ");
- }
- if (ch > 0xFFFF) { // need to split into surrogates?
- ch -= 0x10000; // to normalize it starting with 0x0
- cbuf[outPtr++] = (char) (0xD800 + (ch >> 10));
- // hmmh. can this ever be 0? (not legal, at least?)
- ch = (0xDC00 | (ch & 0x03FF));
- // Room for second part?
- if (outPtr >= len) { // nope
- mSurrogate = (char) ch;
- break main_loop;
- }
- } else { // in 16-bit range... just need validity checks
- if (ch < 0xE000) {
- reportInvalid(ch, outPtr-start, "(a surrogate char) ");
- } else if (ch >= 0xFFFE) {
- reportInvalid(ch, outPtr-start, "");
- }
- }
- } else if (ch == 0x2028 && mXml11) { // LSEP
- ch = CONVERT_LSEP_TO;
- }
- }
- cbuf[outPtr++] = (char) ch;
- if (mBytePtr >= mByteBufferEnd) {
- break main_loop;
- }
- }
-
- len = outPtr - start;
- mCharCount += len;
- return len;
- }
-
- /*
- ////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////
- */
-
- private void reportUnexpectedEOF(int gotBytes, int needed)
- throws IOException
- {
- int bytePos = mByteCount + gotBytes;
- int charPos = mCharCount;
-
- throw new CharConversionException("Unexpected EOF in the middle of a 4-byte UTF-32 char: got "
- +gotBytes+", needed "+needed
- +", at char #"+charPos+", byte #"+bytePos+")");
- }
-
- private void reportInvalid(int value, int offset, String msg)
- throws IOException
- {
- int bytePos = mByteCount + mBytePtr - 1;
- int charPos = mCharCount + offset;
-
- throw new CharConversionException("Invalid UTF-32 character 0x"
- +Integer.toHexString(value)
- +msg+" at char #"+charPos+", byte #"+bytePos+")");
- }
-
- /**
- * @param available Number of "unused" bytes in the input buffer
- *
- * @return True, if enough bytes were read to allow decoding of at least
- * one full character; false if EOF was encountered instead.
- */
- private boolean loadMore(int available)
- throws IOException
- {
- mByteCount += (mByteBufferEnd - available);
-
- // Bytes that need to be moved to the beginning of buffer?
- if (available > 0) {
- /* 11-Nov-2008, TSa: can only move if we own the buffer; otherwise
- * we are stuck with the data.
- */
- if (mBytePtr > 0 && canModifyBuffer()) {
- for (int i = 0; i < available; ++i) {
- mByteBuffer[i] = mByteBuffer[mBytePtr+i];
- }
- mBytePtr = 0;
- mByteBufferEnd = available;
- }
- } else {
- /* Ok; here we can actually reasonably expect an EOF,
- * so let's do a separate read right away:
- */
- int count = readBytes();
- if (count < 1) {
- if (count < 0) { // -1
- freeBuffers(); // to help GC?
- return false;
- }
- // 0 count is no good; let's err out
- reportStrangeStream();
- }
- }
-
- /* Need at least 4 bytes; if we don't get that many, it's an
- * error.
- */
- while (mByteBufferEnd < 4) {
- int count = readBytesAt(mByteBufferEnd);
- if (count < 1) {
- if (count < 0) { // -1, EOF... no good!
- freeBuffers(); // to help GC?
- reportUnexpectedEOF(mByteBufferEnd, 4);
- }
- // 0 count is no good; let's err out
- reportStrangeStream();
- }
- }
- return true;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF8Reader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF8Reader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF8Reader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF8Reader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,420 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.XmlConsts;
-
-/**
- * Optimized Reader that reads UTF-8 encoded content from an input stream.
- * In addition to doing (hopefully) optimal conversion, it can also take
- * array of "pre-read" (leftover) bytes; this is necessary when preliminary
- * stream/reader is trying to figure out XML encoding.
- */
-public final class UTF8Reader
- extends BaseReader
-{
- boolean mXml11 = false;
-
- char mSurrogate = NULL_CHAR;
-
- /**
- * Total read character count; used for error reporting purposes
- */
- int mCharCount = 0;
-
- /**
- * Total read byte count; used for error reporting purposes
- */
- int mByteCount = 0;
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- public UTF8Reader(ReaderConfig cfg, InputStream in, byte[] buf, int ptr, int len,
- boolean recycleBuffer)
- {
- super(cfg, in, buf, ptr, len, recycleBuffer);
- }
-
- public void setXmlCompliancy(int xmlVersion)
- {
- mXml11 = (xmlVersion == XmlConsts.XML_V_11);
- }
-
- /*
- ////////////////////////////////////////
- // Public API
- ////////////////////////////////////////
- */
-
- public int read(char[] cbuf, int start, int len)
- throws IOException
- {
- // Let's first ensure there's enough room...
- if (start < 0 || (start+len) > cbuf.length) {
- reportBounds(cbuf, start, len);
- }
- // Already EOF?
- if (mByteBuffer == null) {
- return -1;
- }
- if (len < 1) { // dummy call?
- return 0;
- }
-
- len += start;
- int outPtr = start;
-
- // Ok, first; do we have a surrogate from last round?
- if (mSurrogate != NULL_CHAR) {
- cbuf[outPtr++] = mSurrogate;
- mSurrogate = NULL_CHAR;
- // No need to load more, already got one char
- } else {
- /* To prevent unnecessary blocking (esp. with network streams),
- * we'll only require decoding of a single char
- */
- int left = (mByteBufferEnd - mBytePtr);
-
- /* So; only need to load more if we can't provide at least
- * one more character. We need not do thorough check here,
- * but let's check the common cases here: either completely
- * empty buffer (left == 0), or one with less than max. byte
- * count for a single char, and starting of a multi-byte
- * encoding (this leaves possibility of a 2/3-byte char
- * that is still fully accessible... but that can be checked
- * by the load method)
- */
-
- if (left < 4) {
- // Need to load more?
- if (left < 1 || mByteBuffer[mBytePtr] < 0) {
- if (!loadMore(left)) { // (legal) EOF?
- return -1;
- }
- }
- }
- }
-
- /* This may look silly, but using a local var is indeed faster
- * (if and when HotSpot properly gets things running) than
- * member variable...
- */
- byte[] buf = mByteBuffer;
- int inPtr = mBytePtr;
- int inBufLen = mByteBufferEnd;
-
- main_loop:
- while (outPtr < len) {
- // At this point we have at least one byte available
- int c = (int) buf[inPtr++];
-
- /* Let's first do the quickie loop for common case; 7-bit
- * ascii:
- */
- if (c >= 0) { // ascii? can probably loop, then
- if (c == 0x7F && mXml11) { // DEL illegal in xml1.1
- int bytePos = mByteCount + inPtr - 1;
- int charPos = mCharCount + (outPtr-start);
- reportInvalidXml11(c, bytePos, charPos);
- }
- cbuf[outPtr++] = (char) c; // ok since MSB is never on
-
- /* Ok, how many such chars could we safely process
- * without overruns? (will combine 2 in-loop comparisons
- * into just one)
- */
- int outMax = (len - outPtr); // max output
- int inMax = (inBufLen - inPtr); // max input
- int inEnd = inPtr + ((inMax < outMax) ? inMax : outMax);
-
- ascii_loop:
- while (true) {
- if (inPtr >= inEnd) {
- break main_loop;
- }
- c = ((int) buf[inPtr++]) & 0xFF;
- if (c >= 0x7F) { // DEL, or multi-byte
- break ascii_loop;
- }
- cbuf[outPtr++] = (char) c;
- }
- if (c == 0x7F) {
- if (mXml11) { // DEL illegal in xml1.1
- int bytePos = mByteCount + inPtr - 1;
- int charPos = mCharCount + (outPtr-start);
- reportInvalidXml11(c, bytePos, charPos);
- } // but not in xml 1.0
- cbuf[outPtr++] = (char) c;
- if(inPtr >= inEnd){
- break main_loop;
- }
- continue main_loop;
- }
- }
-
- int needed;
-
- // Ok; if we end here, we got multi-byte combination
- if ((c & 0xE0) == 0xC0) { // 2 bytes (0x0080 - 0x07FF)
- c = (c & 0x1F);
- needed = 1;
- } else if ((c & 0xF0) == 0xE0) { // 3 bytes (0x0800 - 0xFFFF)
- c = (c & 0x0F);
- needed = 2;
- } else if ((c & 0xF8) == 0xF0) {
- // 4 bytes; double-char BS, with surrogates and all...
- c = (c & 0x0F);
- needed = 3;
- } else {
- reportInvalidInitial(c & 0xFF, outPtr-start);
- // never gets here...
- needed = 1;
- }
- /* Do we have enough bytes? If not, let's just push back the
- * byte and leave, since we have already gotten at least one
- * char decoded. This way we will only block (with read from
- * input stream) when absolutely necessary.
- */
- if ((inBufLen - inPtr) < needed) {
- --inPtr;
- break main_loop;
- }
-
- int d = (int) buf[inPtr++];
- if ((d & 0xC0) != 0x080) {
- reportInvalidOther(d & 0xFF, outPtr-start);
- }
- c = (c << 6) | (d & 0x3F);
-
- if (needed > 1) { // needed == 1 means 2 bytes total
- d = buf[inPtr++]; // 3rd byte
- if ((d & 0xC0) != 0x080) {
- reportInvalidOther(d & 0xFF, outPtr-start);
- }
- c = (c << 6) | (d & 0x3F);
- if (needed > 2) { // 4 bytes? (need surrogates)
- d = buf[inPtr++];
- if ((d & 0xC0) != 0x080) {
- reportInvalidOther(d & 0xFF, outPtr-start);
- }
- c = (c << 6) | (d & 0x3F);
- if (c > XmlConsts.MAX_UNICODE_CHAR) {
- reportInvalid(c, outPtr-start,
- "(above "+Integer.toHexString(XmlConsts.MAX_UNICODE_CHAR)+") ");
- }
- /* Ugh. Need to mess with surrogates. Ok; let's inline them
- * there, then, if there's room: if only room for one,
- * need to save the surrogate for the rainy day...
- */
- c -= 0x10000; // to normalize it starting with 0x0
- cbuf[outPtr++] = (char) (0xD800 + (c >> 10));
- // hmmh. can this ever be 0? (not legal, at least?)
- c = (0xDC00 | (c & 0x03FF));
-
- // Room for second part?
- if (outPtr >= len) { // nope
- mSurrogate = (char) c;
- break main_loop;
- }
- // sure, let's fall back to normal processing:
- } else {
- /* Otherwise, need to check that 3-byte chars are
- * legal ones (should not expand to surrogates;
- * 0xFFFE and 0xFFFF are illegal)
- */
- if (c >= 0xD800) {
- // But first, let's check max chars:
- if (c < 0xE000) {
- reportInvalid(c, outPtr-start, "(a surrogate character) ");
- } else if (c >= 0xFFFE) {
- reportInvalid(c, outPtr-start, "");
- }
- } else if (mXml11 && c == 0x2028) { // LSEP?
- /* 10-May-2006, TSa: Since LSEP is "non-associative",
- * it needs additional handling. One way to do
- * this is to convert preceding \r to \n. This
- * should be implemented better when integrating
- * decoder and tokenizer.
- */
- if (outPtr > start && cbuf[outPtr-1] == '\r') {
- cbuf[outPtr-1] = '\n';
- }
- c = CONVERT_LSEP_TO;
- }
- }
- } else { // (needed == 1)
- if (mXml11) { // high-order ctrl char detection...
- if (c <= 0x9F) {
- if (c == 0x85) { // NEL, let's convert?
- c = CONVERT_NEL_TO;
- } else if (c >= 0x7F) { // DEL, ctrl chars
- int bytePos = mByteCount + inPtr - 1;
- int charPos = mCharCount + (outPtr-start);
- reportInvalidXml11(c, bytePos, charPos);
- }
- }
- }
- }
- cbuf[outPtr++] = (char) c;
- if (inPtr >= inBufLen) {
- break main_loop;
- }
- }
-
- mBytePtr = inPtr;
- len = outPtr - start;
- mCharCount += len;
- return len;
- }
-
- /*
- ////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////
- */
-
- private void reportInvalidInitial(int mask, int offset)
- throws IOException
- {
- // input (byte) ptr has been advanced by one, by now:
- int bytePos = mByteCount + mBytePtr - 1;
- int charPos = mCharCount + offset + 1;
-
- throw new CharConversionException("Invalid UTF-8 start byte 0x"
- +Integer.toHexString(mask)
- +" (at char #"+charPos+", byte #"+bytePos+")");
- }
-
- private void reportInvalidOther(int mask, int offset)
- throws IOException
- {
- int bytePos = mByteCount + mBytePtr - 1;
- int charPos = mCharCount + offset;
-
- throw new CharConversionException("Invalid UTF-8 middle byte 0x"
- +Integer.toHexString(mask)
- +" (at char #"+charPos+", byte #"+bytePos+")");
- }
-
- private void reportUnexpectedEOF(int gotBytes, int needed)
- throws IOException
- {
- int bytePos = mByteCount + gotBytes;
- int charPos = mCharCount;
-
- throw new CharConversionException("Unexpected EOF in the middle of a multi-byte char: got "
- +gotBytes+", needed "+needed
- +", at char #"+charPos+", byte #"+bytePos+")");
- }
-
- private void reportInvalid(int value, int offset, String msg)
- throws IOException
- {
- int bytePos = mByteCount + mBytePtr - 1;
- int charPos = mCharCount + offset;
-
- throw new CharConversionException("Invalid UTF-8 character 0x"
- +Integer.toHexString(value)+msg
- +" at char #"+charPos+", byte #"+bytePos+")");
- }
-
- /**
- * @param available Number of "unused" bytes in the input buffer
- *
- * @return True, if enough bytes were read to allow decoding of at least
- * one full character; false if EOF was encountered instead.
- */
- private boolean loadMore(int available)
- throws IOException
- {
- mByteCount += (mByteBufferEnd - available);
-
- // Bytes that need to be moved to the beginning of buffer?
- if (available > 0) {
- /* 11-Nov-2008, TSa: can only move if we own the buffer; otherwise
- * we are stuck with the data.
- */
- if (mBytePtr > 0 && canModifyBuffer()) {
- for (int i = 0; i < available; ++i) {
- mByteBuffer[i] = mByteBuffer[mBytePtr+i];
- }
- mBytePtr = 0;
- mByteBufferEnd = available;
- }
- } else {
- /* Ok; here we can actually reasonably expect an EOF,
- * so let's do a separate read right away:
- */
- int count = readBytes();
- if (count < 1) {
- if (count < 0) { // -1
- freeBuffers(); // to help GC?
- return false;
- }
- // 0 count is no good; let's err out
- reportStrangeStream();
- }
- }
-
- /* We now have at least one byte... and that allows us to
- * calculate exactly how many bytes we need!
- */
- int c = (int) mByteBuffer[mBytePtr];
- if (c >= 0) { // single byte (ascii) char... cool, can return
- return true;
- }
-
- // Ok, a multi-byte char, let's check how many bytes we'll need:
- int needed;
- if ((c & 0xE0) == 0xC0) { // 2 bytes (0x0080 - 0x07FF)
- needed = 2;
- } else if ((c & 0xF0) == 0xE0) { // 3 bytes (0x0800 - 0xFFFF)
- needed = 3;
- } else if ((c & 0xF8) == 0xF0) {
- // 4 bytes; double-char BS, with surrogates and all...
- needed = 4;
- } else {
- reportInvalidInitial(c & 0xFF, 0);
- // never gets here... but compiler whines without this:
- needed = 1;
- }
-
- /* And then we'll just need to load up to that many bytes;
- * if an EOF is hit, that'll be an error. But we need not do
- * actual decoding here, just load enough bytes.
- */
- while ((mBytePtr + needed) > mByteBufferEnd) {
- int count = readBytesAt(mByteBufferEnd);
- if (count < 1) {
- if (count < 0) { // -1, EOF... no good!
- freeBuffers();
- reportUnexpectedEOF(mByteBufferEnd, needed);
- }
- // 0 count is no good; let's err out
- reportStrangeStream();
- }
- }
- return true;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF8Writer.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF8Writer.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/UTF8Writer.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/UTF8Writer.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,423 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.*;
-
-import com.ctc.wstx.api.WriterConfig;
-
-/**
- * Specialized buffering UTF-8 writer used by
- * {@link com.ctc.wstx.sw.XmlWriter}.
- * The main reason for custom version is to allow for efficient
- * buffer recycling; the second benefit is that encoder has less
- * overhead for short content encoding (compared to JDK default
- * codecs).
- */
-public final class UTF8Writer
- extends Writer
- implements CompletelyCloseable
-{
- private final static int DEFAULT_BUF_LEN = 4000;
-
- final static int SURR1_FIRST = 0xD800;
- final static int SURR1_LAST = 0xDBFF;
- final static int SURR2_FIRST = 0xDC00;
- final static int SURR2_LAST = 0xDFFF;
-
- final WriterConfig mConfig;
-
- final boolean mAutoCloseOutput;
-
- final OutputStream mOut;
-
- byte[] mOutBuffer;
-
- final int mOutBufferLast;
-
- int mOutPtr;
-
- /**
- * When outputting chars from BMP, surrogate pairs need to be coalesced.
- * To do this, both pairs must be known first; and since it is possible
- * pairs may be split, we need temporary storage for the first half
- */
- int mSurrogate = 0;
-
- public UTF8Writer(WriterConfig cfg, OutputStream out, boolean autoclose)
- {
- mConfig = cfg;
- mAutoCloseOutput = autoclose;
- mOut = out;
- mOutBuffer = (mConfig == null) ? new byte[DEFAULT_BUF_LEN] : cfg.allocFullBBuffer(DEFAULT_BUF_LEN);
- /* Max. expansion for a single char (in unmodified UTF-8) is
- * 4 bytes (or 3 depending on how you view it -- 4 when recombining
- * surrogate pairs)
- */
- mOutBufferLast = mOutBuffer.length - 4;
- mOutPtr = 0;
- }
-
- /*
- ////////////////////////////////////////////////////////
- // CompletelyCloseable impl
- ////////////////////////////////////////////////////////
- */
-
- public void closeCompletely() throws IOException
- {
- _close(true);
- }
-
- /*
- ////////////////////////////////////////////////////////
- // java.io.Writer implementation
- ////////////////////////////////////////////////////////
- */
-
- /* !!! 30-Nov-2006, TSa: Due to co-variance between Appendable and
- * Writer, this would not compile with javac 1.5, in 1.4 mode
- * (source and target set to "1.4". Not a huge deal, but since
- * the base impl is just fine, no point in overriding it.
- */
- /*
- public Writer append(char c)
- throws IOException
- // note: this is a JDK 1.5 method
- {
- write(c);
- return this;
- }
- */
-
- public void close() throws IOException
- {
- _close(mAutoCloseOutput);
- }
-
- public void flush()
- throws IOException
- {
- if (mOutPtr > 0 && mOutBuffer != null) {
- mOut.write(mOutBuffer, 0, mOutPtr);
- mOutPtr = 0;
- }
- mOut.flush();
- }
-
- public void write(char[] cbuf)
- throws IOException
- {
- write(cbuf, 0, cbuf.length);
- }
-
- public void write(char[] cbuf, int off, int len)
- throws IOException
- {
- if (len < 2) {
- if (len == 1) {
- write(cbuf[off]);
- }
- return;
- }
-
- // First: do we have a leftover surrogate to deal with?
- if (mSurrogate > 0) {
- char second = cbuf[off++];
- --len;
- write(_convertSurrogate(second));
- // will have at least one more char
- }
-
- int outPtr = mOutPtr;
- byte[] outBuf = mOutBuffer;
- int outBufLast = mOutBufferLast; // has 4 'spare' bytes
-
- // All right; can just loop it nice and easy now:
- len += off; // len will now be the end of input buffer
-
- output_loop:
- for (; off < len; ) {
- /* First, let's ensure we can output at least 4 bytes
- * (longest UTF-8 encoded codepoint):
- */
- if (outPtr >= outBufLast) {
- mOut.write(outBuf, 0, outPtr);
- outPtr = 0;
- }
-
- int c = cbuf[off++];
- // And then see if we have an Ascii char:
- if (c < 0x80) { // If so, can do a tight inner loop:
- outBuf[outPtr++] = (byte)c;
- // Let's calc how many ascii chars we can copy at most:
- int maxInCount = (len - off);
- int maxOutCount = (outBufLast - outPtr);
-
- if (maxInCount > maxOutCount) {
- maxInCount = maxOutCount;
- }
- maxInCount += off;
- ascii_loop:
- while (true) {
- if (off >= maxInCount) { // done with max. ascii seq
- continue output_loop;
- }
- c = cbuf[off++];
- if (c >= 0x80) {
- break ascii_loop;
- }
- outBuf[outPtr++] = (byte) c;
- }
- }
-
- // Nope, multi-byte:
- if (c < 0x800) { // 2-byte
- outBuf[outPtr++] = (byte) (0xc0 | (c >> 6));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- } else { // 3 or 4 bytes
- // Surrogates?
- if (c < SURR1_FIRST || c > SURR2_LAST) {
- outBuf[outPtr++] = (byte) (0xe0 | (c >> 12));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- continue;
- }
- // Yup, a surrogate:
- if (c > SURR1_LAST) { // must be from first range
- mOutPtr = outPtr;
- throwIllegal(c);
- }
- mSurrogate = c;
- // and if so, followed by another from next range
- if (off >= len) { // unless we hit the end?
- break;
- }
- c = _convertSurrogate(cbuf[off++]);
- if (c > 0x10FFFF) { // illegal, as per RFC 3629
- mOutPtr = outPtr;
- throwIllegal(c);
- }
- outBuf[outPtr++] = (byte) (0xf0 | (c >> 18));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 12) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- }
- }
- mOutPtr = outPtr;
- }
-
- public void write(int c)
- throws IOException
- {
- // First; do we have a left over surrogate?
- if (mSurrogate > 0) {
- c = _convertSurrogate(c);
- // If not, do we start with a surrogate?
- } else if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- // Illegal to get second part without first:
- if (c > SURR1_LAST) {
- throwIllegal(c);
- }
- // First part just needs to be held for now
- mSurrogate = c;
- return;
- }
-
- if (mOutPtr >= mOutBufferLast) { // let's require enough room, first
- mOut.write(mOutBuffer, 0, mOutPtr);
- mOutPtr = 0;
- }
-
- if (c < 0x80) { // ascii
- mOutBuffer[mOutPtr++] = (byte) c;
- } else {
- int ptr = mOutPtr;
- if (c < 0x800) { // 2-byte
- mOutBuffer[ptr++] = (byte) (0xc0 | (c >> 6));
- mOutBuffer[ptr++] = (byte) (0x80 | (c & 0x3f));
- } else if (c <= 0xFFFF) { // 3 bytes
- mOutBuffer[ptr++] = (byte) (0xe0 | (c >> 12));
- mOutBuffer[ptr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- mOutBuffer[ptr++] = (byte) (0x80 | (c & 0x3f));
- } else { // 4 bytes
- if (c > 0x10FFFF) { // illegal, as per RFC 3629
- throwIllegal(c);
- }
- mOutBuffer[ptr++] = (byte) (0xf0 | (c >> 18));
- mOutBuffer[ptr++] = (byte) (0x80 | ((c >> 12) & 0x3f));
- mOutBuffer[ptr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- mOutBuffer[ptr++] = (byte) (0x80 | (c & 0x3f));
- }
- mOutPtr = ptr;
- }
- }
-
- public void write(String str)
- throws IOException
- {
- write(str, 0, str.length());
- }
-
- public void write(String str, int off, int len)
- throws IOException
- {
- if (len < 2) {
- if (len == 1) {
- write(str.charAt(off));
- }
- return;
- }
-
- // First: do we have a leftover surrogate to deal with?
- if (mSurrogate > 0) {
- char second = str.charAt(off++);
- --len;
- write(_convertSurrogate(second));
- // will have at least one more char (case of 1 char was checked earlier on)
- }
-
- int outPtr = mOutPtr;
- byte[] outBuf = mOutBuffer;
- int outBufLast = mOutBufferLast; // has 4 'spare' bytes
-
- // All right; can just loop it nice and easy now:
- len += off; // len will now be the end of input buffer
-
- output_loop:
- for (; off < len; ) {
- /* First, let's ensure we can output at least 4 bytes
- * (longest UTF-8 encoded codepoint):
- */
- if (outPtr >= outBufLast) {
- mOut.write(outBuf, 0, outPtr);
- outPtr = 0;
- }
-
- int c = str.charAt(off++);
- // And then see if we have an Ascii char:
- if (c < 0x80) { // If so, can do a tight inner loop:
- outBuf[outPtr++] = (byte)c;
- // Let's calc how many ascii chars we can copy at most:
- int maxInCount = (len - off);
- int maxOutCount = (outBufLast - outPtr);
-
- if (maxInCount > maxOutCount) {
- maxInCount = maxOutCount;
- }
- maxInCount += off;
- ascii_loop:
- while (true) {
- if (off >= maxInCount) { // done with max. ascii seq
- continue output_loop;
- }
- c = str.charAt(off++);
- if (c >= 0x80) {
- break ascii_loop;
- }
- outBuf[outPtr++] = (byte) c;
- }
- }
-
- // Nope, multi-byte:
- if (c < 0x800) { // 2-byte
- outBuf[outPtr++] = (byte) (0xc0 | (c >> 6));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- } else { // 3 or 4 bytes
- // Surrogates?
- if (c < SURR1_FIRST || c > SURR2_LAST) {
- outBuf[outPtr++] = (byte) (0xe0 | (c >> 12));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- continue;
- }
- // Yup, a surrogate:
- if (c > SURR1_LAST) { // must be from first range
- mOutPtr = outPtr;
- throwIllegal(c);
- }
- mSurrogate = c;
- // and if so, followed by another from next range
- if (off >= len) { // unless we hit the end?
- break;
- }
- c = _convertSurrogate(str.charAt(off++));
- if (c > 0x10FFFF) { // illegal, as per RFC 3629
- mOutPtr = outPtr;
- throwIllegal(c);
- }
- outBuf[outPtr++] = (byte) (0xf0 | (c >> 18));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 12) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | ((c >> 6) & 0x3f));
- outBuf[outPtr++] = (byte) (0x80 | (c & 0x3f));
- }
- }
- mOutPtr = outPtr;
- }
-
- /*
- ////////////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////////////
- */
-
- private final void _close(boolean forceClosing)
- throws IOException
- {
- byte[] buf = mOutBuffer;
- if (buf != null) {
- mOutBuffer = null;
- if (mOutPtr > 0) {
- mOut.write(buf, 0, mOutPtr);
- mOutPtr = 0;
- }
- if (mConfig != null) {
- mConfig.freeFullBBuffer(buf);
- }
- }
-
- if (forceClosing) {
- mOut.close();
- }
-
- /* Let's 'flush' orphan surrogate, no matter what; but only
- * after cleanly closing everything else.
- */
- int code = mSurrogate;
- if (code > 0) {
- mSurrogate = 0;
- throwIllegal(code);
- }
- }
-
- /**
- * Method called to calculate UTF codepoint, from a surrogate pair.
- */
- private final int _convertSurrogate(int secondPart)
- throws IOException
- {
- int firstPart = mSurrogate;
- mSurrogate = 0;
-
- // Ok, then, is the second part valid?
- if (secondPart < SURR2_FIRST || secondPart > SURR2_LAST) {
- throw new IOException("Broken surrogate pair: first char 0x"+Integer.toHexString(firstPart)+", second 0x"+Integer.toHexString(secondPart)+"; illegal combination");
- }
- return 0x10000 + ((firstPart - SURR1_FIRST) << 10) + (secondPart - SURR2_FIRST);
- }
-
- private void throwIllegal(int code)
- throws IOException
- {
- if (code > 0x10FFFF) { // over max?
- throw new IOException("Illegal character point (0x"+Integer.toHexString(code)+") to output; max is 0x10FFFF as per RFC 3629");
- }
- if (code >= SURR1_FIRST) {
- if (code <= SURR1_LAST) { // Unmatched first part (closing without second part?)
- throw new IOException("Unmatched first part of surrogate pair (0x"+Integer.toHexString(code)+")");
- }
- throw new IOException("Unmatched second part of surrogate pair (0x"+Integer.toHexString(code)+")");
- }
-
- // should we ever get this?
- throw new IOException("Illegal character point (0x"+Integer.toHexString(code)+") to output");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputData.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputData.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputData.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputData.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,469 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import com.ctc.wstx.util.XmlChars;
-
-/**
- * Base class used by readers (specifically, by
- * {@link com.ctc.wstx.sr.StreamScanner}, and its sub-classes)
- * to encapsulate input buffer portion of the class. Philosophically
- * this should probably be done via containment (composition), not
- * sub-classing but for performance reason, this "core" class is generally
- * extended from instead.
- *
- * Main reason for the input data portion to be factored out of main
- * class is that this way it can also be passed to nested input handling
- * Objects, which can then manipulate input buffers of the caller,
- * efficiently.
- */
-public class WstxInputData
-{
- // // // Some well-known chars:
-
- /**
- * Null-character is used as return value from some method(s), since
- * it is not a legal character in an XML document.
- */
- public final static char CHAR_NULL = '\u0000';
- public final static char INT_NULL = 0;
-
- public final static char CHAR_SPACE = (char) 0x0020;
- public final static char INT_SPACE = 0x0020;
-
- /**
- * This constant defines the highest Unicode character allowed
- * in XML content.
- */
- public final static int MAX_UNICODE_CHAR = 0x10FFFF;
-
- /*
- ////////////////////////////////////////////////////
- // Character validity constants, structs
- ////////////////////////////////////////////////////
- */
-
- /**
- * We will only use validity array for first 256 characters, mostly
- * because after those characters it's easier to do fairly simple
- * block checks.
- */
- private final static int VALID_CHAR_COUNT = 0x100;
-
- // These are the same for both 1.0 and 1.1...
-// private final static int FIRST_VALID_FOR_FIRST = 0x0041; // 'A'
-// private final static int FIRST_VALID_FOR_REST = 0x002D; // '.'
-
- private final static byte NAME_CHAR_INVALID_B = (byte) 0;
- private final static byte NAME_CHAR_ALL_VALID_B = (byte) 1;
- private final static byte NAME_CHAR_VALID_NONFIRST_B = (byte) -1;
-
- private final static byte[] sCharValidity = new byte[VALID_CHAR_COUNT];
-
- static {
- /* First, since all valid-as-first chars are also valid-as-other chars,
- * we'll initialize common chars:
- */
- sCharValidity['_'] = NAME_CHAR_ALL_VALID_B;
- for (int i = 0, last = ('z' - 'a'); i <= last; ++i) {
- sCharValidity['A' + i] = NAME_CHAR_ALL_VALID_B;
- sCharValidity['a' + i] = NAME_CHAR_ALL_VALID_B;
- }
- // not all are fully valid, but
- for (int i = 0xC0; i < VALID_CHAR_COUNT; ++i) {
- sCharValidity[i] = NAME_CHAR_ALL_VALID_B;
- }
- // ... now we can 'revert' ones not fully valid:
- sCharValidity[0xD7] = NAME_CHAR_INVALID_B;
- sCharValidity[0xF7] = NAME_CHAR_INVALID_B;
-
- /* And then we can proceed with ones only valid-as-other.
- */
- sCharValidity['-'] = NAME_CHAR_VALID_NONFIRST_B;
- sCharValidity['.'] = NAME_CHAR_VALID_NONFIRST_B;
- sCharValidity[0xB7] = NAME_CHAR_VALID_NONFIRST_B;
- for (int i = '0'; i <= '9'; ++i) {
- sCharValidity[i] = NAME_CHAR_VALID_NONFIRST_B;
- }
- }
-
- /**
- * Public identifiers only use 7-bit ascii range.
- */
- private final static int VALID_PUBID_CHAR_COUNT = 0x80;
- private final static byte[] sPubidValidity = new byte[VALID_PUBID_CHAR_COUNT];
-// private final static byte PUBID_CHAR_INVALID_B = (byte) 0;
- private final static byte PUBID_CHAR_VALID_B = (byte) 1;
- static {
- for (int i = 0, last = ('z' - 'a'); i <= last; ++i) {
- sPubidValidity['A' + i] = PUBID_CHAR_VALID_B;
- sPubidValidity['a' + i] = PUBID_CHAR_VALID_B;
- }
- for (int i = '0'; i <= '9'; ++i) {
- sPubidValidity[i] = PUBID_CHAR_VALID_B;
- }
-
- // 3 main white space types are valid
- sPubidValidity[0x0A] = PUBID_CHAR_VALID_B;
- sPubidValidity[0x0D] = PUBID_CHAR_VALID_B;
- sPubidValidity[0x20] = PUBID_CHAR_VALID_B;
-
- // And many of punctuation/separator ascii chars too:
- sPubidValidity['-'] = PUBID_CHAR_VALID_B;
- sPubidValidity['\''] = PUBID_CHAR_VALID_B;
- sPubidValidity['('] = PUBID_CHAR_VALID_B;
- sPubidValidity[')'] = PUBID_CHAR_VALID_B;
- sPubidValidity['+'] = PUBID_CHAR_VALID_B;
- sPubidValidity[','] = PUBID_CHAR_VALID_B;
- sPubidValidity['.'] = PUBID_CHAR_VALID_B;
- sPubidValidity['/'] = PUBID_CHAR_VALID_B;
- sPubidValidity[':'] = PUBID_CHAR_VALID_B;
- sPubidValidity['='] = PUBID_CHAR_VALID_B;
- sPubidValidity['?'] = PUBID_CHAR_VALID_B;
- sPubidValidity[';'] = PUBID_CHAR_VALID_B;
- sPubidValidity['!'] = PUBID_CHAR_VALID_B;
- sPubidValidity['*'] = PUBID_CHAR_VALID_B;
- sPubidValidity['#'] = PUBID_CHAR_VALID_B;
- sPubidValidity['@'] = PUBID_CHAR_VALID_B;
- sPubidValidity['$'] = PUBID_CHAR_VALID_B;
- sPubidValidity['_'] = PUBID_CHAR_VALID_B;
- sPubidValidity['%'] = PUBID_CHAR_VALID_B;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Configuration
- ////////////////////////////////////////////////////
- */
-
- /**
- * Flag that indicates whether XML content is to be treated as per
- * XML 1.1 specification or not (if not, it'll use xml 1.0).
- */
- protected boolean mXml11 = false;
-
- /*
- ////////////////////////////////////////////////////
- // Current input data
- ////////////////////////////////////////////////////
- */
-
- /**
- * Current buffer from which data is read; generally data is read into
- * buffer from input source, but not always (especially when using nested
- * input contexts when expanding parsed entity references etc).
- */
- protected char[] mInputBuffer;
-
- /**
- * Pointer to next available character in buffer
- */
- protected int mInputPtr = 0;
-
- /**
- * Index of character after last available one in the buffer.
- */
- protected int mInputEnd = 0;
-
- /*
- ////////////////////////////////////////////////////
- // Current input location information
- ////////////////////////////////////////////////////
- */
-
- /**
- * Number of characters that were contained in previous blocks
- * (blocks that were already processed prior to the current buffer).
- */
- protected long mCurrInputProcessed = 0L;
-
- /**
- * Current row location of current point in input buffer, starting
- * from 1
- */
- protected int mCurrInputRow = 1;
-
- /**
- * Current index of the first character of the current row in input
- * buffer. Needed to calculate column position, if necessary; benefit
- * of not having column itself is that this only has to be updated
- * once per line.
- */
- protected int mCurrInputRowStart = 0;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////////
- */
-
- protected WstxInputData() {
- }
-
- /**
- * Note: Only public due to sub-classes needing to call this on
- * base class instance from different package (confusing?)
- */
- public void copyBufferStateFrom(WstxInputData src)
- {
- mInputBuffer = src.mInputBuffer;
- mInputPtr = src.mInputPtr;
- mInputEnd = src.mInputEnd;
-
- mCurrInputProcessed = src.mCurrInputProcessed;
- mCurrInputRow = src.mCurrInputRow;
- mCurrInputRowStart = src.mCurrInputRowStart;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public/package API, character classes
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method that can be used to check whether specified character
- * is a valid first character of an XML 1.0/1.1 name; except that
- * colon (:) is not recognized as a start char here: caller has
- * to verify it separately (since it generally affects namespace
- * mapping of a qualified name).
- */
- protected final boolean isNameStartChar(char c)
- {
- /* First, let's handle 7-bit ascii range (identical between xml
- * 1.0 and 1.1)
- */
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- return true;
- }
- if (c < 0x41) { // before 'A' just white space
- return false;
- }
- return (c <= 0x5A) || (c == '_'); // 'A' - 'Z' and '_' are ok
- }
- /* Ok, otherwise need to use a big honking bit sets... which
- * differ between 1.0 and 1.1
- */
- return mXml11 ? XmlChars.is11NameStartChar(c) : XmlChars.is10NameStartChar(c);
- }
-
- /**
- * Method that can be used to check whether specified character
- * is a valid character of an XML 1.0/1.1 name as any other char than
- * the first one; except that colon (:) is not recognized as valid here:
- * caller has to verify it separately (since it generally affects namespace
- * mapping of a qualified name).
- */
- protected final boolean isNameChar(char c)
- {
- // First, let's handle 7-bit ascii range
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- return true;
- }
- if (c <= 0x5A) {
- if (c >= 0x41) { // 'A' - 'Z' ok too
- return true;
- }
- // As are 0-9, '.' and '-'
- return (c >= 0x30 && c <= 0x39) || (c == '.') || (c == '-');
- }
- return (c == 0x5F); // '_' is ok too
- }
- return mXml11 ? XmlChars.is11NameChar(c) : XmlChars.is10NameChar(c);
- }
-
- public final static boolean isNameStartChar(char c, boolean nsAware, boolean xml11)
- {
- /* First, let's handle 7-bit ascii range (identical between xml
- * 1.0 and 1.1)
- */
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- return true;
- }
- if (c < 0x41) { // before 'A' just white space (and colon)
- if (c == ':' && !nsAware) {
- return true;
- }
- return false;
- }
- return (c <= 0x5A) || (c == '_'); // 'A' - 'Z' and '_' are ok
- }
- /* Ok, otherwise need to use a big honking bit sets... which
- * differ between 1.0 and 1.1
- */
- return xml11 ? XmlChars.is11NameStartChar(c) : XmlChars.is10NameStartChar(c);
- }
-
- public final static boolean isNameChar(char c, boolean nsAware, boolean xml11)
- {
- // First, let's handle 7-bit ascii range
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- return true;
- }
- if (c <= 0x5A) {
- if (c >= 0x41) { // 'A' - 'Z' ok too
- return true;
- }
- // As are 0-9, '.' and '-'
- return (c >= 0x30 && c <= 0x39) || (c == '.') || (c == '-')
- || (c == ':' && !nsAware);
- }
- return (c == 0x5F); // '_' is ok too
- }
- return xml11 ? XmlChars.is11NameChar(c) : XmlChars.is10NameChar(c);
- }
-
- /**
- * Method that can be called to check whether given String contains
- * any characters that are not legal XML names.
- *
- * @return Index of the first illegal xml name characters, if any;
- * -1 if the name is completely legal
- */
- public final static int findIllegalNameChar(String name, boolean nsAware, boolean xml11)
- {
- int len = name.length();
- if (len < 1) {
- return -1;
- }
-
- char c = name.charAt(0);
-
- // First char legal?
- if (c <= 0x7A) { // 'z' or earlier
- if (c < 0x61) { // 'a' - 'z' (0x61 - 0x7A) are ok
- if (c < 0x41) { // before 'A' just white space (except colon)
- if (c != ':' || nsAware) { // ':' == 0x3A
- return 0;
- }
- } else if ((c > 0x5A) && (c != '_')) {
- // 'A' - 'Z' and '_' are ok
- return 0;
- }
- }
- } else {
- if (xml11) {
- if (!XmlChars.is11NameStartChar(c)) {
- return 0;
- }
- } else {
- if (!XmlChars.is10NameStartChar(c)) {
- return 0;
- }
- }
- }
-
- for (int i = 1; i < len; ++i) {
- c = name.charAt(i);
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- continue;
- }
- if (c <= 0x5A) {
- if (c >= 0x41) { // 'A' - 'Z' ok too
- continue;
- }
- // As are 0-9, '.' and '-'
- if ((c >= 0x30 && c <= 0x39) || (c == '.') || (c == '-')) {
- continue;
- }
- // And finally, colon, in non-ns-aware mode
- if (c == ':' && !nsAware) { // ':' == 0x3A
- continue;
- }
- } else if (c == 0x5F) { // '_' is ok too
- continue;
- }
- } else {
- if (xml11) {
- if (XmlChars.is11NameChar(c)) {
- continue;
- }
- } else {
- if (XmlChars.is10NameChar(c)) {
- continue;
- }
- }
- }
- return i;
- }
-
- return -1;
- }
-
- public final static int findIllegalNmtokenChar(String nmtoken, boolean nsAware, boolean xml11)
- {
- int len = nmtoken.length();
- // No special handling for the first char, just the loop
- for (int i = 1; i < len; ++i) {
- char c = nmtoken.charAt(i);
- if (c <= 0x7A) { // 'z' or earlier
- if (c >= 0x61) { // 'a' - 'z' are ok
- continue;
- }
- if (c <= 0x5A) {
- if (c >= 0x41) { // 'A' - 'Z' ok too
- continue;
- }
- // As are 0-9, '.' and '-'
- if ((c >= 0x30 && c <= 0x39) || (c == '.') || (c == '-')) {
- continue;
- }
- // And finally, colon, in non-ns-aware mode
- if (c == ':' && !nsAware) { // ':' == 0x3A
- continue;
- }
- } else if (c == 0x5F) { // '_' is ok too
- continue;
- }
- } else {
- if (xml11) {
- if (XmlChars.is11NameChar(c)) {
- continue;
- }
- } else {
- if (XmlChars.is10NameChar(c)) {
- continue;
- }
- }
- }
- return i;
- }
- return -1;
- }
-
- public final static boolean isSpaceChar(char c)
- {
- return (c <= CHAR_SPACE);
- }
-
- public static String getCharDesc(char c)
- {
- int i = (int) c;
- if (Character.isISOControl(c)) {
- return "(CTRL-CHAR, code "+i+")";
- }
- if (i > 255) {
- return "'"+c+"' (code "+i+" / 0x"+Integer.toHexString(i)+")";
- }
- return "'"+c+"' (code "+i+")";
- }
-
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputLocation.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputLocation.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputLocation.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputLocation.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,176 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import java.io.Serializable;
-import javax.xml.stream.Location;
-
-import org.codehaus.stax2.XMLStreamLocation2;
-
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Basic implementation of {@link Location}, used by Wstx readers.
- */
-public class WstxInputLocation
- implements Serializable, XMLStreamLocation2
-{
- private static final long serialVersionUID = 1L;
-
- private final static WstxInputLocation sEmptyLocation
- = new WstxInputLocation(null, "", "", -1, -1, -1);
-
- /**
- * Enclosing (parent) input location; location from which current
- * location is derived.
- */
- final protected WstxInputLocation mContext;
-
- final protected String mPublicId, mSystemId;
-
- final protected int mCharOffset;
- final protected int mCol, mRow;
-
- transient protected String mDesc = null;
-
- /**
- * @param ctxt Enclosing input location, if any
- */
- public WstxInputLocation(WstxInputLocation ctxt,
- String pubId, String sysId,
- int charOffset, int row, int col)
- {
- mContext = ctxt;
- mPublicId = pubId;
- mSystemId = sysId;
- /* Overflow? Can obviously only handle limited range of overflows,
- * but let's do that at least?
- */
- mCharOffset = (charOffset < 0) ? Integer.MAX_VALUE : charOffset;
- mCol = col;
- mRow = row;
- }
-
- public static WstxInputLocation getEmptyLocation() {
- return sEmptyLocation;
- }
-
- public int getCharacterOffset() { return mCharOffset; }
- public int getColumnNumber() { return mCol; }
- public int getLineNumber() { return mRow; }
-
- public String getPublicId() { return mPublicId; }
- public String getSystemId() { return mSystemId; }
-
- /*
- ////////////////////////////////////////////////////////
- // StAX 2 API:
- ////////////////////////////////////////////////////////
- */
-
- public XMLStreamLocation2 getContext() { return mContext; }
-
- /*
- ////////////////////////////////////////////////////////
- // Overridden standard methods
- ////////////////////////////////////////////////////////
- */
-
- public String toString()
- {
- if (mDesc == null) {
- StringBuffer sb;
- if (mContext != null) {
- sb = new StringBuffer(200);
- } else {
- sb = new StringBuffer(80);
- }
- appendDesc(sb);
- mDesc = sb.toString();
- }
- return mDesc;
- }
-
- public int hashCode() {
- return mCharOffset ^ mRow ^ mCol + (mCol << 3);
- }
-
- public boolean equals(Object o) {
- if (!(o instanceof Location)) {
- return false;
- }
- Location other = (Location) o;
- // char offset should be good enough, without row/col:
- if (other.getCharacterOffset() != getCharacterOffset()) {
- return false;
- }
- String otherPub = other.getPublicId();
- if (otherPub == null) {
- otherPub = "";
- }
- if (!otherPub.equals(mPublicId)) {
- return false;
- }
- String otherSys = other.getSystemId();
- if (otherSys == null) {
- otherSys = "";
- }
- return otherSys.equals(mSystemId);
- }
-
- /*
- ////////////////////////////////////////////////////////
- // Internal methods:
- ////////////////////////////////////////////////////////
- */
-
- private void appendDesc(StringBuffer sb)
- {
- String srcId;
-
- if (mSystemId != null) {
- sb.append("[row,col,system-id]: ");
- srcId = mSystemId;
- } else if (mPublicId != null) {
- sb.append("[row,col,public-id]: ");
- srcId = mPublicId;
- } else {
- sb.append("[row,col {unknown-source}]: ");
- srcId = null;
- }
- sb.append('[');
- sb.append(mRow);
- sb.append(',');
- sb.append(mCol);
-
- // Uncomment for testing, to see the char offset:
- //sb.append(" #").append(mCharOffset);
- //sb.append("{").append(System.identityHashCode(this)).append("}");
-
- if (srcId != null) {
- sb.append(',');
- sb.append('"');
- sb.append(srcId);
- sb.append('"');
- }
- sb.append(']');
- if (mContext != null) {
- StringUtil.appendLF(sb);
- sb.append(" from ");
- mContext.appendDesc(sb);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputSource.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputSource.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/WstxInputSource.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/WstxInputSource.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,230 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.io;
-
-import java.io.IOException;
-import java.net.URL;
-
-import javax.xml.stream.XMLStreamException;
-
-/**
- * Interface that defines API actual parsers (stream readers)
- * can use to read input from various input sources.
- * Needed to abstract out details of getting input from primary input
- * files, secondary (potentially cached) referenced documents, and from
- * parsed entities, as well as for allowing hierarchic location
- * information for error reporting.
- */
-public abstract class WstxInputSource
-{
- /**
- * Parent in input source stack
- */
- protected final WstxInputSource mParent;
-
- /**
- * Name/id of the entity that was expanded to produce this input source;
- * null if not entity-originated. Used for catching recursive expansions
- * of entities.
- */
- protected final String mFromEntity;
-
- /**
- * Scope of the reader when this entity was initially expanded. Snapshot
- * that will generally be used by the reader to match scoping
- * limitations, such as proper nesting entity expansion with respect
- * to element and declaration nesting.
- */
- protected int mScopeId = 0;
-
- /*
- //////////////////////////////////////////////////////////
- // Life-cycle:
- //////////////////////////////////////////////////////////
- */
-
- protected WstxInputSource(WstxInputSource parent, String fromEntity)
- {
- mParent = parent;
- mFromEntity = fromEntity;
- }
-
- /**
- * Method that can be called to override originally defined source.
- *
- * @param url New base URL to set; may be null.
- *
- * @since 4.0
- */
- public abstract void overrideSource(URL url);
-
- /*
- //////////////////////////////////////////////////////////
- // Basic accessors:
- //////////////////////////////////////////////////////////
- */
-
- public final WstxInputSource getParent() {
- return mParent;
- }
-
- /**
- * Method that checks if this input source expanded from the specified
- * entity id, directly or by ancestor.
- *
- * Note that entity ids are expected to have been interned (using
- * whatever uniqueness mechanism used), and thus can be simply
- * identity checked.
- */
- public boolean isOrIsExpandedFrom(String entityId)
- {
- if (entityId != null) { // should always be true
- WstxInputSource curr = this;
- while (curr != null) {
- if (entityId == curr.mFromEntity) {
- return true;
- }
- curr = curr.mParent;
- }
- }
- return false;
- }
-
- /**
- * @return True, if this input source was directly expanded from an
- * internal entity (general, parsed); false if not (from external
- * entity, DTD ext. subset, main document)
- */
- public abstract boolean fromInternalEntity();
-
- /*
- //////////////////////////////////////////////////////////
- // Location info:
- //////////////////////////////////////////////////////////
- */
-
- public abstract URL getSource();
-
- public abstract String getPublicId();
-
- public abstract String getSystemId();
-
- /**
- * Method usually called to get a parent location for another input
- * source. Works since at this point context (line, row, chars) information
- * has already been saved to this object.
- */
- protected abstract WstxInputLocation getLocation();
-
- public abstract WstxInputLocation getLocation(long total, int row, int col);
-
- public String getEntityId() { return mFromEntity; }
-
- public int getScopeId() { return mScopeId; }
-
- /*
- //////////////////////////////////////////////////////////
- // Actual input handling
- //////////////////////////////////////////////////////////
- */
-
- /**
- * Method called by Reader when current input has changed to come
- * from this input source. Should reset/initialize input location
- * information Reader keeps, for error messages to work ok.
- *
- * @param reader Reader whose data structures are to be used for
- * returning data read
- * @param currScopeId
- */
- public final void initInputLocation(WstxInputData reader, int currScopeId) {
- mScopeId = currScopeId;
- doInitInputLocation(reader);
- }
-
- protected abstract void doInitInputLocation(WstxInputData reader);
-
- /**
- * Method called to read at least one more char from input source, and
- * update input data appropriately.
- *
- * @return Number of characters read from the input source (at least 1),
- * if it had any input; -1 if input source has no more input.
- */
- public abstract int readInto(WstxInputData reader)
- throws IOException, XMLStreamException;
-
- /**
- * Method called by reader when it has to have at least specified number
- * of consequtive input characters in its buffer, and it currently does
- * not have. If so, it asks input source to do whatever it has to do
- * to try to get more data, if possible (including moving stuff in
- * input buffer if necessary and possible).
- *
- * @return True if input source was able to provide specific number of
- * characters or more; false if not. In latter case, source is free
- * to return zero or more characters any way.
- */
- public abstract boolean readMore(WstxInputData reader, int minAmount)
- throws IOException, XMLStreamException;
-
- /**
- * Method Reader calls when this input source is being stored, when
- * a nested input source gets used instead (due to entity expansion).
- * Needs to get location info from Reader and store it in this Object.
- */
- public abstract void saveContext(WstxInputData reader);
-
- /**
- * Method Reader calls when this input source is resumed as the
- * current source. Needs to update Reader's input location data
- * used for error messages etc.
- */
- public abstract void restoreContext(WstxInputData reader);
-
- /**
- * Method reader calls for this input source when it has encountered
- * EOF. This may or may not close the underlying stream/reader; what
- * happens depends on configuration
- */
- public abstract void close() throws IOException;
-
- /**
- * Method reader MAY call to force full closing of the underlying
- * input stream(s)/reader(s). No checks are done regarding configuration,
- * but input source object is to deal gracefully with multiple calls
- * (ie. it's not an error for reader to call this more than once).
- */
- public abstract void closeCompletely() throws IOException;
-
- /*
- //////////////////////////////////////////////////////////
- // Overridden standard methods:
- //////////////////////////////////////////////////////////
- */
-
- public String toString() {
- StringBuffer sb = new StringBuffer(80);
- sb.append("
- * Note about id context provider interface: while it'd be nice to
- * separate that part out, it is unfortunately closely tied to the
- * validation process. Hence it's directly implemented by this class.
- */
-public final class GenericMsvValidator
- extends XMLValidator
- implements com.sun.msv.grammar.IDContextProvider2
-{
- /*
- ///////////////////////////////////////////////////////////////////////
- // Configuration
- ///////////////////////////////////////////////////////////////////////
- */
-
- final protected XMLValidationSchema mParentSchema;
-
- final protected ValidationContext mContext;
-
- final protected DocumentDeclaration mVGM;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // State, helper objects
- ///////////////////////////////////////////////////////////////////////
- */
-
- final protected ArrayList mAcceptors = new ArrayList();
-
- protected Acceptor mCurrAcceptor = null;
-
- final protected TextAccumulator mTextAccumulator = new TextAccumulator();
-
- /**
- * Map that contains information about element id (values of attributes
- * or textual content with type ID) declarations and references
- */
- protected ElementIdMap mIdDefs;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // State, positions
- ///////////////////////////////////////////////////////////////////////
- */
-
- protected String mCurrAttrPrefix;
-
- protected String mCurrAttrLocalName;
-
- /**
- * Sometimes a problem object has to be temporarily
- * stored, and only reported later on. This happens
- * when exceptions can not be thrown via code outside
- * of Woodstox (like validation methods in MSV that do
- * callbacks).
- */
- protected XMLValidationProblem mProblem;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Helper objects
- ///////////////////////////////////////////////////////////////////////
- */
-
- final StringRef mErrorRef = new StringRef();
-
- /**
- * StartTagInfo instance need not be thread-safe, and it is not immutable
- * so let's reuse one instance during a single validation.
- */
- final StartTagInfo mStartTag = new StartTagInfo("", "", "", null, (IDContextProvider2) null);
-
- /**
- * This object provides limited access to attribute values of the
- * currently validated element.
- */
- final AttributeProxy mAttributeProxy;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Construction, configuration
- ///////////////////////////////////////////////////////////////////////
- */
-
- public GenericMsvValidator(XMLValidationSchema parent, ValidationContext ctxt,
- DocumentDeclaration vgm)
- {
- mParentSchema = parent;
- mContext = ctxt;
- mVGM = vgm;
-
- mCurrAcceptor = mVGM.createAcceptor();
- mAttributeProxy = new AttributeProxy(ctxt);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // IDContextProvider2 implementation:
- //
- // Core RelaxNG ValidationContext implementation
- // (org.relaxng.datatype.ValidationContext, base interface
- // of the id provider context)
- ///////////////////////////////////////////////////////////////////////
- */
-
- public String getBaseUri()
- {
- return mContext.getBaseUri();
- }
-
- public boolean isNotation(String notationName)
- {
- return mContext.isNotationDeclared(notationName);
- }
-
- public boolean isUnparsedEntity(String entityName)
- {
- return mContext.isUnparsedEntityDeclared(entityName);
- }
-
- public String resolveNamespacePrefix(String prefix)
- {
- return mContext.getNamespaceURI(prefix);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // IDContextProvider2 implementation, extensions over
- // core ValidationContext
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- *
- * Note: we have to throw a dummy marker exception, which merely
- * signals that a validation problem is to be reported.
- * This is obviously messy, but has to do for now.
- */
- public void onID(Datatype datatype, StringToken idToken)
- throws IllegalArgumentException
- {
- if (mIdDefs == null) {
- mIdDefs = new ElementIdMap();
- }
-
- int idType = datatype.getIdType();
- Location loc = mContext.getValidationLocation();
- PrefixedName elemPName = getElementPName();
- PrefixedName attrPName = getAttrPName();
-
- if (idType == Datatype.ID_TYPE_ID) {
- String idStr = idToken.literal.trim();
- ElementId eid = mIdDefs.addDefined(idStr, loc, elemPName, attrPName);
- // We can detect dups by checking if Location is the one we passed:
- if (eid.getLocation() != loc) {
- mProblem = new XMLValidationProblem(loc, "Duplicate id '"+idStr+"', first declared at "+eid.getLocation());
- mProblem.setReporter(this);
- }
- } else if (idType == Datatype.ID_TYPE_IDREF) {
- String idStr = idToken.literal.trim();
- mIdDefs.addReferenced(idStr, loc, elemPName, attrPName);
- } else if (idType == Datatype.ID_TYPE_IDREFS) {
- StringTokenizer tokens = new StringTokenizer(idToken.literal);
- while (tokens.hasMoreTokens()) {
- mIdDefs.addReferenced(tokens.nextToken(), loc, elemPName, attrPName);
- }
- } else { // sanity check
- throw new IllegalStateException("Internal error: unexpected ID datatype: "+datatype);
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XMLValidator implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- public XMLValidationSchema getSchema() {
- return mParentSchema;
- }
-
- /**
- * Method called to update information about the newly encountered (start)
- * element. At this point namespace information has been resolved, but
- * no DTD validation has been done. Validator is to do these validations,
- * including checking for attribute value (and existence) compatibility.
- */
- public void validateElementStart(String localName, String uri, String prefix)
- throws XMLStreamException
- {
-
- /* [WSTX-200]: If sub-tree we were to validate has ended, we
- * have no current acceptor, and must quite. Ideally we would
- * really handle this more cleanly but...
- */
- if (mCurrAcceptor == null) {
- return;
- }
-
- // Very first thing: do we have text collected?
- if (mTextAccumulator.hasText()) {
- doValidateText(mTextAccumulator);
- }
-
- /* 31-Mar-2006, TSa: MSV seems to require empty String for empty/no
- * namespace, not null.
- */
- if (uri == null) {
- uri = "";
- }
-
- /* Do we need to properly fill it? Or could we just put local name?
- * Looking at code, I do believe it's only used for error reporting
- * purposes...
- */
- //String qname = (prefix == null || prefix.length() == 0) ? localName : (prefix + ":" +localName);
- String qname = localName;
- mStartTag.reinit(uri, localName, qname, mAttributeProxy, this);
-
- mCurrAcceptor = mCurrAcceptor.createChildAcceptor(mStartTag, mErrorRef);
- /* As per documentation, the side-effect of getting the error message
- * is that we also get a recoverable non-null acceptor... thus, should
- * never (?) see null acceptor being returned
- */
- if (mErrorRef.str != null) {
- reportError(mErrorRef);
- }
- if (mProblem != null) { // pending problems (to throw exception on)?
- XMLValidationProblem p = mProblem;
- mProblem = null;
- mContext.reportProblem(p);
- }
- mAcceptors.add(mCurrAcceptor);
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix, String value)
- throws XMLStreamException
- {
- mCurrAttrLocalName = localName;
- mCurrAttrPrefix = prefix;
- if (mCurrAcceptor != null) {
-
- String qname = localName; // for now, let's assume we don't need prefixed version
- DatatypeRef typeRef = null; // for now, let's not care
-
- /* 31-Mar-2006, TSa: MSV seems to require empty String for empty/no
- * namespace, not null.
- */
- if (uri == null) {
- uri = "";
- }
-
- if (!mCurrAcceptor.onAttribute2(uri, localName, qname, value, this, mErrorRef, typeRef)
- || mErrorRef.str != null) {
- reportError(mErrorRef);
- }
- if (mProblem != null) { // pending problems (to throw exception on)?
- XMLValidationProblem p = mProblem;
- mProblem = null;
- mContext.reportProblem(p);
- }
- }
- /* No normalization done by RelaxNG, is there? (at least nothing
- * visible to callers that is)
- */
- return null;
- }
-
- public String validateAttribute(String localName, String uri,
- String prefix,
- char[] valueChars, int valueStart,
- int valueEnd)
- throws XMLStreamException
- {
- int len = valueEnd - valueStart;
- /* This is very sub-optimal... but MSV doesn't deal with char
- * arrays.
- */
- return validateAttribute(localName, uri, prefix,
- new String(valueChars, valueStart, len));
- }
-
- public int validateElementAndAttributes()
- throws XMLStreamException
- {
- // Not handling any attributes
- mCurrAttrLocalName = mCurrAttrPrefix = "";
- if (mCurrAcceptor != null) {
- /* start tag info is still intact here (only attributes sent
- * since child acceptor was created)
- */
- if (!mCurrAcceptor.onEndAttributes(mStartTag, mErrorRef)
- || mErrorRef.str != null) {
- reportError(mErrorRef);
- }
-
- int stringChecks = mCurrAcceptor.getStringCareLevel();
- switch (stringChecks) {
- case Acceptor.STRING_PROHIBITED: // only WS
- return XMLValidator.CONTENT_ALLOW_WS;
- case Acceptor.STRING_IGNORE: // anything (mixed content models)
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- case Acceptor.STRING_STRICT: // validatable (data-oriented)
- return XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT;
- default:
- throw new IllegalArgumentException("Internal error: unexpected string care level value return by MSV: "+stringChecks);
- }
- }
-
- // If no acceptor, we are recovering, no need or use to validate text
- return CONTENT_ALLOW_ANY_TEXT;
- }
-
- /**
- * @return Validation state that should be effective for the parent
- * element state
- */
- public int validateElementEnd(String localName, String uri, String prefix)
- throws XMLStreamException
- {
- // Very first thing: do we have text collected?
- /* 27-Feb-2009, TSa: [WSTX-191]: Actually MSV expects us to call
- * validation anyway, in case there might be restriction(s) on
- * textual content. Otherwise we'll get an error.
- */
- doValidateText(mTextAccumulator);
-
- /* [WSTX-200]: need to avoid problems when doing sub-tree
- * validation... not a proper solution, but has to do for
- * now
- */
- int lastIx = mAcceptors.size()-1;
- if (lastIx < 0) {
- return XMLValidator.CONTENT_ALLOW_WS;
- }
-
- Acceptor acc = (Acceptor)mAcceptors.remove(lastIx);
- if (acc != null) { // may be null during error recovery? or not?
- if (!acc.isAcceptState(mErrorRef) || mErrorRef.str != null) {
- reportError(mErrorRef);
- }
- }
- if (lastIx == 0) { // root closed
- mCurrAcceptor = null;
- } else {
- mCurrAcceptor = (Acceptor) mAcceptors.get(lastIx-1);
- }
- if (mCurrAcceptor != null && acc != null) {
- if (!mCurrAcceptor.stepForward(acc, mErrorRef)
- || mErrorRef.str != null) {
- reportError(mErrorRef);
- }
- int stringChecks = mCurrAcceptor.getStringCareLevel();
- switch (stringChecks) {
- case Acceptor.STRING_PROHIBITED: // only WS
- return XMLValidator.CONTENT_ALLOW_WS;
- case Acceptor.STRING_IGNORE: // anything (mixed content models)
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- case Acceptor.STRING_STRICT: // validatable (data-oriented)
- return XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT;
- default:
- throw new IllegalArgumentException("Internal error: unexpected string care level value return by MSV: "+stringChecks);
- }
- }
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- }
-
- public void validateText(String text, boolean lastTextSegment)
- throws XMLStreamException
- {
- /* If we got here, then it's likely we do need to call onText2().
- * (not guaranteed, though; in case of multiple parallel validators,
- * only one of them may actually be interested)
- */
- mTextAccumulator.addText(text);
- if (lastTextSegment) {
- doValidateText(mTextAccumulator);
- }
- }
-
- public void validateText(char[] cbuf, int textStart, int textEnd,
- boolean lastTextSegment)
- throws XMLStreamException
- {
- /* If we got here, then it's likely we do need to call onText().
- * (not guaranteed, though; in case of multiple parallel validators,
- * only one of them may actually be interested)
- */
- mTextAccumulator.addText(cbuf, textStart, textEnd);
- if (lastTextSegment) {
- doValidateText(mTextAccumulator);
- }
- }
-
- public void validationCompleted(boolean eod)
- throws XMLStreamException
- {
- /* Ok, so, we should verify that there are no undefined
- * IDREF/IDREFS references. But only if we hit EOF, not
- * if validation was cancelled.
- */
- if (eod) {
- if (mIdDefs != null) {
- ElementId ref = mIdDefs.getFirstUndefined();
- if (ref != null) { // problem!
- String msg = "Undefined ID '"+ref.getId()
- +"': referenced from element <"
- +ref.getElemName()+">, attribute '"
- +ref.getAttrName()+"'";
- reportError(msg, ref.getLocation());
- }
- }
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Attribute info access
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // Access to type info
-
- public String getAttributeType(int index)
- {
- // !!! TBI
- return null;
- }
-
- public int getIdAttrIndex()
- {
- // !!! TBI
- return -1;
- }
-
- public int getNotationAttrIndex()
- {
- // !!! TBI
- return -1;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////////////////
- */
-
- PrefixedName getElementPName()
- {
- return PrefixedName.valueOf(mContext.getCurrentElementName());
- }
-
- PrefixedName getAttrPName()
- {
- return new PrefixedName(mCurrAttrPrefix, mCurrAttrLocalName);
- }
-
- void doValidateText(TextAccumulator textAcc)
- throws XMLStreamException
- {
- if (mCurrAcceptor != null) {
- String str = textAcc.getAndClear();
- DatatypeRef typeRef = null;
- if (!mCurrAcceptor.onText2(str, this, mErrorRef, typeRef)
- || mErrorRef.str != null) {
- reportError(mErrorRef);
- }
- }
- }
-
- private void reportError(StringRef errorRef)
- throws XMLStreamException
- {
- String msg = errorRef.str;
- errorRef.str = null;
- if (msg == null) {
- msg = "Unknown reason";
- }
- reportError(msg);
- }
-
- private void reportError(String msg)
- throws XMLStreamException
- {
- reportError(msg, mContext.getValidationLocation());
- }
-
- private void reportError(String msg, Location loc)
- throws XMLStreamException
- {
- XMLValidationProblem prob = new XMLValidationProblem(loc, msg, XMLValidationProblem.SEVERITY_ERROR);
- prob.setReporter(this);
- mContext.reportProblem(prob);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/msv/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/msv/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/msv/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/msv/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,5 +0,0 @@
-
- * Note: most of the SAX features are not configurable as of yet.
- * However, effort is made to recognize all existing standard features
- * and properties, to allow using code to figure out existing
- * capabilities automatically.
- */
-public class WstxSAXParserFactory
- extends SAXParserFactory
-{
- protected final WstxInputFactory mStaxFactory;
-
- /**
- * Sax feature that determines whether namespace declarations need
- * to be also reported as attributes or not.
- */
- protected boolean mFeatNsPrefixes = false;
-
- public WstxSAXParserFactory()
- {
- this(new WstxInputFactory());
- }
-
- /**
- * @since 4.0.8
- */
- public WstxSAXParserFactory(WstxInputFactory f)
- {
- mStaxFactory = f;
- /* defaults should be fine... except that for some weird
- * reason, by default namespace support is defined to be off
- */
- setNamespaceAware(true);
- }
-
- public boolean getFeature(String name)
- throws SAXNotRecognizedException, SAXNotSupportedException
- {
- SAXFeature stdFeat = SAXFeature.findByUri(name);
-
- if (stdFeat == SAXFeature.EXTERNAL_GENERAL_ENTITIES) {
- return mStaxFactory.getConfig().willSupportExternalEntities();
- } else if (stdFeat == SAXFeature.EXTERNAL_PARAMETER_ENTITIES) {
- return mStaxFactory.getConfig().willSupportExternalEntities();
- } else if (stdFeat == SAXFeature.IS_STANDALONE) {
- // Not known at this point...
- return false;
- } else if (stdFeat == SAXFeature.LEXICAL_HANDLER_PARAMETER_ENTITIES) {
- // !!! TODO:
- return false;
- } else if (stdFeat == SAXFeature.NAMESPACES) {
- return mStaxFactory.getConfig().willSupportNamespaces();
- } else if (stdFeat == SAXFeature.NAMESPACE_PREFIXES) {
- return mFeatNsPrefixes;
- } else if (stdFeat == SAXFeature.RESOLVE_DTD_URIS) {
- // !!! TODO:
- return false;
- } else if (stdFeat == SAXFeature.STRING_INTERNING) {
- return mStaxFactory.getConfig().willInternNames();
- } else if (stdFeat == SAXFeature.UNICODE_NORMALIZATION_CHECKING) {
- return false;
- } else if (stdFeat == SAXFeature.USE_ATTRIBUTES2) {
- return true;
- } else if (stdFeat == SAXFeature.USE_LOCATOR2) {
- return true;
- } else if (stdFeat == SAXFeature.USE_ENTITY_RESOLVER2) {
- return true;
- } else if (stdFeat == SAXFeature.VALIDATION) {
- return mStaxFactory.getConfig().willValidateWithDTD();
- } else if (stdFeat == SAXFeature.XMLNS_URIS) {
- /* !!! TODO: default value should be false... but not sure
- * if implementing that mode makes sense
- */
- return true;
- } else if (stdFeat == SAXFeature.XML_1_1) {
- return true;
- } else {
- throw new SAXNotRecognizedException("Feature '"+name+"' not recognized");
- }
- }
-
- public SAXParser newSAXParser()
- {
- return new WstxSAXParser(mStaxFactory, mFeatNsPrefixes);
- }
-
- public void setFeature(String name, boolean value)
- throws SAXNotRecognizedException, SAXNotSupportedException
- {
- boolean invalidValue = false;
- boolean readOnly = false;
- SAXFeature stdFeat = SAXFeature.findByUri(name);
-
- if (stdFeat == SAXFeature.EXTERNAL_GENERAL_ENTITIES) {
- mStaxFactory.getConfig().doSupportExternalEntities(value);
- } else if (stdFeat == SAXFeature.EXTERNAL_PARAMETER_ENTITIES) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.IS_STANDALONE) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.LEXICAL_HANDLER_PARAMETER_ENTITIES) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.NAMESPACES) {
- mStaxFactory.getConfig().doSupportNamespaces(value);
- } else if (stdFeat == SAXFeature.NAMESPACE_PREFIXES) {
- mFeatNsPrefixes = value;
- } else if (stdFeat == SAXFeature.RESOLVE_DTD_URIS) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.STRING_INTERNING) {
- invalidValue = !value;
- } else if (stdFeat == SAXFeature.UNICODE_NORMALIZATION_CHECKING) {
- invalidValue = value;
- } else if (stdFeat == SAXFeature.USE_ATTRIBUTES2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.USE_LOCATOR2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.USE_ENTITY_RESOLVER2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.VALIDATION) {
- mStaxFactory.getConfig().doValidateWithDTD(value);
- } else if (stdFeat == SAXFeature.XMLNS_URIS) {
- invalidValue = !value;
- } else if (stdFeat == SAXFeature.XML_1_1) {
- readOnly = true;
- } else {
- throw new SAXNotRecognizedException("Feature '"+name+"' not recognized");
- }
-
- // Trying to modify read-only properties?
- if (readOnly) {
- throw new SAXNotSupportedException("Feature '"+name+"' is read-only, can not be modified");
- }
- if (invalidValue) {
- throw new SAXNotSupportedException("Trying to set invalid value for feature '"+name+"', '"+value+"'");
- }
- }
-}
-
-
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sax/WstxSAXParser.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sax/WstxSAXParser.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sax/WstxSAXParser.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sax/WstxSAXParser.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1378 +0,0 @@
-/*
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sax;
-
-import java.io.*;
-import java.net.URL;
-import javax.xml.XMLConstants;
-import javax.xml.parsers.SAXParser;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamConstants;
-import javax.xml.stream.XMLStreamException;
-
-import org.xml.sax.*;
-import org.xml.sax.helpers.DefaultHandler;
-import org.xml.sax.ext.Attributes2;
-import org.xml.sax.ext.DeclHandler;
-import org.xml.sax.ext.LexicalHandler;
-import org.xml.sax.ext.Locator2;
-
-//import org.codehaus.stax2.DTDInfo;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.dtd.DTDEventListener;
-import com.ctc.wstx.exc.WstxIOException;
-import com.ctc.wstx.io.DefaultInputResolver;
-import com.ctc.wstx.io.InputBootstrapper;
-import com.ctc.wstx.io.ReaderBootstrapper;
-import com.ctc.wstx.io.StreamBootstrapper;
-import com.ctc.wstx.sr.*;
-import com.ctc.wstx.stax.WstxInputFactory;
-import com.ctc.wstx.util.ExceptionUtil;
-import com.ctc.wstx.util.URLUtil;
-
-/**
- * This class implements parser part of JAXP and SAX interfaces; and
- * effectively offers an alternative to using Stax input factory /
- * stream reader combination.
- */
-public class WstxSAXParser
- extends SAXParser
- implements Parser // SAX1
- ,XMLReader // SAX2
- ,Attributes2 // SAX2
- ,Locator2 // SAX2
- ,DTDEventListener // Woodstox-internal
-{
- final static boolean FEAT_DEFAULT_NS_PREFIXES = false;
-
- /**
- * We will need the factory reference mostly for constructing
- * underlying stream reader we use.
- */
- protected final WstxInputFactory mStaxFactory;
-
- protected final ReaderConfig mConfig;
-
- protected boolean mFeatNsPrefixes;
-
- /**
- * Since the stream reader would mostly be just a wrapper around
- * the underlying scanner (its main job is to implement Stax
- * interface), we can and should just use the scanner. In effect,
- * this class is then a replacement of BasicStreamReader, when
- * using SAX interfaces.
- */
- protected BasicStreamReader mScanner;
-
- protected AttributeCollector mAttrCollector;
-
- protected InputElementStack mElemStack;
-
- // // // Info from xml declaration
-
- protected String mEncoding;
- protected String mXmlVersion;
- protected boolean mStandalone;
-
- // // // Listeners attached:
-
- protected ContentHandler mContentHandler;
- protected DTDHandler mDTDHandler;
- private EntityResolver mEntityResolver;
- private ErrorHandler mErrorHandler;
-
- private LexicalHandler mLexicalHandler;
- private DeclHandler mDeclHandler;
-
- // // // State:
-
- /**
- * Number of attributes accessible via {@link Attributes} and
- * {@link Attributes2} interfaces, for the current start element.
- *
- * Note: does not include namespace declarations, even they are to
- * be reported as attributes.
- */
- protected int mAttrCount;
-
- /**
- * Need to keep track of number of namespaces, if namespace declarations
- * are to be reported along with attributes (see
- * {@link #mFeatNsPrefixes}).
- */
- protected int mNsCount = 0;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- *
- * NOTE: this was a protected constructor for versions 4.0
- * and 3.2; changed to public in 4.1
- */
- public WstxSAXParser(WstxInputFactory sf, boolean nsPrefixes)
- {
- mStaxFactory = sf;
- mFeatNsPrefixes = nsPrefixes;
- mConfig = sf.createPrivateConfig();
- mConfig.doSupportDTDs(true);
- /* Lazy parsing is a tricky thing: although most of the time
- * it's useless with SAX, it is actually necessary to be able
- * to properly model internal DTD subsets, for example. So,
- * we can not really easily determine defaults.
- */
- ResolverProxy r = new ResolverProxy();
- /* SAX doesn't distinguish between DTD (ext. subset, PEs) and
- * entity (external general entities) resolvers, so let's
- * assign them both:
- */
- mConfig.setDtdResolver(r);
- mConfig.setEntityResolver(r);
- mConfig.setDTDEventListener(this);
-
- /* These settings do not make sense as generic defaults, but
- * are helpful when using some test frameworks. Specifically,
- * - DTD caching may remove calls to resolvers, changing
- * observed behavior
- * - Using min. segment length of 1 will force flushing of
- * all content before entity expansion, which will
- * completely serialize entity resolution calls wrt.
- * CHARACTERS events.
- */
- // !!! TEST
- //mConfig.setShortestReportedTextSegment(1);
- //mConfig.doCacheDTDs(false);
- }
-
- /*
- * This constructor is provided for two main use cases: testing,
- * and introspection via SAX classes (as opposed to JAXP-based
- * introspection).
- */
- public WstxSAXParser()
- {
- this(new WstxInputFactory(), FEAT_DEFAULT_NS_PREFIXES);
- }
-
- public final Parser getParser()
- {
- return this;
- }
-
- public final XMLReader getXMLReader()
- {
- return this;
- }
-
- /**
- * Accessor used to allow configuring all standard Stax configuration
- * settings that the underlying reader uses.
- *
- * @since 4.0.8
- */
- public final ReaderConfig getStaxConfig() {
- return mConfig;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Configuration, SAXParser
- ///////////////////////////////////////////////////////////////////////
- */
-
- public boolean isNamespaceAware() {
- return mConfig.willSupportNamespaces();
- }
-
- public boolean isValidating() {
- return mConfig.willValidateWithDTD();
- }
-
- public Object getProperty(String name)
- throws SAXNotRecognizedException, SAXNotSupportedException
- {
- SAXProperty prop = SAXProperty.findByUri(name);
- if (prop == SAXProperty.DECLARATION_HANDLER) {
- return mDeclHandler;
- } else if (prop == SAXProperty.DOCUMENT_XML_VERSION) {
- return mXmlVersion;
- } else if (prop == SAXProperty.DOM_NODE) {
- return null;
- } else if (prop == SAXProperty.LEXICAL_HANDLER) {
- return mLexicalHandler;
- } else if (prop == SAXProperty.XML_STRING) {
- return null;
- }
-
- throw new SAXNotRecognizedException("Property '"+name+"' not recognized");
- }
-
- public void setProperty(String name, Object value)
- throws SAXNotRecognizedException, SAXNotSupportedException
- {
- SAXProperty prop = SAXProperty.findByUri(name);
- if (prop == SAXProperty.DECLARATION_HANDLER) {
- mDeclHandler = (DeclHandler) value;
- return;
- } else if (prop == SAXProperty.DOCUMENT_XML_VERSION) {
- ; // read-only
- } else if (prop == SAXProperty.DOM_NODE) {
- ; // read-only
- } else if (prop == SAXProperty.LEXICAL_HANDLER) {
- mLexicalHandler = (LexicalHandler) value;
- return;
- } else if (prop == SAXProperty.XML_STRING) {
- ; // read-only
- } else {
- throw new SAXNotRecognizedException("Property '"+name+"' not recognized");
- }
-
- // Trying to modify read-only properties?
- throw new SAXNotSupportedException("Property '"+name+"' is read-only, can not be modified");
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Overrides, SAXParser
- ///////////////////////////////////////////////////////////////////////
- */
-
- /* Have to override some methods from SAXParser; JDK
- * implementation is sucky, as it tries to override
- * many things it really should not...
- */
-
- public void parse(InputSource is, HandlerBase hb)
- throws SAXException, IOException
- {
- if (hb != null) {
- /* Ok: let's ONLY set if there are no explicit sets... not
- * extremely clear, but JDK tries to set them always so
- * let's at least do damage control.
- */
- if (mContentHandler == null) {
- setDocumentHandler(hb);
- }
- if (mEntityResolver == null) {
- setEntityResolver(hb);
- }
- if (mErrorHandler == null) {
- setErrorHandler(hb);
- }
- if (mDTDHandler == null) {
- setDTDHandler(hb);
- }
- }
- parse(is);
- }
-
- public void parse(InputSource is, DefaultHandler dh)
- throws SAXException, IOException
- {
- if (dh != null) {
- /* Ok: let's ONLY set if there are no explicit sets... not
- * extremely clear, but JDK tries to set them always so
- * let's at least do damage control.
- */
- if (mContentHandler == null) {
- setContentHandler(dh);
- }
- if (mEntityResolver == null) {
- setEntityResolver(dh);
- }
- if (mErrorHandler == null) {
- setErrorHandler(dh);
- }
- if (mDTDHandler == null) {
- setDTDHandler(dh);
- }
- }
- parse(is);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XLMReader (SAX2) implementation: cfg access
- ///////////////////////////////////////////////////////////////////////
- */
-
- public ContentHandler getContentHandler()
- {
- return mContentHandler;
- }
-
- public DTDHandler getDTDHandler()
- {
- return mDTDHandler;
- }
-
- public EntityResolver getEntityResolver()
- {
- return mEntityResolver;
- }
-
- public ErrorHandler getErrorHandler()
- {
- return mErrorHandler;
- }
-
- public boolean getFeature(String name)
- throws SAXNotRecognizedException
- {
- SAXFeature stdFeat = SAXFeature.findByUri(name);
-
- if (stdFeat == SAXFeature.EXTERNAL_GENERAL_ENTITIES) {
- return mConfig.willSupportExternalEntities();
- } else if (stdFeat == SAXFeature.EXTERNAL_PARAMETER_ENTITIES) {
- return mConfig.willSupportExternalEntities();
- } else if (stdFeat == SAXFeature.IS_STANDALONE) {
- return mStandalone;
- } else if (stdFeat == SAXFeature.LEXICAL_HANDLER_PARAMETER_ENTITIES) {
- // !!! TODO:
- return false;
- } else if (stdFeat == SAXFeature.NAMESPACES) {
- return mConfig.willSupportNamespaces();
- } else if (stdFeat == SAXFeature.NAMESPACE_PREFIXES) {
- return !mConfig.willSupportNamespaces();
- } else if (stdFeat == SAXFeature.RESOLVE_DTD_URIS) {
- // !!! TODO:
- return false;
- } else if (stdFeat == SAXFeature.STRING_INTERNING) {
- return true;
- } else if (stdFeat == SAXFeature.UNICODE_NORMALIZATION_CHECKING) {
- return false;
- } else if (stdFeat == SAXFeature.USE_ATTRIBUTES2) {
- return true;
- } else if (stdFeat == SAXFeature.USE_LOCATOR2) {
- return true;
- } else if (stdFeat == SAXFeature.USE_ENTITY_RESOLVER2) {
- return true;
- } else if (stdFeat == SAXFeature.VALIDATION) {
- return mConfig.willValidateWithDTD();
- } else if (stdFeat == SAXFeature.XMLNS_URIS) {
- /* !!! TODO: default value should be false... but not sure
- * if implementing that mode makes sense
- */
- return true;
- } else if (stdFeat == SAXFeature.XML_1_1) {
- return true;
- }
-
- throw new SAXNotRecognizedException("Feature '"+name+"' not recognized");
- }
-
- // Already implemented for SAXParser
- //public Object getProperty(String name)
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XLMReader (SAX2) implementation: cfg changing
- ///////////////////////////////////////////////////////////////////////
- */
-
- public void setContentHandler(ContentHandler handler)
- {
- mContentHandler = handler;
- }
-
- public void setDTDHandler(DTDHandler handler)
- {
- mDTDHandler = handler;
- }
-
- public void setEntityResolver(EntityResolver resolver)
- {
- mEntityResolver = resolver;
- }
-
- public void setErrorHandler(ErrorHandler handler)
- {
- mErrorHandler = handler;
- }
-
- public void setFeature(String name, boolean value)
- throws SAXNotRecognizedException, SAXNotSupportedException
- {
- boolean invalidValue = false;
- boolean readOnly = false;
- SAXFeature stdFeat = SAXFeature.findByUri(name);
-
- if (stdFeat == SAXFeature.EXTERNAL_GENERAL_ENTITIES) {
- mConfig.doSupportExternalEntities(value);
- } else if (stdFeat == SAXFeature.EXTERNAL_PARAMETER_ENTITIES) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.IS_STANDALONE) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.LEXICAL_HANDLER_PARAMETER_ENTITIES) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.NAMESPACES) {
- mConfig.doSupportNamespaces(value);
- } else if (stdFeat == SAXFeature.NAMESPACE_PREFIXES) {
- mFeatNsPrefixes = value;
- } else if (stdFeat == SAXFeature.RESOLVE_DTD_URIS) {
- // !!! TODO
- } else if (stdFeat == SAXFeature.STRING_INTERNING) {
- invalidValue = !value;
- } else if (stdFeat == SAXFeature.UNICODE_NORMALIZATION_CHECKING) {
- invalidValue = value;
- } else if (stdFeat == SAXFeature.USE_ATTRIBUTES2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.USE_LOCATOR2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.USE_ENTITY_RESOLVER2) {
- readOnly = true;
- } else if (stdFeat == SAXFeature.VALIDATION) {
- mConfig.doValidateWithDTD(value);
- } else if (stdFeat == SAXFeature.XMLNS_URIS) {
- invalidValue = !value;
- } else if (stdFeat == SAXFeature.XML_1_1) {
- readOnly = true;
- } else {
- throw new SAXNotRecognizedException("Feature '"+name+"' not recognized");
- }
-
- // Trying to modify read-only properties?
- if (readOnly) {
- throw new SAXNotSupportedException("Feature '"+name+"' is read-only, can not be modified");
- }
- if (invalidValue) {
- throw new SAXNotSupportedException("Trying to set invalid value for feature '"+name+"', '"+value+"'");
- }
- }
-
- // Already implemented for SAXParser
- //public void setProperty(String name, Object value)
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XLMReader (SAX2) implementation: parsing
- ///////////////////////////////////////////////////////////////////////
- */
-
- public void parse(InputSource input)
- throws SAXException
- {
- mScanner = null;
- String systemId = input.getSystemId();
- ReaderConfig cfg = mConfig;
- URL srcUrl = null;
-
- // Let's figure out input, first, before sending start-doc event
- InputStream is = null;
- Reader r = input.getCharacterStream();
- if (r == null) {
- is = input.getByteStream();
- if (is == null) {
- if (systemId == null) {
- throw new SAXException("Invalid InputSource passed: neither character or byte stream passed, nor system id specified");
- }
- try {
- srcUrl = URLUtil.urlFromSystemId(systemId);
- is = URLUtil.inputStreamFromURL(srcUrl);
- } catch (IOException ioe) {
- SAXException saxe = new SAXException(ioe);
- ExceptionUtil.setInitCause(saxe, ioe);
- throw saxe;
- }
- }
- }
-
- if (mContentHandler != null) {
- mContentHandler.setDocumentLocator(this);
- mContentHandler.startDocument();
- }
-
- /* Note: since we are reusing the same config instance, need to
- * make sure state is not carried forward. Thus:
- */
- cfg.resetState();
-
- try {
- String inputEnc = input.getEncoding();
- String publicId = input.getPublicId();
-
- // Got an InputStream and encoding? Can create a Reader:
- if (r == null && (inputEnc != null && inputEnc.length() > 0)) {
- r = DefaultInputResolver.constructOptimizedReader(cfg, is, false, inputEnc);
- }
- InputBootstrapper bs;
- if (r != null) {
- bs = ReaderBootstrapper.getInstance(publicId, systemId, r, inputEnc);
- // false -> not for event reader; false -> no auto-closing
- mScanner = (BasicStreamReader) mStaxFactory.createSR(cfg, systemId, bs, false, false);
- } else {
- bs = StreamBootstrapper.getInstance(publicId, systemId, is);
- mScanner = (BasicStreamReader) mStaxFactory.createSR(cfg, systemId, bs, false, false);
- }
-
- // Need to get xml declaration stuff out now:
- {
- String enc2 = mScanner.getEncoding();
- if (enc2 == null) {
- enc2 = mScanner.getCharacterEncodingScheme();
- }
- mEncoding = enc2;
- }
- mXmlVersion = mScanner.getVersion();
- mStandalone = mScanner.standaloneSet();
- mAttrCollector = mScanner.getAttributeCollector();
- mElemStack = mScanner.getInputElementStack();
- fireEvents();
- } catch (IOException io) {
- throwSaxException(io);
- } catch (XMLStreamException strex) {
- throwSaxException(strex);
- } finally {
- if (mContentHandler != null) {
- mContentHandler.endDocument();
- }
- // Could try holding onto the buffers, too... but
- // maybe it's better to allow them to be reclaimed, if
- // needed by GC
- if (mScanner != null) {
- BasicStreamReader sr = mScanner;
- mScanner = null;
- try {
- sr.close();
- } catch (XMLStreamException sex) { }
- }
- if (r != null) {
- try {
- r.close();
- } catch (IOException ioe) { }
- }
- if (is != null) {
- try {
- is.close();
- } catch (IOException ioe) { }
- }
- }
- }
-
- public void parse(String systemId)
- throws SAXException
- {
- InputSource src = new InputSource(systemId);
- parse(src);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Parsing loop, helper methods
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * This is the actual "tight event loop" that will send all events
- * between start and end document events. Although we could
- * use the stream reader here, there's not much as it mostly
- * just forwards requests to the scanner: and so we can as well
- * just copy the little code stream reader's next() method has.
- */
- private final void fireEvents()
- throws IOException, SAXException, XMLStreamException
- {
- // First we are in prolog:
- int type;
-
- /* Need to enable lazy parsing, to get DTD start events before
- * its content events. Plus, can skip more efficiently too.
- */
- mConfig.doParseLazily(false);
-
- while ((type = mScanner.next()) != XMLStreamConstants.START_ELEMENT) {
- fireAuxEvent(type, false);
- }
-
- // Now just starting the tree, need to process the START_ELEMENT
- fireStartTag();
-
- int depth = 1;
- while (true) {
- type = mScanner.next();
- if (type == XMLStreamConstants.START_ELEMENT) {
- fireStartTag();
- ++depth;
- } else if (type == XMLStreamConstants.END_ELEMENT) {
- mScanner.fireSaxEndElement(mContentHandler);
- if (--depth < 1) {
- break;
- }
- } else if (type == XMLStreamConstants.CHARACTERS) {
- mScanner.fireSaxCharacterEvents(mContentHandler);
- } else {
- fireAuxEvent(type, true);
- }
- }
-
- // And then epilog:
- while (true) {
- type = mScanner.next();
- if (type == XMLStreamConstants.END_DOCUMENT) {
- break;
- }
- if (type == XMLStreamConstants.SPACE) {
- // Not to be reported via SAX interface (which may or may not
- // be different from Stax)
- continue;
- }
- fireAuxEvent(type, false);
- }
- }
-
- private final void fireAuxEvent(int type, boolean inTree)
- throws IOException, SAXException, XMLStreamException
- {
- switch (type) {
- case XMLStreamConstants.COMMENT:
- mScanner.fireSaxCommentEvent(mLexicalHandler);
- break;
- case XMLStreamConstants.CDATA:
- if (mLexicalHandler != null) {
- mLexicalHandler.startCDATA();
- mScanner.fireSaxCharacterEvents(mContentHandler);
- mLexicalHandler.endCDATA();
- } else {
- mScanner.fireSaxCharacterEvents(mContentHandler);
- }
- break;
- case XMLStreamConstants.DTD:
- if (mLexicalHandler != null) {
- /* Note: this is bit tricky, since calling getDTDInfo() will
- * trigger full reading of the subsets... but we need to
- * get some info first, to be able to send dtd-start event,
- * and only then get the rest. Thus, need to call separate
- * accessors first:
- */
- String rootName = mScanner.getDTDRootName();
- String sysId = mScanner.getDTDSystemId();
- String pubId = mScanner.getDTDPublicId();
- mLexicalHandler.startDTD(rootName, pubId, sysId);
- // Ok, let's get rest (if any) read:
- try {
- /*DTDInfo dtdInfo =*/ mScanner.getDTDInfo();
- } catch (WrappedSaxException wse) {
- throw wse.getSaxException();
- }
- mLexicalHandler.endDTD();
- }
- break;
- case XMLStreamConstants.PROCESSING_INSTRUCTION:
- mScanner.fireSaxPIEvent(mContentHandler);
- break;
- case XMLStreamConstants.SPACE:
- // With SAX, only to be sent as an event if inside the
- // tree, not from within prolog/epilog
- if (inTree) {
- mScanner.fireSaxSpaceEvents(mContentHandler);
- }
- break;
- case XMLStreamConstants.ENTITY_REFERENCE:
- /* Only occurs in non-entity-expanding mode; so effectively
- * we are skipping the entity?
- */
- if (mContentHandler != null) {
- mContentHandler.skippedEntity(mScanner.getLocalName());
- }
- break;
- default:
- if (type == XMLStreamConstants.END_DOCUMENT) {
- throwSaxException("Unexpected end-of-input in "+(inTree ? "tree" : "prolog"));
- }
- throw new RuntimeException("Internal error: unexpected type, "+type);
- }
- }
-
- private final void fireStartTag()
- throws SAXException
- {
- mAttrCount = mAttrCollector.getCount();
- if (mFeatNsPrefixes) {
- /* 15-Dec-2006, TSa: Note: apparently namespace bindings that
- * are added via defaulting are only visible via element
- * stack. Thus, we MUST access things via element stack,
- * not attribute collector; even though latter seems like
- * the more direct route. See
- * {@link NsInputElementStack#addNsBinding} for the method
- * that injects such special namespace bindings (yes, it's
- * a hack, afterthought)
- */
- //mNsCount = mAttrCollector.getNsCount();
- mNsCount = mElemStack.getCurrentNsCount();
- }
- mScanner.fireSaxStartElement(mContentHandler, this);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Parser (SAX1) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- // Already implemented for XMLReader:
- //public void parse(InputSource source)
- //public void parse(String systemId)
- //public void setEntityResolver(EntityResolver resolver)
- //public void setErrorHandler(ErrorHandler handler)
-
- public void setDocumentHandler(DocumentHandler handler)
- {
- setContentHandler(new DocHandlerWrapper(handler));
- }
-
- public void setLocale(java.util.Locale locale)
- {
- // Not supported, let's just ignore
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Attributes (SAX2) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- public int getIndex(String qName)
- {
- if (mElemStack == null) {
- return -1;
- }
- int ix = mElemStack.findAttributeIndex(null, qName);
- // !!! In ns-as-attrs mode, should also match ns decls?
- return ix;
- }
-
- public int getIndex(String uri, String localName)
- {
- if (mElemStack == null) {
- return -1;
- }
- int ix = mElemStack.findAttributeIndex(uri, localName);
- // !!! In ns-as-attrs mode, should also match ns decls?
- return ix;
- }
-
- public int getLength()
- {
- return mAttrCount + mNsCount;
- }
-
- public String getLocalName(int index)
- {
- if (index < mAttrCount) {
- return (index < 0) ? null : mAttrCollector.getLocalName(index);
- }
- index -= mAttrCount;
- if (index < mNsCount) {
- /* As discussed in
- * Data structure contains two separate areas; main hash area (with
- * size
- * Note: public only so that it can be called by unit tests.
- */
- public void reset()
- {
- if (mNsCount > 0) {
- mNamespaceBuilder.reset();
- mDefaultNsDeclared = false;
- mNsCount = 0;
- }
-
- /* No need to clear attr name, or NS prefix Strings; they are
- * canonicalized and will be referenced by symbol table in any
- * case... so we can save trouble of cleaning them up. This Object
- * will get GC'ed soon enough, after parser itself gets disposed of.
- */
- if (mAttrCount > 0) {
- mValueBuilder.reset();
- mAttrCount = 0;
- if (mXmlIdAttrIndex >= 0) {
- mXmlIdAttrIndex = XMLID_IX_NONE;
- }
- }
- /* Note: attribute values will be cleared later on, when validating
- * namespaces. This so that we know how much to clean up; and
- * occasionally can also just avoid clean up (when resizing)
- */
- }
-
- /**
- * Method that can be called to force space normalization (remove
- * leading/trailing spaces, replace non-spaces white space with
- * spaces, collapse spaces to one) on specified attribute.
- * Currently called by {@link InputElementStack} to force
- * normalization of Xml:id attribute
- */
- public void normalizeSpacesInValue(int index)
- {
- // StringUtil has a method, but it works on char arrays...
- char[] attrCB = mValueBuilder.getCharBuffer();
- String normValue = StringUtil.normalizeSpaces
- (attrCB, getValueStartOffset(index), getValueStartOffset(index+1));
- if (normValue != null) {
- mAttributes[index].setValue(normValue);
- }
- }
-
- /*
- ///////////////////////////////////////////////
- // Public accesors (for stream reader)
- ///////////////////////////////////////////////
- */
-
- /**
- * @return Number of namespace declarations collected, including
- * possible default namespace declaration
- */
- protected int getNsCount() {
- return mNsCount;
- }
-
- public boolean hasDefaultNs() {
- return mDefaultNsDeclared;
- }
-
- // // // Direct access to attribute/NS prefixes/localnames/URI
-
- public final int getCount() {
- return mAttrCount;
- }
-
- /**
- * @return Number of attributes that were explicitly specified; may
- * be less than the total count due to attributes created using
- * attribute default values
- */
- public int getSpecifiedCount() {
- return mNonDefCount;
- }
-
- public String getNsPrefix(int index) {
- if (index < 0 || index >= mNsCount) {
- throwIndex(index);
- }
- // for NS decls, local name is stored in prefix
- return mNamespaces[index].mLocalName;
- }
-
- public String getNsURI(int index) {
- if (index < 0 || index >= mNsCount) {
- throwIndex(index);
- }
- return mNamespaces[index].mNamespaceURI;
- }
-
- // // // Direct access to attribute/NS prefixes/localnames/URI
-
- public String getPrefix(int index) {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- return mAttributes[index].mPrefix;
- }
-
- public String getLocalName(int index) {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- return mAttributes[index].mLocalName;
- }
-
- public String getURI(int index) {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- return mAttributes[index].mNamespaceURI;
- }
-
- public QName getQName(int index) {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- return mAttributes[index].getQName();
- }
-
- /**
- *
- * Note: the main reason this method is defined at this level, and
- * made final, is performance. JIT may be able to fully inline this
- * method, even when reference is via this base class. This is important
- * since this is likely to be the most often called method of the
- * collector instances.
- */
- public final String getValue(int index)
- {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- String full = mValueBuilder.getAllValues();
- Attribute attr = mAttributes[index];
- ++index;
- if (index < mAttrCount) { // not last
- int endOffset = mAttributes[index].mValueStartOffset;
- return attr.getValue(full, endOffset);
- }
- // last can be optimized bit more:
- return attr.getValue(full);
- }
-
- public String getValue(String nsURI, String localName)
- {
- // Primary hit?
- int hashSize = mAttrHashSize;
- if (hashSize == 0) { // sanity check, for 'no attributes'
- return null;
- }
- int hash = localName.hashCode();
- if (nsURI != null) {
- if (nsURI.length() == 0) {
- nsURI = null;
- } else {
- hash ^= nsURI.hashCode();
- }
- }
- int ix = mAttrMap[hash & (hashSize-1)];
- if (ix == 0) { // nothing in here; no spills either
- return null;
- }
- --ix;
-
- // Is primary candidate match?
- if (mAttributes[ix].hasQName(nsURI, localName)) {
- return getValue(ix);
- }
-
- /* Nope, need to traverse spill list, which has 2 entries for
- * each spilled attribute id; first for hash value, second index.
- */
- for (int i = hashSize, len = mAttrSpillEnd; i < len; i += 2) {
- if (mAttrMap[i] != hash) {
- continue;
- }
- /* Note: spill indexes are not off-by-one, since there's no need
- * to mask 0
- */
- ix = mAttrMap[i+1];
- if (mAttributes[ix].hasQName(nsURI, localName)) {
- return getValue(ix);
- }
- }
-
- return null;
- }
-
- public int findIndex(String localName) {
- return findIndex(null, localName);
- }
-
- public int findIndex(String nsURI, String localName)
- {
- /* Note: most of the code is from getValue().. could refactor
- * code, performance is bit of concern (one more method call
- * if index access was separate).
- * See comments on that method, for logics.
- */
-
- // Primary hit?
- int hashSize = mAttrHashSize;
- if (hashSize == 0) { // sanity check, for 'no attributes'
- return -1;
- }
- int hash = localName.hashCode();
- if (nsURI != null) {
- if (nsURI.length() == 0) {
- nsURI = null;
- } else {
- hash ^= nsURI.hashCode();
- }
- }
- int ix = mAttrMap[hash & (hashSize-1)];
- if (ix == 0) { // nothing in here; no spills either
- return -1;
- }
- --ix;
-
- // Is primary candidate match?
- if (mAttributes[ix].hasQName(nsURI, localName)) {
- return ix;
- }
-
- /* Nope, need to traverse spill list, which has 2 entries for
- * each spilled attribute id; first for hash value, second index.
- */
- for (int i = hashSize, len = mAttrSpillEnd; i < len; i += 2) {
- if (mAttrMap[i] != hash) {
- continue;
- }
- /* Note: spill indexes are not off-by-one, since there's no need
- * to mask 0
- */
- ix = mAttrMap[i+1];
- if (mAttributes[ix].hasQName(nsURI, localName)) {
- return ix;
- }
- }
- return -1;
- }
-
- public final boolean isSpecified(int index) {
- return (index < mNonDefCount);
- }
-
- public final int getXmlIdAttrIndex() {
- return mXmlIdAttrIndex;
- }
-
- /*
- //////////////////////////////////////////////////////
- // Type-safe accessors to support TypedXMLStreamReader
- //////////////////////////////////////////////////////
- */
-
- /**
- * Method called to decode the whole attribute value as a single
- * typed value.
- * Decoding is done using the decoder provided.
- */
- public final void decodeValue(int index, TypedValueDecoder tvd)
- throws IllegalArgumentException
- {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- /* Should be faster to pass the char array even if we might
- * have a String
- */
- // Either way, need to trim before passing:
- char[] buf = mValueBuilder.getCharBuffer();
- int start = mAttributes[index].mValueStartOffset;
- int end = getValueStartOffset(index+1);
-
- while (true) {
- if (start >= end) {
- tvd.handleEmptyValue();
- return;
- }
- if (!StringUtil.isSpace(buf[start])) {
- break;
- }
- ++start;
- }
- // Trailing space?
- while (--end > start && StringUtil.isSpace(buf[end])) { }
- tvd.decode(buf, start, end+1);
- }
-
- /**
- * Method called to decode the attribute value that consists of
- * zero or more space-separated tokens.
- * Decoding is done using the decoder provided.
- * @return Number of tokens decoded
- */
- public final int decodeValues(int index, TypedArrayDecoder tad,
- InputProblemReporter rep)
- throws XMLStreamException
- {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- // Char[] faster than String... and no need to trim here:
- return decodeValues(tad, rep,
- mValueBuilder.getCharBuffer(),
- mAttributes[index].mValueStartOffset,
- getValueStartOffset(index+1));
- }
-
- public final byte[] decodeBinary(int index, Base64Variant v, CharArrayBase64Decoder dec,
- InputProblemReporter rep)
- throws XMLStreamException
- {
- if (index < 0 || index >= mAttrCount) {
- throwIndex(index);
- }
- /* No point in trying to use String representation, even if one
- * available, faster to process from char[]
- */
- Attribute attr = mAttributes[index];
- char[] cbuf = mValueBuilder.getCharBuffer();
- int start = attr.mValueStartOffset;
- int end = getValueStartOffset(index+1);
- int len = end-start;
- dec.init(v, true, cbuf, start, len, null);
- try {
- return dec.decodeCompletely();
- } catch (IllegalArgumentException iae) {
- // Need to convert to a checked stream exception
- String lexical = new String(cbuf, start, len);
- throw new TypedXMLStreamException(lexical, iae.getMessage(), rep.getLocation(), iae);
- }
- }
-
- private final static int decodeValues(TypedArrayDecoder tad,
- InputProblemReporter rep,
- final char[] buf, int ptr, final int end)
- throws XMLStreamException
- {
- int start = ptr;
- int count = 0;
-
- try {
- decode_loop:
- while (ptr < end) {
- // First, any space to skip?
- while (buf[ptr] <= INT_SPACE) {
- if (++ptr >= end) {
- break decode_loop;
- }
- }
- // Then let's figure out non-space char (token)
- start = ptr;
- ++ptr;
- while (ptr < end && buf[ptr] > INT_SPACE) {
- ++ptr;
- }
- int tokenEnd = ptr;
- ++ptr; // to skip trailing space (or, beyond end)
- // Ok, decode... any more room?
- ++count;
- if (tad.decodeValue(buf, start, tokenEnd)) {
- if (!checkExpand(tad)) {
- break;
- }
- }
- }
- } catch (IllegalArgumentException iae) {
- // Need to convert to a checked stream exception
- Location loc = rep.getLocation();
- String lexical = new String(buf, start, (ptr-start));
- throw new TypedXMLStreamException(lexical, iae.getMessage(), loc, iae);
- }
- return count;
- }
-
- /**
- * Internal method used to see if we can expand the buffer that
- * the array decoder has. Bit messy, but simpler than having
- * separately typed instances; and called rarely so that performance
- * downside of instanceof is irrelevant.
- */
- private final static boolean checkExpand(TypedArrayDecoder tad)
- {
- if (tad instanceof ValueDecoderFactory.BaseArrayDecoder) {
- ((ValueDecoderFactory.BaseArrayDecoder) tad).expand();
- return true;
- }
- return false;
- }
-
- /*
- ///////////////////////////////////////////////
- // Accessors for accessing helper objects
- ///////////////////////////////////////////////
- */
-
- /**
- * Method for getting start pointer within shared value buffer
- * for given attribute. It is also the same as end pointer
- * for preceding attribute, if any.
- */
- protected int getValueStartOffset(int index)
- {
- if (index < mAttrCount) {
- return mAttributes[index].mValueStartOffset;
- }
- return mValueBuilder.getCharSize();
- }
-
- protected char[] getSharedValueBuffer()
- {
- return mValueBuilder.getCharBuffer();
- }
-
- /**
- * Method called to resolve and initialize specified collected
- * namespace declaration
- *
- * @return Attribute that contains specified namespace declaration
- */
- protected Attribute resolveNamespaceDecl(int index, boolean internURI)
- {
- Attribute ns = mNamespaces[index];
- String full = mNamespaceBuilder.getAllValues();
- String uri;
-
- if (mNsCount == 0) {
- uri = full;
- } else {
- ++index;
- if (index < mNsCount) { // not last
- int endOffset = mNamespaces[index].mValueStartOffset;
- uri = ns.getValue(full, endOffset);
- } else { // is last
- uri = ns.getValue(full);
- }
- }
- if (internURI && uri.length() > 0) {
- uri = sInternCache.intern(uri);
- }
- ns.mNamespaceURI = uri;
- return ns;
- }
-
- /**
- * Method needed by event creating code, to build a non-transient
- * attribute container, to use with XMLEvent objects (specifically
- * implementation of StartElement event).
- */
- public ElemAttrs buildAttrOb()
- {
- int count = mAttrCount;
- if (count == 0) {
- return null;
- }
- /* If we have actual attributes, let's first just create the
- * raw array that has all attribute information:
- */
- String[] raw = new String[count << 2];
- for (int i = 0; i < count; ++i) {
- Attribute attr = mAttributes[i];
- int ix = (i << 2);
- raw[ix] = attr.mLocalName;
- raw[ix+1] = attr.mNamespaceURI;
- raw[ix+2] = attr.mPrefix;
- raw[ix+3] = getValue(i);
- }
-
- // Do we have a "short" list?
- if (count < LONG_ATTR_LIST_LEN) {
- return new ElemAttrs(raw, mNonDefCount);
- }
-
- // Ok, nope; we need to also pass the Map information...
- /* 02-Feb-2009, TSa: Must make a copy of the Map array now,
- * otherwise could get overwritten.
- */
- int amapLen = mAttrMap.length;
- int[] amap = new int[amapLen];
- // TODO: JDK 1.6 has Arrays.copyOf(), should use with Woodstox 6
- System.arraycopy(mAttrMap, 0, amap, 0, amapLen);
- return new ElemAttrs(raw, mNonDefCount,
- amap, mAttrHashSize, mAttrSpillEnd);
- }
-
- protected void validateAttribute(int index, XMLValidator vld)
- throws XMLStreamException
- {
- Attribute attr = mAttributes[index];
- String normValue = vld.validateAttribute
- (attr.mLocalName, attr.mNamespaceURI, attr.mPrefix,
- mValueBuilder.getCharBuffer(),
- getValueStartOffset(index),
- getValueStartOffset(index+1));
-
- if (normValue != null) {
- attr.setValue(normValue);
- }
- }
-
- /*
- ///////////////////////////////////////////////
- // Attribute, namespace decl building
- ///////////////////////////////////////////////
- */
-
- /**
- * Low-level accessor method that attribute validation code may call
- * for certain types of attributes; generally only for id and idref/idrefs
- * attributes. It returns the underlying 'raw' attribute value buffer
- * for direct access.
- */
- public final TextBuilder getAttrBuilder(String attrPrefix, String attrLocalName)
- {
- /* Ok: we have parsed prefixed-name of a regular
- * attribute. So let's initialize the instance...
- */
- if (mAttrCount == 0) {
- if (mAttributes == null) {
- allocBuffers();
- }
- mAttributes[0] = new Attribute(attrPrefix, attrLocalName, 0);
- } else {
- int valueStart = mValueBuilder.getCharSize();
- if (mAttrCount >= mAttributes.length) {
- mAttributes = (Attribute[]) DataUtil.growArrayBy50Pct(mAttributes);
- }
- Attribute curr = mAttributes[mAttrCount];
- if (curr == null) {
- mAttributes[mAttrCount] = new Attribute(attrPrefix, attrLocalName, valueStart);
- } else {
- curr.reset(attrPrefix, attrLocalName, valueStart);
- }
- }
- ++mAttrCount;
- // 25-Sep-2006, TSa: Need to keep track of xml:id attribute?
- if (attrLocalName == mXmlIdLocalName) {
- if (attrPrefix == mXmlIdPrefix) {
- if (mXmlIdAttrIndex != XMLID_IX_DISABLED) {
- mXmlIdAttrIndex = mAttrCount - 1;
- }
- }
- }
- /* Can't yet create attribute map by name, since we only know
- * name prefix, not necessarily matching URI.
- */
- return mValueBuilder;
- }
-
- /**
- * Method called by validator to insert an attribute that has a default
- * value and wasn't yet included in collector's attribute set.
- *
- * @return Index of the newly added attribute, if added; -1 to indicate
- * this was a duplicate
- */
- public int addDefaultAttribute(String localName, String uri, String prefix,
- String value)
- {
- int attrIndex = mAttrCount;
- if (attrIndex < 1) {
- /* had no explicit attributes... better initialize now, then.
- * Let's just use hash area of 4, and
- */
- initHashArea();
- }
-
- /* Ok, first, since we do want to verify that we can not accidentally
- * add duplicates, let's first try to add entry to Map, since that
- * will catch dups.
- */
- int hash = localName.hashCode();
- if (uri != null && uri.length() > 0) {
- hash ^= uri.hashCode();
- }
- int index = hash & (mAttrHashSize - 1);
- int[] map = mAttrMap;
- if (map[index] == 0) { // whoa, have room...
- map[index] = attrIndex+1; // add 1 to get 1-based index (0 is empty marker)
- } else { // nah, collision...
- int currIndex = map[index]-1; // Index of primary collision entry
- int spillIndex = mAttrSpillEnd;
- map = spillAttr(uri, localName, map, currIndex, spillIndex,
- attrIndex, hash, mAttrHashSize);
- if (map == null) { // dup!
- return -1; // could return negation (-(index+1)) of the prev index?
- }
- map[++spillIndex] = attrIndex; // no need to specifically avoid 0
- mAttrMap = map;
- mAttrSpillEnd = ++spillIndex;
- }
-
- /* Can reuse code; while we don't really need the builder,
- * we need to add/reset attribute
- */
- getAttrBuilder(prefix, localName);
- Attribute attr = mAttributes[mAttrCount-1];
- attr.mNamespaceURI = uri;
- attr.setValue(value);
- // attribute count has been updated; index is one less than count
- return (mAttrCount-1);
- }
-
- /**
- * Low-level mutator method that attribute validation code may call
- * for certain types of attributes, when it wants to handle the whole
- * validation and normalization process by itself. It is generally
- * only called for id and idref/idrefs attributes, as those values
- * are usually normalized.
- */
- public final void setNormalizedValue(int index, String value)
- {
- mAttributes[index].setValue(value);
- }
-
- /**
- * @return null if the default namespace URI has been already declared
- * for the current element; TextBuilder to add URI to if not.
- */
- public TextBuilder getDefaultNsBuilder()
- {
- if (mDefaultNsDeclared) {
- return null;
- }
- mDefaultNsDeclared = true;
- return getNsBuilder(null);
- }
-
- /**
- * @return null if prefix has been already declared; TextBuilder to
- * add value to if not.
- */
- public TextBuilder getNsBuilder(String prefix)
- {
- // first: must verify that it's not a dup
- if (mNsCount == 0) {
- if (mNamespaces == null) {
- mNamespaces = new Attribute[EXP_NS_COUNT];
- }
- mNamespaces[0] = new Attribute(null, prefix, 0);
- } else {
- int len = mNsCount;
- /* Ok: must ensure that there are no duplicate namespace
- * declarations (ie. decls with same prefix being bound)
- */
- if (prefix != null) { // null == default ns
- for (int i = 0; i < len; ++i) {
- // note: for ns decls, bound prefix is in 'local name'
- if (prefix == mNamespaces[i].mLocalName) {
- return null;
- }
- }
- }
- if (len >= mNamespaces.length) {
- mNamespaces = (Attribute[]) DataUtil.growArrayBy50Pct(mNamespaces);
- }
- int uriStart = mNamespaceBuilder.getCharSize();
- Attribute curr = mNamespaces[len];
- if (curr == null) {
- mNamespaces[len] = new Attribute(null, prefix, uriStart);
- } else {
- curr.reset(null, prefix, uriStart);
- }
- }
- ++mNsCount;
- return mNamespaceBuilder;
- }
-
- /**
- * Method called to resolve namespace URIs from attribute prefixes.
- *
- * Note: public only so that it can be called by unit tests.
- *
- * @param rep Reporter to use for reporting well-formedness problems
- * @param ns Namespace prefix/URI mappings active for this element
- *
- * @return Index of xml:id attribute, if any, -1 if not
- */
- public int resolveNamespaces(InputProblemReporter rep, StringVector ns)
- throws XMLStreamException
- {
- int attrCount = mAttrCount;
-
- /* Let's now set number of 'real' attributes, to allow figuring
- * out number of attributes created via default value expansion
- */
- mNonDefCount = attrCount;
-
- if (attrCount < 1) {
- // Checked if doing access by FQN:
- mAttrHashSize = mAttrSpillEnd = 0;
- // And let's just bail out, too...
- return mXmlIdAttrIndex;
- }
- for (int i = 0; i < attrCount; ++i) {
- Attribute attr = mAttributes[i];
- String prefix = attr.mPrefix;
- // Attributes' ns URI is null after reset, so can skip setting "no namespace"
- if (prefix != null) {
- if (prefix == "xml") {
- attr.mNamespaceURI = XMLConstants.XML_NS_URI;
- } else {
- String uri = ns.findLastFromMap(prefix);
- if (uri == null) {
- rep.throwParseError(ErrorConsts.ERR_NS_UNDECLARED_FOR_ATTR,
- prefix, attr.mLocalName);
- }
- attr.mNamespaceURI = uri;
- }
- }
- }
-
- /* Ok, finally, let's create attribute map, to allow efficient
- * access by prefix+localname combination. Could do it on-demand,
- * but this way we can check for duplicates right away.
- */
- int[] map = mAttrMap;
-
- /* What's minimum size to contain at most 80% full hash area,
- * plus 1/8 spill area (12.5% spilled entries, two ints each)?
- */
- int hashCount = 4;
- {
- int min = attrCount + (attrCount >> 2); // == 80% fill rate
- /* Need to get 2^N size that can contain all elements, with
- * 80% fill rate
- */
- while (hashCount < min) {
- hashCount += hashCount; // 2x
- }
- // And then add the spill area
- mAttrHashSize = hashCount;
- min = hashCount + (hashCount >> 4); // 12.5 x 2 ints
- if (map == null || map.length < min) {
- map = new int[min];
- } else {
- /* Need to clear old hash entries (if any). But note that
- * spilled entries we can leave alone -- they are just ints,
- * and get overwritten if and as needed
- */
- Arrays.fill(map, 0, hashCount, 0);
- }
- }
-
- {
- int mask = hashCount-1;
- int spillIndex = hashCount;
-
- // Ok, array's fine, let's hash 'em in!
- for (int i = 0; i < attrCount; ++i) {
- Attribute attr = mAttributes[i];
- String name = attr.mLocalName;
- int hash = name.hashCode();
- String uri = attr.mNamespaceURI;
- if (uri != null) {
- hash ^= uri.hashCode();
- }
- int index = hash & mask;
- // Hash slot available?
- if (map[index] == 0) {
- map[index] = i+1; // since 0 is marker
- } else {
- int currIndex = map[index]-1;
- /* nope, need to spill; let's extract most of that code to
- * a separate method for clarity (and maybe it'll be
- * easier to inline by JVM too)
- */
- map = spillAttr(uri, name, map, currIndex, spillIndex,
- attrCount, hash, hashCount);
- if (map == null) {
- throwDupAttr(rep, currIndex);
- // never returns here...
- } else { // let's use else to keep FindBugs happy
- map[++spillIndex] = i; // no need to specifically avoid 0
- ++spillIndex;
- }
- }
- }
- mAttrSpillEnd = spillIndex;
- }
- mAttrMap = map;
- return mXmlIdAttrIndex;
- }
-
- /*
- ///////////////////////////////////////////////
- // Package/core methods:
- ///////////////////////////////////////////////
- */
-
- protected void throwIndex(int index) {
- throw new IllegalArgumentException("Invalid index "+index+"; current element has only "+getCount()+" attributes");
- }
-
- /**
- * Method that basically serializes the specified (read-in) attribute
- * using Writers provided. Serialization is done by
- * writing out (fully-qualified) name
- * of the attribute, followed by the equals sign and quoted value.
- */
-
- public void writeAttribute(int index, XmlWriter xw)
- throws IOException, XMLStreamException
- {
- // Note: here we assume index checks have been done by caller
- Attribute attr = mAttributes[index];
- String ln = attr.mLocalName;
- String prefix = attr.mPrefix;
- if (prefix == null || prefix.length() == 0) {
- xw.writeAttribute(ln, getValue(index));
- } else {
- xw.writeAttribute(prefix, ln, getValue(index));
- }
- }
-
- /**
- * Method called to initialize buffers that need not be immediately
- * initialized
- */
- protected final void allocBuffers()
- {
- if (mAttributes == null) {
- mAttributes = new Attribute[8];
- }
- if (mValueBuilder == null) {
- mValueBuilder = new TextBuilder(EXP_ATTR_COUNT);
- }
- }
-
- /*
- ///////////////////////////////////////////////
- // Internal methods:
- ///////////////////////////////////////////////
- */
-
- /**
- * @return Null, if attribute is a duplicate (to indicate error);
- * map itself, or resized version, otherwise.
- */
- private int[] spillAttr(String uri, String name,
- int[] map, int currIndex, int spillIndex, int attrCount,
- int hash, int hashCount)
- {
- // Do we have a dup with primary entry?
- /* Can do equality comp for local name, as they
- * are always canonicalized:
- */
- Attribute oldAttr = mAttributes[currIndex];
- if (oldAttr.mLocalName == name) {
- // URIs may or may not be interned though:
- String currURI = oldAttr.mNamespaceURI;
- if (currURI == uri || (currURI != null && currURI.equals(uri))) {
- return null;
- }
- }
-
- /* Is there room to spill into? (need to 2 int spaces; one for hash,
- * the other for index)
- */
- if ((spillIndex + 1)>= map.length) {
- // Let's just add room for 4 spills...
- map = DataUtil.growArrayBy(map, 8);
- }
- // Let's first ensure we aren't adding a dup:
- for (int j = hashCount; j < spillIndex; j += 2) {
- if (map[j] == hash) {
- currIndex = map[j+1];
- Attribute attr = mAttributes[currIndex];
- if (oldAttr.mLocalName == name) {
- String currURI = attr.mNamespaceURI;
- if (currURI == uri || (currURI != null && currURI.equals(uri))) {
- return null;
- }
- }
- }
- }
- map[spillIndex] = hash;
- return map;
- }
-
- /**
- * Method called to ensure hash area will be properly set up in
- * cases where initially no room was needed, but default attribute(s)
- * is being added.
- */
- private void initHashArea()
- {
- /* Let's use small hash area of size 4, and one spill; don't
- * want too big (need to clear up room), nor too small (only
- * collisions)
- */
- mAttrHashSize = mAttrSpillEnd = 4;
- if (mAttrMap == null || mAttrMap.length < mAttrHashSize) {
- mAttrMap = new int[mAttrHashSize+1];
- }
- mAttrMap[0] = mAttrMap[1] = mAttrMap[2] = mAttrMap[3] = 0;
- allocBuffers();
- }
-
- /**
- * Method that can be used to get the specified attribute value,
- * by getting it written using Writer passed in. Can potentially
- * save one String allocation, since no (temporary) Strings need
- * to be created.
- */
- /*
- protected final void writeValue(int index, Writer w)
- throws IOException
- {
- mValueBuilder.getEntry(index, w);
- }
- */
-
- protected void throwDupAttr(InputProblemReporter rep, int index)
- throws XMLStreamException
- {
- rep.throwParseError("Duplicate attribute '"+getQName(index)+"'.");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/Attribute.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/Attribute.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/Attribute.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/Attribute.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,153 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import javax.xml.namespace.QName;
-
-import com.ctc.wstx.compat.QNameCreator;
-
-/**
- * Container for information collected regarding a single element
- * attribute instance. Used for both regular explicit attributes
- * and values added via attribute value defaulting.
- *
- * This class is not exposed outside of the package and is considered
- * part of internal implementation.
- *
- * @since 4.1
- */
-final class Attribute
-{
- // // // Name information
-
- protected String mLocalName;
-
- protected String mPrefix;
-
- protected String mNamespaceURI;
-
- // // // Value information
-
- /**
- * Numeric offset within text builder that denotes pointer
- * to the first character of the value for this attribute
- * (or namespace). End offset is derived by looking at
- * start pointer of the following attribute; or total
- * length for the last entry
- */
- protected int mValueStartOffset;
-
- /**
- * Value as a String iff it has been requested once; stored
- * here in case it will be accessed again.
- */
- protected String mReusableValue;
-
- /*
- //////////////////////////////////////////////////
- // Life-cycle
- //////////////////////////////////////////////////
- */
-
- public Attribute(String prefix, String localName, int valueStart)
- {
- mLocalName = localName;
- mPrefix = prefix;
- mValueStartOffset = valueStart;
- }
-
- public void reset(String prefix, String localName, int valueStart)
- {
- mLocalName = localName;
- mPrefix = prefix;
- mValueStartOffset = valueStart;
- mNamespaceURI = null;
- mReusableValue = null;
- }
-
- /**
- * Method called to inject specific value for this attribute.
- */
- public void setValue(String value) {
- mReusableValue = value;
- }
-
- /*
- //////////////////////////////////////////////////
- // Accessors
- //////////////////////////////////////////////////
- */
-
- /**
- * @param uri Namespace URI of the attribute, if any; MUST be
- * given as null if no namespace
- * @param localName Local name to match. Note: is NOT guaranteed
- * to have been interned
- *
- * @return True if qualified name of this attribute is the same
- * as what arguments describe
- */
- protected boolean hasQName(String uri, String localName)
- {
- if (localName != mLocalName && !localName.equals(mLocalName)) {
- return false;
- }
- if (mNamespaceURI == uri) {
- return true;
- }
- if (uri == null) {
- return (mNamespaceURI == null) || mNamespaceURI.length() == 0;
- }
- return (mNamespaceURI != null && uri.equals(mNamespaceURI));
- }
-
- public QName getQName()
- {
- if (mPrefix == null) {
- if (mNamespaceURI == null) {
- return new QName(mLocalName);
- }
- return new QName(mNamespaceURI, mLocalName);
- }
- String uri = mNamespaceURI;
- if (uri == null) { // Some QName impls (older JDKs) don't like nulls
- uri = "";
- }
- // For [WSTX-174] need to use indirection:
- return QNameCreator.create(uri, mLocalName, mPrefix);
- }
-
- /**
- * Method called if this attribute is the last one with value
- * in the buffer. If so, end value is implied
- */
- public String getValue(String allValues)
- {
- if (mReusableValue == null) {
- mReusableValue = (mValueStartOffset == 0) ?
- allValues : allValues.substring(mValueStartOffset);
- }
- return mReusableValue;
- }
-
- public String getValue(String allValues, int endOffset)
- {
- if (mReusableValue == null) {
- mReusableValue = allValues.substring(mValueStartOffset, endOffset);
- }
- return mReusableValue;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/BasicStreamReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/BasicStreamReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/BasicStreamReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/BasicStreamReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,5512 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.io.*;
-import java.text.MessageFormat;
-import java.util.Map;
-
-import org.xml.sax.Attributes;
-import org.xml.sax.ContentHandler;
-import org.xml.sax.SAXException;
-import org.xml.sax.ext.LexicalHandler;
-
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.AttributeInfo;
-import org.codehaus.stax2.DTDInfo;
-import org.codehaus.stax2.LocationInfo;
-import org.codehaus.stax2.XMLStreamLocation2;
-import org.codehaus.stax2.XMLStreamProperties;
-import org.codehaus.stax2.XMLStreamReader2;
-import org.codehaus.stax2.typed.TypedXMLStreamException;
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.api.WstxInputProperties;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.dtd.MinimalDTDReader;
-import com.ctc.wstx.ent.EntityDecl;
-import com.ctc.wstx.exc.WstxException;
-import com.ctc.wstx.io.*;
-import com.ctc.wstx.util.DefaultXmlSymbolTable;
-import com.ctc.wstx.util.TextBuffer;
-import com.ctc.wstx.util.TextBuilder;
-
-/**
- * Partial implementation of {@link XMLStreamReader2} consisting of
- * all functionality other than DTD-validation-specific parts, and
- * Typed Access API (Stax2 v3.0), which are implemented at
- * sub-classes.
- *
- *
- * @author Tatu Saloranta
- */
-public abstract class BasicStreamReader
- extends StreamScanner
- implements StreamReaderImpl, DTDInfo, LocationInfo
-{
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Constants
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // Standalone values:
-
- final static int DOC_STANDALONE_UNKNOWN = 0;
- final static int DOC_STANDALONE_YES = 1;
- final static int DOC_STANDALONE_NO = 2;
-
- // // // Main state consts:
-
- final static int STATE_PROLOG = 0; // Before root element
- final static int STATE_TREE = 1; // Parsing actual XML tree
- final static int STATE_EPILOG = 2; // After root element has been closed
- final static int STATE_MULTIDOC_HACK = 3; // State "between" multiple documents (in multi-doc mode)
- final static int STATE_CLOSED = 4; // After reader has been closed
-
- // // // Tokenization state consts:
-
- // no idea as to what comes next (unknown type):
- final static int TOKEN_NOT_STARTED = 0;
-
- // token type figured out, but not long enough:
- final static int TOKEN_STARTED = 1;
-
- /* minimum token length returnable achieved; only used for CDATA and
- * CHARACTERS events which allow fragments to be returned
- */
- final static int TOKEN_PARTIAL_SINGLE = 2;
-
- /* a single physical event has been succesfully tokenized; as with
- * partial, only used with CDATA and CHARACTERS (meaningless for others,
- * which should only use TOKEN_FULL_COALESCED, TOKEN_NOT_STARTED or
- * TOKEN_STARTED.
- */
- final static int TOKEN_FULL_SINGLE = 3;
-
- /* all adjacent (text) events have been tokenized and coalesced (for
- * CDATA and CHARACTERS), or that the full event has been parsed (for
- * others)
- */
- final static int TOKEN_FULL_COALESCED = 4;
-
- // // // Bit masks used for quick type comparisons
-
- /**
- * This mask covers all types for which basic {@link #getText} method
- * can be called.
- */
- final protected static int MASK_GET_TEXT =
- (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE)
- | (1 << COMMENT) | (1 << DTD) | (1 << ENTITY_REFERENCE);
-
- /**
- * This mask covers all types for which extends
- * Initially set to fully tokenized, since it's the virtual
- * START_DOCUMENT event that we fully know by now (parsed by
- * bootstrapper)
- */
- protected int mTokenState = TOKEN_FULL_COALESCED;
-
- /**
- * Threshold value that defines tokenization state that needs to be
- * achieved to "finish" current logical text segment (which
- * may consist of adjacent CDATA and text segments; or be a complete
- * physical segment; or just even a fragment of such a segment)
- */
- protected final int mStTextThreshold;
-
- /// Flag that indicates current start element is an empty element
- protected boolean mStEmptyElem = false;
-
- /**
- * Main parsing/tokenization state (STATE_xxx)
- */
- protected int mParseState;
-
- /**
- * Current state of the stream, ie token value returned by
- * {@link #getEventType}. Needs to be initialized to START_DOCUMENT,
- * since that's the state it starts in.
- */
- protected int mCurrToken = START_DOCUMENT;
-
- /**
- * Additional information sometimes stored (when generating dummy
- * events in multi-doc mode, for example) temporarily when
- * {@link #mCurrToken} is already populated.
- */
- protected int mSecondaryToken = START_DOCUMENT;
-
- /**
- * Status of current (text) token's "whitespaceness", that is,
- * whether it is or is not all white space.
- */
- protected int mWsStatus;
-
- /**
- * Flag that indicates that textual content (CDATA, CHARACTERS) is to
- * be validated within current element's scope. Enabled if one of
- * validators returns {@link XMLValidator#CONTENT_ALLOW_VALIDATABLE_TEXT},
- * and will prevent lazy parsing of text.
- */
- protected boolean mValidateText = false;
-
- /**
- * Counter used for determining whether we are to try to heuristically
- * "intern" white space that seems to be used for indentation purposes
- */
- protected int mCheckIndentation;
-
- /**
- * Due to the way Stax API does not allow throwing stream exceptions
- * from many methods for which Woodstox would need to throw one
- * (especially
- * Note: method name is rather confusing (compare to {@link #getEncoding}).
- */
- public String getCharacterEncodingScheme() {
- return mDocXmlEncoding;
- }
-
- /**
- * As per Stax (1.0) specs, needs to return whatever parser determined
- * the encoding was, if it was able to figure it out. If not (there are
- * cases where this can not be found; specifically when being passed a
- * {@link Reader}), it should return null.
- */
- public String getEncoding() {
- return mDocInputEncoding;
- }
-
- public String getVersion()
- {
- if (mDocXmlVersion == XmlConsts.XML_V_10) {
- return XmlConsts.XML_V_10_STR;
- }
- if (mDocXmlVersion == XmlConsts.XML_V_11) {
- return XmlConsts.XML_V_11_STR;
- }
- return null; // unknown
- }
-
- public boolean isStandalone() {
- return mDocStandalone == DOC_STANDALONE_YES;
- }
-
- public boolean standaloneSet() {
- return mDocStandalone != DOC_STANDALONE_UNKNOWN;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Public API, configuration
- ///////////////////////////////////////////////////////////////////////
- */
-
- public Object getProperty(String name)
- {
- /* 18-Nov-2008, TSa: As per [WSTX-50], should report the
- * actual Base URL. It can be overridden by matching
- * setProperty, but if not, is set to actual source
- * of content being parsed.
- */
- if (WstxInputProperties.P_BASE_URL.equals(name)) {
- return mInput.getSource();
- }
- /* 23-Apr-2008, TSa: Let's NOT throw IllegalArgumentException
- * for unknown property; JavaDocs do not suggest it needs
- * to be done (different from that of XMLInputFactory
- * and XMLStreamWriter specification)
- */
- return mConfig.safeGetProperty(name);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XMLStreamReader, current state
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // Attribute access:
-
- public int getAttributeCount() {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.getCount();
- }
-
- public String getAttributeLocalName(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.getLocalName(index);
- }
-
- public QName getAttributeName(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.getQName(index);
- }
-
- public String getAttributeNamespace(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- // Internally it's marked as null, externally need to see ""
- String uri = mAttrCollector.getURI(index);
- return (uri == null) ? XmlConsts.ATTR_NO_NS_URI : uri;
- }
-
- public String getAttributePrefix(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- // Internally it's marked as null, externally need to see ""
- String p = mAttrCollector.getPrefix(index);
- return (p == null) ? XmlConsts.ATTR_NO_PREFIX : p;
- }
-
- public String getAttributeType(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- // Attr. collector doesn't know it, elem stack does:
- return mElementStack.getAttributeType(index);
- }
-
- public String getAttributeValue(int index) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.getValue(index);
- }
-
- public String getAttributeValue(String nsURI, String localName) {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.getValue(nsURI, localName);
- }
-
- /**
- * From StAX specs:
- *
- * 05-Apr-2004, TSa: Could try to determine status when text is actually
- * read. That'd prevent double reads... but would it slow down that
- * one reading so that net effect would be negative?
- */
- public boolean isWhiteSpace()
- {
- if (mCurrToken == CHARACTERS || mCurrToken == CDATA) {
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- if (mWsStatus == ALL_WS_UNKNOWN) {
- mWsStatus = mTextBuffer.isAllWhitespace() ?
- ALL_WS_YES : ALL_WS_NO;
- }
- return mWsStatus == ALL_WS_YES;
- }
- return (mCurrToken == SPACE);
- }
-
- public void require(int type, String nsUri, String localName)
- throws XMLStreamException
- {
- int curr = mCurrToken;
-
- /* There are some special cases; specifically, CDATA
- * is sometimes reported as CHARACTERS. Let's be lenient by
- * allowing both 'real' and reported types, for now.
- */
- if (curr != type) {
- if (curr == CDATA) {
- if (mCfgCoalesceText || mCfgReportTextAsChars) {
- curr = CHARACTERS;
- }
- } else if (curr == SPACE) {
- // Hmmh. Should we require it to be empty or something?
- //curr = CHARACTERS;
- // For now, let's not change the check
- }
- }
-
- if (type != curr) {
- throwParseError("Expected type "+tokenTypeDesc(type)
- +", current type "
- +tokenTypeDesc(curr));
- }
-
- if (localName != null) {
- if (curr != START_ELEMENT && curr != END_ELEMENT
- && curr != ENTITY_REFERENCE) {
- throwParseError("Expected non-null local name, but current token not a START_ELEMENT, END_ELEMENT or ENTITY_REFERENCE (was "+tokenTypeDesc(mCurrToken)+")");
- }
- String n = getLocalName();
- if (n != localName && !n.equals(localName)) {
- throwParseError("Expected local name '"+localName+"'; current local name '"+n+"'.");
- }
- }
- if (nsUri != null) {
- if (curr != START_ELEMENT && curr != END_ELEMENT) {
- throwParseError("Expected non-null NS URI, but current token not a START_ELEMENT or END_ELEMENT (was "+tokenTypeDesc(curr)+")");
- }
- String uri = mElementStack.getNsURI();
- // No namespace?
- if (nsUri.length() == 0) {
- if (uri != null && uri.length() > 0) {
- throwParseError("Expected empty namespace, instead have '"+uri+"'.");
- }
- } else {
- if ((nsUri != uri) && !nsUri.equals(uri)) {
- throwParseError("Expected namespace '"+nsUri+"'; have '"
- +uri+"'.");
- }
- }
- }
- // Ok, fine, all's good
- }
-
- /*
- ////////////////////////////////////////////////////
- // XMLStreamReader, iterating
- ////////////////////////////////////////////////////
- */
-
- public final int next()
- throws XMLStreamException
- {
- /* 24-Sep-2006, TSa: We may have deferred an exception that occurred
- * during parsing of the previous event. If so, now it needs to
- * be thrown.
- */
- if (mPendingException != null) {
- XMLStreamException strEx = mPendingException;
- mPendingException = null;
- throw strEx;
- }
-
- /* Note: can not yet accurately record the location, since the
- * previous event might not yet be completely finished...
- */
- if (mParseState == STATE_TREE) {
- int type = nextFromTree();
- mCurrToken = type;
- if (mTokenState < mStTextThreshold) { // incomplete?
- /* Can remain incomplete if lazy parsing is enabled,
- * and this is not a validatable text segment; otherwise
- * must finish
- */
- if (!mCfgLazyParsing ||
- (mValidateText && (type == CHARACTERS || type == CDATA))) {
- finishToken(false);
- }
- }
-
- /* Special cases -- sometimes (when coalescing text, or
- * when specifically configured to do so), CDATA and SPACE are
- * to be reported as CHARACTERS, although we still will
- * internally keep track of the real type.
- */
- if (type == CDATA) {
- if (mValidateText) {
- mElementStack.validateText(mTextBuffer, false);
- }
- if (mCfgCoalesceText || mCfgReportTextAsChars) {
- return CHARACTERS;
- }
- /*
- } else if (type == SPACE) {
- //if (mValidateText) { throw new IllegalStateException("Internal error: trying to validate SPACE event"); }
- */
- } else if (type == CHARACTERS) {
- if (mValidateText) {
- /* We may be able to determine that there will be
- * no more text coming for this element: but only
- * seeing the end tag marker ("
- * Note: as per StAX 1.0 specs, this method does NOT close the underlying
- * input reader. That is, unless the new StAX2 property
- * {@link org.codehaus.stax2.XMLInputFactory2#P_AUTO_CLOSE_INPUT} is
- * set to true.
- */
- public void close()
- throws XMLStreamException
- {
- if (mParseState != STATE_CLOSED) {
- mParseState = STATE_CLOSED;
- /* Let's see if we should notify factory that symbol table
- * has new entries, and may want to reuse this symbol table
- * instead of current root.
- */
- if (mCurrToken != END_DOCUMENT) {
- mCurrToken = mSecondaryToken = END_DOCUMENT;
- if (mSymbols.isDirty()) {
- mOwner.updateSymbolTable(mSymbols);
- }
- }
- /* Hmmh. Actually, we need to close all the dependant input
- * sources, first, and then also call close()
- * on the root input source object; it
- * will only do real close if that was enabled earlier.
- * The root input source also prevents multiple close() calls
- * for the underlying source, so we need not check that here.
- */
- closeAllInput(false);
- // And finally, can now recycle low-level (text) buffers
- mTextBuffer.recycle(true);
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // XMLStreamReader2 (StAX2) implementation
- ////////////////////////////////////////////////////
- */
-
- // // // StAX2, per-reader configuration
-
- public Object getFeature(String name)
- {
- // No readable features defined yet...
- throw new IllegalArgumentException(MessageFormat.format(ErrorConsts.ERR_UNKNOWN_FEATURE, new Object[] { name }));
- }
-
- public void setFeature(String name, Object value)
- {
- // Base-class has no settable features at this point.
- throw new IllegalArgumentException(MessageFormat.format(ErrorConsts.ERR_UNKNOWN_FEATURE, new Object[] { name }));
- }
-
- // NOTE: getProperty() defined in Stax 1.0 interface
-
- public boolean isPropertySupported(String name) {
- // !!! TBI: not all these properties are really supported
- return mConfig.isPropertySupported(name);
- }
-
- /**
- * @param name Name of the property to set
- * @param value Value to set property to.
- *
- * @return True, if the specified property was succesfully
- * set to specified value; false if its value was not changed
- */
- public boolean setProperty(String name, Object value)
- {
- boolean ok = mConfig.setProperty(name, value);
- /* To make [WSTX-50] work fully dynamically (i.e. allow
- * setting BASE_URL after stream reader has been constructed)
- * need to force
- */
- if (ok && WstxInputProperties.P_BASE_URL.equals(name)) {
- // Easiest to just access from config: may come in as a String etc
- mInput.overrideSource(mConfig.getBaseURL());
- }
- return ok;
- }
-
- // // // StAX2, additional traversal methods
-
- public void skipElement() throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- int nesting = 1; // need one more end elements than start elements
-
- while (true) {
- int type = next();
- if (type == START_ELEMENT) {
- ++nesting;
- } else if (type == END_ELEMENT) {
- if (--nesting == 0) {
- break;
- }
- }
- }
- }
-
- // // // StAX2, additional attribute access
-
- public AttributeInfo getAttributeInfo() throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- /* Although attribute collector knows about specific parsed
- * information, the element stack has DTD-derived information (if
- * any)... and knows how to call attribute collector when necessary.
- */
- return mElementStack;
- }
-
- // // // StAX2, Additional DTD access
-
- /**
- * Since this class implements {@link DTDInfo}, method can just
- * return
- * TODO: try to optimize to allow completely streaming pass-through:
- * currently will still read all data in memory buffers before
- * outputting
- *
- * @param w Writer to use for writing textual contents
- * @param preserveContents If true, reader has to preserve contents
- * so that further calls to
- * Note: DTD-handling sub-classes need to override this method.
- */
- public Object getProcessedDTD() {
- return null;
- }
-
- public String getDTDRootName() {
- if (mRootPrefix == null) {
- return mRootLName;
- }
- return mRootPrefix + ":" + mRootLName;
- }
-
- public String getDTDPublicId() {
- return mDtdPublicId;
- }
-
- public String getDTDSystemId() {
- return mDtdSystemId;
- }
-
- /**
- * @return Internal subset portion of the DOCTYPE declaration, if any;
- * empty String if none
- */
- public String getDTDInternalSubset() {
- if (mCurrToken != DTD) {
- return null;
- }
- return mTextBuffer.contentsAsString();
- }
-
- /**
- * Internal method used by implementation
- */
- private char[] getDTDInternalSubsetArray() {
- /* Note: no checks for current state, but only because it's
- * an internal method and callers are known to ensure it's ok
- * to call this
- */
- return mTextBuffer.contentsAsArray();
- }
-
- // // StAX2, v2.0
-
- /**
- * Sub-class will override this method
- */
- public DTDValidationSchema getProcessedDTDSchema() {
- return null;
- }
-
- /*
- ////////////////////////////////////////////////////
- // LocationInfo implementation (StAX 2)
- ////////////////////////////////////////////////////
- */
-
- // // // First, the "raw" offset accessors:
-
- public long getStartingByteOffset() {
- /* 15-Apr-2005, TSa: No way to reliably keep track of byte offsets,
- * at least for variable-length encodings... so let's just
- * return -1 for now
- */
- return -1L;
- }
-
- public long getStartingCharOffset() {
- return mTokenInputTotal;
- }
-
- public long getEndingByteOffset() throws XMLStreamException
- {
- /* 15-Apr-2005, TSa: No way to reliably keep track of byte offsets,
- * at least for variable-length encodings... so let's just
- * return -1 for now
- */
- return -1;
- }
-
- public long getEndingCharOffset() throws XMLStreamException
- {
- // Need to get to the end of the token, if not there yet
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- return mCurrInputProcessed + mInputPtr;
- }
-
- // // // and then the object-based access methods:
-
- public final Location getLocation() {
- return getStartLocation();
- }
-
- // public XMLStreamLocation2 getStartLocation() // from base class
- // public XMLStreamLocation2 getCurrentLocation() // - "" -
-
- public final XMLStreamLocation2 getEndLocation()
- throws XMLStreamException
- {
- // Need to get to the end of the token, if not there yet
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- // And then we just need the current location!
- return getCurrentLocation();
- }
-
- /*
- ////////////////////////////////////////////////////
- // Stax2 validation
- ////////////////////////////////////////////////////
- */
-
- public XMLValidator validateAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- // Not implemented by the basic reader:
- return null;
- }
-
- public XMLValidator stopValidatingAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- // Not implemented by the basic reader:
- return null;
- }
-
- public XMLValidator stopValidatingAgainst(XMLValidator validator)
- throws XMLStreamException
- {
- // Not implemented by the basic reader:
- return null;
- }
-
- public ValidationProblemHandler setValidationProblemHandler(ValidationProblemHandler h)
- {
- // Not implemented by the basic reader:
- return null;
- }
-
- /*
- //////////////////////////////////////////////////////
- // StreamReaderImpl implementation
- //////////////////////////////////////////////////////
- */
-
- public EntityDecl getCurrentEntityDecl() {
- return mCurrEntity;
- }
-
- /**
- * Method called by {@link com.ctc.wstx.evt.DefaultEventAllocator}
- * to get double-indirection necessary for constructing start element
- * events.
- *
- * @return Null, if stream does not point to start element; whatever
- * callback returns otherwise.
- */
- public Object withStartElement(ElemCallback cb, Location loc)
- {
- if (mCurrToken != START_ELEMENT) {
- return null;
- }
- return cb.withStartElement(loc, getName(),
- mElementStack.createNonTransientNsContext(loc),
- mAttrCollector.buildAttrOb(),
- mStEmptyElem);
- }
-
- public boolean isNamespaceAware() {
- return mCfgNsEnabled;
- }
-
- /**
- * Method needed by classes (like stream writer implementations)
- * that want to have efficient direct access to element stack
- * implementation
- */
- public InputElementStack getInputElementStack() {
- return mElementStack;
- }
-
- /**
- * Method needed by classes (like stream writer implementations)
- * that want to have efficient direct access to attribute collector
- * Object, for optimal attribute name and value access.
- */
- public AttributeCollector getAttributeCollector() {
- return mAttrCollector;
- }
-
- /*
- //////////////////////////////////////////////////////
- // Support for SAX XMLReader implementation
- //////////////////////////////////////////////////////
- */
-
- public void fireSaxStartElement(ContentHandler h, Attributes attrs)
- throws SAXException
- {
- if (h != null) {
- // First; any ns declarations?
- int nsCount = mElementStack.getCurrentNsCount();
- for (int i = 0; i < nsCount; ++i) {
- String prefix = mElementStack.getLocalNsPrefix(i);
- String uri = mElementStack.getLocalNsURI(i);
- h.startPrefixMapping((prefix == null) ? "" : prefix, uri);
- }
-
- // Then start-elem event itself:
- String uri = mElementStack.getNsURI();
- // Sax requires "" (not null) for ns uris...
- h.startElement((uri == null) ? "" : uri,
- mElementStack.getLocalName(), getPrefixedName(), attrs);
- }
- }
-
- public void fireSaxEndElement(ContentHandler h)
- throws SAXException
- {
- if (h != null) {
- /* Order of events is reversed (wrt. start-element): first
- * the end tag event, then unbound prefixes
- */
- String uri = mElementStack.getNsURI();
- // Sax requires "" (not null) for ns uris...
- h.endElement((uri == null) ? "" : uri,
- mElementStack.getLocalName(), getPrefixedName());
- // Any expiring ns declarations?
- int nsCount = mElementStack.getCurrentNsCount();
- for (int i = 0; i < nsCount; ++i) {
- String prefix = mElementStack.getLocalNsPrefix(i);
- //String nsUri = mElementStack.getLocalNsURI(i);
- h.endPrefixMapping((prefix == null) ? "" : prefix);
- }
- }
- }
-
- public void fireSaxCharacterEvents(ContentHandler h)
- throws XMLStreamException, SAXException
- {
- if (h != null) {
- if (mPendingException != null) {
- XMLStreamException sex = mPendingException;
- mPendingException = null;
- throw sex;
- }
- /* Let's not defer errors; SAXTest implies
- * it's expected errors are thrown right away
- */
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- mTextBuffer.fireSaxCharacterEvents(h);
- }
- }
-
- public void fireSaxSpaceEvents(ContentHandler h)
- throws XMLStreamException, SAXException
- {
- if (h != null) {
- if (mTokenState < mStTextThreshold) {
- finishToken(false); // no error deferring
- }
- mTextBuffer.fireSaxSpaceEvents(h);
- }
- }
-
- public void fireSaxCommentEvent(LexicalHandler h)
- throws XMLStreamException, SAXException
- {
- if (h != null) {
- if (mTokenState < mStTextThreshold) {
- finishToken(false); // no error deferring
- }
- mTextBuffer.fireSaxCommentEvent(h);
- }
- }
-
- public void fireSaxPIEvent(ContentHandler h)
- throws XMLStreamException, SAXException
- {
- if (h != null) {
- if (mTokenState < mStTextThreshold) {
- finishToken(false); // no error deferring
- }
- h.processingInstruction(mCurrName, mTextBuffer.contentsAsString());
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, config access
- ////////////////////////////////////////////////////
- */
-
- protected final boolean hasConfigFlags(int flags) {
- return (mConfigFlags & flags) == flags;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, parsing helper methods
- ////////////////////////////////////////////////////
- */
-
- /**
- * @return Null, if keyword matches ok; String that contains erroneous
- * keyword if not.
- */
- protected String checkKeyword(char c, String expected)
- throws XMLStreamException
- {
- int ptr = 0;
- int len = expected.length();
-
- while (expected.charAt(ptr) == c && ++ptr < len) {
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- int ci = getNext();
- if (ci < 0) { // EOF
- break;
- }
- c = (char) ci;
- }
- }
-
- if (ptr == len) {
- // Probable match... but let's make sure keyword is finished:
- int i = peekNext();
- if (i < 0 || (!isNameChar((char) i) && i != ':')) {
- return null;
- }
- // Nope, continues, need to find the rest:
- }
-
- StringBuffer sb = new StringBuffer(expected.length() + 16);
- sb.append(expected.substring(0, ptr));
- if (ptr < len) {
- sb.append(c);
- }
-
- while (true) {
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else {
- int ci = getNext();
- if (ci < 0) { // EOF
- break;
- }
- c = (char) ci;
- }
- if (!isNameChar(c)) {
- // Let's push it back then
- --mInputPtr;
- break;
- }
- sb.append(c);
- }
-
- return sb.toString();
- }
-
- protected void checkCData()
- throws XMLStreamException
- {
- String wrong = checkKeyword(getNextCharFromCurrent(SUFFIX_IN_CDATA), "CDATA");
- if (wrong != null) {
- throwParseError("Unrecognized XML directive '"+wrong+"'; expected 'CDATA'.");
- }
- // Plus, need the bracket too:
- char c = getNextCharFromCurrent(SUFFIX_IN_CDATA);
- if (c != '[') {
- throwUnexpectedChar(c, "excepted '[' after '= 3
- && (ch = resolveSimpleEntity(true)) != 0) {
- // Ok, fine, c is whatever it is
- ;
- } else { // full entity just changes buffer...
- ch = fullyResolveEntity(false);
- if (ch == 0) {
- // need to skip output, thusly (expanded to new input source)
- continue;
- }
- }
- if (ch <= 0xFFFF) {
- c = (char) ch;
- } else {
- ch -= 0x10000;
- if (outPtr >= outLen) {
- outBuf = tb.bufferFull(1);
- outLen = outBuf.length;
- }
- outBuf[outPtr++] = (char) ((ch >> 10) + 0xD800);
- c = (char) ((ch & 0x3FF) + 0xDC00);
- }
- }
- } else if (c == '<') {
- throwParseError("Unexpected '<' "+SUFFIX_IN_ATTR_VALUE);
- }
-
- // Ok, let's just add char in, whatever it was
- if (outPtr >= outLen) {
- outBuf = tb.bufferFull(1);
- outLen = outBuf.length;
- }
- outBuf[outPtr++] = c;
- }
-
- // Fine; let's tell TextBuild we're done:
- tb.setBufferSize(outPtr);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, parsing prolog (before root) and
- // epilog
- /////////////////////////////////////////////////////
- */
-
- /**
- * Method called to find type of next token in prolog; either reading
- * just enough information to know the type (lazy parsing), or the
- * full contents (non-lazy)
- *
- * @return True if we hit EOI, false otherwise
- */
- private boolean nextFromProlog(boolean isProlog)
- throws XMLStreamException
- {
- int i;
-
- // First, do we need to finish currently open token?
- if (mTokenState < mStTextThreshold) {
- mTokenState = TOKEN_FULL_COALESCED;
- i = skipToken();
- // note: skipToken() updates the start location
- } else {
- // Need to update the start location...
- mTokenInputTotal = mCurrInputProcessed + mInputPtr;
- mTokenInputRow = mCurrInputRow;
- mTokenInputCol = mInputPtr - mCurrInputRowStart;
- i = getNext();
- }
-
- // Any white space to parse or skip?
- if (i <= CHAR_SPACE && i >= 0) {
- // Need to return as an event?
- if (hasConfigFlags(CFG_REPORT_PROLOG_WS)) {
- mCurrToken = SPACE;
- if (readSpacePrimary((char) i, true)) {
- /* no need to worry about coalescing, since CDATA is not
- * allowed at this level...
- */
- mTokenState = TOKEN_FULL_COALESCED;
- } else {
- if (mCfgLazyParsing) {
- /* Let's not even bother checking if it's
- * "long enough"; shouldn't usually matter, but few
- * apps care to get multiple adjacent SPACE events...
- */
- mTokenState = TOKEN_STARTED;
- } else {
- readSpaceSecondary(true);
- mTokenState = TOKEN_FULL_COALESCED;
- }
- }
- return false;
- }
- // If not, can skip it right away
- --mInputPtr; // to handle linefeeds gracefully
- i = getNextAfterWS();
- if (i >= 0) {
- // ... after which location has to be reset properly:
- /* 11-Apr-2005, TSa: But note that we need to "move back"
- * column and total offset values by one, to compensate
- * for the char that was read (row can not have changed,
- * since it's non-WS, and thus non-lf/cr char)
- */
- mTokenInputTotal = mCurrInputProcessed + mInputPtr - 1;
- mTokenInputRow = mCurrInputRow;
- mTokenInputCol = mInputPtr - mCurrInputRowStart - 1;
- }
- }
-
- // Did we hit EOI?
- if (i < 0) {
- handleEOF(isProlog);
- mParseState = STATE_CLOSED;
- return true;
- }
-
- // Now we better have a lt...
- if (i != '<') {
- throwUnexpectedChar(i, (isProlog ? SUFFIX_IN_PROLOG : SUFFIX_IN_EPILOG)
- +"; expected '<'");
- }
-
- // And then it should be easy to figure out type:
- char c = getNextChar(isProlog ? SUFFIX_IN_PROLOG : SUFFIX_IN_EPILOG);
-
- if (c == '?') { // proc. inst
- mCurrToken = readPIPrimary();
- } else if (c == '!') { // DOCTYPE or comment (or CDATA, but not legal here)
- // Need to figure out bit more first...
- nextFromPrologBang(isProlog);
- } else if (c == '/') { // end tag not allowed...
- if (isProlog) {
- throwParseError("Unexpected character combination ' clean'n recycle
- // It's ok to get EOF from epilog but not from prolog
- if (isProlog) {
- throwUnexpectedEOF(SUFFIX_IN_PROLOG);
- }
- return mCurrToken;
- }
-
- /**
- * Method called if a root-level element is found after the main
- * root element was closed. This is legal in multi-doc parsing
- * mode (and in fragment mode), but not in the default single-doc
- * mode.
- *
- * @return Token to return
- */
- private int handleExtraRoot(char c)
- throws XMLStreamException
- {
- if (!mConfig.inputParsingModeDocuments()) {
- /* Has to be single-doc mode, since fragment mode
- * should never get here (since fragment mode never has epilog
- * or prolog modes)
- */
- throwParseError("Illegal to have multiple roots (start tag in epilog?).");
- }
- // Need to push back the char, since it is the first char of elem name
- --mInputPtr;
- return handleMultiDocStart(START_ELEMENT);
- }
-
- /**
- * Method called when an event was encountered that indicates document
- * boundary in multi-doc mode. Needs to trigger dummy
- * END_DOCUMENT/START_DOCUMENT event combination, followed by the
- * handling of the original event.
- *
- * @return Event type to return
- */
- protected int handleMultiDocStart(int nextEvent)
- {
- mParseState = STATE_MULTIDOC_HACK;
- mTokenState = TOKEN_FULL_COALESCED; // this is a virtual event after all...
- mSecondaryToken = nextEvent;
- return END_DOCUMENT;
- }
-
- /**
- * Method called to get the next event when we are "multi-doc hack" mode,
- * during which extra END_DOCUMENT/START_DOCUMENT events need to be
- * returned.
- */
- private int nextFromMultiDocState()
- throws XMLStreamException
- {
- if (mCurrToken == END_DOCUMENT) {
- /* Ok; this is the initial step; need to advance: need to parse
- * xml declaration if that was the cause, otherwise just clear
- * up values.
- */
- if (mSecondaryToken == START_DOCUMENT) {
- handleMultiDocXmlDecl();
- } else { // Nah, DOCTYPE or start element... just need to clear decl info:
- mDocXmlEncoding = null;
- mDocXmlVersion = XmlConsts.XML_V_UNKNOWN;
- mDocStandalone = DOC_STANDALONE_UNKNOWN;
- }
- return START_DOCUMENT;
- }
- if (mCurrToken == START_DOCUMENT) {
- mParseState = STATE_PROLOG; // yup, we are now officially in prolog again...
-
- // Had an xml decl (ie. "real" START_DOCUMENT event)
- if (mSecondaryToken == START_DOCUMENT) { // was a real xml decl
- nextFromProlog(true);
- return mCurrToken;
- }
- // Nah, start elem or DOCTYPE
- if (mSecondaryToken == START_ELEMENT) {
- handleRootElem(getNextChar(SUFFIX_IN_ELEMENT));
- return START_ELEMENT;
- }
- if (mSecondaryToken == DTD) {
- mStDoctypeFound = true;
- startDTD();
- return DTD;
- }
- }
- throw new IllegalStateException("Internal error: unexpected state; current event "
- +tokenTypeDesc(mCurrToken)+", sec. state: "+tokenTypeDesc(mSecondaryToken));
- }
-
- protected void handleMultiDocXmlDecl()
- throws XMLStreamException
- {
- // Let's default these first
- mDocStandalone = DOC_STANDALONE_UNKNOWN;
- mDocXmlEncoding = null;
-
- char c = getNextInCurrAfterWS(SUFFIX_IN_XML_DECL);
- String wrong = checkKeyword(c, XmlConsts.XML_DECL_KW_VERSION);
- if (wrong != null) {
- throwParseError(ErrorConsts.ERR_UNEXP_KEYWORD, wrong, XmlConsts.XML_DECL_KW_VERSION);
- }
- c = skipEquals(XmlConsts.XML_DECL_KW_VERSION, SUFFIX_IN_XML_DECL);
- TextBuffer tb = mTextBuffer;
- tb.resetInitialized();
- parseQuoted(XmlConsts.XML_DECL_KW_VERSION, c, tb);
-
- if (tb.equalsString(XmlConsts.XML_V_10_STR)) {
- mDocXmlVersion = XmlConsts.XML_V_10;
- mXml11 = false;
- } else if (tb.equalsString(XmlConsts.XML_V_11_STR)) {
- mDocXmlVersion = XmlConsts.XML_V_11;
- mXml11 = true;
- } else {
- mDocXmlVersion = XmlConsts.XML_V_UNKNOWN;
- mXml11 = false;
- throwParseError("Unexpected xml version '"+tb.toString()+"'; expected '"+XmlConsts.XML_V_10_STR+"' or '"+XmlConsts.XML_V_11_STR+"'");
- }
-
- c = getNextInCurrAfterWS(SUFFIX_IN_XML_DECL);
-
- if (c != '?') { // '?' signals end...
- if (c == 'e') { // encoding
- wrong = checkKeyword(c, XmlConsts.XML_DECL_KW_ENCODING);
- if (wrong != null) {
- throwParseError(ErrorConsts.ERR_UNEXP_KEYWORD, wrong, XmlConsts.XML_DECL_KW_ENCODING);
- }
- c = skipEquals(XmlConsts.XML_DECL_KW_ENCODING, SUFFIX_IN_XML_DECL);
- tb.resetWithEmpty();
- parseQuoted(XmlConsts.XML_DECL_KW_ENCODING, c, tb);
- mDocXmlEncoding = tb.toString();
- /* should we verify encoding at this point? let's not, for now;
- * since it's for information only, first declaration from
- * bootstrapper is used for the whole stream.
- */
- c = getNextInCurrAfterWS(SUFFIX_IN_XML_DECL);
- } else if (c != 's') {
- throwUnexpectedChar(c, " in xml declaration; expected either 'encoding' or 'standalone' pseudo-attribute");
- }
-
- // Standalone?
- if (c == 's') {
- wrong = checkKeyword(c, XmlConsts.XML_DECL_KW_STANDALONE);
- if (wrong != null) {
- throwParseError(ErrorConsts.ERR_UNEXP_KEYWORD, wrong, XmlConsts.XML_DECL_KW_STANDALONE);
- }
- c = skipEquals(XmlConsts.XML_DECL_KW_STANDALONE, SUFFIX_IN_XML_DECL);
- tb.resetWithEmpty();
- parseQuoted(XmlConsts.XML_DECL_KW_STANDALONE, c, tb);
- if (tb.equalsString(XmlConsts.XML_SA_YES)) {
- mDocStandalone = DOC_STANDALONE_YES;
- } else if (tb.equalsString(XmlConsts.XML_SA_NO)) {
- mDocStandalone = DOC_STANDALONE_NO;
- } else {
- throwParseError("Unexpected xml '"+XmlConsts.XML_DECL_KW_STANDALONE+"' pseudo-attribute value '"
- +tb.toString()+"'; expected '"+XmlConsts.XML_SA_YES+"' or '"+
- XmlConsts.XML_SA_NO+"'");
- }
- c = getNextInCurrAfterWS(SUFFIX_IN_XML_DECL);
- }
- }
-
- if (c != '?') {
- throwUnexpectedChar(c, " in xml declaration; expected '?>' as the end marker");
- }
- c = getNextCharFromCurrent(SUFFIX_IN_XML_DECL);
- if (c != '>') {
- throwUnexpectedChar(c, " in xml declaration; expected '>' to close the declaration");
- }
- }
-
- /**
- * Method that checks that input following is of form
- * '[S]* '=' [S]*' (as per XML specs, production #25).
- * Will push back non-white space characters as necessary, in
- * case no equals char is encountered.
- */
- protected final char skipEquals(String name, String eofMsg)
- throws XMLStreamException
- {
- char c = getNextInCurrAfterWS(eofMsg);
- if (c != '=') {
- throwUnexpectedChar(c, " in xml declaration; expected '=' to follow pseudo-attribute '"+name+"'");
- }
- // trailing space?
- return getNextInCurrAfterWS(eofMsg);
- }
-
- /**
- * Method called to parse quoted xml declaration pseudo-attribute values.
- * Works similar to attribute value parsing, except no entities can be
- * included, and in general need not be as picky (since caller is to
- * verify contents). One exception is that we do check for linefeeds
- * and lt chars, since they generally would indicate problems and
- * are useful to catch early on (can happen if a quote is missed etc)
- *
- * Note: since it'll be called at most 3 times per document, this method
- * is not optimized too much.
- */
- protected final void parseQuoted(String name, char quoteChar, TextBuffer tbuf)
- throws XMLStreamException
- {
- if (quoteChar != '"' && quoteChar != '\'') {
- throwUnexpectedChar(quoteChar, " in xml declaration; waited ' or \" to start a value for pseudo-attribute '"+name+"'");
- }
- char[] outBuf = tbuf.getCurrentSegment();
- int outPtr = 0;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextChar(SUFFIX_IN_XML_DECL);
-
- if (c == quoteChar) {
- break;
- }
- if (c < CHAR_SPACE || c == '<') {
- throwUnexpectedChar(c, SUFFIX_IN_XML_DECL);
- } else if (c == CHAR_NULL) {
- throwNullChar();
- }
- if (outPtr >= outBuf.length) {
- outBuf = tbuf.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- }
- tbuf.setCurrentLength(outPtr);
- }
-
- /**
- * Called after character sequence '<!' has been found; expectation is
- * that it'll either be DOCTYPE declaration (if we are in prolog and
- * haven't yet seen one), or a comment. CDATA is not legal here;
- * it would start same way otherwise.
- */
- private void nextFromPrologBang(boolean isProlog)
- throws XMLStreamException
- {
- int i = getNext();
- if (i < 0) {
- throwUnexpectedEOF(SUFFIX_IN_PROLOG);
- }
- if (i == 'D') { // Doctype declaration?
- String keyw = checkKeyword('D', "DOCTYPE");
- if (keyw != null) {
- throwParseError("Unrecognized XML directive '
- *
- *. And we have already read the DOCTYPE token.
- */
-
- char c = getNextInCurrAfterWS(SUFFIX_IN_DTD);
- if (mCfgNsEnabled) {
- String str = parseLocalName(c);
- c = getNextChar(SUFFIX_IN_DTD);
- if (c == ':') { // Ok, got namespace and local name
- mRootPrefix = str;
- mRootLName = parseLocalName(getNextChar(SUFFIX_EOF_EXP_NAME));
- } else if (c <= CHAR_SPACE || c == '[' || c == '>') {
- // ok to get white space or '[', or closing '>'
- --mInputPtr; // pushback
- mRootPrefix = null;
- mRootLName = str;
- } else {
- throwUnexpectedChar(c, " in DOCTYPE declaration; expected '[' or white space.");
- }
- } else {
- mRootLName = parseFullName(c);
- mRootPrefix = null;
- }
-
- // Ok, fine, what next?
- c = getNextInCurrAfterWS(SUFFIX_IN_DTD);
- if (c != '[' && c != '>') {
- String keyw = null;
-
- if (c == 'P') {
- keyw = checkKeyword(getNextChar(SUFFIX_IN_DTD), "UBLIC");
- if (keyw != null) {
- keyw = "P" + keyw;
- } else {
- if (!skipWS(getNextChar(SUFFIX_IN_DTD))) {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected a space between PUBLIC keyword and public id");
- }
- c = getNextCharFromCurrent(SUFFIX_IN_DTD);
- if (c != '"' && c != '\'') {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected a public identifier.");
- }
- mDtdPublicId = parsePublicId(c, SUFFIX_IN_DTD);
- if (mDtdPublicId.length() == 0) {
- // According to XML specs, this isn't illegal?
- // however, better report it as empty, not null.
- //mDtdPublicId = null;
- }
- if (!skipWS(getNextChar(SUFFIX_IN_DTD))) {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected a space between public and system identifiers");
- }
- c = getNextCharFromCurrent(SUFFIX_IN_DTD);
- if (c != '"' && c != '\'') {
- throwParseError(SUFFIX_IN_DTD+"; expected a system identifier.");
- }
- mDtdSystemId = parseSystemId(c, mNormalizeLFs, SUFFIX_IN_DTD);
- if (mDtdSystemId.length() == 0) {
- // According to XML specs, this isn't illegal?
- // however, better report it as empty, not null.
- //mDtdSystemId = null;
- }
- }
- } else if (c == 'S') {
- mDtdPublicId = null;
- keyw = checkKeyword(getNextChar(SUFFIX_IN_DTD), "YSTEM");
- if (keyw != null) {
- keyw = "S" + keyw;
- } else {
- c = getNextInCurrAfterWS(SUFFIX_IN_DTD);
- if (c != '"' && c != '\'') {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected a system identifier.");
- }
- mDtdSystemId = parseSystemId(c, mNormalizeLFs, SUFFIX_IN_DTD);
- if (mDtdSystemId.length() == 0) {
- // According to XML specs, this isn't illegal?
- mDtdSystemId = null;
- }
- }
- } else {
- if (!isNameStartChar(c)) {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected keywords 'PUBLIC' or 'SYSTEM'.");
- } else {
- --mInputPtr;
- keyw = checkKeyword(c, "SYSTEM"); // keyword passed in doesn't matter
- }
- }
-
- if (keyw != null) { // error:
- throwParseError("Unexpected keyword '"+keyw+"'; expected 'PUBLIC' or 'SYSTEM'");
- }
-
- // Ok, should be done with external DTD identifier:
- c = getNextInCurrAfterWS(SUFFIX_IN_DTD);
- }
-
- if (c == '[') { // internal subset
- ;
- } else {
- if (c != '>') {
- throwUnexpectedChar(c, SUFFIX_IN_DTD+"; expected closing '>'.");
- }
- }
-
- /* Actually, let's just push whatever char it is, back; this way
- * we can lazily initialize text buffer with DOCTYPE declaration
- * if/as necessary, even if there's no internal subset.
- */
- --mInputPtr; // pushback
- mTokenState = TOKEN_STARTED;
- }
-
- /**
- * This method gets called to handle remainder of DOCTYPE declaration,
- * essentially the optional internal subset. This class implements the
- * basic "ignore it" functionality, but can optionally still store copy
- * of the contents to the read buffer.
- *
- * NOTE: Since this default implementation will be overridden by
- * some sub-classes, make sure you do NOT change the method signature.
- *
- * @param copyContents If true, will copy contents of the internal
- * subset of DOCTYPE declaration
- * in the text buffer; if false, will just completely ignore the
- * subset (if one found).
- */
- protected void finishDTD(boolean copyContents)
- throws XMLStreamException
- {
- /* We know there are no spaces, as this char was read and pushed
- * back earlier...
- */
- char c = getNextChar(SUFFIX_IN_DTD);
- if (c == '[') {
- // Do we need to get contents as text too?
- if (copyContents) {
- ((BranchingReaderSource) mInput).startBranch(mTextBuffer, mInputPtr, mNormalizeLFs);
- }
-
- try {
- MinimalDTDReader.skipInternalSubset(this, mInput, mConfig);
- } finally {
- /* Let's close branching in any and every case (may allow
- * graceful recovery in error cases in future
- */
- if (copyContents) {
- /* Need to "push back" ']' got in the succesful case
- * (that's -1 part below);
- * in error case it'll just be whatever last char was.
- */
- ((BranchingReaderSource) mInput).endBranch(mInputPtr-1);
- }
- }
-
- // And then we need closing '>'
- c = getNextCharAfterWS(SUFFIX_IN_DTD_INTERNAL);
- }
-
- if (c != '>') {
- throwUnexpectedChar(c, "; expected '>' to finish DOCTYPE declaration.");
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, main parsing (inside root)
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method called to parse beginning of the next event within
- * document tree, and return its type.
- */
- private final int nextFromTree()
- throws XMLStreamException
- {
- int i;
-
- // First, do we need to finish currently open token?
- if (mTokenState < mStTextThreshold) {
- // No need to update state... will get taken care of
- /* 03-Mar-2006, TSa: Let's add a sanity check here, temporarily,
- * to ensure we never skip any textual content when it is
- * to be validated
- */
- if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT) {
- if (mCurrToken == CHARACTERS || mCurrToken == CDATA) { // should never happen
- throwParseError("Internal error: skipping validatable text");
- }
- }
- i = skipToken();
- // note: skipToken() updates the start location
- } else {
- /* Start/end elements are never unfinished (ie. are always
- * completely read in)
- */
- if (mCurrToken == START_ELEMENT) {
- // Start tag may be an empty tag:
- if (mStEmptyElem) {
- // and if so, we'll then get 'virtual' close tag:
- mStEmptyElem = false;
- // ... and location info is correct already
- /* 27-Feb-2009, TSa: but we do have to handle validation
- * of the end tag now
- */
- int vld = mElementStack.validateEndElement();
- mVldContent = vld;
- mValidateText = (vld == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT);
- return END_ELEMENT;
- }
- } else if (mCurrToken == END_ELEMENT) {
- // Close tag removes current element from stack
- if (!mElementStack.pop()) { // false if root closed
- // if so, we'll get to epilog, unless in fragment mode
- if (!mConfig.inputParsingModeFragment()) {
- return closeContentTree();
- }
- // in fragment mode, fine, we'll just continue
- }
- } else if (mCurrToken == CDATA && mTokenState <= TOKEN_PARTIAL_SINGLE) {
- /* Just returned a partial CDATA... that's ok, just need to
- * know we won't get opening marker etc.
- * The tricky part here is just to ensure there's at least
- * one character; if not, need to just discard the empty
- * 'event' (note that it is possible to have an initial
- * empty CDATA event for truly empty CDATA block; but not
- * partial ones!). Let's just read it like a new
- * CData section first:
- */
- // First, need to update the start location...
- mTokenInputTotal = mCurrInputProcessed + mInputPtr;
- mTokenInputRow = mCurrInputRow;
- mTokenInputCol = mInputPtr - mCurrInputRowStart;
- char c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextChar(SUFFIX_IN_CDATA);
- if (readCDataPrimary(c)) { // got it all!
- /* note: can not be in coalescing mode at this point;
- * as we can never have partial cdata without unfinished
- * token
- */
- // ... still need to have gotten at least 1 char though:
- if (mTextBuffer.size() > 0) {
- return CDATA;
- }
- // otherwise need to continue and parse the next event
- } else {
- /* Hmmh. Have to verify we get at least one char from
- * CData section; if so, we are good to go for now;
- * if not, need to get that damn char first:
- */
- if (mTextBuffer.size() == 0
- && readCDataSecondary(mCfgLazyParsing
- ? 1 : mShortestTextSegment)) {
- // Ok, all of it read
- if (mTextBuffer.size() > 0) {
- // And had some contents
- mTokenState = TOKEN_FULL_SINGLE;
- return CDATA;
- }
- // if nothing read, we'll just fall back (see below)
- } else { // good enough!
- mTokenState = TOKEN_PARTIAL_SINGLE;
- return CDATA;
- }
- }
-
- /* If we get here, it was the end of the section, without
- * any more text inside CDATA, so let's just continue
- */
- }
- // Once again, need to update the start location info:
- mTokenInputTotal = mCurrInputProcessed + mInputPtr;
- mTokenInputRow = mCurrInputRow;
- mTokenInputCol = mInputPtr - mCurrInputRowStart;
- i = getNext();
- }
-
- if (i < 0) {
- /* 07-Oct-2005, TSa: May be ok in fragment mode (not otherwise),
- * but we can just check if element stack has anything, as that
- * handles all cases
- */
- if (!mElementStack.isEmpty()) {
- throwUnexpectedEOF();
- }
- return handleEOF(false);
- }
-
- /* 26-Aug-2004, TSa: We have to deal with entities, usually, if
- * they are the next thing; even in non-expanding mode there
- * are entities and then there are entities... :-)
- * Let's start with char entities; they can be expanded right away.
- */
- while (i == '&') {
- mWsStatus = ALL_WS_UNKNOWN;
-
- /* 30-Aug-2004, TSa: In some contexts entities are not
- * allowed in any way, shape or form:
- */
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- /* May be char entity, general entity; whatever it is it's
- * invalid!
- */
- reportInvalidContent(ENTITY_REFERENCE);
- }
-
- /* Need to call different methods based on whether we can do
- * automatic entity expansion or not:
- */
- int ch = mCfgReplaceEntities ?
- fullyResolveEntity(true) : resolveCharOnlyEntity(true);
-
- if (ch != 0) {
- /* Char-entity... need to initialize text output buffer, then;
- * independent of whether it'll be needed or not.
- */
- /* 30-Aug-2004, TSa: In some contexts only white space is
- * accepted...
- */
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- // As per xml specs, only straight white space is legal
- if (ch > CHAR_SPACE) {
- /* 21-Sep-2008, TSa: Used to also require a call to
- * 'mElementStack.reallyValidating', if only ws
- * allowed, to cover the case where non-typing-dtd
- * was only used to discover SPACE type. But
- * now that we have CONTENT_ALLOW_WS_NONSTRICT,
- * shouldn't be needed.
- */
- //if (mVldContent < XMLValidator.CONTENT_ALLOW_WS || mElementStack.reallyValidating()) {
- reportInvalidContent(CHARACTERS);
- }
- }
- TextBuffer tb = mTextBuffer;
- tb.resetInitialized();
- if (ch <= 0xFFFF) {
- tb.append((char) ch);
- } else {
- ch -= 0x10000;
- tb.append((char) ((ch >> 10) + 0xD800));
- tb.append((char) ((ch & 0x3FF) + 0xDC00));
- }
- mTokenState = TOKEN_STARTED;
- return CHARACTERS;
- }
-
- /* Nope; was a general entity... in auto-mode, it's now been
- * expanded; in non-auto, need to figure out entity itself.
- */
- if (!mCfgReplaceEntities|| mCfgTreatCharRefsAsEntities) {
- if (!mCfgTreatCharRefsAsEntities) {
- final EntityDecl ed = resolveNonCharEntity();
- // Note: ed may still be null at this point
- mCurrEntity = ed;
- }
- // Note: ed may still be null at this point
- mTokenState = TOKEN_FULL_COALESCED;
- /*
- // let's not worry about non-parsed entities, since this is unexpanded mode
- // ... although it'd be an error either way? Should we report it?
- if (ed != null && !ed.isParsed()) {
- throwParseError("Reference to unparsed entity '"+ed.getName()+"' from content not allowed.");
- }
- */
- return ENTITY_REFERENCE;
- }
-
- // Otherwise automatic expansion fine; just need the next char:
- i = getNextChar(SUFFIX_IN_DOC);
- }
-
- if (i == '<') { // Markup
- // And then it should be easy to figure out type:
- char c = getNextChar(SUFFIX_IN_ELEMENT);
- if (c == '?') { // proc. inst
- // 30-Aug-2004, TSa: Not legal for EMPTY elements
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- reportInvalidContent(PROCESSING_INSTRUCTION);
- }
- return readPIPrimary();
- }
-
- if (c == '!') { // CDATA or comment
- // Need to figure out bit more first...
- int type = nextFromTreeCommentOrCData();
- // 30-Aug-2004, TSa: Not legal for EMPTY elements
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- reportInvalidContent(type);
- }
- return type;
- }
- if (c == '/') { // always legal (if name matches etc)
- readEndElem();
- return END_ELEMENT;
- }
-
- if (c == ':' || isNameStartChar(c)) {
- /* Note: checking for EMPTY content type is done by the
- * validator, no need to check here
- */
- handleStartElem(c);
- return START_ELEMENT;
- }
- if (c == '[') {
- throwUnexpectedChar(c, " in content after '<' (malformed = mShortestTextSegment) {
- mTokenState = TOKEN_PARTIAL_SINGLE;
- } else {
- mTokenState = TOKEN_STARTED;
- }
- }
- return CHARACTERS;
- }
-
- /**
- * Method called when advacing stream past the end tag that closes
- * the root element of the open document.
- * Document can be either the singular one, in regular mode, or one of
- * possibly multiple, in multi-doc mode: this method is never called
- * in fragment mode. Method needs to update state properly and
- * parse following epilog event (if any).
- *
- * @return Event following end tag of the root elemennt, if any;
- * END_DOCUMENT otherwis.e
- */
- private int closeContentTree()
- throws XMLStreamException
- {
- mParseState = STATE_EPILOG;
- // this call will update the location too...
- if (nextFromProlog(false)) {
- mSecondaryToken = 0;
- }
- /* 10-Apr-2006, TSa: Let's actually try to update
- * SymbolTable here (after main xml tree); caller
- * may not continue parsing after this.
- */
- if (mSymbols.isDirty()) {
- mOwner.updateSymbolTable(mSymbols);
- }
- /* May be able to recycle, but not certain; and
- * definitely can not just clean contents (may
- * contain space(s) read)
- */
- mTextBuffer.recycle(false);
- return mCurrToken;
- }
-
- /**
- * Method that takes care of parsing of start elements; including
- * full parsing of namespace declarations and attributes, as well as
- * namespace resolution.
- */
- private final void handleStartElem(char c)
- throws XMLStreamException
- {
- mTokenState = TOKEN_FULL_COALESCED;
- boolean empty;
-
- if (mCfgNsEnabled) {
- String str = parseLocalName(c);
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_EOF_EXP_NAME);
- if (c == ':') { // Ok, got namespace and local name
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_EOF_EXP_NAME);
- mElementStack.push(str, parseLocalName(c));
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- } else {
- mElementStack.push(null, str);
- // c is fine as
- }
- /* Enough about element name itself; let's then parse attributes
- * and namespace declarations. Split into another method for clarity,
- * and so that maybe JIT has easier time to optimize it separately.
- */
- /* 04-Jul-2005, TSa: But hold up: we can easily check for a fairly
- * common case of no attributes showing up, and us getting the
- * closing '>' right away. Let's do that, since it can save
- * a call to a rather long method.
- */
- empty = (c == '>') ? false : handleNsAttrs(c);
- } else { // Namespace handling not enabled:
- mElementStack.push(null, parseFullName(c));
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- empty = (c == '>') ? false : handleNonNsAttrs(c);
- }
- if (!empty) {
- ++mCurrDepth; // needed to match nesting with entity expansion
- }
- mStEmptyElem = empty;
-
- /* 27-Feb-2009, TSa: [WSTX-191]: We used to validate virtual
- * end element here for empty elements, but it really should
- * occur later on when actually returning that end element.
- */
- int vld = mElementStack.resolveAndValidateElement();
- mVldContent = vld;
- mValidateText = (vld == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT);
- }
-
- /**
- * @return True if this is an empty element; false if not
- */
- private final boolean handleNsAttrs(char c)
- throws XMLStreamException
- {
- AttributeCollector ac = mAttrCollector;
-
- while (true) {
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- } else if (c != '/' && c != '>') {
- throwUnexpectedChar(c, " excepted space, or '>' or \"/>\"");
- }
-
- if (c == '/') {
- c = getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c != '>') {
- throwUnexpectedChar(c, " expected '>'");
- }
- return true;
- } else if (c == '>') {
- return false;
- } else if (c == '<') {
- throwParseError("Unexpected '<' character in element (missing closing '>'?)");
- }
-
- String prefix, localName;
- String str = parseLocalName(c);
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_EOF_EXP_NAME);
- if (c == ':') { // Ok, got namespace and local name
- prefix = str;
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_EOF_EXP_NAME);
- localName = parseLocalName(c);
- } else {
- --mInputPtr; // pushback
- prefix = null;
- localName = str;
- }
-
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- }
- if (c != '=') {
- throwUnexpectedChar(c, " expected '='");
- }
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- }
-
- // And then a quote:
- if (c != '"' && c != '\'') {
- throwUnexpectedChar(c, SUFFIX_IN_ELEMENT+" Expected a quote");
- }
-
- // And then the actual value
- int startLen = -1;
- TextBuilder tb;
-
- if (prefix == sPrefixXmlns) { // non-default namespace declaration
- tb = ac.getNsBuilder(localName);
- // returns null if it's a dupe:
- if (null == tb) {
- throwParseError("Duplicate declaration for namespace prefix '"+localName+"'.");
- }
- startLen = tb.getCharSize();
- } else if (localName == sPrefixXmlns && prefix == null) {
- tb = ac.getDefaultNsBuilder();
- // returns null if default ns was already declared
- if (null == tb) {
- throwParseError("Duplicate default namespace declaration.");
- }
- } else {
- tb = ac.getAttrBuilder(prefix, localName);
- }
- parseAttrValue(c, tb);
-
- /* 19-Jul-2004, TSa: Need to check that non-default namespace
- * URI is NOT empty, as per XML namespace specs, #2,
- * ("...In such declarations, the namespace name may not
- * be empty.")
- */
- /* (note: startLen is only set to first char position for
- * non-default NS declarations, see above...)
- */
- /* 04-Feb-2005, TSa: Namespaces 1.1 does allow this, though,
- * so for xml 1.1 documents we need to allow it
- */
- if (!mXml11) {
- if (startLen >= 0 && tb.getCharSize() == startLen) { // is empty!
- throwParseError(ErrorConsts.ERR_NS_EMPTY);
- }
- }
-
- // and then we need to iterate some more
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- }
- // never gets here
- }
-
- /**
- * @return True if this is an empty element; false if not
- */
- private final boolean handleNonNsAttrs(char c)
- throws XMLStreamException
- {
- AttributeCollector ac = mAttrCollector;
-
- while (true) {
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- } else if (c != '/' && c != '>') {
- throwUnexpectedChar(c, " excepted space, or '>' or \"/>\"");
- }
- if (c == '/') {
- c = getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c != '>') {
- throwUnexpectedChar(c, " expected '>'");
- }
- return true;
- } else if (c == '>') {
- return false;
- } else if (c == '<') {
- throwParseError("Unexpected '<' character in element (missing closing '>'?)");
- }
-
- String name = parseFullName(c);
- TextBuilder tb = ac.getAttrBuilder(null, name);
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- }
- if (c != '=') {
- throwUnexpectedChar(c, " expected '='");
- }
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_ELEMENT, c);
- }
-
- // And then a quote:
- if (c != '"' && c != '\'') {
- throwUnexpectedChar(c, SUFFIX_IN_ELEMENT+" Expected a quote");
- }
-
- // And then the actual value
- parseAttrValue(c, tb);
- // and then we need to iterate some more
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_ELEMENT);
- }
- // never gets here
- }
-
- /**
- * Method called to completely read a close tag, and update element
- * stack appropriately (including checking that tag matches etc).
- */
- protected final void readEndElem()
- throws XMLStreamException
- {
- mTokenState = TOKEN_FULL_COALESCED; // will be read completely
-
- if (mElementStack.isEmpty()) {
- // Let's just offline this for clarity
- reportExtraEndElem();
- return; // never gets here
- }
-
- char c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- // Quick check first; missing name?
- if (!isNameStartChar(c) && c != ':') {
- if (c <= CHAR_SPACE) { // space
- throwUnexpectedChar(c, "; missing element name?");
- }
- throwUnexpectedChar(c, "; expected an element name.");
- }
-
- /* Ok, now; good thing is we know exactly what to compare
- * against...
- */
- String expPrefix = mElementStack.getPrefix();
- String expLocalName = mElementStack.getLocalName();
-
- // Prefix to match?
- if (expPrefix != null && expPrefix.length() > 0) {
- int len = expPrefix.length();
- int i = 0;
-
- while (true){
- if (c != expPrefix.charAt(i)) {
- reportWrongEndPrefix(expPrefix, expLocalName, i);
- return; // never gets here
- }
- if (++i >= len) {
- break;
- }
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- }
- // And then we should get a colon
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- if (c != ':') {
- reportWrongEndPrefix(expPrefix, expLocalName, i);
- return;
- }
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- }
-
- // Ok, then, does the local name match?
- int len = expLocalName.length();
- int i = 0;
-
- while (true){
- if (c != expLocalName.charAt(i)) {
- // Not a match...
- reportWrongEndElem(expPrefix, expLocalName, i);
- return; // never gets here
- }
- if (++i >= len) {
- break;
- }
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- }
-
- // Let's see if end element still continues, however?
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CLOSE_ELEMENT);
- if (c <= CHAR_SPACE) {
- c = getNextInCurrAfterWS(SUFFIX_IN_CLOSE_ELEMENT, c);
- } else if (c == '>') {
- ;
- } else if (c == ':' || isNameChar(c)) {
- reportWrongEndElem(expPrefix, expLocalName, len);
- }
-
- // Ok, fine, match ok; now we just need the closing gt char.
- if (c != '>') {
- throwUnexpectedChar(c, SUFFIX_IN_CLOSE_ELEMENT+" Expected '>'.");
- }
-
- // Finally, let's let validator detect if things are ok
- int vld = mElementStack.validateEndElement();
- mVldContent = vld;
- mValidateText = (vld == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT);
-
- // Plus verify WFC that start and end tags came from same entity
- /* 13-Feb-2006, TSa: Are we about to close an element that
- * started within a parent element?
- * That's a GE/element nesting WFC violation...
- */
- if (mCurrDepth == mInputTopDepth) {
- handleGreedyEntityProblem(mInput);
- }
- --mCurrDepth;
- }
-
- private void reportExtraEndElem()
- throws XMLStreamException
- {
- String name = parseFNameForError();
- throwParseError("Unbalanced close tag ; no open start tag.");
- }
-
- private void reportWrongEndPrefix(String prefix, String localName, int done)
- throws XMLStreamException
- {
- --mInputPtr; // pushback
- String fullName = prefix + ":" + localName;
- String rest = parseFNameForError();
- String actName = fullName.substring(0, done) + rest;
- throwParseError("Unexpected close tag ; expected .");
- }
-
- private void reportWrongEndElem(String prefix, String localName, int done)
- throws XMLStreamException
- {
- --mInputPtr; // pushback
- String fullName;
- if (prefix != null && prefix.length() > 0) {
- fullName = prefix + ":" + localName;
- done += 1 + prefix.length();
- } else {
- fullName = localName;
- }
- String rest = parseFNameForError();
- String actName = fullName.substring(0, done) + rest;
- throwParseError("Unexpected close tag ; expected .");
- }
-
- /**
- *
- * Note: According to StAX 1.0, coalesced text events are always to be
- * returned as CHARACTERS, never as CDATA. And since at this point we
- * don't really know if there's anything to coalesce (but there may
- * be), let's convert CDATA if necessary.
- */
- private int nextFromTreeCommentOrCData()
- throws XMLStreamException
- {
- char c = getNextCharFromCurrent(SUFFIX_IN_DOC);
- if (c == '[') {
- checkCData();
- /* Good enough; it is a CDATA section... but let's just also
- * parse the easy ("free") stuff:
- */
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CDATA);
- readCDataPrimary(c); // sets token state appropriately...
- return CDATA;
- }
- if (c == '-' && getNextCharFromCurrent(SUFFIX_IN_DOC) == '-') {
- mTokenState = TOKEN_STARTED;
- return COMMENT;
- }
- throwParseError("Unrecognized XML directive; expected CDATA or comment ('
- * Note: this method is to accurately update the location information
- * to reflect where the next event will start (or, in case of EOF, where
- * EOF was encountered, ie. where event would start, if there was one).
- *
- * @return Next character after node has been skipped, or -1 if EOF
- * follows
- */
- private int skipToken()
- throws XMLStreamException
- {
- int result;
-
- main_switch:
- switch (mCurrToken) {
- case CDATA:
- {
- /* 30-Aug-2004, TSa: Need to be careful here: we may
- * actually have finished with CDATA, but are just
- * coalescing... if so, need to skip first part of
- * skipping
- */
- if (mTokenState <= TOKEN_PARTIAL_SINGLE) {
- // Skipping CDATA is easy; just need to spot closing ]]>
- skipCommentOrCData(SUFFIX_IN_CDATA, ']', false);
- }
- result = getNext();
- // ... except if coalescing, may need to skip more:
- if (mCfgCoalesceText) {
- result = skipCoalescedText(result);
- }
- }
- break;
-
- case COMMENT:
- skipCommentOrCData(SUFFIX_IN_COMMENT, '-', true);
- result = 0;
- break;
-
- case CHARACTERS:
- {
- result = skipTokenText(getNext());
- // ... except if coalescing, need to skip more:
- if (mCfgCoalesceText) {
- result = skipCoalescedText(result);
- }
- }
- break;
-
- case DTD:
- finishDTD(false);
- result = 0;
- break;
-
- case PROCESSING_INSTRUCTION:
- while (true) {
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_PROC_INSTR);
- if (c == '?') {
- do {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_PROC_INSTR);
- } while (c == '?');
- if (c == '>') {
- result = 0;
- break main_switch;
- }
- }
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
- }
- // never gets in here
-
- case SPACE:
-
- while (true) {
- // Fairly easy to skip through white space...
- while (mInputPtr < mInputEnd) {
- char c = mInputBuffer[mInputPtr++];
- if (c > CHAR_SPACE) { // non-EOF non-WS?
- result = c;
- break main_switch;
- }
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- }
- if (!loadMore()) {
- result = -1;
- break main_switch;
- }
- }
- // never gets in here
-
- case ENTITY_REFERENCE: // these should never end up in here...
- case ENTITY_DECLARATION:
- case NOTATION_DECLARATION:
- case START_DOCUMENT:
- case END_DOCUMENT:
- // As are start/end document
- throw new IllegalStateException("skipToken() called when current token is "+tokenTypeDesc(mCurrToken));
-
- case ATTRIBUTE:
- case NAMESPACE:
- // These two are never returned by this class
- case START_ELEMENT:
- case END_ELEMENT:
- /* Never called for elements tokens; start token handled
- * differently, end token always completely read in the first place
- */
-
- default:
- throw new IllegalStateException("Internal error: unexpected token "+tokenTypeDesc(mCurrToken));
-
- }
-
- /* Ok; now we have 3 possibilities; result is:
- *
- * + 0 -> could reliably read the prev event, now need the
- * following char/EOF
- * + -1 -> hit EOF; can return it
- * + something else -> this is the next char, return it.
- *
- * In first 2 cases, next event start offset is the current location;
- * in third case, it needs to be backtracked by one char
- */
- if (result < 1) {
- mTokenInputRow = mCurrInputRow;
- mTokenInputTotal = mCurrInputProcessed + mInputPtr;
- mTokenInputCol = mInputPtr - mCurrInputRowStart;
- return (result < 0) ? result : getNext();
- }
-
- // Ok, need to offset location, and return whatever we got:
- mTokenInputRow = mCurrInputRow;
- mTokenInputTotal = mCurrInputProcessed + mInputPtr - 1;
- mTokenInputCol = mInputPtr - mCurrInputRowStart - 1;
- return result;
- }
-
- private void skipCommentOrCData(String errorMsg, char endChar, boolean preventDoubles)
- throws XMLStreamException
- {
- /* Let's skip all chars except for double-ending chars in
- * question (hyphen for comments, right brack for cdata)
- */
- while (true) {
- char c;
- do {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextCharFromCurrent(errorMsg);
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
- } while (c != endChar);
-
- // Now, we may be getting end mark; first need second marker char:.
- c = getNextChar(errorMsg);
- if (c == endChar) { // Probably?
- // Now; we should be getting a '>', most likely.
- c = getNextChar(errorMsg);
- if (c == '>') {
- break;
- }
- if (preventDoubles) { // if not, it may be a problem...
- throwParseError("String '--' not allowed in comment (missing '>'?)");
- }
- // Otherwise, let's loop to see if there is end
- while (c == endChar) {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextCharFromCurrent(errorMsg);
- }
- if (c == '>') {
- break;
- }
- }
-
- // No match, did we get a linefeed?
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
-
- // Let's continue from beginning, then
- }
- }
-
- /**
- * Method called to skip past all following text and CDATA segments,
- * until encountering something else (including a general entity,
- * which may in turn expand to text).
- *
- * @return Character following all the skipped text and CDATA segments,
- * if any; or -1 to denote EOF
- */
- private int skipCoalescedText(int i)
- throws XMLStreamException
- {
- while (true) {
- // Ok, plain text or markup?
- if (i == '<') { // markup, maybe CDATA?
- // Need to distinguish "= 3
- && resolveSimpleEntity(true) != 0) {
- ;
- } else {
- i = fullyResolveEntity(true);
- /* Either way, it's just fine; we don't care about
- * returned single-char value.
- */
- }
- } else {
- /* Can only skip character entities; others need to
- * be returned separately.
- */
- if (resolveCharOnlyEntity(true) == 0) {
- /* Now points to the char after ampersand, and we need
- * to return the ampersand itself
- */
- return i;
- }
- }
- } else if (i < CHAR_SPACE) {
- if (i == '\r' || i == '\n') {
- skipCRLF((char) i);
- } else if (i < 0) { // EOF
- return i;
- } else if (i != '\t') {
- throwInvalidSpace(i);
- }
-
- }
-
- // Hmmh... let's do quick looping here:
- while (mInputPtr < mInputEnd) {
- char c = mInputBuffer[mInputPtr++];
- if (c < CHAR_FIRST_PURE_TEXT) { // need to check it
- i = c;
- continue main_loop;
- }
- }
-
- i = getNext();
- }
- // never gets here...
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, parsing
- ////////////////////////////////////////////////////
- */
-
- protected void ensureFinishToken()
- throws XMLStreamException
- {
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- }
-
- protected void safeEnsureFinishToken()
- {
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- }
-
- protected void safeFinishToken()
- {
- try {
- /* 24-Sep-2006, TSa: Let's try to reduce number of unchecked
- * (wrapped) exceptions we throw, and defer some. For now,
- * this is only for CHARACTERS (since it's always legal to
- * split CHARACTERS segment); could be expanded in future.
- */
- boolean deferErrors = (mCurrToken == CHARACTERS);
- finishToken(deferErrors);
- } catch (XMLStreamException strex) {
- throwLazyError(strex);
- }
- }
-
- /**
- * Method called to read in contents of the token completely, if not
- * yet read. Generally called when caller needs to access anything
- * other than basic token type (except for elements), text contents
- * or such.
- *
- * @param deferErrors Flag to enable storing an exception to a
- * variable, instead of immediately throwing it. If true, will
- * just store the exception; if false, will not store, just throw.
- */
- protected void finishToken(boolean deferErrors)
- throws XMLStreamException
- {
- switch (mCurrToken) {
- case CDATA:
- if (mCfgCoalesceText) {
- readCoalescedText(mCurrToken, deferErrors);
- } else {
- if (readCDataSecondary(mShortestTextSegment)) {
- mTokenState = TOKEN_FULL_SINGLE;
- } else {
- mTokenState = TOKEN_PARTIAL_SINGLE;
- }
- }
- return;
-
- case CHARACTERS:
- if (mCfgCoalesceText) {
- /* 21-Sep-2005, TSa: It is often possible to optimize
- * here: if we get '<' NOT followed by '!', it can not
- * be CDATA, and thus we are done.
- */
- if (mTokenState == TOKEN_FULL_SINGLE
- && (mInputPtr + 1) < mInputEnd
- && mInputBuffer[mInputPtr+1] != '!') {
- mTokenState = TOKEN_FULL_COALESCED;
- return;
- }
- readCoalescedText(mCurrToken, deferErrors);
- } else {
- if (readTextSecondary(mShortestTextSegment, deferErrors)) {
- mTokenState = TOKEN_FULL_SINGLE;
- } else {
- mTokenState = TOKEN_PARTIAL_SINGLE;
- }
- }
- return;
-
- case SPACE:
- {
- /* Only need to ensure there's no non-whitespace text
- * when parsing 'real' ignorable white space (in validating
- * mode, but that's implicit here)
- */
- boolean prolog = (mParseState != STATE_TREE);
- readSpaceSecondary(prolog);
- mTokenState = TOKEN_FULL_COALESCED;
- }
- return;
-
- case COMMENT:
- readComment();
- mTokenState = TOKEN_FULL_COALESCED;
- return;
-
- case DTD:
-
- /* 05-Jan-2006, TSa: Although we shouldn't have to use finally
- * here, it's probably better to do that for robustness
- * (specifically, in case of a parsing problem, we don't want
- * to remain in 'DTD partially read' case -- it's better
- * to get in panic mode and skip the rest)
- */
- try {
- finishDTD(true);
- } finally {
- mTokenState = TOKEN_FULL_COALESCED;
- }
- return;
-
- case PROCESSING_INSTRUCTION:
- readPI();
- mTokenState = TOKEN_FULL_COALESCED;
- return;
-
- case START_ELEMENT:
- case END_ELEMENT: // these 2 should never end up in here...
- case ENTITY_REFERENCE:
- case ENTITY_DECLARATION:
- case NOTATION_DECLARATION:
- case START_DOCUMENT:
- case END_DOCUMENT:
- throw new IllegalStateException("finishToken() called when current token is "+tokenTypeDesc(mCurrToken));
-
- case ATTRIBUTE:
- case NAMESPACE:
- // These two are never returned by this class
- default:
- }
-
- throw new IllegalStateException("Internal error: unexpected token "+tokenTypeDesc(mCurrToken));
- }
-
- private void readComment()
- throws XMLStreamException
- {
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
- int ptr = mInputPtr;
- int start = ptr;
-
- // Let's first see if we can just share input buffer:
- while (ptr < inputLen) {
- char c = inputBuf[ptr++];
- if (c > '-') {
- continue;
- }
-
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(ptr);
- } else if (c == '\r') {
- if (!mNormalizeLFs && ptr < inputLen) {
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- markLF(ptr);
- } else {
- --ptr; // pushback
- break;
- }
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '-') {
- // Ok; need to get '->', can not get '--'
-
- if ((ptr + 1) >= inputLen) {
- /* Can't check next 2, let's push '-' back, for rest of
- * code to take care of
- */
- --ptr;
- break;
- }
-
- if (inputBuf[ptr] != '-') {
- // Can't skip, might be LF/CR
- continue;
- }
- // Ok; either get '>' or error:
- c = inputBuf[ptr+1];
- if (c != '>') {
- throwParseError("String '--' not allowed in comment (missing '>'?)");
- }
- mTextBuffer.resetWithShared(inputBuf, start, ptr-start-1);
- mInputPtr = ptr + 2;
- return;
- }
- }
-
- mInputPtr = ptr;
- mTextBuffer.resetWithCopy(inputBuf, start, ptr-start);
- readComment2(mTextBuffer);
- }
-
- private void readComment2(TextBuffer tb)
- throws XMLStreamException
- {
- /* Output pointers; calls will also ensure that the buffer is
- * not shared, AND has room for at least one more char
- */
- char[] outBuf = mTextBuffer.getCurrentSegment();
- int outPtr = mTextBuffer.getCurrentSegmentSize();
- int outLen = outBuf.length;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_COMMENT);
-
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- if (skipCRLF(c)) { // got 2 char LF
- if (!mNormalizeLFs) {
- if (outPtr >= outLen) { // need more room?
- outBuf = mTextBuffer.finishCurrentSegment();
- outLen = outBuf.length;
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- }
- // And let's let default output the 2nd char
- c = '\n';
- } else if (mNormalizeLFs) { // just \r, but need to convert
- c = '\n'; // For Mac text
- }
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '-') { // Ok; need to get '->', can not get '--'
- c = getNextCharFromCurrent(SUFFIX_IN_COMMENT);
- if (c == '-') { // Ok, has to be end marker then:
- // Either get '>' or error:
- c = getNextCharFromCurrent(SUFFIX_IN_COMMENT);
- if (c != '>') {
- throwParseError(ErrorConsts.ERR_HYPHENS_IN_COMMENT);
- }
- break;
- }
-
- /* Not the end marker; let's just output the first hyphen,
- * push the second char back , and let main
- * code handle it.
- */
- c = '-';
- --mInputPtr;
- }
-
- // Need more room?
- if (outPtr >= outLen) {
- outBuf = mTextBuffer.finishCurrentSegment();
- outLen = outBuf.length;
- outPtr = 0;
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
- }
-
- // Ok, all done, then!
- mTextBuffer.setCurrentLength(outPtr);
- }
-
- /**
- * Method that reads the primary part of a PI, ie. target, and also
- * skips white space between target and data (if any data)
- *
- * @return Usually
- * When the method is called, it's expected that the first character
- * has been read as is in the current input buffer just before current
- * pointer
- *
- * @param c First character in the CDATA segment (possibly part of end
- * marker for empty segments
- *
- * @return True if the whole CDATA segment was completely read; this
- * happens only if lt-char is hit; false if it's possible that
- * it wasn't read (ie. end-of-buffer or entity encountered).
- */
- private final boolean readCDataPrimary(char c)
- throws XMLStreamException
- {
- mWsStatus = (c <= CHAR_SPACE) ? ALL_WS_UNKNOWN : ALL_WS_NO;
-
- int ptr = mInputPtr;
- int inputLen = mInputEnd;
- char[] inputBuf = mInputBuffer;
- int start = ptr-1;
-
- while (true) {
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(ptr);
- } else if (c == '\r') {
- if (ptr >= inputLen) { // can't peek?
- --ptr;
- break;
- }
- if (mNormalizeLFs) { // can we do in-place Mac replacement?
- if (inputBuf[ptr] == '\n') { // nope, 2 char lf
- --ptr;
- break;
- }
- inputBuf[ptr-1] = '\n'; // yup
- } else {
- // No LF normalization... can we just skip it?
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- }
- markLF(ptr);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == ']') {
- // Ok; need to get one or more ']'s, then '>'
- if ((ptr + 1) >= inputLen) { // not enough room? need to push it back
- --ptr;
- break;
- }
-
- // Needs to be followed by another ']'...
- if (inputBuf[ptr] == ']') {
- ++ptr;
- inner_loop:
- while (true) {
- if (ptr >= inputLen) {
- /* Need to push back last 2 right brackets; it may
- * be end marker divided by input buffer boundary
- */
- ptr -= 2;
- break inner_loop;
- }
- c = inputBuf[ptr++];
- if (c == '>') { // Ok, got it!
- mInputPtr = ptr;
- ptr -= (start+3);
- mTextBuffer.resetWithShared(inputBuf, start, ptr);
- mTokenState = TOKEN_FULL_SINGLE;
- return true;
- }
- if (c != ']') {
- // Need to re-check this char (may be linefeed)
- --ptr;
- break inner_loop;
- }
- // Fall through to next round
- }
- }
- }
-
- if (ptr >= inputLen) { // end-of-buffer?
- break;
- }
- c = inputBuf[ptr++];
- }
-
- mInputPtr = ptr;
-
- /* If we end up here, we either ran out of input, or hit something
- * which would leave 'holes' in buffer... fine, let's return then;
- * we can still update shared buffer copy: would be too early to
- * make a copy since caller may not even be interested in the
- * stuff.
- */
- int len = ptr - start;
- mTextBuffer.resetWithShared(inputBuf, start, len);
- if (mCfgCoalesceText ||
- (mTextBuffer.size() < mShortestTextSegment)) {
- mTokenState = TOKEN_STARTED;
- } else {
- mTokenState = TOKEN_PARTIAL_SINGLE;
- }
- return false;
- }
-
- /**
- * @return True if the whole CData section was completely read (we
- * hit the end marker); false if a shorter segment was returned.
- */
- protected boolean readCDataSecondary(int shortestSegment)
- throws XMLStreamException
- {
- // Input pointers
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
- int inputPtr = mInputPtr;
-
- /* Output pointers; calls will also ensure that the buffer is
- * not shared, AND has room for one more char
- */
- char[] outBuf = mTextBuffer.getCurrentSegment();
- int outPtr = mTextBuffer.getCurrentSegmentSize();
-
- while (true) {
- if (inputPtr >= inputLen) {
- loadMore(SUFFIX_IN_CDATA);
- inputBuf = mInputBuffer;
- inputPtr = mInputPtr;
- inputLen = mInputEnd;
- }
- char c = inputBuf[inputPtr++];
-
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(inputPtr);
- } else if (c == '\r') {
- mInputPtr = inputPtr;
- if (skipCRLF(c)) { // got 2 char LF
- if (!mNormalizeLFs) {
- // Special handling, to output 2 chars at a time:
- outBuf[outPtr++] = c;
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
- // And let's let default output the 2nd char, either way
- c = '\n';
- } else if (mNormalizeLFs) { // just \r, but need to convert
- c = '\n'; // For Mac text
- }
- /* Since skipCRLF() needs to peek(), buffer may have
- * changed, even if there was no CR+LF.
- */
- inputPtr = mInputPtr;
- inputBuf = mInputBuffer;
- inputLen = mInputEnd;
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == ']') {
- // Ok; need to get ']>'
- mInputPtr = inputPtr;
- if (checkCDataEnd(outBuf, outPtr)) {
- return true;
- }
- inputPtr = mInputPtr;
- inputBuf = mInputBuffer;
- inputLen = mInputEnd;
-
- outBuf = mTextBuffer.getCurrentSegment();
- outPtr = mTextBuffer.getCurrentSegmentSize();
- continue; // need to re-process last (non-bracket) char
- }
-
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- TextBuffer tb = mTextBuffer;
- // Perhaps we have now enough to return?
- if (!mCfgCoalesceText) {
- tb.setCurrentLength(outBuf.length);
- if (tb.size() >= shortestSegment) {
- mInputPtr = inputPtr;
- return false;
- }
- }
- // If not, need more buffer space:
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- }
- // never gets here
- }
-
- /**
- * Method that will check, given the starting ']', whether there is
- * ending ']]>' (including optional extra ']'s); if so, will updated
- * output buffer with extra ]s, if not, will make sure input and output
- * are positioned for further checking.
- *
- * @return True, if we hit the end marker; false if not.
- */
- private boolean checkCDataEnd(char[] outBuf, int outPtr)
- throws XMLStreamException
- {
- int bracketCount = 0;
- char c;
- do {
- ++bracketCount;
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CDATA);
- } while (c == ']');
-
- boolean match = (bracketCount >= 2 && c == '>');
- if (match) {
- bracketCount -= 2;
- }
- while (bracketCount > 0) {
- --bracketCount;
- outBuf[outPtr++] = ']';
- if (outPtr >= outBuf.length) {
- /* Can't really easily return, even if we have enough
- * stuff here, since we've more than one char...
- */
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
- mTextBuffer.setCurrentLength(outPtr);
- // Match? Can break, then:
- if (match) {
- return true;
- }
- // No match, need to push the last char back and admit defeat...
- --mInputPtr;
- return false;
- }
-
- /**
- * Method called to read in consequtive beginning parts of a text
- * segment, up to either end of the segment (lt char) or until
- * first 'hole' in text (buffer end, 2-char lf to convert, entity).
- *
- * When the method is called, it's expected that the first character
- * has been read as is in the current input buffer just before current
- * pointer
- *
- * @param c First character of the text segment
- *
- * @return True if the whole text segment was completely read; this
- * happens only if lt-char is hit; false if it's possible that
- * it wasn't read (ie. end-of-buffer or entity encountered).
- */
- private final boolean readTextPrimary(char c)
- throws XMLStreamException
- {
- int ptr = mInputPtr;
- int start = ptr-1;
-
- // First: can we heuristically canonicalize ws used for indentation?
- if (c <= CHAR_SPACE) {
- int len = mInputEnd;
- /* Even without indentation removal, it's good idea to
- * 'convert' \r or \r\n into \n (by replacing or skipping first
- * char): this may allow reusing the buffer.
- * But note that conversion MUST be enabled -- this is toggled
- * by code that includes internal entities, to prevent replacement
- * of CRs from int. general entities, as applicable.
- */
- do {
- // We'll need at least one char, no matter what:
- if (ptr < len && mNormalizeLFs) {
- if (c == '\r') {
- c = '\n';
- if (mInputBuffer[ptr] == c) {
- /* Ok, whatever happens, can 'skip' \r, to
- * point to following \n:
- */
- ++start;
- // But if that's buffer end, can't skip that
- if (++ptr >= len) {
- break;
- }
- } else {
- mInputBuffer[start] = c;
- }
- } else if (c != '\n') {
- break;
- }
- markLF(ptr);
- if (mCheckIndentation > 0) {
- ptr = readIndentation(c, ptr);
- if (ptr < 0) { // success!
- return true;
- }
- }
- // If we got this far, we skipped a lf, need to read next char
- c = mInputBuffer[ptr++];
- }
- } while (false);
-
- // can we figure out indentation?
- mWsStatus = ALL_WS_UNKNOWN;
- } else {
- mWsStatus = ALL_WS_NO;
- }
-
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
-
- // Let's first see if we can just share input buffer:
- while (true) {
- if (c < CHAR_FIRST_PURE_TEXT) {
- if (c == '<') {
- mInputPtr = --ptr;
- mTextBuffer.resetWithShared(inputBuf, start, ptr-start);
- return true;
- }
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(ptr);
- } else if (c == '\r') {
- if (ptr >= inputLen) { // can't peek?
- --ptr;
- break;
- }
- if (mNormalizeLFs) { // can we do in-place Mac replacement?
- if (inputBuf[ptr] == '\n') { // nope, 2 char lf
- --ptr;
- break;
- }
- /* This would otherwise be risky (may modify value
- * of a shared entity value), but since DTDs are
- * cached/accessed based on properties including
- * lf-normalization there's no harm in 'fixing' it
- * in place.
- */
- inputBuf[ptr-1] = '\n'; // yup
- } else {
- // No LF normalization... can we just skip it?
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- }
- markLF(ptr);
- } else if (c != '\t') {
- // Should consume invalid char, but not include in result
- mInputPtr = ptr;
- mTextBuffer.resetWithShared(inputBuf, start, ptr-start-1);
- /* Let's defer exception, provided we got at least
- * one valid character (if not, better throw
- * exception right away)
- */
- boolean deferErrors = (ptr - start) > 1;
- mPendingException = throwInvalidSpace(c, deferErrors);
- return true;
- }
- } else if (c == '&') {
- // Let's push it back and break
- --ptr;
- break;
- } else if (c == '>') {
- // Let's see if we got ']]>'?
- if ((ptr - start) >= 3) {
- if (inputBuf[ptr-3] == ']' && inputBuf[ptr-2] == ']') {
- /* Let's include ']]' in there, not '>' (since that
- * makes it non-wellformed): but need to consume
- * that char nonetheless
- */
- mInputPtr = ptr;
- mTextBuffer.resetWithShared(inputBuf, start, ptr-start-1);
- mPendingException = throwWfcException(ErrorConsts.ERR_BRACKET_IN_TEXT, true);
- return true; // and we are fully done
- }
- }
- }
- } // if (char in lower code range)
-
- if (ptr >= inputLen) { // end-of-buffer?
- break;
- }
- c = inputBuf[ptr++];
- }
- mInputPtr = ptr;
-
- /* If we end up here, we either ran out of input, or hit something
- * which would leave 'holes' in buffer... fine, let's return then;
- * we can still update shared buffer copy: would be too early to
- * make a copy since caller may not even be interested in the
- * stuff.
- */
- mTextBuffer.resetWithShared(inputBuf, start, ptr - start);
- return false;
- }
-
- /**
- *
- * @param deferErrors Flag to enable storing an exception to a
- * variable, instead of immediately throwing it. If true, will
- * just store the exception; if false, will not store, just throw.
- *
- * @return True if the text segment was completely read ('<' was hit,
- * or in non-entity-expanding mode, a non-char entity); false if
- * it may still continue
- */
- protected final boolean readTextSecondary(int shortestSegment, boolean deferErrors)
- throws XMLStreamException
- {
- /* Output pointers; calls will also ensure that the buffer is
- * not shared, AND has room for at least one more char
- */
- char[] outBuf = mTextBuffer.getCurrentSegment();
- int outPtr = mTextBuffer.getCurrentSegmentSize();
- int inputPtr = mInputPtr;
- char[] inputBuffer = mInputBuffer;
- int inputLen = mInputEnd;
-
- while (true) {
- if (inputPtr >= inputLen) {
- /* 07-Oct-2005, TSa: Let's not throw an exception for EOF from
- * here -- in fragment mode, it shouldn't be thrown, and in
- * other modes we might as well first return text, and only
- * then throw an exception: no need to do that yet.
- */
- mInputPtr = inputPtr;
- if (!loadMore()) {
- break;
- }
- inputPtr = mInputPtr;
- inputBuffer = mInputBuffer;
- inputLen = mInputEnd;
- }
- char c = inputBuffer[inputPtr++];
-
- // Most common case is we don't have special char, thus:
- if (c < CHAR_FIRST_PURE_TEXT) {
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(inputPtr);
- } else if (c == '\r') {
- mInputPtr = inputPtr;
- if (skipCRLF(c)) { // got 2 char LF
- if (!mNormalizeLFs) {
- // Special handling, to output 2 chars at a time:
- outBuf[outPtr++] = c;
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
- // And let's let default output the 2nd char
- c = '\n';
- } else if (mNormalizeLFs) { // just \r, but need to convert
- c = '\n'; // For Mac text
- }
- /* note: skipCRLF() may change ptr and len, but since
- * it does not close input source, it won't change
- * actual buffer object:
- */
- //inputBuffer = mInputBuffer;
- inputLen = mInputEnd;
- inputPtr = mInputPtr;
- } else if (c != '\t') {
- mTextBuffer.setCurrentLength(outPtr);
- mInputPtr = inputPtr;
- mPendingException = throwInvalidSpace(c, deferErrors);
- break;
- }
- } else if (c == '<') { // end is nigh!
- mInputPtr = inputPtr-1;
- break;
- } else if (c == '&') {
- mInputPtr = inputPtr;
- int ch;
- if (mCfgReplaceEntities) { // can we expand all entities?
- if ((inputLen - inputPtr) >= 3
- && (ch = resolveSimpleEntity(true)) != 0) {
- // Ok, it's fine then
- } else {
- ch = fullyResolveEntity(true);
- if (ch == 0) {
- // Input buffer changed, nothing to output quite yet:
- inputBuffer = mInputBuffer;
- inputLen = mInputEnd;
- inputPtr = mInputPtr;
- continue;
- }
- // otherwise char is now fine...
- }
- } else {
- /* Nope, can only expand char entities; others need
- * to be separately handled.
- */
- ch = resolveCharOnlyEntity(true);
- if (ch == 0) { // some other entity...
- /* can't expand; underlying pointer now points to
- * char after ampersand, need to rewind
- */
- --mInputPtr;
- break;
- }
- // .. otherwise we got char we needed
- }
- if (ch <= 0xFFFF) {
- c = (char) ch;
- } else {
- ch -= 0x10000;
- // need more room?
- if (outPtr >= outBuf.length) {
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = (char) ((ch >> 10) + 0xD800);
- c = (char) ((ch & 0x3FF) + 0xDC00);
- }
- inputPtr = mInputPtr;
- // not quite sure why this is needed... but it is:
- inputLen = mInputEnd;
- } else if (c == '>') {
- // Let's see if we got ']]>'?
- /* 21-Apr-2005, TSa: But we can NOT check the output buffer
- * as it contains _expanded_ stuff... only input side.
- * For now, 98% accuracy has to do, as we may not be able
- * to access previous buffer's contents. But at least we
- * won't produce false positives from entity expansion
- */
- if (inputPtr > 2) { // can we do it here?
- // Since mInputPtr has been advanced, -1 refers to '>'
- if (inputBuffer[inputPtr-3] == ']'
- && inputBuffer[inputPtr-2] == ']') {
- mInputPtr = inputPtr;
- /* We have already added ']]' into output buffer...
- * should be ok, since only with '>' does it become
- * non-wellformed.
- */
- mTextBuffer.setCurrentLength(outPtr);
- mPendingException = throwWfcException(ErrorConsts.ERR_BRACKET_IN_TEXT, deferErrors);
- break;
- }
- } else {
- /* 21-Apr-2005, TSa: No good way to verify it,
- * at this point. Should come back and think of how
- * to properly handle this (rare) possibility.
- */
- ;
- }
- }
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- TextBuffer tb = mTextBuffer;
- // Perhaps we have now enough to return?
- tb.setCurrentLength(outBuf.length);
- if (tb.size() >= shortestSegment) {
- mInputPtr = inputPtr;
- return false;
- }
- // If not, need more buffer space:
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- }
-
- mTextBuffer.setCurrentLength(outPtr);
- return true;
- }
-
- /**
- * Method called to try to parse and canonicalize white space that
- * has a good chance of being white space with somewhat regular
- * structure; specifically, something that looks like typical
- * indentation.
- *
- * Note: Caller guarantees that there will be at least 2 characters
- * available in the input buffer. And method has to ensure that if
- * it does not find a match, it will return pointer value such
- * that there is at least one valid character remaining.
- *
- * @return -1, if the content was determined to be canonicalizable
- * (indentation) white space; and thus fully parsed. Otherwise
- * pointer (value to set to mInputPtr) to the next character
- * to process (not processed by this method)
- */
- private final int readIndentation(char c, int ptr)
- throws XMLStreamException
- {
- /* We need to verify that:
- * (a) we can read enough contiguous data to do determination
- * (b) sequence is a linefeed, with either zero or more following
- * spaces, or zero or more tabs; and followed by non-directive
- * tag (start/end tag)
- * and if so, we can use a canonical shared representation of
- * this even.
- */
- final int inputLen = mInputEnd;
- final char[] inputBuf = mInputBuffer;
- int start = ptr-1;
- final char lf = c;
-
- // Note: caller guarantees at least one more char in the input buffer
- ws_loop:
- do { // dummy loop to allow for break (which indicates failure)
- c = inputBuf[ptr++];
- if (c == ' ' || c == '\t') { // indentation?
- // Need to limit to maximum
- int lastIndCharPos = (c == ' ') ? TextBuffer.MAX_INDENT_SPACES : TextBuffer.MAX_INDENT_TABS;
- lastIndCharPos += ptr;
- if (lastIndCharPos > inputLen) {
- lastIndCharPos = inputLen;
- }
-
- inner_loop:
- while (true) {
- if (ptr >= lastIndCharPos) { // overflow; let's backtrack
- --ptr;
- break ws_loop;
- }
- char d = inputBuf[ptr++];
- if (d != c) {
- if (d == '<') { // yup, got it!
- break inner_loop;
- }
- --ptr; // caller needs to reprocess it
- break ws_loop; // nope, blew it
- }
- }
- // This means we had success case; let's fall through
- } else if (c != '<') { // nope, can not be
- --ptr; // simpler if we just push it back; needs to be processed later on
- break ws_loop;
- }
-
- // Ok; we got '<'... just need any other char than '!'...
- if (ptr < inputLen && inputBuf[ptr] != '!') {
- // Voila!
- mInputPtr = --ptr; // need to push back that '<' too
- mTextBuffer.resetWithIndentation(ptr - start - 1, c);
- // One more thing: had a positive match, need to note it
- if (mCheckIndentation < INDENT_CHECK_MAX) {
- mCheckIndentation += INDENT_CHECK_START;
- }
- mWsStatus = ALL_WS_YES;
- return -1;
- }
- // Nope: need to push '<' back, then
- --ptr;
- } while (false);
-
- // Ok, nope... caller can/need to take care of it:
- /* Also, we may need to subtract indentation check count to possibly
- * disable this check if it doesn't seem to work.
- */
- --mCheckIndentation;
- /* Also; if lf we got was \r, need to convert it now (this
- * method only gets called in lf converting mode)
- * (and yes, it is safe to modify input buffer at this point;
- * see calling method for details)
- */
- if (lf == '\r') {
- inputBuf[start] = '\n';
- }
- return ptr;
- }
-
- /**
- * Reading whitespace should be very similar to reading normal text;
- * although couple of simplifications can be made. Further, since this
- * method is very unlikely to be of much performance concern, some
- * optimizations are left out, where it simplifies code.
- *
- * @param c First white space characters; known to contain white space
- * at this point
- * @param prologWS If true, is reading white space outside XML tree,
- * and as such can get EOF. If false, should not get EOF, nor be
- * followed by any other char than <
- *
- * @return True if the whole white space segment was read; false if
- * something prevented that (end of buffer, replaceable 2-char lf)
- */
- private final boolean readSpacePrimary(char c, boolean prologWS)
- throws XMLStreamException
- {
- int ptr = mInputPtr;
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
- int start = ptr-1;
-
- // Let's first see if we can just share input buffer:
- while (true) {
- /* 30-Aug-2006, TSa: Let's not check for validity errors yet,
- * even if we could detect problems at this point.
- * This because it's not always
- * an error (in dtd-aware, non-validating mode); but also since
- * that way we can first return all space we got, and only
- * indicate error when next token is to be accessed.
- */
- if (c > CHAR_SPACE) { // End of whitespace
- mInputPtr = --ptr;
- mTextBuffer.resetWithShared(mInputBuffer, start, ptr-start);
- return true;
- }
-
- if (c == '\n') {
- markLF(ptr);
- } else if (c == '\r') {
- if (ptr >= mInputEnd) { // can't peek?
- --ptr;
- break;
- }
- if (mNormalizeLFs) { // can we do in-place Mac replacement?
- if (inputBuf[ptr] == '\n') { // nope, 2 char lf
- --ptr;
- break;
- }
- inputBuf[ptr-1] = '\n'; // yup
- } else {
- // No LF normalization... can we just skip it?
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- }
- markLF(ptr);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- if (ptr >= inputLen) { // end-of-buffer?
- break;
- }
- c = inputBuf[ptr++];
- }
-
- mInputPtr = ptr;
-
- /* Ok, couldn't read it completely, let's just return whatever
- * we did get as shared data
- */
- mTextBuffer.resetWithShared(inputBuf, start, ptr - start);
- return false;
- }
-
- /**
- * This is very similar to readSecondaryText(); called when we need
- * to read in rest of (ignorable) white space segment.
- *
- * @param prologWS True if the ignorable white space is within prolog
- * (or epilog); false if it's within xml tree.
- */
- private void readSpaceSecondary(boolean prologWS)
- throws XMLStreamException
- {
- /* Let's not bother optimizing input. However, we can easily optimize
- * output, since it's easy to do, yet has more effect on performance
- * than localizing input variables.
- */
- char[] outBuf = mTextBuffer.getCurrentSegment();
- int outPtr = mTextBuffer.getCurrentSegmentSize();
-
- while (true) {
- if (mInputPtr >= mInputEnd) {
- /* 07-Oct-2005, TSa: Let's not throw an exception yet --
- * can return SPACE, and let exception be thrown
- * when trying to fetch next event.
- */
- if (!loadMore()) {
- break;
- }
- }
- char c = mInputBuffer[mInputPtr];
- if (c > CHAR_SPACE) { // end of WS?
- break;
- }
- ++mInputPtr;
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- if (skipCRLF(c)) {
- if (!mNormalizeLFs) {
- // Special handling, to output 2 chars at a time:
- outBuf[outPtr++] = c;
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
- c = '\n';
- } else if (mNormalizeLFs) {
- c = '\n'; // For Mac text
- }
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
-
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- }
- mTextBuffer.setCurrentLength(outPtr);
- }
-
- /**
- * Method called to read the contents of the current CHARACTERS
- * event, and write all contents using the specified Writer.
- *
- * @param w Writer to use for writing out textual content parsed
- *
- * @return Total number of characters written using the writer
- */
- private int readAndWriteText(Writer w)
- throws IOException, XMLStreamException
- {
- mTokenState = TOKEN_FULL_SINGLE; // we'll read it all
-
- /* We should be able to mostly just use the input buffer at this
- * point; exceptions being two-char linefeeds (when converting
- * to single ones) and entities (which likewise can expand or
- * shrink), both of which require flushing and/or single byte
- * output.
- */
- int start = mInputPtr;
- int count = 0;
-
- main_loop:
- while (true) {
- char c;
- // Reached the end of buffer? Need to flush, then
- if (mInputPtr >= mInputEnd) {
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- c = getNextChar(SUFFIX_IN_TEXT);
- start = mInputPtr-1; // needs to be prior to char we got
- } else {
- c = mInputBuffer[mInputPtr++];
- }
- // Most common case is we don't have a special char, thus:
- if (c < CHAR_FIRST_PURE_TEXT) {
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- char d;
- if (mInputPtr >= mInputEnd) {
- /* If we can't peek easily, let's flush past stuff
- * and load more... (have to flush, since new read
- * will overwrite inbut buffers)
- */
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- d = getNextChar(SUFFIX_IN_TEXT);
- start = mInputPtr; // to mark 'no past content'
- } else {
- d = mInputBuffer[mInputPtr++];
- }
- if (d == '\n') {
- if (mNormalizeLFs) {
- /* Let's flush content prior to 2-char LF, and
- * start the new segment on the second char...
- * this way, no mods are needed for the buffer,
- * AND it'll also work on split 2-char lf!
- */
- int len = mInputPtr - 2 - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- start = mInputPtr-1; // so '\n' is the first char
- } else {
- ; // otherwise it's good as is
- }
- } else { // not 2-char... need to replace?
- --mInputPtr;
- if (mNormalizeLFs) {
- mInputBuffer[mInputPtr-1] = '\n';
- }
- }
- markLF();
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '<') { // end is nigh!
- break main_loop;
- } else if (c == '&') {
- /* Have to flush all stuff, since entities pretty much
- * force it; input buffer won't be contiguous
- */
- int len = mInputPtr - 1 - start; // -1 to remove ampersand
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- int ch;
- if (mCfgReplaceEntities) { // can we expand all entities?
- if ((mInputEnd - mInputPtr) < 3
- || (ch = resolveSimpleEntity(true)) == 0) {
- ch = fullyResolveEntity(true);
- }
- } else {
- ch = resolveCharOnlyEntity(true);
- if (ch == 0) { // some other entity...
- /* can't expand, so, let's just bail out... but
- * let's also ensure no text is added twice, as
- * all prev text was just flushed, but resolve
- * may have moved input buffer around.
- */
- start = mInputPtr;
- break main_loop;
- }
- }
- if (ch != 0) {
- if (ch <= 0xFFFF) {
- c = (char) ch;
- } else {
- ch -= 0x10000;
- w.write((char) ((ch >> 10) + 0xD800));
- c = (char) ((ch & 0x3FF) + 0xDC00);
- }
- w.write(c);
- ++count;
- }
- start = mInputPtr;
- } else if (c == '>') { // did we get ']]>'?
- /* 21-Apr-2005, TSa: But we can NOT check the output buffer
- * (see comments in readTextSecondary() for details)
- */
- if (mInputPtr >= 2) { // can we do it here?
- if (mInputBuffer[mInputPtr-2] == ']'
- && mInputBuffer[mInputPtr-1] == ']') {
- // Anything to flush?
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- }
- throwParseError(ErrorConsts.ERR_BRACKET_IN_TEXT);
- }
- } else {
- ; // !!! TBI: how to check past boundary?
- }
- } else if (c == CHAR_NULL) {
- throwNullChar();
- }
- }
- } // while (true)
-
- /* Need to push back '<' or '&', whichever caused us to
- * get out...
- */
- --mInputPtr;
-
- // Anything left to flush?
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- return count;
- }
-
- /**
- * Method called to read the contents of the current (possibly partially
- * read) CDATA
- * event, and write all contents using the specified Writer.
- *
- * @param w Writer to use for writing out textual content parsed
- *
- * @return Total number of characters written using the writer for
- * the current CDATA event
- */
- private int readAndWriteCData(Writer w)
- throws IOException, XMLStreamException
- {
- mTokenState = TOKEN_FULL_SINGLE; // we'll read it all
-
- /* Ok; here we can basically have 2 modes; first the big loop to
- * gather all data up until a ']'; and then another loop to see
- * if ']' is part of ']]>', and after this if no end marker found,
- * go back to the first part.
- */
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(SUFFIX_IN_CDATA);
- int count = 0;
-
- main_loop:
- while (true) {
- int start = mInputPtr-1;
-
- quick_loop:
- while (true) {
- if (c > CHAR_CR_LF_OR_NULL) {
- if (c == ']') {
- break quick_loop;
- }
- } else {
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- char d;
- if (mInputPtr >= mInputEnd) {
- /* If we can't peek easily, let's flush past stuff
- * and load more... (have to flush, since new read
- * will overwrite inbut buffers)
- */
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- d = getNextChar(SUFFIX_IN_CDATA);
- start = mInputPtr; // to mark 'no past content'
- } else {
- d = mInputBuffer[mInputPtr++];
- }
- if (d == '\n') {
- if (mNormalizeLFs) {
- /* Let's flush content prior to 2-char LF, and
- * start the new segment on the second char...
- * this way, no mods are needed for the buffer,
- * AND it'll also work on split 2-char lf!
- */
- int len = mInputPtr - 2 - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- start = mInputPtr-1; // so '\n' is the first char
- } else {
- // otherwise it's good as is
- }
- } else { // not 2-char... need to replace?
- --mInputPtr;
- if (mNormalizeLFs) {
- mInputBuffer[mInputPtr-1] = '\n';
- }
- }
- markLF();
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
- }
- // Reached the end of buffer? Need to flush, then
- if (mInputPtr >= mInputEnd) {
- int len = mInputPtr - start;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- start = 0;
- c = getNextChar(SUFFIX_IN_CDATA);
- } else {
- c = mInputBuffer[mInputPtr++];
- }
- } // while (true)
-
- // Anything to flush once we hit ']'?
- {
- /* -1 since the last char in there (a '[') is NOT to be
- * output at this point
- */
- int len = mInputPtr - start - 1;
- if (len > 0) {
- w.write(mInputBuffer, start, len);
- count += len;
- }
- }
-
- /* Ok; we only get this far when we hit a ']'. We got one,
- * so let's see if we can find at least one more bracket,
- * immediately followed by '>'...
- */
- int bracketCount = 0;
- do {
- ++bracketCount;
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_CDATA);
- } while (c == ']');
-
- boolean match = (bracketCount >= 2 && c == '>');
- if (match) {
- bracketCount -= 2;
- }
- while (bracketCount > 0) {
- --bracketCount;
- w.write(']');
- ++count;
- }
- if (match) {
- break main_loop;
- }
- /* Otherwise we'll just loop; now c is properly set to be
- * the next char as well.
- */
- } // while (true)
-
- return count;
- }
-
- /**
- * @return Number of characters written to Writer during the call
- */
- private int readAndWriteCoalesced(Writer w, boolean wasCData)
- throws IOException, XMLStreamException
- {
- mTokenState = TOKEN_FULL_COALESCED;
- int count = 0;
-
- /* Ok, so what do we have next? CDATA, CHARACTERS, or something
- * else?
- */
- main_loop:
- while (true) {
- if (mInputPtr >= mInputEnd) {
- if (!loadMore()) {
- /* Shouldn't normally happen, but let's just let
- * caller deal with it...
- */
- break main_loop;
- }
- }
- // Let's peek, ie. not advance it yet
- char c = mInputBuffer[mInputPtr];
- if (c == '<') { // CDATA, maybe?
- // Need to distinguish " CHAR_SPACE) {
- return false;
- }
- while (true) {
- // Linefeed?
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- if (mInputPtr >= mInputEnd) {
- // Let's see if current source has more
- if (!loadMoreFromCurrent()) {
- return true;
- }
- }
- c = mInputBuffer[mInputPtr];
- if (c > CHAR_SPACE) { // not WS? Need to return
- return true;
- }
- ++mInputPtr;
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Abstract method implementations
- ///////////////////////////////////////////////////////////////////////
- */
-
- // @Override
- protected EntityDecl findEntity(String id, Object arg)
- throws XMLStreamException
- {
- EntityDecl ed = (EntityDecl) mConfig.findCustomInternalEntity(id);
- if (ed == null && mGeneralEntities != null) {
- ed = (EntityDecl) mGeneralEntities.get(id);
- }
- /* 05-Mar-2006, TSa: Externally declared entities are illegal
- * if we were declared as "standalone='yes'"...
- */
- if (mDocStandalone == DOC_STANDALONE_YES) {
- if (ed != null && ed.wasDeclaredExternally()) {
- throwParseError(ErrorConsts.ERR_WF_ENTITY_EXT_DECLARED, ed.getName(), null);
- }
- }
- return ed;
- }
-
- protected void handleUndeclaredEntity(String id)
- throws XMLStreamException
- {
- throwParseError(((mDocStandalone == DOC_STANDALONE_YES) ?
- ErrorConsts.ERR_WF_GE_UNDECLARED_SA :
- ErrorConsts.ERR_WF_GE_UNDECLARED),
- id, null);
- }
-
- protected void handleIncompleteEntityProblem(WstxInputSource closing)
- throws XMLStreamException
- {
- String top = mElementStack.isEmpty() ? "[ROOT]" : mElementStack.getTopElementDesc();
- throwParseError("Unexpected end of entity expansion for entity &{0}; was expecting a close tag for element <{1}>",
- closing.getEntityId(), top);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Internal methods, validation, error handling and reporting
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * This problem gets reported if an entity tries to expand to
- * a close tag matching start tag that did not came from the same
- * entity (but from parent).
- */
- protected void handleGreedyEntityProblem(WstxInputSource input)
- throws XMLStreamException
- {
- String top = mElementStack.isEmpty() ? "[ROOT]" : mElementStack.getTopElementDesc();
- throwParseError("Improper GE/element nesting: entity &"
- +input.getEntityId()+" contains closing tag for <"+top+">");
- }
-
- private void throwNotTextual(int type)
- {
- throw new IllegalStateException("Not a textual event ("
- +tokenTypeDesc(mCurrToken)+")");
- }
-
- private void throwNotTextXxx(int type)
- {
- throw new IllegalStateException("getTextXxx() methods can not be called on "
- +tokenTypeDesc(mCurrToken));
- }
-
- protected void throwNotTextualOrElem(int type)
- {
- throw new IllegalStateException(MessageFormat.format(ErrorConsts.ERR_STATE_NOT_ELEM_OR_TEXT, new Object[] { tokenTypeDesc(type) }));
- }
-
-
- /**
- * Method called when we get an EOF within content tree
- */
- protected void throwUnexpectedEOF()
- throws WstxException
- {
- throwUnexpectedEOF("; was expecting a close tag for element <"+mElementStack.getTopElementDesc()+">");
- }
-
- /**
- * Method called to report a problem with
- */
- protected XMLStreamException _constructUnexpectedInTyped(int nextToken)
- {
- if (nextToken == START_ELEMENT) {
- return _constructTypeException("Element content can not contain child START_ELEMENT when using Typed Access methods", null);
- }
- return _constructTypeException("Expected a text token, got "+tokenTypeDesc(nextToken), null);
- }
-
- protected TypedXMLStreamException _constructTypeException(String msg, String lexicalValue)
- {
- return new TypedXMLStreamException(lexicalValue, msg, getStartLocation());
- }
-
- /**
- * Stub method implemented by validating parsers, to report content
- * that's not valid for current element context. Defined at this
- * level since some such problems need to be caught at low-level;
- * however, details of error reports are not needed here.
- */
- protected void reportInvalidContent(int evtType)
- throws XMLStreamException
- {
- // should never happen; sub-class has to override:
- throwParseError("Internal error: sub-class should override method");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/CompactNsContext.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/CompactNsContext.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/CompactNsContext.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/CompactNsContext.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,250 +0,0 @@
-package com.ctc.wstx.sr;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.util.ArrayList;
-import java.util.Iterator;
-
-import javax.xml.XMLConstants;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-
-import org.codehaus.stax2.ri.EmptyIterator;
-import org.codehaus.stax2.ri.SingletonIterator;
-// This is unfortunate dependency, but...
-import org.codehaus.stax2.ri.evt.NamespaceEventImpl;
-
-import com.ctc.wstx.util.BaseNsContext;
-
-/**
- * Simple implementation of separate non-transient namespace context
- * object. Created for start-element event by transient namespace
- * instance updated by stream reader.
- *
- * Note about implementation: Location information is only needed (and
- * only needs to passed) if access is made via extended interface; one
- * that can return information about actual Namespace event objects.
- */
-public final class CompactNsContext
- extends BaseNsContext
-{
- final Location mLocation;
-
- /**
- * Array that contains 2 Strings for each declared default namespace
- * (including default namespace declarations); first is the prefix,
- * second URI.
- */
- final String[] mNamespaces;
-
- /**
- * Number of entries in {@link #mNamespaces} (which is twice the number
- * of bindings)
- */
- final int mNsLength;
-
- /**
- * Index of first namespace pair in mNamespaces that is declared
- * in scope of element for which this context was constructed. May be
- * equal to {@link #mNsLength} (which indicates there are no local
- * bindings).
- */
- final int mFirstLocalNs;
-
- /**
- * List only needed to support List accessor from start-element event;
- * created lazily if/as needed.
- */
- transient ArrayList mNsList;
-
- public CompactNsContext(Location loc, String defaultNsURI,
- String[] namespaces, int nsLen,
- int firstLocal)
- {
- mLocation = loc;
- mNamespaces = namespaces;
- mNsLength = nsLen;
- mFirstLocalNs = firstLocal;
- }
-
- /**
- * @param prefix Non-null, non-empty prefix (base-class verifies these
- * constraints) to find namespace URI for.
- */
- public String doGetNamespaceURI(String prefix)
- {
- /* Let's search from beginning towards end; this way we'll first
- * find the innermost (or, in case of same-level declaration, last)
- * declaration for prefix.
- */
- // (note: default namespace will be there too)
- String[] ns = mNamespaces;
- if (prefix.length() == 0) {
- for (int i = mNsLength-2; i >= 0; i -= 2) {
- if (ns[i] == null) {
- return ns[i+1];
- }
- }
- return null; // default ns not bound
- }
- for (int i = mNsLength-2; i >= 0; i -= 2) {
- if (prefix.equals(ns[i])) {
- return ns[i+1];
- }
- }
- return null;
- }
-
- public String doGetPrefix(String nsURI)
- {
- // Note: base class checks for 'known' problems and prefixes:
-
- String[] ns = mNamespaces;
- int len = mNsLength;
-
- main_loop:
- for (int i = len-1; i > 0; i -= 2) {
- if (nsURI.equals(ns[i])) {
- /* 29-Sep-2004, TSa: Actually, need to make sure that this
- * declaration is not masked by a later declaration.
- * This happens when same prefix is declared on a later
- * entry (ie. for child element)
- */
- String prefix = ns[i-1];
- for (int j = i+1; j < len; j += 2) {
- // Prefixes are interned, can do straight equality check
- if (ns[j] == prefix) {
- continue main_loop; // was masked!
- }
- }
- String uri = ns[i-1];
- /* 19-Mar-2006, TSa: Empty namespaces are represented by
- * null prefixes; but need to be represented as empty
- * strings (to distinguish from unbound URIs).
- */
- return (uri == null) ? "" : uri;
- }
- }
- return null;
- }
-
- public Iterator doGetPrefixes(String nsURI)
- {
- // Note: base class checks for 'known' problems and prefixes:
-
- String[] ns = mNamespaces;
- int len = mNsLength;
- String first = null;
- ArrayList all = null;
-
- main_loop:
- for (int i = len-1; i > 0; i -= 2) {
- String currNS = ns[i];
- if (currNS == nsURI || currNS.equals(nsURI)) {
- /* 29-Sep-2004, TSa: Need to ensure it's not masked by
- * a later ns declaration in a child element.
- */
- String prefix = ns[i-1];
- for (int j = i+1; j < len; j += 2) {
- // Prefixes are interned, can do straight equality check
- if (ns[j] == prefix) {
- continue main_loop; // was masked, need to ignore
- }
- }
- /* 19-Mar-2006, TSa: Empty namespaces are represented by
- * null prefixes; but need to be represented as empty
- * strings (to distinguish from unbound URIs).
- */
- if (prefix == null) {
- prefix = "";
- }
- if (first == null) {
- first = prefix;
- } else {
- if (all == null) {
- all = new ArrayList();
- all.add(first);
- }
- all.add(prefix);
- }
- }
- }
- if (all != null) {
- return all.iterator();
- }
- if (first != null) {
- return new SingletonIterator(first);
- }
- return EmptyIterator.getInstance();
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Extended API, needed by Wstx classes
- ///////////////////////////////////////////////////////
- */
-
- public Iterator getNamespaces()
- {
- if (mNsList == null) {
- int firstLocal = mFirstLocalNs;
- int len = mNsLength - firstLocal;
- if (len == 0) { // can this happen?
- return EmptyIterator.getInstance();
- }
- if (len == 2) { // only one NS
- return new SingletonIterator(NamespaceEventImpl.constructNamespace
- (mLocation,
- mNamespaces[firstLocal],
- mNamespaces[firstLocal+1]));
- }
- ArrayList l = new ArrayList(len >> 1);
- String[] ns = mNamespaces;
- for (len = mNsLength; firstLocal < len;
- firstLocal += 2) {
- l.add(NamespaceEventImpl.constructNamespace(mLocation, ns[firstLocal],
- ns[firstLocal+1]));
- }
- mNsList = l;
- }
- return mNsList.iterator();
- }
-
- /**
- * Method called by {@link com.ctc.wstx.evt.CompactStartElement}
- * to output all 'local' namespace declarations active in current
- * namespace scope, if any. Local means that declaration was done in
- * scope of current element, not in a parent element.
- */
- public void outputNamespaceDeclarations(Writer w) throws IOException
- {
- String[] ns = mNamespaces;
- for (int i = mFirstLocalNs, len = mNsLength; i < len; i += 2) {
- w.write(' ');
- w.write(XMLConstants.XMLNS_ATTRIBUTE);
- String prefix = ns[i];
- if (prefix != null && prefix.length() > 0) {
- w.write(':');
- w.write(prefix);
- }
- w.write("=\"");
- w.write(ns[i+1]);
- w.write('"');
- }
- }
-
- public void outputNamespaceDeclarations(XMLStreamWriter w) throws XMLStreamException
- {
- String[] ns = mNamespaces;
- for (int i = mFirstLocalNs, len = mNsLength; i < len; i += 2) {
- String nsURI = ns[i+1];
- String prefix = ns[i];
- if (prefix != null && prefix.length() > 0) {
- w.writeNamespace(prefix, nsURI);
- } else {
- w.writeDefaultNamespace(nsURI);
- }
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ElemAttrs.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ElemAttrs.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ElemAttrs.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ElemAttrs.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,228 +0,0 @@
-package com.ctc.wstx.sr;
-
-import javax.xml.namespace.QName;
-
-/**
- * Container class that is constructed with enough raw attribute information,
- * to be able to lazily construct full attribute objects, to be accessed
- * via Iterator, or fully-qualified name.
- *
- * Implementation note: code for using Map-like structure is unfortunately
- * cut'n pasted from {@link AttributeCollector}. Problem
- * with refactoring is that it's 90% the same code, but not 100%.
- *
- * Although instances of this class are constructed by stream readers,
- * it is actually used by element event objects.
- */
-public final class ElemAttrs
-{
- //private final static int OFFSET_LOCAL_NAME = 0;
- private final static int OFFSET_NS_URI = 1;
- //private final static int OFFSET_NS_PREFIX = 2;
- //private final static int OFFSET_VALUE = 3;
-
- /**
- * Array that contains 4 Strings for each attribute;
- * localName, URI, prefix, value. Can be used to lazily construct
- * structure(s) needed to return Iterator for accessing all
- * attributes.
- */
- private final String[] mRawAttrs;
-
- /**
- * Raw offset (in
- * Currently the limit is 4 attributes; 1, 2 or 3 attribute lists are
- * considered short, 4 or more 'long'.
- *
- * @param rawAttrs Array that contains 4 Strings for each attribute;
- * localName, URI, prefix, value. Can be used to lazily construct
- * structure(s) needed to return Iterator for accessing all
- * attributes.
- * @param defOffset Index of the first default attribute, if any;
- * number of all attributes if none
- */
- public ElemAttrs(String[] rawAttrs, int defOffset)
- {
- mRawAttrs = rawAttrs;
- mAttrMap = null;
- mAttrHashSize = 0;
- mAttrSpillEnd = 0;
- mDefaultOffset = (defOffset << 2);
- }
-
- /**
- * Method called to create "long" attribute list; list that has
- * a few entries, and efficient access by fully-qualified name should
- * not be done by linear search.
- *
- * @param rawAttrs Array that contains 4 Strings for each attribute;
- * localName, URI, prefix, value. Can be used to lazily construct
- * structure(s) needed to return Iterator for accessing all
- * attributes.
- */
- public ElemAttrs(String[] rawAttrs, int defOffset,
- int[] attrMap, int hashSize, int spillEnd)
- {
- mRawAttrs = rawAttrs;
- mDefaultOffset = (defOffset << 2);
- mAttrMap = attrMap;
- mAttrHashSize = hashSize;
- mAttrSpillEnd = spillEnd;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public API
- ////////////////////////////////////////////////////
- */
-
- public String[] getRawAttrs() {
- return mRawAttrs;
- }
-
- public int findIndex(QName name)
- {
- // Do we have a Map to do lookup against?
- if (mAttrMap != null) { // yup
- return findMapIndex(name.getNamespaceURI(), name.getLocalPart());
- }
-
- // Nope, linear search:
- String ln = name.getLocalPart();
- String uri = name.getNamespaceURI();
- boolean defaultNs = (uri == null || uri.length() == 0);
- String[] raw = mRawAttrs;
-
- for (int i = 0, len = raw.length; i < len; i += 4) {
- if (!ln.equals(raw[i])) {
- continue;
- }
- String thisUri = raw[i+OFFSET_NS_URI];
- if (defaultNs) {
- if (thisUri == null || thisUri.length() == 0) {
- return i;
- }
- } else { // non-default NS
- if (thisUri != null &&
- (thisUri == uri || thisUri.equals(uri))) {
- return i;
- }
- }
- }
- return -1;
- }
-
- public int getFirstDefaultOffset() {
- return mDefaultOffset;
- }
-
- public boolean isDefault(int ix) {
- return (ix >= mDefaultOffset);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////
- */
-
- /**
- *
- * Note: this method is very similar to
- * {@link com.ctc.wstx.sr.AttributeCollector#getAttrValue}; basically
- * most of it was cut'n pasted. Would be nice to refactor, but it's
- * bit hard to do that since data structures are not 100% identical
- * (mostly attribute storage, not Map structure itself).
- */
- private final int findMapIndex(String nsURI, String localName)
- {
- // Primary hit?
- int hash = localName.hashCode();
- if (nsURI == null) {
- nsURI = ""; // just to simplify comparisons -- array contains nulls
- } else if (nsURI.length() > 0) {
- hash ^= nsURI.hashCode();
- }
- int ix = mAttrMap[hash & (mAttrHashSize - 1)];
- if (ix == 0) { // nothing in here; no spills either
- return -1;
- }
- // Index is "one off" (since 0 indicates 'null), 4 Strings per attr
- ix = (ix - 1) << 2;
-
- // Is primary candidate match?
- String[] raw = mRawAttrs;
- String thisName = raw[ix];
- /* Equality first, since although equals() checks that too, it's
- * very likely to match (if interning Strings), and we can save
- * a method call.
- */
- if (thisName == localName || thisName.equals(localName)) {
- String thisURI = raw[ix+OFFSET_NS_URI];
- if (thisURI == nsURI) {
- return ix;
- }
- if (thisURI == null) {
- if (nsURI.length() == 0) {
- return ix;
- }
- } else if (thisURI.equals(nsURI)) {
- return ix;
- }
- }
-
- /* Nope, need to traverse spill list, which has 2 entries for
- * each spilled attribute id; first for hash value, second index.
- */
- for (int i = mAttrHashSize, len = mAttrSpillEnd; i < len; i += 2) {
- if (mAttrMap[i] != hash) {
- continue;
- }
- /* Note: spill indexes are not off-by-one, since there's no need
- * to mask 0
- */
- ix = mAttrMap[i+1] << 2; // ... but there are 4 Strings for each attr
- thisName = raw[ix];
- if (thisName == localName || thisName.equals(localName)) {
- String thisURI = raw[ix+1];
- if (thisURI == nsURI) {
- return ix;
- }
- if (thisURI == null) {
- if (nsURI.length() == 0) {
- return ix;
- }
- } else if (thisURI.equals(nsURI)) {
- return ix;
- }
- }
- }
-
- return -1;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ElemCallback.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ElemCallback.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ElemCallback.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ElemCallback.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,18 +0,0 @@
-package com.ctc.wstx.sr;
-
-import javax.xml.stream.Location;
-import javax.xml.namespace.QName;
-
-import com.ctc.wstx.util.BaseNsContext;
-
-/**
- * Abstract base class that defines set of simple callbacks to be
- * called by the stream reader, passing information about element
- * that the stream currently points to, if any.
- */
-public abstract class ElemCallback
-{
- public abstract Object withStartElement(Location loc, QName name,
- BaseNsContext nsCtxt, ElemAttrs attrs,
- boolean wasEmpty);
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/Element.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/Element.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/Element.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/Element.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,93 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-/**
- * Container for information collected regarding a single
- * (start) element instance.
- *
- * This class is not exposed outside of the package and is considered
- * part of internal implementation.
- *
- * @since 4.1
- */
-final class Element
-{
- // // // Element name
-
- protected String mLocalName;
-
- /**
- * Prefix this element has, if any; null if none
- */
- protected String mPrefix;
-
- /**
- * Namespace this element is in
- */
- protected String mNamespaceURI;
-
- /**
- * Default namespace for this element.
- */
- protected String mDefaultNsURI;
-
- // // // Namespace support
-
- /**
- * Offset within namespace array, maintained by
- * {@link InputElementStack} that owns this element.
- */
- protected int mNsOffset;
-
- // // // Back links to parent element(s)
-
- /**
- * Parent element, if any; null for root
- */
- protected Element mParent;
-
- /*
- /////////////////////////////////////////////////////////
- // Life-cycle
- /////////////////////////////////////////////////////////
- */
-
- public Element(Element parent, int nsOffset, String prefix, String ln)
- {
- mParent = parent;
- mNsOffset = nsOffset;
- mPrefix = prefix;
- mLocalName = ln;
- }
-
- public void reset(Element parent, int nsOffset, String prefix, String ln)
- {
- mParent = parent;
- mNsOffset = nsOffset;
- mPrefix = prefix;
- mLocalName = ln;
- }
-
- /**
- * Method called to temporarily "store" this Element for later reuse.
- */
- public void relink(Element next)
- {
- mParent = next;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/InputElementStack.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/InputElementStack.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/InputElementStack.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/InputElementStack.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1049 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.util.*;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.AttributeInfo;
-import org.codehaus.stax2.ri.EmptyIterator;
-import org.codehaus.stax2.ri.SingletonIterator;
-import org.codehaus.stax2.validation.ValidationContext;
-import org.codehaus.stax2.validation.XMLValidator;
-import org.codehaus.stax2.validation.XMLValidationProblem;
-import org.codehaus.stax2.validation.XMLValidationSchema;
-import org.codehaus.stax2.validation.ValidatorPair;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.api.WstxInputProperties;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.compat.QNameCreator;
-import com.ctc.wstx.dtd.DTDValidatorBase; // unfortunate dependency
-import com.ctc.wstx.util.*;
-
-/**
- * Shared base class that defines API stream reader uses to communicate
- * with the element stack implementation, independent of whether it's
- * operating in namespace-aware or non-namespace modes.
- * Element stack class is used for storing nesting information about open
- * elements, and for namespace-aware mode, also information about
- * namespaces active (including default namespace), during parsing of
- * XML input.
- *
- * This class also implements {@link NamespaceContext}, since it has all
- * the information necessary, so parser can just return element stack
- * instance as necesary.
- */
-public final class InputElementStack
- implements AttributeInfo, NamespaceContext, ValidationContext
-{
- final static int ID_ATTR_NONE = -1;
-
- /*
- //////////////////////////////////////////////////
- // Configuration
- //////////////////////////////////////////////////
- */
-
- protected final boolean mNsAware;
-
- protected final AttributeCollector mAttrCollector;
-
- protected final ReaderConfig mConfig;
-
- protected InputProblemReporter mReporter = null;
-
- /**
- * Object that will need to be consulted about namespace bindings,
- * since it has some knowledge about default namespace declarations
- * (has default attribute value expansion).
- */
- protected NsDefaultProvider mNsDefaultProvider;
-
- /*
- //////////////////////////////////////////////////
- // Element, namespace information
- //////////////////////////////////////////////////
- */
-
- protected int mDepth = 0;
-
- /**
- * Vector that contains all currently active namespaces; one String for
- * prefix, another for matching URI. Does also include default name
- * spaces (at most one per level).
- */
- protected final StringVector mNamespaces = new StringVector(64);
-
- /**
- * Currently open element, if any; null outside root element.
- */
- protected Element mCurrElement;
-
- protected boolean mMayHaveNsDefaults = false;
-
- /*
- //////////////////////////////////////////////////
- // Element validation (optional), attribute typing
- //////////////////////////////////////////////////
- */
-
- /**
- * Optional validator object that will get called if set,
- * and that can validate xml content. Note that it is possible
- * that this is set to a proxy object that calls multiple
- * validators in sequence.
- */
- protected XMLValidator mValidator = null;
-
- /**
- * Index of the attribute with type of ID, if known (most likely
- * due to Xml:id support); -1 if not available, or no ID attribute
- * for current element.
- */
- protected int mIdAttrIndex = ID_ATTR_NONE;
-
- /*
- //////////////////////////////////////////////////
- // Simple 1-slot QName cache; used for improving
- // efficiency of code that uses QNames extensively
- // (like StAX Event API implementation)
- //////////////////////////////////////////////////
- */
-
- protected String mLastLocalName = null;
- protected String mLastPrefix = null;
- protected String mLastNsURI = null;
-
- protected QName mLastName = null;
-
- /*
- /////////////////////////////////////////////////////
- // Other simple caching
- /////////////////////////////////////////////////////
- */
-
- // Non-transient NamespaceContext caching; mostly for event API
-
- /**
- * Last potentially shareable NamespaceContext created by
- * this stack. This reference is cleared each time bindings
- * change (either due to a start element with new bindings, or due
- * to the matching end element that closes scope of such binding(s)).
- */
- protected BaseNsContext mLastNsContext = null;
-
- // Chain of reusable Element instances
-
- protected Element mFreeElement = null;
-
- /*
- //////////////////////////////////////////////////
- // Life-cycle (create, update state)
- //////////////////////////////////////////////////
- */
-
- protected InputElementStack(ReaderConfig cfg, boolean nsAware)
- {
- mConfig = cfg;
- mNsAware = nsAware;
- mAttrCollector = new AttributeCollector(cfg, nsAware);
- }
-
- protected void connectReporter(InputProblemReporter rep)
- {
- mReporter = rep;
- }
-
- protected XMLValidator addValidator(XMLValidator vld)
- {
- if (mValidator == null) {
- mValidator = vld;
- } else {
- mValidator = new ValidatorPair(mValidator, vld);
- }
- return vld;
- }
-
- /**
- * Method called to connect the automatically handled DTD validator
- * (one detected from DOCTYPE, loaded and completely handled by
- * the stream reader without application calling validation methods).
- * Handled separately, since its behaviour is potentially different
- * from that of explicitly added validators.
- */
- protected void setAutomaticDTDValidator(XMLValidator validator, NsDefaultProvider nsDefs)
- {
- mNsDefaultProvider = nsDefs;
- addValidator(validator);
- }
-
- /*
- //////////////////////////////////////////////////
- // Start/stop validation
- //////////////////////////////////////////////////
- */
-
- public XMLValidator validateAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- /* Should we first check if we maybe already have a validator
- * for the schema?
- */
- return addValidator(schema.createValidator(this));
- }
-
-
-
- public XMLValidator stopValidatingAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- XMLValidator[] results = new XMLValidator[2];
- if (ValidatorPair.removeValidator(mValidator, schema, results)) { // found
- XMLValidator found = results[0];
- mValidator = results[1];
- found.validationCompleted(false);
- return found;
- }
- return null;
- }
-
- public XMLValidator stopValidatingAgainst(XMLValidator validator)
- throws XMLStreamException
- {
- XMLValidator[] results = new XMLValidator[2];
- if (ValidatorPair.removeValidator(mValidator, validator, results)) { // found
- XMLValidator found = results[0];
- mValidator = results[1];
- found.validationCompleted(false);
- return found;
- }
- return null;
- }
-
- /*
- //////////////////////////////////////////////////
- // Accessors:
- //////////////////////////////////////////////////
- */
-
- /**
- * This is a method called by the reader to ensure that we have at
- * least one 'real' validator. This is only needed to ensure that
- * validation problems that the reader can detect (illegal textual
- * content) can be reported as validity errors. Since the validator
- * API does not have a good way to cleanly deal with such a possibility,
- * the check is rather fragile, but should work for now: essentially
- * we need at least one validator object that either is not a sub-class
- * of
- * !!! TODO: remove need for this method (and method itself) with
- * Woodstox 4.0, by adding necessary support in Stax2 XMLValidator
- * interface.
- */
- protected boolean reallyValidating()
- {
- if (mValidator == null) { // no validators, no validation
- // (although, should never get called if no validators)
- return false;
- }
- if (!(mValidator instanceof DTDValidatorBase)) {
- // note: happens for validator pair, for one
- return true;
- }
- return ((DTDValidatorBase) mValidator).reallyValidating();
- }
-
- /**
- * Method called by {@link BasicStreamReader}, to retrieve the
- * attribute collector it needs for some direct access.
- */
- public final AttributeCollector getAttrCollector() {
- return mAttrCollector;
- }
-
- /**
- * Method called to construct a non-transient NamespaceContext instance;
- * generally needed when creating events to return from event-based
- * iterators.
- */
- public BaseNsContext createNonTransientNsContext(Location loc)
- {
- // Have an instance we can reuse? Great!
- if (mLastNsContext != null) {
- return mLastNsContext;
- }
-
- // No namespaces declared at this point? Easy, as well:
- int totalNsSize = mNamespaces.size();
- if (totalNsSize < 1) {
- return (mLastNsContext = EmptyNamespaceContext.getInstance());
- }
-
- // Otherwise, we need to create a new non-empty context:
- int localCount = getCurrentNsCount() << 1;
- BaseNsContext nsCtxt = new CompactNsContext
- (loc, getDefaultNsURI(),
- mNamespaces.asArray(), totalNsSize,
- totalNsSize - localCount);
- /* And it can be shared if there are no new ('local', ie. included
- * within this start element) bindings -- if there are, underlying
- * array might be shareable, but offsets wouldn't be)
- */
- if (localCount == 0) {
- mLastNsContext = nsCtxt;
- }
- return nsCtxt;
-}
-
- /**
- * Method called by the stream reader to add new (start) element
- * into the stack in namespace-aware mode; called when a start element
- * is encountered during parsing, but only in ns-aware mode.
- */
- public final void push(String prefix, String localName)
- {
- ++mDepth;
- String defaultNs = (mCurrElement == null) ?
- XmlConsts.DEFAULT_NAMESPACE_URI : mCurrElement.mDefaultNsURI;
-
- if (mFreeElement == null) {
- mCurrElement = new Element(mCurrElement, mNamespaces.size(), prefix, localName);
- } else {
- Element newElem = mFreeElement;
- mFreeElement = newElem.mParent;
- newElem.reset(mCurrElement, mNamespaces.size(), prefix, localName);
- mCurrElement = newElem;
- }
- mCurrElement.mDefaultNsURI = defaultNs;
- mAttrCollector.reset();
-
- /* 20-Feb-2006, TSa: Hmmh. Namespace default provider unfortunately
- * needs an advance warning...
- */
- if (mNsDefaultProvider != null) {
- mMayHaveNsDefaults = mNsDefaultProvider.mayHaveNsDefaults(prefix, localName);
- }
- }
-
- /**
- * Method called by the stream reader to remove the topmost (start)
- * element from the stack;
- * called when an end element is encountered during parsing.
- *
- * @return True if stack has more elements; false if not (that is,
- * root element closed)
- */
- public final boolean pop()
- throws XMLStreamException
- {
- if (mCurrElement == null) {
- throw new IllegalStateException("Popping from empty stack");
- }
- --mDepth;
-
- Element child = mCurrElement;
- Element parent = child.mParent;
- mCurrElement = parent;
-
- // Let's do simple recycling of Element instances...
- child.relink(mFreeElement);
- mFreeElement = child;
-
- // Need to purge namespaces?
- int nsCount = mNamespaces.size() - child.mNsOffset;
- if (nsCount > 0) { // 2 entries for each NS mapping:
- mLastNsContext = null; // let's invalidate ns ctxt too, if we had one
- mNamespaces.removeLast(nsCount);
- }
- return (parent != null);
- }
-
- /**
- * Method called to resolve element and attribute namespaces (in
- * namespace-aware mode), and do optional validation using pluggable
- * validator object.
- *
- * @return Text content validation state that should be effective
- * for the fully resolved element context
- */
- public int resolveAndValidateElement()
- throws XMLStreamException
- {
- if (mDepth == 0) { // just a simple sanity check
- throw new IllegalStateException("Calling validate() on empty stack.");
- }
- AttributeCollector ac = mAttrCollector;
-
- // Any namespace declarations?
- {
- int nsCount = ac.getNsCount();
- if (nsCount > 0) {
- /* let's first invalidate old (possibly) shared ns ctxt too,
- * if we had one; new one can be created at a later point
- */
- mLastNsContext = null;
-
- boolean internNsUris = mConfig.willInternNsURIs();
- for (int i = 0; i < nsCount; ++i) {
- Attribute ns = ac.resolveNamespaceDecl(i, internNsUris);
- String nsUri = ns.mNamespaceURI;
- // note: for namespaces, prefix is stored as local name
- String prefix = ns.mLocalName;
-
- /* 18-Jul-2004, TSa: Need to check that 'xml' and 'xmlns'
- * prefixes are not re-defined (and 'xmlns' not even
- * defined to its correct ns).
- */
- if (prefix == "xmlns") {
- // xmlns can never be declared, even to its correct URI
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XMLNS);
- } else if (prefix == "xml") {
- // whereas xml is ok, as long as it's same URI:
- if (!nsUri.equals(XMLConstants.XML_NS_URI)) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XML,
- nsUri, null);
- }
- /* 09-Feb-2006, TSa: Hmmh. Now, should this explicit
- * xml declaration be visible to the app? SAX API
- * seem to ignore it.
- */
- //mNamespaces.addStrings(prefix, nsUri);
- } else { // ok, valid prefix, so far
- /* 17-Mar-2006, TSa: Unbinding default NS needs to
- * result in null being added:
- */
- if (nsUri == null || nsUri.length() == 0) {
- nsUri = XmlConsts.DEFAULT_NAMESPACE_URI;
- }
- // The default ns binding needs special handling:
- if (prefix == null) {
- mCurrElement.mDefaultNsURI = nsUri;
- }
-
- /* But then let's ensure that URIs matching xml
- * and xmlns are not being bound to anything else
- */
- if (internNsUris) { // identity comparison is ok:
- if (nsUri == XMLConstants.XML_NS_URI) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XML_URI, prefix, null);
- } else if (nsUri == XMLConstants.XMLNS_ATTRIBUTE_NS_URI) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XMLNS_URI);
- }
- } else { // need to check equals()
- if (nsUri.equals(XMLConstants.XML_NS_URI)) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XML_URI, prefix, null);
- } else if (nsUri.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_REDECL_XMLNS_URI);
- }
- }
- /* and at any rate, binding needs to be added, to
- * be visible to the app (including def ns):
- */
- mNamespaces.addStrings(prefix, nsUri);
- }
- }
- }
- }
-
- /* 20-Feb-2006, TSa: Any attribute defaults for namespace declaration
- * pseudo-attributes?
- */
- if (mMayHaveNsDefaults) {
- mNsDefaultProvider.checkNsDefaults(this);
- }
-
- // Then, let's set element's namespace, if any:
- String prefix = mCurrElement.mPrefix;
- String ns;
-
- if (prefix == null) { // use default NS, if any
- ns = mCurrElement.mDefaultNsURI;
- } else if (prefix == "xml") {
- ns = XMLConstants.XML_NS_URI;
- } else {
- // Need to find namespace with the prefix:
- ns = mNamespaces.findLastFromMap(prefix);
- /* 07-Sep-2007, TSa: "no namespace" should now be indicated
- * by an empty string, however, due to historical reasons
- * let's be bit defensive and allow nulls for the same too
- */
- if (ns == null || ns.length() == 0) {
- mReporter.throwParseError(ErrorConsts.ERR_NS_UNDECLARED, prefix, null);
- }
- }
- mCurrElement.mNamespaceURI = ns;
-
- // And finally, resolve attributes' namespaces too:
- int xmlidIx = ac.resolveNamespaces(mReporter, mNamespaces);
- mIdAttrIndex = xmlidIx;
-
- XMLValidator vld = mValidator;
- /* If we have no validator(s), nothing more to do,
- * except perhaps little bit of Xml:id handling:
- */
- if (vld == null) { // no validator in use
- if (xmlidIx >= 0) { // need to normalize xml:id, still?
- ac.normalizeSpacesInValue(xmlidIx);
- }
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- }
-
- // Otherwise need to call relevant validation methods.
-
- /* First, a call to check if the element itself may be acceptable
- * within structure:
- */
- vld.validateElementStart
- (mCurrElement.mLocalName, mCurrElement.mNamespaceURI, mCurrElement.mPrefix);
-
- // Then attributes, if any:
- int attrLen = ac.getCount();
- if (attrLen > 0) {
- for (int i = 0; i < attrLen; ++i) {
- ac.validateAttribute(i, mValidator);
- }
- }
-
- /* And finally let's wrap things up to see what textual content
- * is allowed as child content, if any:
- */
- return mValidator.validateElementAndAttributes();
- }
-
- /**
- * Method called after parsing (but before returning) end element,
- * to allow for pluggable validators to verify correctness of
- * the content model for the closing element.
- *
- * @return Validation state that should be effective for the parent
- * element state
- */
- public int validateEndElement()
- throws XMLStreamException
- {
- if (mValidator == null) { // should never be null if we get here
- return XMLValidator.CONTENT_ALLOW_ANY_TEXT;
- }
- int result = mValidator.validateElementEnd
- (mCurrElement.mLocalName, mCurrElement.mNamespaceURI, mCurrElement.mPrefix);
- if (mDepth == 1) { // root closing
- mValidator.validationCompleted(true);
- }
- return result;
- }
-
- /*
- ///////////////////////////////////////////////////
- // AttributeInfo methods (StAX2)
- ///////////////////////////////////////////////////
- */
-
- public final int getAttributeCount()
- {
- return mAttrCollector.getCount();
- }
-
- public final int findAttributeIndex(String nsURI, String localName)
- {
- return mAttrCollector.findIndex(nsURI, localName);
- }
-
- /**
- * Default implementation just indicates it does not know of such
- * attributes; this because that requires DTD information that only
- * some implementations have.
- */
- public final int getIdAttributeIndex()
- {
- if (mIdAttrIndex >= 0) {
- return mIdAttrIndex;
- }
- return (mValidator == null) ? -1 : mValidator.getIdAttrIndex();
- }
-
- /**
- * Default implementation just indicates it does not know of such
- * attributes; this because that requires DTD information that only
- * some implementations have.
- */
- public final int getNotationAttributeIndex()
- {
- return (mValidator == null) ? -1 :
- mValidator.getNotationAttrIndex();
- }
-
- /*
- ///////////////////////////////////////////////////
- // Implementation of NamespaceContext:
- ///////////////////////////////////////////////////
- */
-
- public final String getNamespaceURI(String prefix)
- {
- if (prefix == null) {
- throw new IllegalArgumentException(ErrorConsts.ERR_NULL_ARG);
- }
- if (prefix.length() == 0) {
- if (mDepth == 0) { // unexpected... but let's not err at this point
- /* 07-Sep-2007, TSa: Default/"no namespace" does map to
- * "URI" of empty String.
- */
- return XmlConsts.DEFAULT_NAMESPACE_URI;
- }
- return mCurrElement.mDefaultNsURI;
- }
- if (prefix.equals(XMLConstants.XML_NS_PREFIX)) {
- return XMLConstants.XML_NS_URI;
- }
- if (prefix.equals(XMLConstants.XMLNS_ATTRIBUTE)) {
- return XMLConstants.XMLNS_ATTRIBUTE_NS_URI;
- }
- /* Ok, need to find the match, if any; starting from end of the
- * list of active namespaces. Note that we can not count on prefix
- * being interned/canonicalized.
- */
- return mNamespaces.findLastNonInterned(prefix);
- }
-
- public final String getPrefix(String nsURI)
- {
- if (nsURI == null || nsURI.length() == 0) {
- throw new IllegalArgumentException("Illegal to pass null/empty prefix as argument.");
- }
- if (nsURI.equals(XMLConstants.XML_NS_URI)) {
- return XMLConstants.XML_NS_PREFIX;
- }
- if (nsURI.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- return XMLConstants.XMLNS_ATTRIBUTE;
- }
-
- /* Ok, need to find the match, if any; starting from end of the
- * list of active namespaces. Note that we can not count on prefix
- * being interned/canonicalized.
- */
- String prefix = null;
-
- // 29-Sep-2004, TSa: Need to check for namespace masking, too...
- String[] strs = mNamespaces.getInternalArray();
- int len = mNamespaces.size();
-
- main_loop:
- for (int index = len-1; index > 0; index -= 2) {
- if (nsURI.equals(strs[index])) {
- // Ok, is prefix masked?
- prefix = strs[index-1];
- for (int j = index+1; j < len; j += 2) {
- if (strs[j] == prefix) { // masked!
- prefix = null;
- continue main_loop;
- }
- }
- // nah, it's good
- // 17-Mar-2006, TSa: ... but default NS has prefix null...
- if (prefix == null) {
- prefix = "";
- }
- break main_loop;
- }
- }
-
- return prefix;
- }
-
- public final Iterator getPrefixes(String nsURI)
- {
- if (nsURI == null || nsURI.length() == 0) {
- throw new IllegalArgumentException("Illegal to pass null/empty prefix as argument.");
- }
- if (nsURI.equals(XMLConstants.XML_NS_URI)) {
- return new SingletonIterator(XMLConstants.XML_NS_PREFIX);
- }
- if (nsURI.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- return new SingletonIterator(XMLConstants.XMLNS_ATTRIBUTE);
- }
-
- // 29-Sep-2004, TSa: Need to check for namespace masking, too...
- String[] strs = mNamespaces.getInternalArray();
- int len = mNamespaces.size();
- ArrayList l = null;
-
- main_loop:
- for (int index = len-1; index > 0; index -= 2) {
- if (nsURI.equals(strs[index])) {
- // Ok, is prefix masked?
- String prefix = strs[index-1];
- for (int j = index+1; j < len; j += 2) {
- if (strs[j] == prefix) { // masked!
- continue main_loop;
- }
- }
- // nah, it's good!
- if (l == null) {
- l = new ArrayList();
- }
- l.add(prefix);
- }
- }
-
- return (l == null) ? EmptyIterator.getInstance() : l.iterator();
- }
-
- /*
- ///////////////////////////////////////////////////
- // ValidationContext
- ///////////////////////////////////////////////////
- */
-
- public final String getXmlVersion()
- {
- return mConfig.isXml11() ? XmlConsts.XML_V_11_STR : XmlConsts.XML_V_10_STR;
- }
-
- // Part of Stax2, see above:
- //public int getAttributeCount();
-
- public String getAttributeLocalName(int index) {
- return getAttrCollector().getLocalName(index);
- }
-
- public String getAttributeNamespace(int index) {
- return getAttrCollector().getURI(index);
- }
-
- public String getAttributePrefix(int index) {
- return getAttrCollector().getPrefix(index);
- }
-
- public String getAttributeValue(int index) {
- return getAttrCollector().getValue(index);
- }
-
- public String getAttributeValue(String nsURI, String localName)
- {
- int ix = findAttributeIndex(nsURI, localName);
- return (ix < 0) ? null : getAttributeValue(ix);
- }
-
- // Part of Stax2, see above:
- //public int findAttributeIndex(String nsURI, String localName);
-
- public boolean isNotationDeclared(String name)
- {
- // !!! TBI
- return false;
- }
-
- public boolean isUnparsedEntityDeclared(String name)
- {
- // !!! TBI
- return false;
- }
-
- public String getBaseUri()
- {
- // !!! TBI
- return null;
- }
-
- public final QName getCurrentElementName()
- {
- if (mDepth == 0) {
- return null;
- }
- String prefix = mCurrElement.mPrefix;
- /* 17-Mar-2006, TSa: We only map prefix to empty String because
- * some QName impls barf on nulls. Otherwise we will always
- * use null to indicate missing prefixes.
- */
- if (prefix == null) {
- prefix = "";
- }
- /* 03-Dec-2004, TSa: Maybe we can just reuse the last QName
- * object created, if we have same data? (happens if
- * state hasn't changed, or we got end element for a leaf
- * element, or repeating leaf elements)
- */
- String nsURI = mCurrElement.mNamespaceURI;
- String ln = mCurrElement.mLocalName;
-
- /* Since we generally intern most Strings, can do identity
- * comparisons here:
- */
- if (ln != mLastLocalName) {
- mLastLocalName = ln;
- mLastPrefix = prefix;
- mLastNsURI = nsURI;
- } else if (prefix != mLastPrefix) {
- mLastPrefix = prefix;
- mLastNsURI = nsURI;
- } else if (nsURI != mLastNsURI) {
- mLastNsURI = nsURI;
- } else {
- return mLastName;
- }
- QName n = QNameCreator.create(nsURI, ln, prefix);
- mLastName = n;
- return n;
- }
-
- // This was defined above for NamespaceContext
- //public String getNamespaceURI(String prefix);
-
- public Location getValidationLocation()
- {
- return mReporter.getLocation();
- }
-
- public void reportProblem(XMLValidationProblem problem)
- throws XMLStreamException
- {
- mReporter.reportValidationProblem(problem);
- }
-
- /**
- * Method called by actual validator instances when attributes with
- * default values have no explicit values for the element; if so,
- * default value needs to be added as if it was parsed from the
- * element.
- */
- public int addDefaultAttribute(String localName, String uri, String prefix,
- String value)
- {
- return mAttrCollector.addDefaultAttribute(localName, uri, prefix, value);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Support for NsDefaultProvider
- ///////////////////////////////////////////////////
- */
-
- public boolean isPrefixLocallyDeclared(String internedPrefix)
- {
- if (internedPrefix != null && internedPrefix.length() == 0) { // default ns
- internedPrefix = null;
- }
-
- int offset = mCurrElement.mNsOffset;
- for (int len = mNamespaces.size(); offset < len; offset += 2) {
- // both interned, can use identity comparison
- String thisPrefix = mNamespaces.getString(offset);
- if (thisPrefix == internedPrefix) {
- return true;
- }
- }
- return false;
- }
-
- /**
- * Callback method called by the namespace default provider. At
- * this point we can trust it to only call this method with somewhat
- * valid arguments (no dups etc).
- */
- public void addNsBinding(String prefix, String uri)
- {
- // Unbind? (xml 1.1...)
- if ((uri == null) || (uri.length() == 0)) {
- uri = null;
- }
-
- // Default ns declaration?
- if ((prefix == null) || (prefix.length() == 0)) {
- prefix = null;
- mCurrElement.mDefaultNsURI = uri;
- }
- mNamespaces.addStrings(prefix, uri);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Support for validation:
- ///////////////////////////////////////////////////
- */
-
- public final void validateText(TextBuffer tb, boolean lastTextSegment)
- throws XMLStreamException
- {
- tb.validateText(mValidator, lastTextSegment);
- }
-
- public final void validateText(String contents, boolean lastTextSegment)
- throws XMLStreamException
- {
- mValidator.validateText(contents, lastTextSegment);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Accessors:
- ///////////////////////////////////////////////////
- */
-
- // // // Generic stack information:
-
- public final boolean isNamespaceAware() {
- return mNsAware;
- }
-
- // // // Generic stack information:
-
- public final boolean isEmpty() {
- return mDepth == 0;
- }
-
- /**
- * @return Number of open elements in the stack; 0 when parser is in
- * prolog/epilog, 1 inside root element and so on.
- */
- public final int getDepth() { return mDepth; }
-
- // // // Information about element at top of stack:
-
- public final String getDefaultNsURI() {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- return mCurrElement.mDefaultNsURI;
- }
-
- public final String getNsURI() {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- return mCurrElement.mNamespaceURI;
- }
-
- public final String getPrefix() {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- return mCurrElement.mPrefix;
- }
-
- public final String getLocalName() {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- return mCurrElement.mLocalName;
- }
-
- public final boolean matches(String prefix, String localName)
- {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- String thisPrefix = mCurrElement.mPrefix;
- if (prefix == null || prefix.length() == 0) { // no name space
- if (thisPrefix != null && thisPrefix.length() > 0) {
- return false;
- }
- } else {
- if (thisPrefix != prefix && !thisPrefix.equals(prefix)) {
- return false;
- }
- }
-
- String thisName = mCurrElement.mLocalName;
- return (thisName == localName) || thisName.equals(localName);
- }
-
- public final String getTopElementDesc()
- {
- if (mDepth == 0) {
- throw new IllegalStateException("Illegal access, empty stack.");
- }
- String name = mCurrElement.mLocalName;
- String prefix = mCurrElement.mPrefix;
- if (prefix == null) { // no name space
- return name;
- }
- return prefix + ":" + name;
- }
-
- // // // Namespace information:
-
- /**
- * @return Number of active prefix/namespace mappings for current scope,
- * including mappings from enclosing elements.
- */
- public final int getTotalNsCount() {
- return mNamespaces.size() >> 1;
- }
-
- /**
- * @return Number of active prefix/namespace mappings for current scope,
- * NOT including mappings from enclosing elements.
- */
- public final int getCurrentNsCount()
- {
- // Need not check for empty stack; should return 0 properly
- return (mNamespaces.size() - mCurrElement.mNsOffset) >> 1;
- }
-
- public final String getLocalNsPrefix(int index)
- {
- int offset = mCurrElement.mNsOffset;
- int localCount = (mNamespaces.size() - offset);
- index <<= 1; // 2 entries, prefix/URI for each NS
- if (index < 0 || index >= localCount) {
- throwIllegalIndex(index >> 1, localCount >> 1);
- }
- return mNamespaces.getString(offset + index);
- }
-
- public final String getLocalNsURI(int index)
- {
- int offset = mCurrElement.mNsOffset;
- int localCount = (mNamespaces.size() - offset);
- index <<= 1; // 2 entries, prefix/URI for each NS
- if (index < 0 || index >= localCount) {
- throwIllegalIndex(index >> 1, localCount >> 1);
- }
- return mNamespaces.getString(offset + index + 1);
- }
-
- private void throwIllegalIndex(int index, int localCount)
- {
- throw new IllegalArgumentException("Illegal namespace index "
- +(index >> 1)
- +"; current scope only has "
- +(localCount >> 1)
- +" namespace declarations.");
- }
-
- // // // DTD-derived attribute information:
-
- /**
- * @return Schema (DTD, RNG, W3C Schema) based type of the attribute
- * in specified index
- */
- public final String getAttributeType(int index)
- {
- if (index == mIdAttrIndex && index >= 0) { // second check to ensure -1 is not passed
- return "ID";
- }
- return (mValidator == null) ? WstxInputProperties.UNKNOWN_ATTR_TYPE :
- mValidator.getAttributeType(index);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/InputProblemReporter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/InputProblemReporter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/InputProblemReporter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/InputProblemReporter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,53 +0,0 @@
-package com.ctc.wstx.sr;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.XMLValidationProblem;
-
-/**
- * Interface implemented by input reader, and used by other components to
- * report problem that are related to current input position.
- */
-public interface InputProblemReporter
-{
- /*
- ////////////////////////////////////////////////////
- // Methods for reporting "hard" errors:
- ////////////////////////////////////////////////////
- */
-
- public void throwParseError(String msg) throws XMLStreamException;
- public void throwParseError(String msg, Object arg, Object arg2)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////////////////////
- // Reporting validation problems
- ///////////////////////////////////////////////////////
- */
-
- public void reportValidationProblem(XMLValidationProblem prob)
- throws XMLStreamException;
- public void reportValidationProblem(String msg)
- throws XMLStreamException;
- public void reportValidationProblem(String msg, Object arg, Object arg2)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////////////////////
- // Methods for reporting other "soft" (recoverable) problems
- ///////////////////////////////////////////////////////
- */
-
- public void reportProblem(Location loc, String probType, String format, Object arg, Object arg2)
- throws XMLStreamException;
-
- /*
- ////////////////////////////////////////////////////
- // Supporting methods needed by reporting
- ////////////////////////////////////////////////////
- */
-
- public Location getLocation();
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/NsDefaultProvider.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/NsDefaultProvider.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/NsDefaultProvider.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/NsDefaultProvider.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,44 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import javax.xml.stream.XMLStreamException;
-
-/**
- * Interface only used by Woodstox core. The main reason for the interface
- * is to reduce coupling with the input element stack and dtd validator
- * instances: while dtd validator needs to be able to inject namespace
- * declarations based on attribute default values, it should not have to
- * know too much about element stack implementation, and vice versa.
- * As a result, this interface defines API input element stack calls
- * on the dtd validator instance. Validator instance then refers to the
- * input element stack base class to do callbacks if and as necessary.
- */
-public interface NsDefaultProvider
-{
- public boolean mayHaveNsDefaults(String elemPrefix, String elemLN);
-
- /**
- * Method called by the input element stack to indicate that
- * it has just added local namespace declarations from the
- * current element, and is about to start resolving element
- * and attribute namespace bindings. This provider instance is
- * to add namespace declarations from attribute defaults, if
- * any, using callbacks to the input element stack.
- */
- public void checkNsDefaults(InputElementStack nsStack)
- throws XMLStreamException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,4 +0,0 @@
-
- * Note that readers in general should only access the configuration info
- * when they are created (from constructor).
- */
-public interface ReaderCreator
-{
- /*
- ///////////////////////////////////////////////////////
- // Methods for accessing configuration info
- ///////////////////////////////////////////////////////
- */
-
- public DTDSubset findCachedDTD(DTDId id);
-
- /*
- ///////////////////////////////////////////////////////
- // Methods for updating information factory has
- ///////////////////////////////////////////////////////
- */
-
- public void updateSymbolTable(SymbolTable t);
-
- public void addCachedDTD(DTDId id, DTDSubset extSubset);
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/StreamReaderImpl.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/StreamReaderImpl.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/StreamReaderImpl.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/StreamReaderImpl.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,28 +0,0 @@
-package com.ctc.wstx.sr;
-
-import javax.xml.stream.*;
-
-import org.codehaus.stax2.XMLStreamReader2;
-
-import com.ctc.wstx.ent.EntityDecl;
-
-/**
- * Interface that defines "internal Woodstox API". It is used to decouple
- * parts of the Woodstox that need to know something more about woodstox
- * stream reader implementation, but not about implementation details.
- * Specifically, there are some simple dependencies from the stream
- * writer; they should only need to refer to this interface.
- */
-public interface StreamReaderImpl
- extends XMLStreamReader2
-{
- public EntityDecl getCurrentEntityDecl();
-
- public Object withStartElement(ElemCallback cb, Location loc);
-
- public boolean isNamespaceAware();
-
- public AttributeCollector getAttributeCollector();
-
- public InputElementStack getInputElementStack();
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/StreamScanner.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/StreamScanner.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/StreamScanner.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/StreamScanner.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,2444 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.io.FileNotFoundException;
-import java.io.IOException;
-import java.net.URL;
-import java.text.MessageFormat;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.Map;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLInputFactory;
-import javax.xml.stream.XMLReporter;
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.XMLReporter2;
-import org.codehaus.stax2.XMLStreamLocation2;
-import org.codehaus.stax2.validation.XMLValidationProblem;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.InputConfigFlags;
-import com.ctc.wstx.cfg.ParsingErrorMsgs;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.dtd.MinimalDTDReader;
-import com.ctc.wstx.ent.EntityDecl;
-import com.ctc.wstx.ent.IntEntity;
-import com.ctc.wstx.exc.*;
-import com.ctc.wstx.io.DefaultInputResolver;
-import com.ctc.wstx.io.WstxInputData;
-import com.ctc.wstx.io.WstxInputLocation;
-import com.ctc.wstx.io.WstxInputSource;
-import com.ctc.wstx.util.ExceptionUtil;
-import com.ctc.wstx.util.SymbolTable;
-import com.ctc.wstx.util.TextBuffer;
-
-/**
- * Abstract base class that defines some basic functionality that all
- * Woodstox reader classes (main XML reader, DTD reader) extend from.
- */
-
-public abstract class StreamScanner
- extends WstxInputData
- implements InputProblemReporter,
- InputConfigFlags, ParsingErrorMsgs
-{
-
- // // // Some well-known chars:
-
- /**
- * Last (highest) char code of the three, LF, CR and NULL
- */
- public final static char CHAR_CR_LF_OR_NULL = (char) 13;
-
- public final static int INT_CR_LF_OR_NULL = 13;
-
- /**
- * Character that allows quick check of whether a char can potentially
- * be some kind of markup, WRT input stream processing;
- * has to contain linefeeds, &, < and > (">" only matters when
- * quoting text, as part of "]]>")
- */
- protected final static char CHAR_FIRST_PURE_TEXT = (char) ('>' + 1);
-
-
- /**
- * First character in Unicode (ie one with lowest id) that is legal
- * as part of a local name (all valid name chars minus ':'). Used
- * for doing quick check for local name end; usually name ends in
- * a whitespace or equals sign.
- */
- protected final static char CHAR_LOWEST_LEGAL_LOCALNAME_CHAR = '-';
-
- /*
- ///////////////////////////////////////////////////////////
- // Character validity constants, structs
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * We will only use validity array for first 256 characters, mostly
- * because after those characters it's easier to do fairly simple
- * block checks.
- */
- private final static int VALID_CHAR_COUNT = 0x100;
-
- private final static byte NAME_CHAR_INVALID_B = (byte) 0;
- private final static byte NAME_CHAR_ALL_VALID_B = (byte) 1;
- private final static byte NAME_CHAR_VALID_NONFIRST_B = (byte) -1;
-
- private final static byte[] sCharValidity = new byte[VALID_CHAR_COUNT];
-
- static {
- /* First, since all valid-as-first chars are also valid-as-other chars,
- * we'll initialize common chars:
- */
- sCharValidity['_'] = NAME_CHAR_ALL_VALID_B;
- for (int i = 0, last = ('z' - 'a'); i <= last; ++i) {
- sCharValidity['A' + i] = NAME_CHAR_ALL_VALID_B;
- sCharValidity['a' + i] = NAME_CHAR_ALL_VALID_B;
- }
- for (int i = 0xC0; i < 0xF6; ++i) { // not all are fully valid, but
- sCharValidity[i] = NAME_CHAR_ALL_VALID_B;
- }
- // ... now we can 'revert' ones not fully valid:
- sCharValidity[0xD7] = NAME_CHAR_INVALID_B;
- sCharValidity[0xF7] = NAME_CHAR_INVALID_B;
-
- /* And then we can proceed with ones only valid-as-other.
- */
- sCharValidity['-'] = NAME_CHAR_VALID_NONFIRST_B;
- sCharValidity['.'] = NAME_CHAR_VALID_NONFIRST_B;
- sCharValidity[0xB7] = NAME_CHAR_VALID_NONFIRST_B;
- for (int i = '0'; i <= '9'; ++i) {
- sCharValidity[i] = NAME_CHAR_VALID_NONFIRST_B;
- }
- }
-
- /**
- * Public identifiers only use 7-bit ascii range.
- */
- private final static int VALID_PUBID_CHAR_COUNT = 0x80;
- private final static byte[] sPubidValidity = new byte[VALID_PUBID_CHAR_COUNT];
-// private final static byte PUBID_CHAR_INVALID_B = (byte) 0;
- private final static byte PUBID_CHAR_VALID_B = (byte) 1;
- static {
- for (int i = 0, last = ('z' - 'a'); i <= last; ++i) {
- sPubidValidity['A' + i] = PUBID_CHAR_VALID_B;
- sPubidValidity['a' + i] = PUBID_CHAR_VALID_B;
- }
- for (int i = '0'; i <= '9'; ++i) {
- sPubidValidity[i] = PUBID_CHAR_VALID_B;
- }
-
- // 3 main white space types are valid
- sPubidValidity[0x0A] = PUBID_CHAR_VALID_B;
- sPubidValidity[0x0D] = PUBID_CHAR_VALID_B;
- sPubidValidity[0x20] = PUBID_CHAR_VALID_B;
-
- // And many of punctuation/separator ascii chars too:
- sPubidValidity['-'] = PUBID_CHAR_VALID_B;
- sPubidValidity['\''] = PUBID_CHAR_VALID_B;
- sPubidValidity['('] = PUBID_CHAR_VALID_B;
- sPubidValidity[')'] = PUBID_CHAR_VALID_B;
- sPubidValidity['+'] = PUBID_CHAR_VALID_B;
- sPubidValidity[','] = PUBID_CHAR_VALID_B;
- sPubidValidity['.'] = PUBID_CHAR_VALID_B;
- sPubidValidity['/'] = PUBID_CHAR_VALID_B;
- sPubidValidity[':'] = PUBID_CHAR_VALID_B;
- sPubidValidity['='] = PUBID_CHAR_VALID_B;
- sPubidValidity['?'] = PUBID_CHAR_VALID_B;
- sPubidValidity[';'] = PUBID_CHAR_VALID_B;
- sPubidValidity['!'] = PUBID_CHAR_VALID_B;
- sPubidValidity['*'] = PUBID_CHAR_VALID_B;
- sPubidValidity['#'] = PUBID_CHAR_VALID_B;
- sPubidValidity['@'] = PUBID_CHAR_VALID_B;
- sPubidValidity['$'] = PUBID_CHAR_VALID_B;
- sPubidValidity['_'] = PUBID_CHAR_VALID_B;
- sPubidValidity['%'] = PUBID_CHAR_VALID_B;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Basic configuration
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Copy of the configuration object passed by the factory.
- * Contains immutable settings for this reader (or in case
- * of DTD parsers, reader that uses it)
- */
- protected final ReaderConfig mConfig;
-
- // // // Various extracted settings:
-
- /**
- * If true, Reader is namespace aware, and should do basic checks
- * (usually enforcing limitations on having colons in names)
- */
- protected final boolean mCfgNsEnabled;
-
- // Extracted standard on/off settings:
-
- /**
- * note: left non-final on purpose: sub-class may need to modify
- * the default value after construction.
- */
- protected boolean mCfgReplaceEntities;
-
- /*
- ///////////////////////////////////////////////////////////
- // Symbol handling, if applicable
- ///////////////////////////////////////////////////////////
- */
-
- final SymbolTable mSymbols;
-
- /**
- * Local full name for the event, if it has one (note: element events
- * do NOT use this variable; those names are stored in element stack):
- * target for processing instructions.
- *
- * Currently used for proc. instr. target, and entity name (at least
- * when current entity reference is null).
- *
- * Note: this variable is generally not cleared, since it comes from
- * a symbol table, ie. this won't be the only reference.
- */
- protected String mCurrName;
-
- /*
- ///////////////////////////////////////////////////////////
- // Input handling
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Currently active input source; contains link to parent (nesting) input
- * sources, if any.
- */
- protected WstxInputSource mInput;
-
- /**
- * Top-most input source this reader can use; due to input source
- * chaining, this is not necessarily the root of all input; for example,
- * external DTD subset reader's root input still has original document
- * input as its parent.
- */
- protected final WstxInputSource mRootInput;
-
- /**
- * Custom resolver used to handle external entities that are to be expanded
- * by this reader (external param/general entity expander)
- */
- XMLResolver mEntityResolver = null;
-
- /**
- * This is the current depth of the input stack (same as what input
- * element stack would return as its depth).
- * It is used to enforce input scope constraints for nesting of
- * elements (for xml reader) and dtd declaration (for dtd reader)
- * with regards to input block (entity expansion) boundaries.
- *
- * Basically this value is compared to {@link #mInputTopDepth}, which
- * indicates what was the depth at the point where the currently active
- * input scope/block was started.
- */
- protected int mCurrDepth = 0;
-
- protected int mInputTopDepth = 0;
-
- /**
- * Flag that indicates whether linefeeds in the input data are to
- * be normalized or not.
- * Xml specs mandate that the line feeds are only normalized
- * when they are from the external entities (main doc, external
- * general/parsed entities), so normalization has to be
- * suppressed when expanding internal general/parsed entities.
- */
- protected boolean mNormalizeLFs;
-
- /*
- ///////////////////////////////////////////////////////////
- // Buffer(s) for local name(s) and text content
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Temporary buffer used if local name can not be just directly
- * constructed from input buffer (name is on a boundary or such).
- */
- protected char[] mNameBuffer = null;
-
- /*
- ///////////////////////////////////////////////////////////
- // Information about starting location of event
- // Reader is pointing to; updated on-demand
- ///////////////////////////////////////////////////////////
- */
-
- // // // Location info at point when current token was started
-
- /**
- * Total number of characters read before start of current token.
- * For big (gigabyte-sized) sizes are possible, needs to be long,
- * unlike pointers and sizes related to in-memory buffers.
- */
- protected long mTokenInputTotal = 0;
-
- /**
- * Input row on which current token starts, 1-based
- */
- protected int mTokenInputRow = 1;
-
- /**
- * Column on input row that current token starts; 0-based (although
- * in the end it'll be converted to 1-based)
- */
- protected int mTokenInputCol = 0;
-
- /*
- ///////////////////////////////////////////////////////////
- // XML document information (from doc decl if one
- // was found) common to all entities (main xml
- // document, external DTD subset)
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Input stream encoding, if known (passed in, or determined by
- * auto-detection); null if not.
- */
- String mDocInputEncoding = null;
-
- /**
- * Character encoding from xml declaration, if any; null if no
- * declaration, or it didn't specify encoding.
- */
- String mDocXmlEncoding = null;
-
- /**
- * XML version as declared by the document; one of constants
- * from {@link XmlConsts} (like {@link XmlConsts#XML_V_10}).
- */
- protected int mDocXmlVersion = XmlConsts.XML_V_UNKNOWN;
-
- /**
- * Cache of internal character entities;
- */
- protected Map mCachedEntities;
-
- /**
- * Flag for whether or not character references should be treated as entities
- */
- protected boolean mCfgTreatCharRefsAsEntities;
-
- /**
- * Entity reference stream currently points to.
- */
- protected EntityDecl mCurrEntity;
-
- /*
- ///////////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Constructor used when creating a complete new (main-level) reader that
- * does not share its input buffers or state with another reader.
- */
- protected StreamScanner(WstxInputSource input, ReaderConfig cfg,
- XMLResolver res)
- {
- super();
- mInput = input;
- // 17-Jun-2004, TSa: Need to know root-level input source
- mRootInput = input;
-
- mConfig = cfg;
- mSymbols = cfg.getSymbols();
- int cf = cfg.getConfigFlags();
- mCfgNsEnabled = (cf & CFG_NAMESPACE_AWARE) != 0;
- mCfgReplaceEntities = (cf & CFG_REPLACE_ENTITY_REFS) != 0;
-
- mNormalizeLFs = mConfig.willNormalizeLFs();
- mInputBuffer = null;
- mInputPtr = mInputEnd = 0;
- mEntityResolver = res;
-
- mCfgTreatCharRefsAsEntities = mConfig.willTreatCharRefsAsEnts();
- mCachedEntities = mCfgTreatCharRefsAsEntities ? new HashMap() : Collections.EMPTY_MAP;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Package API
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method that returns location of the last character returned by this
- * reader; that is, location "one less" than the currently pointed to
- * location.
- */
- protected WstxInputLocation getLastCharLocation()
- {
- return mInput.getLocation(mCurrInputProcessed + mInputPtr - 1,
- mCurrInputRow,
- mInputPtr - mCurrInputRowStart);
- }
-
- protected URL getSource() {
- return mInput.getSource();
- }
-
- protected String getSystemId() {
- return mInput.getSystemId();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Partial LocationInfo implementation (not implemented
- // by this base class, but is by some sub-classes)
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Returns location of last properly parsed token; as per StAX specs,
- * apparently needs to be the end of current event, which is the same
- * as the start of the following event (or EOF if that's next).
- */
- public abstract Location getLocation();
-
- public XMLStreamLocation2 getStartLocation()
- {
- // note: +1 is used as columns are 1-based...
- return mInput.getLocation(mTokenInputTotal, mTokenInputRow,
- mTokenInputCol + 1);
- }
-
- public XMLStreamLocation2 getCurrentLocation()
- {
- return mInput.getLocation(mCurrInputProcessed + mInputPtr,
- mCurrInputRow,
- mInputPtr - mCurrInputRowStart + 1);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // InputProblemReporter implementation
- ///////////////////////////////////////////////////////////
- */
-
- public WstxException throwWfcException(String msg, boolean deferErrors)
- throws WstxException
- {
- WstxException ex = constructWfcException(msg);
- if (!deferErrors) {
- throw ex;
- }
- return ex;
- }
-
- public void throwParseError(String msg) throws XMLStreamException
- {
- throwParseError(msg, null, null);
- }
-
- /**
- * Throws generic parse error with specified message and current parsing
- * location.
- *
- * Note: public access only because core code in other packages needs
- * to access it.
- */
- public void throwParseError(String format, Object arg, Object arg2)
- throws XMLStreamException
- {
- String msg = (arg != null || arg2 != null) ?
- MessageFormat.format(format, new Object[] { arg, arg2 }) : format;
- throw constructWfcException(msg);
- }
-
- public void reportProblem(String probType, String format, Object arg, Object arg2)
- throws XMLStreamException
- {
- XMLReporter rep = mConfig.getXMLReporter();
- if (rep != null) {
- _reportProblem(rep, probType,
- MessageFormat.format(format, new Object[] { arg, arg2 }), null);
- }
- }
-
- public void reportProblem(Location loc, String probType,
- String format, Object arg, Object arg2)
- throws XMLStreamException
- {
- XMLReporter rep = mConfig.getXMLReporter();
- if (rep != null) {
- String msg = (arg != null || arg2 != null) ?
- MessageFormat.format(format, new Object[] { arg, arg2 }) : format;
- _reportProblem(rep, probType, msg, loc);
- }
- }
-
- protected void _reportProblem(XMLReporter rep, String probType, String msg, Location loc)
- throws XMLStreamException
- {
- if (loc == null) {
- loc = getLastCharLocation();
- }
- _reportProblem(rep, new XMLValidationProblem(loc, msg, XMLValidationProblem.SEVERITY_ERROR, probType));
- }
-
- protected void _reportProblem(XMLReporter rep, XMLValidationProblem prob)
- throws XMLStreamException
- {
- if (rep != null) {
- Location loc = prob.getLocation();
- if (loc == null) {
- loc = getLastCharLocation();
- prob.setLocation(loc);
- }
- // Backwards-compatibility fix: add non-null type, if missing:
- if (prob.getType() == null) {
- prob.setType(ErrorConsts.WT_VALIDATION);
- }
- // [WSTX-154]: was catching and dropping thrown exception: shouldn't.
- // [WTSX-157]: need to support XMLReporter2
- if (rep instanceof XMLReporter2) {
- ((XMLReporter2) rep).report(prob);
- } else {
- rep.report(prob.getMessage(), prob.getType(), prob, loc);
- }
- }
- }
-
- /**
- *
- * Note: this is the base implementation used for implementing
- *
- * Note: public only because some implementations are on different
- * package.
- */
- public final WstxInputSource getCurrentInput() {
- return mInput;
- }
-
- protected final int inputInBuffer() {
- return mInputEnd - mInputPtr;
- }
-
- protected final int getNext()
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- if (!loadMore()) {
- return -1;
- }
- }
- return (int) mInputBuffer[mInputPtr++];
- }
-
- /**
- * Similar to {@link #getNext}, but does not advance pointer
- * in input buffer.
- *
- * Note: this method only peeks within current input source;
- * it does not close it and check nested input source (if any).
- * This is necessary when checking keywords, since they can never
- * cross input block boundary.
- */
- protected final int peekNext()
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- if (!loadMoreFromCurrent()) {
- return -1;
- }
- }
- return (int) mInputBuffer[mInputPtr];
- }
-
- protected final char getNextChar(String errorMsg)
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- loadMore(errorMsg);
- }
- return mInputBuffer[mInputPtr++];
- }
-
- /**
- * Similar to {@link #getNextChar}, but will not read more characters
- * from parent input source(s) if the current input source doesn't
- * have more content. This is often needed to prevent "runaway" content,
- * such as comments that start in an entity but do not have matching
- * close marker inside entity; XML specification specifically states
- * such markup is not legal.
- */
- protected final char getNextCharFromCurrent(String errorMsg)
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- loadMoreFromCurrent(errorMsg);
- }
- return mInputBuffer[mInputPtr++];
- }
-
- /**
- * Method that will skip through zero or more white space characters,
- * and return either the character following white space, or -1 to
- * indicate EOF (end of the outermost input source)/
- */
- protected final int getNextAfterWS()
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- if (!loadMore()) {
- return -1;
- }
- }
- char c = mInputBuffer[mInputPtr++];
- while (c <= CHAR_SPACE) {
- // Linefeed?
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- // Still a white space?
- if (mInputPtr >= mInputEnd) {
- if (!loadMore()) {
- return -1;
- }
- }
- c = mInputBuffer[mInputPtr++];
- }
- return (int) c;
- }
-
- protected final char getNextCharAfterWS(String errorMsg)
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- loadMore(errorMsg);
- }
-
- char c = mInputBuffer[mInputPtr++];
- while (c <= CHAR_SPACE) {
- // Linefeed?
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
-
- // Still a white space?
- if (mInputPtr >= mInputEnd) {
- loadMore(errorMsg);
- }
- c = mInputBuffer[mInputPtr++];
- }
- return c;
- }
-
- protected final char getNextInCurrAfterWS(String errorMsg)
- throws XMLStreamException
- {
- return getNextInCurrAfterWS(errorMsg, getNextCharFromCurrent(errorMsg));
- }
-
- protected final char getNextInCurrAfterWS(String errorMsg, char c)
- throws XMLStreamException
- {
- while (c <= CHAR_SPACE) {
- // Linefeed?
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
-
- // Still a white space?
- if (mInputPtr >= mInputEnd) {
- loadMoreFromCurrent(errorMsg);
- }
- c = mInputBuffer[mInputPtr++];
- }
- return c;
- }
-
- /**
- * Method called when a CR has been spotted in input; checks if next
- * char is LF, and if so, skips it. Note that next character has to
- * come from the current input source, to qualify; it can never come
- * from another (nested) input source.
- *
- * @return True, if passed in char is '\r' and next one is '\n'.
- */
- protected final boolean skipCRLF(char c)
- throws XMLStreamException
- {
- boolean result;
-
- if (c == '\r' && peekNext() == '\n') {
- ++mInputPtr;
- result = true;
- } else {
- result = false;
- }
- ++mCurrInputRow;
- mCurrInputRowStart = mInputPtr;
- return result;
- }
-
- protected final void markLF() {
- ++mCurrInputRow;
- mCurrInputRowStart = mInputPtr;
- }
-
- protected final void markLF(int inputPtr) {
- ++mCurrInputRow;
- mCurrInputRowStart = inputPtr;
- }
-
- /**
- * Method to push back last character read; can only be called once,
- * that is, no more than one char can be guaranteed to be succesfully
- * returned.
- */
- protected final void pushback() { --mInputPtr; }
-
- /*
- ///////////////////////////////////////////////////////////
- // Sub-class overridable input handling methods
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method called when an entity has been expanded (new input source
- * has been created). Needs to initialize location information and change
- * active input source.
- *
- * @param entityId Name of the entity being expanded
- */
- protected void initInputSource(WstxInputSource newInput, boolean isExt,
- String entityId)
- throws XMLStreamException
- {
- mInput = newInput;
- // Let's make sure new input will be read next time input is needed:
- mInputPtr = 0;
- mInputEnd = 0;
- /* Plus, reset the input location so that'll be accurate for
- * error reporting etc.
- */
- mInputTopDepth = mCurrDepth;
- mInput.initInputLocation(this, mCurrDepth);
-
- /* 21-Feb-2006, TSa: Linefeeds are NOT normalized when expanding
- * internal entities (XML, 2.11)
- */
- if (isExt) {
- mNormalizeLFs = true;
- } else {
- mNormalizeLFs = false;
- }
- }
-
- /**
- * Method that will try to read one or more characters from currently
- * open input sources; closing input sources if necessary.
- *
- * @return true if reading succeeded (or may succeed), false if
- * we reached EOF.
- */
- protected boolean loadMore()
- throws XMLStreamException
- {
- WstxInputSource input = mInput;
- do {
- /* Need to make sure offsets are properly updated for error
- * reporting purposes, and do this now while previous amounts
- * are still known.
- */
- mCurrInputProcessed += mInputEnd;
- mCurrInputRowStart -= mInputEnd;
- int count;
- try {
- count = input.readInto(this);
- if (count > 0) {
- return true;
- }
- input.close();
- } catch (IOException ioe) {
- throw constructFromIOE(ioe);
- }
- if (input == mRootInput) {
- /* Note: no need to check entity/input nesting in this
- * particular case, since it will be handled by higher level
- * parsing code (results in an unexpected EOF)
- */
- return false;
- }
- WstxInputSource parent = input.getParent();
- if (parent == null) { // sanity check!
- throwNullParent(input);
- }
- /* 13-Feb-2006, TSa: Ok, do we violate a proper nesting constraints
- * with this input block closure?
- */
- if (mCurrDepth != input.getScopeId()) {
- handleIncompleteEntityProblem(input);
- }
-
- mInput = input = parent;
- input.restoreContext(this);
- mInputTopDepth = input.getScopeId();
- /* 21-Feb-2006, TSa: Since linefeed normalization needs to be
- * suppressed for internal entity expansion, we may need to
- * change the state...
- */
- if (!mNormalizeLFs) {
- mNormalizeLFs = !input.fromInternalEntity();
- }
- // Maybe there are leftovers from that input in buffer now?
- } while (mInputPtr >= mInputEnd);
-
- return true;
- }
-
- protected final boolean loadMore(String errorMsg)
- throws XMLStreamException
- {
- if (!loadMore()) {
- throwUnexpectedEOF(errorMsg);
- }
- return true;
- }
-
- protected boolean loadMoreFromCurrent()
- throws XMLStreamException
- {
- // Need to update offsets properly
- mCurrInputProcessed += mInputEnd;
- mCurrInputRowStart -= mInputEnd;
- try {
- int count = mInput.readInto(this);
- return (count > 0);
- } catch (IOException ie) {
- throw constructFromIOE(ie);
- }
- }
-
- protected final boolean loadMoreFromCurrent(String errorMsg)
- throws XMLStreamException
- {
- if (!loadMoreFromCurrent()) {
- throwUnexpectedEOB(errorMsg);
- }
- return true;
- }
-
- /**
- * Method called to make sure current main-level input buffer has at
- * least specified number of characters available consequtively,
- * without having to call {@link #loadMore}. It can only be called
- * when input comes from main-level buffer; further, call can shift
- * content in input buffer, so caller has to flush any data still
- * pending. In short, caller has to know exactly what it's doing. :-)
- *
- * Note: method does not check for any other input sources than the
- * current one -- if current source can not fulfill the request, a
- * failure is indicated.
- *
- * @return true if there's now enough data; false if not (EOF)
- */
- protected boolean ensureInput(int minAmount)
- throws XMLStreamException
- {
- int currAmount = mInputEnd - mInputPtr;
- if (currAmount >= minAmount) {
- return true;
- }
- try {
- return mInput.readMore(this, minAmount);
- } catch (IOException ie) {
- throw constructFromIOE(ie);
- }
- }
-
- protected void closeAllInput(boolean force)
- throws XMLStreamException
- {
- WstxInputSource input = mInput;
- while (true) {
- try {
- if (force) {
- input.closeCompletely();
- } else {
- input.close();
- }
- } catch (IOException ie) {
- throw constructFromIOE(ie);
- }
- if (input == mRootInput) {
- break;
- }
- WstxInputSource parent = input.getParent();
- if (parent == null) { // sanity check!
- throwNullParent(input);
- }
- mInput = input = parent;
- }
- }
-
- protected void throwNullParent(WstxInputSource curr)
- {
- throw new IllegalStateException(ErrorConsts.ERR_INTERNAL);
- //throw new IllegalStateException("Internal error: null parent for input source '"+curr+"'; should never occur (should have stopped at root input '"+mRootInput+"').");
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Entity resolution
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method that tries to resolve a character entity, or (if caller so
- * specifies), a pre-defined internal entity (lt, gt, amp, apos, quot).
- * It will succeed iff:
- *
- * Note: On entry we are guaranteed there are at least 3 more characters
- * in this buffer; otherwise we shouldn't be called.
- *
- * @param checkStd If true, will check pre-defined internal entities
- * (gt, lt, amp, apos, quot); if false, will only check actual
- * character entities.
- *
- * @return (Valid) character value, if entity is a character reference,
- * and could be resolved from current input buffer (does not span
- * buffer boundary); null char (code 0) if not (either non-char
- * entity, or spans input buffer boundary).
- */
- protected int resolveSimpleEntity(boolean checkStd)
- throws XMLStreamException
- {
- char[] buf = mInputBuffer;
- int ptr = mInputPtr;
- char c = buf[ptr++];
-
- // Numeric reference?
- if (c == '#') {
- c = buf[ptr++];
- int value = 0;
- int inputLen = mInputEnd;
- if (c == 'x') { // hex
- while (ptr < inputLen) {
- c = buf[ptr++];
- if (c == ';') {
- break;
- }
- value = value << 4;
- if (c <= '9' && c >= '0') {
- value += (c - '0');
- } else if (c >= 'a' && c <= 'f') {
- value += (10 + (c - 'a'));
- } else if (c >= 'A' && c <= 'F') {
- value += (10 + (c - 'A'));
- } else {
- mInputPtr = ptr; // so error points to correct char
- throwUnexpectedChar(c, "; expected a hex digit (0-9a-fA-F).");
- }
- /* Need to check for overflow; easiest to do right as
- * it happens...
- */
- if (value > MAX_UNICODE_CHAR) {
- reportUnicodeOverflow();
- }
- }
- } else { // numeric (decimal)
- while (c != ';') {
- if (c <= '9' && c >= '0') {
- value = (value * 10) + (c - '0');
- // Overflow?
- if (value > MAX_UNICODE_CHAR) {
- reportUnicodeOverflow();
- }
- } else {
- mInputPtr = ptr; // so error points to correct char
- throwUnexpectedChar(c, "; expected a decimal number.");
- }
- if (ptr >= inputLen) {
- break;
- }
- c = buf[ptr++];
- }
- }
- /* We get here either if we got it all, OR if we ran out of
- * input in current buffer.
- */
- if (c == ';') { // got the full thing
- mInputPtr = ptr;
- validateChar(value);
- return value;
- }
-
- /* If we ran out of input, need to just fall back, gets
- * resolved via 'full' resolution mechanism.
- */
- } else if (checkStd) {
- /* Caller may not want to resolve these quite yet...
- * (when it wants separate events for non-char entities)
- */
- if (c == 'a') { // amp or apos?
- c = buf[ptr++];
-
- if (c == 'm') { // amp?
- if (buf[ptr++] == 'p') {
- if (ptr < mInputEnd && buf[ptr++] == ';') {
- mInputPtr = ptr;
- return '&';
- }
- }
- } else if (c == 'p') { // apos?
- if (buf[ptr++] == 'o') {
- int len = mInputEnd;
- if (ptr < len && buf[ptr++] == 's') {
- if (ptr < len && buf[ptr++] == ';') {
- mInputPtr = ptr;
- return '\'';
- }
- }
- }
- }
- } else if (c == 'g') { // gt?
- if (buf[ptr++] == 't' && buf[ptr++] == ';') {
- mInputPtr = ptr;
- return '>';
- }
- } else if (c == 'l') { // lt?
- if (buf[ptr++] == 't' && buf[ptr++] == ';') {
- mInputPtr = ptr;
- return '<';
- }
- } else if (c == 'q') { // quot?
- if (buf[ptr++] == 'u' && buf[ptr++] == 'o') {
- int len = mInputEnd;
- if (ptr < len && buf[ptr++] == 't') {
- if (ptr < len && buf[ptr++] == ';') {
- mInputPtr = ptr;
- return '"';
- }
- }
- }
- }
- }
- return 0;
- }
-
- /**
- * Method called to resolve character entities, and only character
- * entities (except that pre-defined char entities -- amp, apos, lt,
- * gt, quote -- MAY be "char entities" in this sense, depending on
- * arguments).
- * Otherwise it is to return the null char; if so,
- * the input pointer will point to the same point as when method
- * entered (char after ampersand), plus the ampersand itself is
- * guaranteed to be in the input buffer (so caller can just push it
- * back if necessary).
- *
- * Most often this method is called when reader is not to expand
- * non-char entities automatically, but to return them as separate
- * events.
- *
- * Main complication here is that we need to do 5-char lookahead. This
- * is problematic if chars are on input buffer boundary. This is ok
- * for the root level input buffer, but not for some nested buffers.
- * However, according to XML specs, such split entities are actually
- * illegal... so we can throw an exception in those cases.
- *
- * @param checkStd If true, will check pre-defined internal entities
- * (gt, lt, amp, apos, quot) as character entities; if false, will only
- * check actual 'real' character entities.
- *
- * @return (Valid) character value, if entity is a character reference,
- * and could be resolved from current input buffer (does not span
- * buffer boundary); null char (code 0) if not (either non-char
- * entity, or spans input buffer boundary).
- */
- protected int resolveCharOnlyEntity(boolean checkStd)
- throws XMLStreamException
- {
- //int avail = inputInBuffer();
- int avail = mInputEnd - mInputPtr;
- if (avail < 6) {
- // split entity, or buffer boundary
- /* Don't want to lose leading '&' (in case we can not expand
- * the entity), so let's push it back first
- */
- --mInputPtr;
- /* Shortest valid reference would be 3 chars ('&a;'); which
- * would only be legal from an expanded entity...
- */
- if (!ensureInput(6)) {
- avail = inputInBuffer();
- if (avail < 3) {
- throwUnexpectedEOF(SUFFIX_IN_ENTITY_REF);
- }
- } else {
- avail = 6;
- }
- // ... and now we can move pointer back as well:
- ++mInputPtr;
- }
-
- /* Ok, now we have one more character to check, and that's enough
- * to determine type decisively.
- */
- char c = mInputBuffer[mInputPtr];
-
- // A char reference?
- if (c == '#') { // yup
- ++mInputPtr;
- return resolveCharEnt(null);
- }
-
- // nope... except may be a pre-def?
- if (checkStd) {
- if (c == 'a') {
- char d = mInputBuffer[mInputPtr+1];
- if (d == 'm') {
- if (avail >= 4
- && mInputBuffer[mInputPtr+2] == 'p'
- && mInputBuffer[mInputPtr+3] == ';') {
- mInputPtr += 4;
- return '&';
- }
- } else if (d == 'p') {
- if (avail >= 5
- && mInputBuffer[mInputPtr+2] == 'o'
- && mInputBuffer[mInputPtr+3] == 's'
- && mInputBuffer[mInputPtr+4] == ';') {
- mInputPtr += 5;
- return '\'';
- }
- }
- } else if (c == 'l') {
- if (avail >= 3
- && mInputBuffer[mInputPtr+1] == 't'
- && mInputBuffer[mInputPtr+2] == ';') {
- mInputPtr += 3;
- return '<';
- }
- } else if (c == 'g') {
- if (avail >= 3
- && mInputBuffer[mInputPtr+1] == 't'
- && mInputBuffer[mInputPtr+2] == ';') {
- mInputPtr += 3;
- return '>';
- }
- } else if (c == 'q') {
- if (avail >= 5
- && mInputBuffer[mInputPtr+1] == 'u'
- && mInputBuffer[mInputPtr+2] == 'o'
- && mInputBuffer[mInputPtr+3] == 't'
- && mInputBuffer[mInputPtr+4] == ';') {
- mInputPtr += 5;
- return '"';
- }
- }
- }
- return 0;
- }
-
- /**
- * Reverse of {@link #resolveCharOnlyEntity}; will only resolve entity
- * if it is NOT a character entity (or pre-defined 'generic' entity;
- * amp, apos, lt, gt or quot). Only used in cases where entities
- * are to be separately returned unexpanded (in non-entity-replacing
- * mode); which means it's never called from dtd handler.
- */
- protected EntityDecl resolveNonCharEntity()
- throws XMLStreamException
- {
- //int avail = inputInBuffer();
- int avail = mInputEnd - mInputPtr;
- if (avail < 6) {
- // split entity, or buffer boundary
- /* Don't want to lose leading '&' (in case we can not expand
- * the entity), so let's push it back first
- */
- --mInputPtr;
-
- /* Shortest valid reference would be 3 chars ('&a;'); which
- * would only be legal from an expanded entity...
- */
- if (!ensureInput(6)) {
- avail = inputInBuffer();
- if (avail < 3) {
- throwUnexpectedEOF(SUFFIX_IN_ENTITY_REF);
- }
- } else {
- avail = 6;
- }
- // ... and now we can move pointer back as well:
- ++mInputPtr;
- }
-
- // We don't care about char entities:
- char c = mInputBuffer[mInputPtr];
- if (c == '#') {
- return null;
- }
-
- /* 19-Aug-2004, TSa: Need special handling for pre-defined
- * entities; they are not counted as 'real' general parsed
- * entities, but more as character entities...
- */
-
- // have chars at least up to mInputPtr+4 by now
- if (c == 'a') {
- char d = mInputBuffer[mInputPtr+1];
- if (d == 'm') {
- if (avail >= 4
- && mInputBuffer[mInputPtr+2] == 'p'
- && mInputBuffer[mInputPtr+3] == ';') {
- // If not automatically expanding:
- //return sEntityAmp;
- // mInputPtr += 4;
- return null;
- }
- } else if (d == 'p') {
- if (avail >= 5
- && mInputBuffer[mInputPtr+2] == 'o'
- && mInputBuffer[mInputPtr+3] == 's'
- && mInputBuffer[mInputPtr+4] == ';') {
- return null;
- }
- }
- } else if (c == 'l') {
- if (avail >= 3
- && mInputBuffer[mInputPtr+1] == 't'
- && mInputBuffer[mInputPtr+2] == ';') {
- return null;
- }
- } else if (c == 'g') {
- if (avail >= 3
- && mInputBuffer[mInputPtr+1] == 't'
- && mInputBuffer[mInputPtr+2] == ';') {
- return null;
- }
- } else if (c == 'q') {
- if (avail >= 5
- && mInputBuffer[mInputPtr+1] == 'u'
- && mInputBuffer[mInputPtr+2] == 'o'
- && mInputBuffer[mInputPtr+3] == 't'
- && mInputBuffer[mInputPtr+4] == ';') {
- return null;
- }
- }
-
- // Otherwise, let's just parse in generic way:
- ++mInputPtr; // since we already read the first letter
- String id = parseEntityName(c);
- mCurrName = id;
-
- return findEntity(id, null);
- }
-
- /**
- * Method that does full resolution of an entity reference, be it
- * character entity, internal entity or external entity, including
- * updating of input buffers, and depending on whether result is
- * a character entity (or one of 5 pre-defined entities), returns
- * char in question, or null character (code 0) to indicate it had
- * to change input source.
- *
- * @param allowExt If true, is allowed to expand external entities
- * (expanding text); if false, is not (expanding attribute value).
- *
- * @return Either single-character replacement (which is NOT to be
- * reparsed), or null char (0) to indicate expansion is done via
- * input source.
- */
- protected int fullyResolveEntity(boolean allowExt)
- throws XMLStreamException
- {
- char c = getNextCharFromCurrent(SUFFIX_IN_ENTITY_REF);
- // Do we have a (numeric) character entity reference?
- if (c == '#') { // numeric
- final StringBuffer originalSurface = new StringBuffer("#");
- int ch = resolveCharEnt(originalSurface);
- if (mCfgTreatCharRefsAsEntities) {
- final char[] originalChars = new char[originalSurface.length()];
- originalSurface.getChars(0, originalSurface.length(), originalChars, 0);
- mCurrEntity = getIntEntity(ch, originalChars);
- return 0;
- }
- return ch;
- }
-
- String id = parseEntityName(c);
-
- // Perhaps we have a pre-defined char reference?
- c = id.charAt(0);
- /*
- * 16-May-2004, TSa: Should custom entities (or ones defined in int/ext subset) override
- * pre-defined settings for these?
- */
- char d = CHAR_NULL;
- if (c == 'a') { // amp or apos?
- if (id.equals("amp")) {
- d = '&';
- } else if (id.equals("apos")) {
- d = '\'';
- }
- } else if (c == 'g') { // gt?
- if (id.length() == 2 && id.charAt(1) == 't') {
- d = '>';
- }
- } else if (c == 'l') { // lt?
- if (id.length() == 2 && id.charAt(1) == 't') {
- d = '<';
- }
- } else if (c == 'q') { // quot?
- if (id.equals("quot")) {
- d = '"';
- }
- }
-
- if (d != CHAR_NULL) {
- if (mCfgTreatCharRefsAsEntities) {
- final char[] originalChars = new char[id.length()];
- id.getChars(0, id.length(), originalChars, 0);
- mCurrEntity = getIntEntity(d, originalChars);
- return 0;
- }
- return d;
- }
-
- final EntityDecl e = expandEntity(id, allowExt, null);
- if (mCfgTreatCharRefsAsEntities) {
- mCurrEntity = e;
- }
- return 0;
- }
-
- /**
- * Returns an entity (possibly from cache) for the argument character using the encoded
- * representation in mInputBuffer[entityStartPos ... mInputPtr-1].
- */
- protected EntityDecl getIntEntity(int ch, final char[] originalChars)
- {
- String cacheKey = new String(originalChars);
-
- IntEntity entity = (IntEntity) mCachedEntities.get(cacheKey);
- if (entity == null) {
- String repl;
- if (ch <= 0xFFFF) {
- repl = Character.toString((char) ch);
- } else {
- StringBuffer sb = new StringBuffer(2);
- ch -= 0x10000;
- sb.append((char) ((ch >> 10) + 0xD800));
- sb.append((char) ((ch & 0x3FF) + 0xDC00));
- repl = sb.toString();
- }
- entity = IntEntity.create(new String(originalChars), repl);
- mCachedEntities.put(cacheKey, entity);
- }
- return entity;
- }
-
-
- /**
- * Helper method that will try to expand a parsed entity (parameter or
- * generic entity).
- *
- * note: called by sub-classes (dtd parser), needs to be protected.
- *
- * @param id Name of the entity being expanded
- * @param allowExt Whether external entities can be expanded or not; if
- * not, and the entity to expand would be external one, an exception
- * will be thrown
- */
- protected EntityDecl expandEntity(String id, boolean allowExt,
- Object extraArg)
- throws XMLStreamException
- {
- mCurrName = id;
-
- EntityDecl ed = findEntity(id, extraArg);
-
- if (ed == null) {
- /* 30-Sep-2005, TSa: As per [WSTX-5], let's only throw exception
- * if we have to resolve it (otherwise it's just best-effort,
- * and null is ok)
- */
- /* 02-Oct-2005, TSa: Plus, [WSTX-4] adds "undeclared entity
- * resolver"
- */
- if (mCfgReplaceEntities) {
- mCurrEntity = expandUnresolvedEntity(id);
- }
- return null;
- }
-
- if (!mCfgTreatCharRefsAsEntities || this instanceof MinimalDTDReader) {
- expandEntity(ed, allowExt);
- }
-
- return ed;
- }
-
- /**
- *
- *
- * note: defined as private for documentation, ie. it's just called
- * from within this class (not sub-classes), from one specific method
- * (see above)
- *
- * @param ed Entity to be expanded
- * @param allowExt Whether external entities are allowed or not.
- */
- private void expandEntity(EntityDecl ed, boolean allowExt)
- throws XMLStreamException
- {
- String id = ed.getName();
-
- /* Very first thing; we can immediately check if expanding
- * this entity would result in infinite recursion:
- */
- if (mInput.isOrIsExpandedFrom(id)) {
- throwRecursionError(id);
- }
-
- /* Should not refer unparsed entities from attribute values
- * or text content (except via notation mechanism, but that's
- * not parsed here)
- */
- if (!ed.isParsed()) {
- throwParseError("Illegal reference to unparsed external entity \"{0}\"", id, null);
- }
-
- // 28-Jun-2004, TSa: Do we support external entity expansion?
- boolean isExt = ed.isExternal();
- if (isExt) {
- if (!allowExt) { // never ok in attribute value...
- throwParseError("Encountered a reference to external parsed entity \"{0}\" when expanding attribute value: not legal as per XML 1.0/1.1 #3.1", id, null);
- }
- if (!mConfig.willSupportExternalEntities()) {
- throwParseError("Encountered a reference to external entity \"{0}\", but stream reader has feature \"{1}\" disabled",
- id, XMLInputFactory.IS_SUPPORTING_EXTERNAL_ENTITIES);
- }
- }
-
- // First, let's give current context chance to save its stuff
- WstxInputSource oldInput = mInput;
- oldInput.saveContext(this);
- WstxInputSource newInput = null;
- try {
- newInput = ed.expand(oldInput, mEntityResolver, mConfig, mDocXmlVersion);
- } catch (FileNotFoundException fex) {
- /* Let's catch and rethrow this just so we get more meaningful
- * description (with input source position etc)
- */
- throwParseError("(was {0}) {1}", fex.getClass().getName(), fex.getMessage());
- } catch (IOException ioe) {
- throw constructFromIOE(ioe);
- }
- /* And then we'll need to make sure new input comes from the new
- * input source
- */
- initInputSource(newInput, isExt, id);
- }
-
- /**
- *
- * note: only called from the local expandEntity() method
- */
- private EntityDecl expandUnresolvedEntity(String id)
- throws XMLStreamException
- {
- XMLResolver resolver = mConfig.getUndeclaredEntityResolver();
- if (resolver != null) {
- /* Ok, we can check for recursion here; but let's only do that
- * if there is any chance that it might get resolved by
- * the special resolver (it must have been resolved this way
- * earlier, too...)
- */
- if (mInput.isOrIsExpandedFrom(id)) {
- throwRecursionError(id);
- }
-
- WstxInputSource oldInput = mInput;
- oldInput.saveContext(this);
- // null, null -> no public or system ids
- int xmlVersion = mDocXmlVersion;
- // 05-Feb-2006, TSa: If xmlVersion not explicitly known, defaults to 1.0
- if (xmlVersion == XmlConsts.XML_V_UNKNOWN) {
- xmlVersion = XmlConsts.XML_V_10;
- }
- WstxInputSource newInput;
- try {
- newInput = DefaultInputResolver.resolveEntityUsing
- (oldInput, id, null, null, resolver, mConfig, xmlVersion);
- if (mCfgTreatCharRefsAsEntities) {
- return new IntEntity(WstxInputLocation.getEmptyLocation(), newInput.getEntityId(),
- newInput.getSource(), new char[]{}, WstxInputLocation.getEmptyLocation());
- }
- } catch (IOException ioe) {
- throw constructFromIOE(ioe);
- }
- if (newInput != null) {
- // true -> is external
- initInputSource(newInput, true, id);
- return null;
- }
- }
- handleUndeclaredEntity(id);
- return null;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Abstract methods for sub-classes to implement
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Abstract method for sub-classes to implement, for finding
- * a declared general or parsed entity.
- *
- * @param id Identifier of the entity to find
- * @param arg Optional argument passed from caller; needed by DTD
- * reader.
- */
- protected abstract EntityDecl findEntity(String id, Object arg)
- throws XMLStreamException;
-
- /**
- * This method gets called if a declaration for an entity was not
- * found in entity expanding mode (enabled by default for xml reader,
- * always enabled for dtd reader).
- */
- protected abstract void handleUndeclaredEntity(String id)
- throws XMLStreamException;
-
- protected abstract void handleIncompleteEntityProblem(WstxInputSource closing)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////////////////////////
- // Basic tokenization
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method that will parse name token (roughly equivalent to XML specs;
- * although bit lenier for more efficient handling); either uri prefix,
- * or local name.
- *
- * Much of complexity in this method has to do with the intention to
- * try to avoid any character copies. In this optimal case algorithm
- * would be fairly simple. However, this only works if all data is
- * already in input buffer... if not, copy has to be made halfway
- * through parsing, and that complicates things.
- *
- * One thing to note is that String returned has been canonicalized
- * and (if necessary) added to symbol table. It can thus be compared
- * against other such (usually id) Strings, with simple equality operator.
- *
- * @param c First character of the name; not yet checked for validity
- *
- * @return Canonicalized name String (which may have length 0, if
- * EOF or non-name-start char encountered)
- */
- protected String parseLocalName(char c)
- throws XMLStreamException
- {
- /* Has to start with letter, or '_' (etc); we won't allow ':' as that
- * is taken as namespace separator; no use trying to optimize
- * heavily as it's 98% likely it is a valid char...
- */
- if (!isNameStartChar(c)) {
- if (c == ':') {
- throwUnexpectedChar(c, " (missing namespace prefix?)");
- }
- throwUnexpectedChar(c, " (expected a name start character)");
- }
-
- int ptr = mInputPtr;
- int hash = (int) c;
- final int inputLen = mInputEnd;
- int startPtr = ptr-1; // already read previous char
- final char[] inputBuf = mInputBuffer;
-
- /* After which there may be zero or more name chars
- * we have to consider
- */
- while (true) {
- if (ptr >= inputLen) {
- /* Ok, identifier may continue past buffer end, need
- * to continue with part 2 (separate method, as this is
- * not as common as having it all in buffer)
- */
- mInputPtr = ptr;
- return parseLocalName2(startPtr, hash);
- }
- // Ok, we have the char... is it a name char?
- c = inputBuf[ptr];
- if (c < CHAR_LOWEST_LEGAL_LOCALNAME_CHAR) {
- break;
- }
- if (!isNameChar(c)) {
- break;
- }
- hash = (hash * 31) + (int) c;
- ++ptr;
- }
- mInputPtr = ptr;
- return mSymbols.findSymbol(mInputBuffer, startPtr, ptr - startPtr, hash);
- }
-
- /**
- * Second part of name token parsing; called when name can continue
- * past input buffer end (so only part was read before calling this
- * method to read the rest).
- *
- * Note that this isn't heavily optimized, on assumption it's not
- * called very often.
- */
- protected String parseLocalName2(int start, int hash)
- throws XMLStreamException
- {
- int ptr = mInputEnd - start;
- // Let's assume fairly short names
- char[] outBuf = getNameBuffer(ptr+8);
-
- if (ptr > 0) {
- System.arraycopy(mInputBuffer, start, outBuf, 0, ptr);
- }
-
- int outLen = outBuf.length;
- while (true) {
- // note: names can not cross input block (entity) boundaries...
- if (mInputPtr >= mInputEnd) {
- if (!loadMoreFromCurrent()) {
- break;
- }
- }
- char c = mInputBuffer[mInputPtr];
- if (c < CHAR_LOWEST_LEGAL_LOCALNAME_CHAR) {
- break;
- }
- if (!isNameChar(c)) {
- break;
- }
- ++mInputPtr;
- if (ptr >= outLen) {
- mNameBuffer = outBuf = expandBy50Pct(outBuf);
- outLen = outBuf.length;
- }
- outBuf[ptr++] = c;
- hash = (hash * 31) + (int) c;
- }
- // Still need to canonicalize the name:
- return mSymbols.findSymbol(outBuf, 0, ptr, hash);
- }
-
- /**
- * Method that will parse 'full' name token; what full means depends on
- * whether reader is namespace aware or not. If it is, full name means
- * local name with no namespace prefix (PI target, entity/notation name);
- * if not, name can contain arbitrary number of colons. Note that
- * element and attribute names are NOT parsed here, so actual namespace
- * prefix separation can be handled properly there.
- *
- * Similar to {@link #parseLocalName}, much of complexity stems from
- * trying to avoid copying name characters from input buffer.
- *
- * Note that returned String will be canonicalized, similar to
- * {@link #parseLocalName}, but without separating prefix/local name.
- *
- * @return Canonicalized name String (which may have length 0, if
- * EOF or non-name-start char encountered)
- */
- protected String parseFullName()
- throws XMLStreamException
- {
- if (mInputPtr >= mInputEnd) {
- loadMoreFromCurrent();
- }
- return parseFullName(mInputBuffer[mInputPtr++]);
- }
-
- protected String parseFullName(char c)
- throws XMLStreamException
- {
- // First char has special handling:
- if (!isNameStartChar(c)) {
- if (c == ':') { // no name.... generally an error:
- if (mCfgNsEnabled) {
- throwNsColonException(parseFNameForError());
- }
- // Ok, that's fine actually
- } else {
- if (c <= CHAR_SPACE) {
- throwUnexpectedChar(c, " (missing name?)");
- }
- throwUnexpectedChar(c, " (expected a name start character)");
- }
- }
-
- int ptr = mInputPtr;
- int hash = (int) c;
- int inputLen = mInputEnd;
- int startPtr = ptr-1; // to account for the first char
-
- /* After which there may be zero or more name chars
- * we have to consider
- */
- while (true) {
- if (ptr >= inputLen) {
- /* Ok, identifier may continue past buffer end, need
- * to continue with part 2 (separate method, as this is
- * not as common as having it all in buffer)
- */
- mInputPtr = ptr;
- return parseFullName2(startPtr, hash);
- }
- c = mInputBuffer[ptr];
- if (c == ':') { // colon only allowed in non-NS mode
- if (mCfgNsEnabled) {
- mInputPtr = ptr;
- throwNsColonException(new String(mInputBuffer, startPtr, ptr - startPtr) + parseFNameForError());
- }
- } else {
- if (c < CHAR_LOWEST_LEGAL_LOCALNAME_CHAR) {
- break;
- }
- if (!isNameChar(c)) {
- break;
- }
- }
- hash = (hash * 31) + (int) c;
- ++ptr;
- }
- mInputPtr = ptr;
- return mSymbols.findSymbol(mInputBuffer, startPtr, ptr - startPtr, hash);
- }
-
- protected String parseFullName2(int start, int hash)
- throws XMLStreamException
- {
- int ptr = mInputEnd - start;
- // Let's assume fairly short names
- char[] outBuf = getNameBuffer(ptr+8);
-
- if (ptr > 0) {
- System.arraycopy(mInputBuffer, start, outBuf, 0, ptr);
- }
-
- int outLen = outBuf.length;
- while (true) {
- /* 06-Sep-2004, TSa: Name tokens are not allowed to continue
- * past entity expansion ranges... that is, all characters
- * have to come from the same input source. Thus, let's only
- * load things from same input level
- */
- if (mInputPtr >= mInputEnd) {
- if (!loadMoreFromCurrent()) {
- break;
- }
- }
- char c = mInputBuffer[mInputPtr];
- if (c == ':') { // colon only allowed in non-NS mode
- if (mCfgNsEnabled) {
- throwNsColonException(new String(outBuf, 0, ptr) + c + parseFNameForError());
- }
- } else if (c < CHAR_LOWEST_LEGAL_LOCALNAME_CHAR) {
- break;
- } else if (!isNameChar(c)) {
- break;
- }
- ++mInputPtr;
-
- if (ptr >= outLen) {
- mNameBuffer = outBuf = expandBy50Pct(outBuf);
- outLen = outBuf.length;
- }
- outBuf[ptr++] = c;
- hash = (hash * 31) + (int) c;
- }
-
- // Still need to canonicalize the name:
- return mSymbols.findSymbol(outBuf, 0, ptr, hash);
- }
-
- /**
- * Method called to read in full name, including unlimited number of
- * namespace separators (':'), for the purpose of displaying name in
- * an error message. Won't do any further validations, and parsing
- * is not optimized: main need is just to get more meaningful error
- * messages.
- */
- protected String parseFNameForError()
- throws XMLStreamException
- {
- StringBuffer sb = new StringBuffer(100);
- while (true) {
- char c;
-
- if (mInputPtr < mInputEnd) {
- c = mInputBuffer[mInputPtr++];
- } else { // can't error here, so let's accept EOF for now:
- int i = getNext();
- if (i < 0) {
- break;
- }
- c = (char) i;
- }
- if (c != ':' && !isNameChar(c)) {
- --mInputPtr;
- break;
- }
- sb.append(c);
- }
- return sb.toString();
- }
-
- protected final String parseEntityName(char c)
- throws XMLStreamException
- {
- String id = parseFullName(c);
- // Needs to be followed by a semi-colon, too.. from same input source:
- if (mInputPtr >= mInputEnd) {
- if (!loadMoreFromCurrent()) {
- throwParseError("Missing semicolon after reference for entity \"{0}\"", id, null);
- }
- }
- c = mInputBuffer[mInputPtr++];
- if (c != ';') {
- throwUnexpectedChar(c, "; expected a semi-colon after the reference for entity '"+id+"'");
- }
- return id;
- }
-
- /**
- * Note: does not check for number of colons, amongst other things.
- * Main idea is to skip through what superficially seems like a valid
- * id, nothing more. This is only done when really skipping through
- * something we do not care about at all: not even whether names/ids
- * would be valid (for example, when ignoring internal DTD subset).
- *
- * @return Length of skipped name.
- */
- protected int skipFullName(char c)
- throws XMLStreamException
- {
- if (!isNameStartChar(c)) {
- --mInputPtr;
- return 0;
- }
-
- /* After which there may be zero or more name chars
- * we have to consider
- */
- int count = 1;
- while (true) {
- c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(SUFFIX_EOF_EXP_NAME);
- if (c != ':' && !isNameChar(c)) {
- break;
- }
- ++count;
- }
- return count;
- }
-
- /**
- * Simple parsing method that parses system ids, which are generally
- * used in entities (from DOCTYPE declaration to internal/external
- * subsets).
- *
- * NOTE: returned String is not canonicalized, on assumption that
- * external ids may be longish, and are not shared all that often, as
- * they are generally just used for resolving paths, if anything.
- *
- * As per xml specs, the contents are actually normalized.
- *
- * NOTE: returned String is not canonicalized, on assumption that
- * external ids may be longish, and are not shared all that often, as
- * they are generally just used for resolving paths, if anything.
- *
- * !!! 05-Sep-2008, tatu: Current implementation is not optimal
- * either performance-wise, or from getting accurate Location
- * for decoding problems. But it works otherwise, and we need
- * to get Woodstox 4.0 out by the end of the year... so it'll
- * do, for now.
- *
- * @return Number of elements decoded (if any were decoded), or
- * -1 to indicate that no more values can be decoded.
- */
- public final int readElementAsArray(TypedArrayDecoder dec)
- throws XMLStreamException
- {
- int type = mCurrToken;
- // First things first: must be acceptable start state:
- if (((1 << type) & MASK_TYPED_ACCESS_ARRAY) == 0) {
- throwNotTextualOrElem(type);
- }
-
- // Are we just starting (START_ELEMENT)?
- if (type == START_ELEMENT) {
- // Empty? Not common, but can short-cut handling if occurs
- if (mStEmptyElem) {
- mStEmptyElem = false;
- mCurrToken = END_ELEMENT;
- return -1;
- }
- // Otherwise let's just find the first text segment
- while (true) {
- type = next();
- if (type == END_ELEMENT) {
- // Simple... no textul content
- return -1;
- }
- if (type == COMMENT || type == PROCESSING_INSTRUCTION) {
- continue;
- }
- if (type == CHARACTERS || type == CDATA) {
- break;
- }
- // otherwise just not legal (how about SPACE, unexpanded entities?)
- throw _constructUnexpectedInTyped(type);
- }
- }
-
- int count = 0;
- while (type != END_ELEMENT) {
- /* Ok then: we will have a valid textual type. Just need to
- * ensure current segment is completed. Plus, for current impl,
- * also need to coalesce to prevent artificial CDATA/text
- * boundary from splitting tokens
- */
- if (type == CHARACTERS || type == CDATA || type == SPACE) {
- if (mTokenState < TOKEN_FULL_SINGLE) {
- readCoalescedText(type, false);
- }
- } else if (type == COMMENT || type == PROCESSING_INSTRUCTION) {
- type = next();
- continue;
- } else {
- throw _constructUnexpectedInTyped(type);
- }
- count += mTextBuffer.decodeElements(dec, this);
- if (!dec.hasRoom()) {
- break;
- }
- type = next();
- }
-
- // If nothing was found, needs to be indicated via -1, not 0
- return (count > 0) ? count : -1;
- }
-
- /*
- ////////////////////////////////////////////////////////
- // TypedXMLStreamReader2 implementation, binary data
- ////////////////////////////////////////////////////////
- */
-
- public final int readElementAsBinary(byte[] resultBuffer, int offset, int maxLength)
- throws XMLStreamException
- {
- return readElementAsBinary(resultBuffer, offset, maxLength, Base64Variants.getDefaultVariant());
- }
-
- public int readElementAsBinary(byte[] resultBuffer, int offset, int maxLength, Base64Variant v)
- throws XMLStreamException
- {
- if (resultBuffer == null) {
- throw new IllegalArgumentException("resultBuffer is null");
- }
- if (offset < 0) {
- throw new IllegalArgumentException("Illegal offset ("+offset+"), must be [0, "+resultBuffer.length+"[");
- }
- if (maxLength < 1 || (offset + maxLength) > resultBuffer.length) {
- if (maxLength == 0) { // special case, allowed, but won't do anything
- return 0;
- }
- throw new IllegalArgumentException("Illegal maxLength ("+maxLength+"), has to be positive number, and offset+maxLength can not exceed"+resultBuffer.length);
- }
-
- final CharArrayBase64Decoder dec = _base64Decoder();
- int type = mCurrToken;
- // First things first: must be acceptable start state:
- if (((1 << type) & MASK_TYPED_ACCESS_BINARY) == 0) {
- if (type == END_ELEMENT) {
- // Minor complication: may have unflushed stuff (non-padded versions)
- if (!dec.hasData()) {
- return -1;
- }
- } else {
- throwNotTextualOrElem(type);
- }
- } else if (type == START_ELEMENT) { // just starting (START_ELEMENT)?
- if (mStEmptyElem) { // empty element? simple...
- mStEmptyElem = false;
- mCurrToken = END_ELEMENT;
- return -1;
- }
- // Otherwise let's just find the first text segment
- while (true) {
- type = next();
- if (type == END_ELEMENT) {
- // Simple... no textual content
- return -1;
- }
- if (type == COMMENT || type == PROCESSING_INSTRUCTION) {
- continue;
- }
- /* 12-Dec-2009, tatu: Important: in coalescing mode we may
- * have incomplete segment that needs to be completed
- */
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- _initBinaryChunks(v, dec, type, true);
- break;
- }
- }
-
- int totalCount = 0;
-
- main_loop:
- while (true) {
- // Ok, decode:
- int count;
- try {
- count = dec.decode(resultBuffer, offset, maxLength);
- } catch (IllegalArgumentException iae) {
- // !!! 26-Sep-2008, tatus: should try to figure out which char (etc) triggered problem to pass with typed exception
- throw _constructTypeException(iae.getMessage(), "");
- }
- offset += count;
- totalCount += count;
- maxLength -= count;
-
- /* And if we filled the buffer we are done. Or, an edge
- * case: reached END_ELEMENT (for non-padded variant)
- */
- if (maxLength < 1 || mCurrToken == END_ELEMENT) {
- break;
- }
- // Otherwise need to advance to the next event
- while (true) {
- type = next();
- if (type == COMMENT || type == PROCESSING_INSTRUCTION
- || type == SPACE) { // space is ignorable too
- continue;
- }
- if (type == END_ELEMENT) {
- /* Just need to verify we don't have partial stuff
- * (missing one to three characters of a full quartet
- * that encodes 1 - 3 bytes). Also: non-padding
- * variants can be in incomplete state, from which
- * data may need to be flushed...
- */
- int left = dec.endOfContent();
- if (left < 0) { // incomplete, error
- throw _constructTypeException("Incomplete base64 triplet at the end of decoded content", "");
- } else if (left > 0) { // 1 or 2 more bytes of data, loop some more
- continue main_loop;
- }
- // Otherwise, no more data, we are done
- break main_loop;
- }
- /* 12-Dec-2009, tatu: Important: in coalescing mode we may
- * have incomplete segment that needs to be completed
- */
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- _initBinaryChunks(v, dec, type, false);
- break;
- }
- }
-
- // If nothing was found, needs to be indicated via -1, not 0
- return (totalCount > 0) ? totalCount : -1;
- }
-
- private final void _initBinaryChunks(Base64Variant v, CharArrayBase64Decoder dec, int type, boolean isFirst)
- throws XMLStreamException
- {
- if (type == CHARACTERS) {
- if (mTokenState < mStTextThreshold) {
- mTokenState = readTextSecondary(MIN_BINARY_CHUNK, false)
- ? TOKEN_FULL_SINGLE : TOKEN_PARTIAL_SINGLE;
- }
- } else if (type == CDATA) {
- if (mTokenState < mStTextThreshold) {
- mTokenState = readCDataSecondary(MIN_BINARY_CHUNK)
- ? TOKEN_FULL_SINGLE : TOKEN_PARTIAL_SINGLE;
- }
- } else {
- throw _constructUnexpectedInTyped(type);
- }
- mTextBuffer.initBinaryChunks(v, dec, isFirst);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // TypedXMLStreamReader2 implementation, scalar attributes
- ///////////////////////////////////////////////////////////
- */
-
- public int getAttributeIndex(String namespaceURI, String localName)
- {
- // Note: cut'n pasted from "getAttributeInfo()"
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mElementStack.findAttributeIndex(namespaceURI, localName);
- }
-
- public boolean getAttributeAsBoolean(int index) throws XMLStreamException
- {
- ValueDecoderFactory.BooleanDecoder dec = _decoderFactory().getBooleanDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public int getAttributeAsInt(int index) throws XMLStreamException
- {
- ValueDecoderFactory.IntDecoder dec = _decoderFactory().getIntDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public long getAttributeAsLong(int index) throws XMLStreamException
- {
- ValueDecoderFactory.LongDecoder dec = _decoderFactory().getLongDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public float getAttributeAsFloat(int index) throws XMLStreamException
- {
- ValueDecoderFactory.FloatDecoder dec = _decoderFactory().getFloatDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public double getAttributeAsDouble(int index) throws XMLStreamException
- {
- ValueDecoderFactory.DoubleDecoder dec = _decoderFactory().getDoubleDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public BigInteger getAttributeAsInteger(int index) throws XMLStreamException
- {
- ValueDecoderFactory.IntegerDecoder dec = _decoderFactory().getIntegerDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public BigDecimal getAttributeAsDecimal(int index) throws XMLStreamException
- {
- ValueDecoderFactory.DecimalDecoder dec = _decoderFactory().getDecimalDecoder();
- getAttributeAs(index, dec);
- return dec.getValue();
- }
-
- public QName getAttributeAsQName(int index) throws XMLStreamException
- {
- ValueDecoderFactory.QNameDecoder dec = _decoderFactory().getQNameDecoder(getNamespaceContext());
- getAttributeAs(index, dec);
- return _verifyQName(dec.getValue());
- }
-
- public void getAttributeAs(int index, TypedValueDecoder tvd) throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- try {
- mAttrCollector.decodeValue(index, tvd);
- } catch (IllegalArgumentException iae) {
- throw _constructTypeException(iae, mAttrCollector.getValue(index));
- }
- }
-
- public int[] getAttributeAsIntArray(int index) throws XMLStreamException
- {
- ValueDecoderFactory.IntArrayDecoder dec = _decoderFactory().getIntArrayDecoder();
- getAttributeAsArray(index, dec);
- return dec.getValues();
- }
-
- public long[] getAttributeAsLongArray(int index) throws XMLStreamException
- {
- ValueDecoderFactory.LongArrayDecoder dec = _decoderFactory().getLongArrayDecoder();
- getAttributeAsArray(index, dec);
- return dec.getValues();
- }
-
- public float[] getAttributeAsFloatArray(int index) throws XMLStreamException
- {
- ValueDecoderFactory.FloatArrayDecoder dec = _decoderFactory().getFloatArrayDecoder();
- getAttributeAsArray(index, dec);
- return dec.getValues();
- }
-
- public double[] getAttributeAsDoubleArray(int index) throws XMLStreamException
- {
- ValueDecoderFactory.DoubleArrayDecoder dec = _decoderFactory().getDoubleArrayDecoder();
- getAttributeAsArray(index, dec);
- return dec.getValues();
- }
-
- /**
- * Method that allows reading contents of an attribute as an array
- * of whitespace-separate tokens, decoded using specified decoder.
- *
- * @return Number of tokens decoded, 0 if none found
- */
- public int getAttributeAsArray(int index, TypedArrayDecoder tad) throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.decodeValues(index, tad, this);
- }
-
- public byte[] getAttributeAsBinary(int index) throws XMLStreamException
- {
- return getAttributeAsBinary(index, Base64Variants.getDefaultVariant());
- }
-
- public byte[] getAttributeAsBinary(int index, Base64Variant v) throws XMLStreamException
- {
- return mAttrCollector.decodeBinary(index, v, _base64Decoder(), this);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal helper methods
- /////////////////////////////////////////////////////
- */
-
- /**
- * Method called to verify validity of the parsed QName element
- * or attribute value. At this point binding of a prefixed name
- * (if qname has a prefix) has been verified, and thereby prefix
- * also must be valid (since there must have been a preceding
- * declaration). But local name might still not be a legal
- * well-formed xml name, so let's verify that.
- */
- protected QName _verifyQName(QName n)
- throws TypedXMLStreamException
- {
- String ln = n.getLocalPart();
- int ix = WstxInputData.findIllegalNameChar(ln, mCfgNsEnabled, mXml11);
- if (ix >= 0) {
- String prefix = n.getPrefix();
- String pname = (prefix != null && prefix.length() > 0) ?
- (prefix + ":" +ln) : ln;
- throw _constructTypeException("Invalid local name \""+ln+"\" (character at #"+ix+" is invalid)", pname);
- }
- return n;
- }
-
- protected ValueDecoderFactory _decoderFactory()
- {
- if (_decoderFactory == null) {
- _decoderFactory = new ValueDecoderFactory();
- }
- return _decoderFactory;
- }
-
- protected CharArrayBase64Decoder _base64Decoder()
- {
- if (_base64Decoder == null) {
- _base64Decoder = new CharArrayBase64Decoder();
- }
- return _base64Decoder;
- }
-
- /**
- * Method called to handle value that has empty String
- * as representation. This will usually either lead to an
- * exception, or parsing to the default value for the
- * type in question (null for nullable types and so on).
- */
- private void _handleEmptyValue(TypedValueDecoder dec)
- throws XMLStreamException
- {
- try { // default action is to throw an exception
- dec.handleEmptyValue();
- } catch (IllegalArgumentException iae) {
- throw _constructTypeException(iae, "");
- }
- }
-
- /**
- * Method called to wrap or convert given conversion-fail exception
- * into a full {@link TypedXMLStreamException},
- *
- * @param iae Problem as reported by converter
- * @param lexicalValue Lexical value (element content, attribute value)
- * that could not be converted succesfully.
- */
- protected TypedXMLStreamException _constructTypeException(IllegalArgumentException iae, String lexicalValue)
- {
- return new TypedXMLStreamException(lexicalValue, iae.getMessage(), getStartLocation(), iae);
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ValidatingStreamReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ValidatingStreamReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/ValidatingStreamReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/ValidatingStreamReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,641 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.io.FileNotFoundException;
-import java.io.IOException;
-import java.net.URI;
-import java.net.URISyntaxException;
-import java.net.URL;
-import java.util.*;
-
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.XMLInputFactory2;
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.io.*;
-import com.ctc.wstx.dtd.DTDId;
-import com.ctc.wstx.dtd.DTDSubset;
-import com.ctc.wstx.dtd.DTDValidatorBase;
-import com.ctc.wstx.dtd.FullDTDReader;
-import com.ctc.wstx.util.URLUtil;
-
-/**
- * Implementation of {@link org.codehaus.stax2.XMLStreamReader2}
- * that builds on {@link TypedStreamReader} and adds full DTD-handling
- * including DTD validation
- *
- * @author Tatu Saloranta
- * @author Benson Margulies
- */
-public class ValidatingStreamReader
- extends TypedStreamReader
-{
- /*
- ///////////////////////////////////////////////////////////////////////
- // Constants for standard StAX properties:
- ///////////////////////////////////////////////////////////////////////
- */
-
- final static String STAX_PROP_ENTITIES = "javax.xml.stream.entities";
-
- final static String STAX_PROP_NOTATIONS = "javax.xml.stream.notations";
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Validation (DTD) information (entities, ...)
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // Note: some members that logically belong here, are actually
- // // // part of superclass
-
- /**
- * Combined DTD set, constructed from parsed internal and external
- * entities (which may have been set via override DTD functionality).
- */
- DTDValidationSchema mDTD = null;
-
- /**
- * Validating reader keeps of automatically created DTD-based
- * validator, since its handling may differ from that of application
- * managed validators.
- */
- XMLValidator mAutoDtdValidator = null;
-
- /**
- * Flag that indicates whether a DTD validator has been automatically
- * set (as per DOCTYPE declaration or override)
- */
- boolean mDtdValidatorSet = false;
-
- /**
- * Custom validation problem handler, if any.
- */
- protected ValidationProblemHandler mVldProbHandler = null;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Life-cycle (ctors)
- ///////////////////////////////////////////////////////////////////////
- */
-
- private ValidatingStreamReader(InputBootstrapper bs,
- BranchingReaderSource input, ReaderCreator owner,
- ReaderConfig cfg, InputElementStack elemStack,
- boolean forER)
- throws XMLStreamException
- {
- super(bs, input, owner, cfg, elemStack, forER);
- }
-
- /**
- * Factory method for constructing readers.
- *
- * @param owner "Owner" of this reader, factory that created the reader;
- * needed for returning updated symbol table information after parsing.
- * @param input Input source used to read the XML document.
- * @param cfg Object that contains reader configuration info.
- * @param bs Bootstrapper to use, for reading xml declaration etc.
- * @param forER True if this reader is to be (configured to be) used by
- * an event reader. Will cause some changes to default settings, as
- * required by contracts Woodstox XMLEventReader implementation has
- * (with respect to lazy parsing, short text segments etc)
- */
- public static ValidatingStreamReader createValidatingStreamReader
- (BranchingReaderSource input, ReaderCreator owner,
- ReaderConfig cfg, InputBootstrapper bs, boolean forER)
- throws XMLStreamException
- {
- ValidatingStreamReader sr = new ValidatingStreamReader
- (bs, input, owner, cfg, createElementStack(cfg), forER);
- return sr;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Public API, configuration
- ///////////////////////////////////////////////////////////////////////
- */
-
- public Object getProperty(String name)
- {
- // DTD-specific properties...
- if (name.equals(STAX_PROP_ENTITIES)) {
- safeEnsureFinishToken();
- if (mDTD == null || !(mDTD instanceof DTDSubset)) {
- return null;
- }
- List l = ((DTDSubset) mDTD).getGeneralEntityList();
- /* Let's make a copy, so that caller can not modify
- * DTD's internal list instance
- */
- return new ArrayList(l);
- }
- if (name.equals(STAX_PROP_NOTATIONS)) {
- safeEnsureFinishToken();
- if (mDTD == null || !(mDTD instanceof DTDSubset)) {
- return null;
- }
- /* Let's make a copy, so that caller can not modify
- * DTD's internal list instance
- */
- List l = ((DTDSubset) mDTD).getNotationList();
- return new ArrayList(l);
- }
- return super.getProperty(name);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XMLStreamReader2 (StAX2) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- // // // StAX2, per-reader configuration
-
- // no additional readable features
- //public Object getFeature(String name)
-
- public void setFeature(String name, Object value)
- {
- // Referring to DTD-related features?
- if (name.equals(FEATURE_DTD_OVERRIDE)) {
- /* !!! 06-Feb-2007, TSa: Null with 4.0 will actually mean
- * 'remove any overrides'; which is different from earlier
- * meaning (which was use a dummy/empty override).
- * Should we throw an exception, or warn, or something...?
- */
- if (value != null && !(value instanceof DTDValidationSchema)) {
- throw new IllegalArgumentException("Value to set for feature "+name+" not of type DTDValidationSchema");
- }
- mConfig.setProperty(XMLInputFactory2.P_DTD_OVERRIDE, (DTDValidationSchema) value);
- } else {
- super.setFeature(name, value);
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // DTDInfo implementation (StAX 2)
- ///////////////////////////////////////////////////////////////////////
- */
-
- public Object getProcessedDTD() {
- return getProcessedDTDSchema();
- }
-
- public DTDValidationSchema getProcessedDTDSchema() {
- DTDValidationSchema dtd = mConfig.getDTDOverride();
- if (dtd == null) {
- dtd = mDTD;
- }
- return mDTD;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Stax2 validation
- ///////////////////////////////////////////////////////////////////////
- */
-
- // @Override
- public XMLValidator validateAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- return mElementStack.validateAgainst(schema);
- }
-
- // @Override
- public XMLValidator stopValidatingAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- return mElementStack.stopValidatingAgainst(schema);
- }
-
- // @Override
- public XMLValidator stopValidatingAgainst(XMLValidator validator)
- throws XMLStreamException
- {
- return mElementStack.stopValidatingAgainst(validator);
- }
-
- // @Override
- public ValidationProblemHandler setValidationProblemHandler(ValidationProblemHandler h)
- {
- ValidationProblemHandler oldH = mVldProbHandler;
- mVldProbHandler = h;
- return oldH;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Private methods, DOCTYPE handling
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * This method gets called to handle remainder of DOCTYPE declaration,
- * essentially the optional internal subset. Internal subset, if such
- * exists, is always read, but whether its contents are added to the
- * read buffer depend on passed-in argument.
- *
- * NOTE: Since this method overrides the default implementation, make
- * sure you do NOT change the method signature.
- *
- * @param copyContents If true, will copy contents of the internal
- * subset of DOCTYPE declaration
- * in the text buffer (in addition to parsing it for actual use); if
- * false, will only do parsing.
- */
- protected void finishDTD(boolean copyContents)
- throws XMLStreamException
- {
- if (!hasConfigFlags(CFG_SUPPORT_DTD)) {
- super.finishDTD(copyContents);
- return;
- }
-
- /* We know there are no spaces, as this char was read and pushed
- * back earlier...
- */
- char c = getNextChar(SUFFIX_IN_DTD);
- DTDSubset intSubset = null;
-
- /* Do we have an internal subset? Note that we have earlier checked
- * that it has to be either '[' or closing '>'.
- */
- if (c == '[') {
- // Do we need to copy the contents of int. subset in the buffer?
- if (copyContents) {
- ((BranchingReaderSource) mInput).startBranch(mTextBuffer, mInputPtr, mNormalizeLFs);
- }
-
- try {
- intSubset = FullDTDReader.readInternalSubset(this, mInput, mConfig,
- hasConfigFlags(CFG_VALIDATE_AGAINST_DTD),
- mDocXmlVersion);
- } finally {
- /* Let's close branching in any and every case (may allow
- * graceful recovery in error cases in future
- */
- if (copyContents) {
- /* Need to "push back" ']' got in the succesful case
- * (that's -1 part below);
- * in error case it'll just be whatever last char was.
- */
- ((BranchingReaderSource) mInput).endBranch(mInputPtr-1);
- }
- }
-
- // And then we need closing '>'
- c = getNextCharAfterWS(SUFFIX_IN_DTD_INTERNAL);
- }
-
- if (c != '>') {
- throwUnexpectedChar(c, "; expected '>' to finish DOCTYPE declaration.");
- }
-
- /* But, then, we also may need to read the external subset, if
- * one was defined:
- */
- /* 19-Sep-2004, TSa: That does not need to be done, however, if
- * there's a DTD override set.
- */
-
- mDTD = mConfig.getDTDOverride();
- if (mDTD != null) {
- // We have earlier override that's already parsed
- } else { // Nope, no override
- DTDSubset extSubset = null;
-
- /* 05-Mar-2006, TSa: If standalone was specified as "yes", we
- * should not rely on any external declarations, so shouldn't
- * we really just skip the external subset?
- */
- /* Alas: SAX (Xerces) still tries to read it... should we
- * do the Right Thing, or follow the leader? For now, let's
- * just follow the wrong example.
- */
-
- //if (mDocStandalone != DOC_STANDALONE_YES) {
- if (true) {
- if (mDtdPublicId != null || mDtdSystemId != null) {
- extSubset = findDtdExtSubset(mDtdPublicId, mDtdSystemId, intSubset);
- }
- }
- if (intSubset == null) {
- mDTD = extSubset;
- } else if (extSubset == null) {
- mDTD = intSubset;
- } else {
- mDTD = intSubset.combineWithExternalSubset(this, extSubset);
- }
- }
-
-
- if (mDTD == null) { // only if specifically overridden not to have any
- mGeneralEntities = null;
- } else {
- if (mDTD instanceof DTDSubset) {
- mGeneralEntities = ((DTDSubset) mDTD).getGeneralEntityMap();
- } else {
- /* Also, let's warn if using non-native DTD implementation,
- * since entities and notations can not be accessed
- */
- _reportProblem(mConfig.getXMLReporter(), ErrorConsts.WT_DT_DECL,
- "Value to set for feature "+FEATURE_DTD_OVERRIDE+" not a native Woodstox DTD implementation (but "+mDTD.getClass()+"): can not access full entity or notation information", null);
- }
- /* 16-Jan-2006, TSa: Actually, we have both fully-validating mode,
- * and non-validating-but-DTD-aware mode. In latter case, we'll
- * still need to add a validator, but just to get type info
- * and to add attribute default values if necessary.
- */
- mAutoDtdValidator = mDTD.createValidator(/*(ValidationContext)*/ mElementStack);
- mDtdValidatorSet = true; // so we won't get nags
- NsDefaultProvider nsDefs = null;
- if (mAutoDtdValidator instanceof DTDValidatorBase) {
- DTDValidatorBase dtdv = (DTDValidatorBase) mAutoDtdValidator;
- dtdv.setAttrValueNormalization(true);
- // Do we have any attribute defaults for 'xmlns' or 'xmlns:*'?
- if (dtdv.hasNsDefaults()) {
- nsDefs = dtdv;
- }
- }
- mElementStack.setAutomaticDTDValidator(mAutoDtdValidator, nsDefs);
- }
- }
-
- /**
- * If there is an error handler established, call it.
- */
- // @Override
- public void reportValidationProblem(XMLValidationProblem prob)
- throws XMLStreamException
- {
- if (mVldProbHandler != null) {
- // Fix for [WSTX-209]
- mVldProbHandler.reportProblem(prob);
- } else {
- super.reportValidationProblem(prob);
- }
- }
-
- /**
- * Method called right before handling the root element, by the base
- * class. This allows for some initialization and checks to be done
- * (not including ones that need access to actual element name)
- */
- protected void initValidation()
- throws XMLStreamException
- {
- if (hasConfigFlags(CFG_VALIDATE_AGAINST_DTD)
- && !mDtdValidatorSet) {
- /* It's ok to miss it, but it may not be what caller wants. Either
- * way, let's pass the info and continue
- */
- reportProblem(null, ErrorConsts.WT_DT_DECL, ErrorConsts.W_MISSING_DTD, null, null);
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Private methods, external subset access
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Method called by
- * Currently supported configuration options fall into two categories. First,
- * all properties from {@link XMLInputFactory} (such as, say,
- * {@link XMLInputFactory#IS_NAMESPACE_AWARE}) are at least recognized, and
- * most are supported. Second, there are additional properties, defined in
- * constant class {@link WstxInputProperties}, that are supported.
- * See {@link WstxInputProperties} for further explanation of these 'custom'
- * properties.
- *
- * @author Tatu Saloranta
- */
-public class WstxInputFactory
- extends XMLInputFactory2
- implements ReaderCreator,
- InputConfigFlags
-{
- /**
- * Let's limit max size to 3/4 of 16k, since this corresponds
- * to 64k main hash index. This should not be too low, but could
- * perhaps be further lowered?
- */
- final static int MAX_SYMBOL_TABLE_SIZE = 12000;
-
- /**
- * Number of generations should not matter as much as raw
- * size... but let's still cap it at some number. 500 generations
- * seems reasonable for flushing (note: does not count uses
- * where no new symbols were added).
- */
- final static int MAX_SYMBOL_TABLE_GENERATIONS = 500;
-
- /*
- ///////////////////////////////////////////////////////////
- // Actual storage of configuration settings
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Current configurations for this factory
- */
- protected final ReaderConfig mConfig;
-
- // // // Stax - mandated objects:
-
- protected XMLEventAllocator mAllocator = null;
-
- // // // Other configuration objects:
-
- protected SimpleCache mDTDCache = null;
-
- /*
- ///////////////////////////////////////////////////////////
- // Objects shared by actual parsers
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * 'Root' symbol table, used for creating actual symbol table instances,
- * but never as is.
- */
- final static SymbolTable mRootSymbols = DefaultXmlSymbolTable.getInstance();
- static {
- /* By default, let's enable intern()ing of names (element, attribute,
- * prefixes) added to symbol table. This is likely to make some
- * access (attr by QName) and comparison of element/attr names
- * more efficient. Although it will add some overhead on adding
- * new symbols to symbol table that should be rather negligible.
- *
- * Also note that always doing intern()ing allows for more efficient
- * access during DTD validation.
- */
- mRootSymbols.setInternStrings(true);
- }
-
- /**
- * Actual current 'parent' symbol table; concrete instances will be
- * created from this instance using
- * Note: parser is only to call this method, if passed-in symbol
- * table was modified, ie new entry/ies were added in addition to
- * whatever was in root table.
- */
- public synchronized void updateSymbolTable(SymbolTable t)
- {
- SymbolTable curr = mSymbols;
- /* Let's only add if table was direct descendant; this prevents
- * siblings from keeping overwriting settings (multiple direct
- * children have additional symbols added)
- */
- if (t.isDirectChildOf(curr)) {
- /* 07-Apr-2006, TSa: Actually, since huge symbol tables
- * might become hindrance more than benefit (either in
- * pathological cases with random names; or with very
- * long running processes), let's actually limit both
- * number of generations, and, more imporantly, maximum
- * size of the symbol table
- */
- if (t.size() > MAX_SYMBOL_TABLE_SIZE ||
- t.version() > MAX_SYMBOL_TABLE_GENERATIONS) {
- // If so, we'll reset from bare defaults
- mSymbols = mRootSymbols;
-//System.err.println("DEBUG: !!!! XXXXX Symbol Table Flush: size: "+t.size()+"; version: "+t.version());
- } else {
- mSymbols.mergeChild(t);
-//System.err.println("Debug: new symbol table: size: "+t.size()+"; version: "+t.version());
- }
- }
-//else System.err.println("Debug: skipping symbol table update");
- }
-
- public synchronized void addCachedDTD(DTDId id, DTDSubset extSubset)
- {
- if (mDTDCache == null) {
- mDTDCache = new SimpleCache(mConfig.getDtdCacheSize());
- }
- mDTDCache.add(id, extSubset);
- }
-
- /*
- /////////////////////////////////////////////////////
- // Stax, XMLInputFactory; factory methods
- /////////////////////////////////////////////////////
- */
-
- // // // Filtered reader factory methods
-
- public XMLEventReader createFilteredReader(XMLEventReader reader, EventFilter filter)
- {
- return new Stax2FilteredEventReader(Stax2EventReaderAdapter.wrapIfNecessary(reader), filter);
- }
-
- public XMLStreamReader createFilteredReader(XMLStreamReader reader, StreamFilter filter)
- throws XMLStreamException
- {
- Stax2FilteredStreamReader fr = new Stax2FilteredStreamReader(reader, filter);
- /* [WSTX-111] As per Stax 1.0 TCK, apparently the filtered
- * reader is expected to be automatically forwarded to the first
- * acceptable event. This is different from the way RI works, but
- * since specs don't say anything about filtered readers, let's
- * consider TCK to be "more formal" for now, and implement that
- * behavior.
- */
- if (!filter.accept(fr)) { // START_DOCUMENT ok?
- // Ok, nope, this should do the trick:
- fr.next();
- }
- return fr;
- }
-
- // // // Event reader factory methods
-
- public XMLEventReader createXMLEventReader(InputStream in)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return new WstxEventReader(createEventAllocator(),
- createSR(null, in, null, true, false));
- }
-
- public XMLEventReader createXMLEventReader(InputStream in, String enc)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return new WstxEventReader(createEventAllocator(),
- createSR(null, in, enc, true, false));
- }
-
- public XMLEventReader createXMLEventReader(Reader r)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return new WstxEventReader(createEventAllocator(),
- createSR(null, r, true, false));
- }
-
- public XMLEventReader createXMLEventReader(javax.xml.transform.Source source)
- throws XMLStreamException
- {
- return new WstxEventReader(createEventAllocator(),
- createSR(source, true));
- }
-
- public XMLEventReader createXMLEventReader(String systemId, InputStream in)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return new WstxEventReader(createEventAllocator(),
- createSR(systemId, in, null, true, false));
- }
-
- public XMLEventReader createXMLEventReader(String systemId, Reader r)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the reader
- return new WstxEventReader(createEventAllocator(),
- createSR(systemId, r, true, false));
- }
-
- public XMLEventReader createXMLEventReader(XMLStreamReader sr)
- throws XMLStreamException
- {
- return new WstxEventReader(createEventAllocator(), Stax2ReaderAdapter.wrapIfNecessary(sr));
- }
-
- // // // Stream reader factory methods
-
- public XMLStreamReader createXMLStreamReader(InputStream in)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return createSR(null, in, null, false, false);
- }
-
- public XMLStreamReader createXMLStreamReader(InputStream in, String enc)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return createSR(null, in, enc, false, false);
- }
-
- public XMLStreamReader createXMLStreamReader(Reader r)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the reader
- return createSR(null, r, false, false);
- }
-
- public XMLStreamReader createXMLStreamReader(javax.xml.transform.Source src)
- throws XMLStreamException
- {
- // false -> not for event. No definition for auto-close; called method will decide
- return createSR(src, false);
- }
-
- public XMLStreamReader createXMLStreamReader(String systemId, InputStream in)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the input stream
- return createSR(systemId, in, null, false, false);
- }
-
- public XMLStreamReader createXMLStreamReader(String systemId, Reader r)
- throws XMLStreamException
- {
- // false for auto-close, since caller has access to the Reader
- return createSR(systemId, r, false, false);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Stax, XMLInputFactory; generic accessors/mutators
- ///////////////////////////////////////////////////////////
- */
-
- public Object getProperty(String name)
- {
- Object ob = mConfig.getProperty(name);
-
- if (ob == null) {
- if (name.equals(XMLInputFactory.ALLOCATOR)) {
- // Event allocator not available via J2ME subset...
- return getEventAllocator();
- }
- }
- return ob;
- }
-
- public void setProperty(String propName, Object value)
- {
- if (!mConfig.setProperty(propName, value)) {
- if (XMLInputFactory.ALLOCATOR.equals(propName)) {
- setEventAllocator((XMLEventAllocator) value);
- }
- }
- }
-
- public XMLEventAllocator getEventAllocator() {
- return mAllocator;
- }
-
- public XMLReporter getXMLReporter() {
- return mConfig.getXMLReporter();
- }
-
- public XMLResolver getXMLResolver() {
- return mConfig.getXMLResolver();
- }
-
- public boolean isPropertySupported(String name) {
- return mConfig.isPropertySupported(name);
- }
-
- public void setEventAllocator(XMLEventAllocator allocator) {
- mAllocator = allocator;
- }
-
- public void setXMLReporter(XMLReporter r) {
- mConfig.setXMLReporter(r);
- }
-
- /**
- * Note: it's preferable to use Wstx-specific
- * {@link ReaderConfig#setEntityResolver}
- * instead, if possible, since this just wraps passed in resolver.
- */
- public void setXMLResolver(XMLResolver r)
- {
- mConfig.setXMLResolver(r);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Stax2 implementation
- ///////////////////////////////////////////////////////////
- */
-
- // // // Stax2, additional factory methods:
-
- public XMLEventReader2 createXMLEventReader(URL src)
- throws XMLStreamException
- {
- /* true for auto-close, since caller has no access to the underlying
- * input stream created from the URL
- */
- return new WstxEventReader(createEventAllocator(),
- createSR(createPrivateConfig(), src, true, true));
- }
-
- public XMLEventReader2 createXMLEventReader(File f)
- throws XMLStreamException
- {
- /* true for auto-close, since caller has no access to the underlying
- * input stream created from the File
- */
- return new WstxEventReader(createEventAllocator(),
- createSR(f, true, true));
- }
-
- public XMLStreamReader2 createXMLStreamReader(URL src)
- throws XMLStreamException
- {
- /* true for auto-close, since caller has no access to the underlying
- * input stream created from the URL
- */
- return createSR(createPrivateConfig(), src, false, true);
- }
-
- /**
- * Convenience factory method that allows for parsing a document
- * stored in the specified file.
- */
- public XMLStreamReader2 createXMLStreamReader(File f)
- throws XMLStreamException
- {
- /* true for auto-close, since caller has no access to the underlying
- * input stream created from the File
- */
- return createSR(f, false, true);
- }
-
- // // // Stax2 "Profile" mutators
-
- public void configureForXmlConformance()
- {
- mConfig.configureForXmlConformance();
- }
-
- public void configureForConvenience()
- {
- mConfig.configureForConvenience();
- }
-
- public void configureForSpeed()
- {
- mConfig.configureForSpeed();
- }
-
- public void configureForLowMemUsage()
- {
- mConfig.configureForLowMemUsage();
- }
-
- public void configureForRoundTripping()
- {
- mConfig.configureForRoundTripping();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Woodstox-specific configuration access
- ///////////////////////////////////////////////////////////
- */
-
- public ReaderConfig getConfig() {
- return mConfig;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods:
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Bottleneck method used for creating ALL full stream reader instances
- * (via other createSR() methods and directly)
- *
- * @param forER True, if the reader is being constructed to be used
- * by an event reader; false if it is not (or the purpose is not known)
- * @param autoCloseInput Whether the underlying input source should be
- * actually closed when encountering EOF, or when
- * Note: defined as public method because it needs to be called by
- * SAX implementation.
- *
- * @param systemId System id used for this reader (if any)
- * @param bs Bootstrapper to use for creating actual underlying
- * physical reader
- * @param forER Flag to indicate whether it will be used via
- * Event API (will affect some configuration settings), true if it
- * will be, false if not (or not known)
- * @param autoCloseInput Whether the underlying input source should be
- * actually closed when encountering EOF, or when
- * Note: only public so that other woodstox components outside of
- * this package can access it.
- */
- public ReaderConfig createPrivateConfig()
- {
- return mConfig.createNonShared(mSymbols.makeChild());
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/stax/WstxOutputFactory.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/stax/WstxOutputFactory.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/stax/WstxOutputFactory.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/stax/WstxOutputFactory.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,372 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.stax;
-
-import java.io.IOException;
-import java.io.OutputStream;
-import java.io.OutputStreamWriter;
-import java.io.Writer;
-
-import javax.xml.stream.XMLEventWriter;
-import javax.xml.stream.XMLOutputFactory;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.transform.Result;
-import javax.xml.transform.dom.DOMResult;
-import javax.xml.transform.sax.SAXResult;
-import javax.xml.transform.stream.StreamResult;
-
-import org.codehaus.stax2.XMLOutputFactory2;
-import org.codehaus.stax2.XMLStreamWriter2;
-import org.codehaus.stax2.io.Stax2Result;
-import org.codehaus.stax2.ri.Stax2EventWriterImpl;
-import org.codehaus.stax2.ri.Stax2WriterAdapter;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.api.WstxOutputProperties;
-import com.ctc.wstx.cfg.OutputConfigFlags;
-import com.ctc.wstx.dom.WstxDOMWrappingWriter;
-import com.ctc.wstx.exc.WstxIOException;
-import com.ctc.wstx.io.CharsetNames;
-import com.ctc.wstx.io.UTF8Writer;
-import com.ctc.wstx.sw.AsciiXmlWriter;
-import com.ctc.wstx.sw.BufferingXmlWriter;
-import com.ctc.wstx.sw.ISOLatin1XmlWriter;
-import com.ctc.wstx.sw.NonNsStreamWriter;
-import com.ctc.wstx.sw.RepairingNsStreamWriter;
-import com.ctc.wstx.sw.SimpleNsStreamWriter;
-import com.ctc.wstx.sw.XmlWriter;
-import com.ctc.wstx.util.URLUtil;
-
-/**
- * Implementation of {@link XMLOutputFactory} for Wstx.
- *
- * TODO:
- *
- * Note: Root namespace context works best if automatic prefix
- * creation ("namespace/prefix repairing" in StAX lingo) is enabled.
- */
- public void setNamespaceContext(NamespaceContext ctxt)
- throws XMLStreamException
- {
- // This is only allowed before root element output:
- if (mState != STATE_PROLOG) {
- throwOutputError("Called setNamespaceContext() after having already output root element.");
- }
-
- mRootNsContext = ctxt;
- mCurrElem.setRootNsContext(ctxt);
- }
-
- public void setPrefix(String prefix, String uri)
- throws XMLStreamException
- {
- if (prefix == null) {
- throw new NullPointerException("Can not pass null 'prefix' value");
- }
- // Are we actually trying to set the default namespace?
- if (prefix.length() == 0) {
- setDefaultNamespace(uri);
- return;
- }
- if (uri == null) {
- throw new NullPointerException("Can not pass null 'uri' value");
- }
-
- /* 25-Sep-2004, TSa: Let's check that "xml" and "xmlns" are not
- * (re-)defined to any other value, nor that value they
- * are bound to are bound to other prefixes.
- */
- /* 01-Apr-2005, TSa: And let's not leave it optional: such
- * bindings should never succeed.
- */
- // ... perhaps it really should be optional though?
- {
- if (prefix.equals(sPrefixXml)) { // prefix "xml"
- if (!uri.equals(XMLConstants.XML_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XML, uri);
- }
- } else if (prefix.equals(sPrefixXmlns)) { // prefix "xmlns"
- if (!uri.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XMLNS, uri);
- }
- } else {
- // Neither of prefixes.. but how about URIs?
- if (uri.equals(XMLConstants.XML_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XML_URI, prefix);
- } else if (uri.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- throwOutputError(ErrorConsts.ERR_NS_REDECL_XMLNS_URI, prefix);
- }
- }
-
- /* 05-Feb-2005, TSa: Also, as per namespace specs; the 'empty'
- * namespace URI can not be bound as a non-default namespace
- * (ie. for any actual prefix)
- */
- /* 04-Feb-2005, TSa: Namespaces 1.1 does allow this, though,
- * so for xml 1.1 documents we need to allow it
- */
- if (!mXml11) {
- if (uri.length() == 0) {
- throwOutputError(ErrorConsts.ERR_NS_EMPTY);
- }
- }
- }
-
- doSetPrefix(prefix, uri);
- }
-
- /**
- * It's assumed calling this method implies caller just wants to add
- * an attribute that does not belong to any namespace; as such no
- * namespace checking or prefix generation is needed.
- */
- public void writeAttribute(String localName, String value)
- throws XMLStreamException
- {
- // No need to set mAnyOutput, nor close the element
- if (!mStartElementOpen && mCheckStructure) {
- reportNwfStructure(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- doWriteAttr(localName, null, null, value);
- }
-
- public abstract void writeAttribute(String nsURI, String localName, String value)
- throws XMLStreamException;
-
- public abstract void writeAttribute(String prefix, String nsURI,
- String localName, String value)
- throws XMLStreamException;
-
- /**
- *
- * Note: It is assumed caller just wants the element to belong to whatever
- * is the current default namespace.
- */
- public void writeEmptyElement(String localName)
- throws XMLStreamException
- {
- checkStartElement(localName, null);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
- mEmptyElement = true;
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, localName);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(localName);
- }
- doWriteStartTag(localName);
-
- }
-
- public void writeEmptyElement(String nsURI, String localName)
- throws XMLStreamException
- {
- writeStartOrEmpty(localName, nsURI);
- mEmptyElement = true;
- }
-
- public void writeEmptyElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- writeStartOrEmpty(prefix, localName, nsURI);
- mEmptyElement = true;
- }
-
- public void writeEndElement()
- throws XMLStreamException
- {
- doWriteEndTag(null, mCfgAutomaticEmptyElems);
- }
-
- /**
- * This method is assumed to just use default namespace (if any),
- * and no further checks should be done.
- */
- public void writeStartElement(String localName)
- throws XMLStreamException
- {
- checkStartElement(localName, null);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
- mEmptyElement = false;
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, localName);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(localName);
- }
-
- doWriteStartTag(localName);
- }
-
- public void writeStartElement(String nsURI, String localName)
- throws XMLStreamException
- {
- writeStartOrEmpty(localName, nsURI);
- mEmptyElement = false;
- }
-
- public void writeStartElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- writeStartOrEmpty(prefix, localName, nsURI);
- mEmptyElement = false;
- }
-
- protected void writeTypedAttribute(String prefix, String nsURI, String localName,
- AsciiValueEncoder enc)
- throws XMLStreamException
- {
- if (!mStartElementOpen) {
- throwOutputError(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- if (mCheckAttrs) { // still need to ensure no duplicate attrs?
- mCurrElem.checkAttrWrite(nsURI, localName);
- }
- try {
- if (mValidator == null) {
- if (prefix == null || prefix.length() == 0) {
- mWriter.writeTypedAttribute(localName, enc);
- } else {
- mWriter.writeTypedAttribute(prefix, localName, enc);
- }
- } else {
- mWriter.writeTypedAttribute
- (prefix, localName, nsURI, enc, mValidator, getCopyBuffer());
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Remaining XMLStreamWriter2 methods (StAX2)
- ////////////////////////////////////////////////////
- */
-
- /**
- * Similar to {@link #writeEndElement}, but never allows implicit
- * creation of empty elements.
- */
- public void writeFullEndElement()
- throws XMLStreamException
- {
- doWriteEndTag(null, false);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Remaining ValidationContext methods (StAX2)
- ////////////////////////////////////////////////////
- */
-
- public QName getCurrentElementName()
- {
- return mCurrElem.getName();
- }
-
- public String getNamespaceURI(String prefix) {
- return mCurrElem.getNamespaceURI(prefix);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Implementations for base-class defined abstract methods
- //////////////////////////////////////////////////////////
- */
-
- /**
- * Method called by {@link javax.xml.stream.XMLEventWriter} implementation
- * (instead of the version
- * that takes no argument), so that we can verify it does match the
- * start element, if necessary
- */
- public void writeEndElement(QName name)
- throws XMLStreamException
- {
- doWriteEndTag(mCheckStructure ? name : null, mCfgAutomaticEmptyElems);
- }
-
- /**
- * Method called to close an open start element, when another
- * main-level element (not namespace declaration or attribute)
- * is being output; except for end element which is handled differently.
- *
- * @param emptyElem If true, the element being closed is an empty
- * element; if false, a separate stand-alone start element.
- */
- protected void closeStartElement(boolean emptyElem)
- throws XMLStreamException
- {
- mStartElementOpen = false;
-
- try {
- if (emptyElem) {
- mWriter.writeStartTagEmptyEnd();
- } else {
- mWriter.writeStartTagEnd();
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- if (mValidator != null) {
- mVldContent = mValidator.validateElementAndAttributes();
- }
-
- // Need bit more special handling for empty elements...
- if (emptyElem) {
- SimpleOutputElement curr = mCurrElem;
- mCurrElem = curr.getParent();
- if (mCurrElem.isRoot()) { // Did we close the root? (isRoot() returns true for the virtual "document node")
- mState = STATE_EPILOG;
- }
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd
- (curr.getLocalName(), curr.getNamespaceURI(), curr.getPrefix());
- }
- if (mPoolSize < MAX_POOL_SIZE) {
- curr.addToPool(mOutputElemPool);
- mOutputElemPool = curr;
- ++mPoolSize;
- }
- }
- }
-
- protected String getTopElementDesc() {
- return mCurrElem.getNameDesc();
- }
-
- /*
- ////////////////////////////////////////////////////
- // Package methods sub-classes may also need
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method that is called to ensure that we can start writing an
- * element, both from structural point of view, and from syntactic
- * (close previously open start element, if any).
- */
- protected void checkStartElement(String localName, String prefix)
- throws XMLStreamException
- {
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- } else if (mState == STATE_PROLOG) {
- verifyRootElement(localName, prefix);
- } else if (mState == STATE_EPILOG) {
- if (mCheckStructure) {
- String name = (prefix == null || prefix.length() == 0) ?
- localName : (prefix + ":" + localName);
- reportNwfStructure(ErrorConsts.WERR_PROLOG_SECOND_ROOT, name);
- }
- /* When outputting a fragment, need to reset this to the
- * tree. No point in trying to verify the root element?
- */
- mState = STATE_TREE;
- }
- }
-
- protected final void doWriteAttr(String localName, String nsURI, String prefix,
- String value)
- throws XMLStreamException
- {
- if (mCheckAttrs) { // still need to ensure no duplicate attrs?
- mCurrElem.checkAttrWrite(nsURI, localName);
- }
-
- if (mValidator != null) {
- /* No need to get it normalized... even if validator does normalize
- * it, we don't use that for anything
- */
- mValidator.validateAttribute(localName, nsURI, prefix, value);
- }
- try {
- int vlen = value.length();
- // Worthwhile to make a local copy?
- if (vlen >= ATTR_MIN_ARRAYCOPY) {
- char[] buf = mCopyBuffer;
- if (buf == null) {
- mCopyBuffer = buf = mConfig.allocMediumCBuffer(DEFAULT_COPYBUFFER_LEN);
- }
- /* Ok, and in unlikely case of attribute values longer than
- * buffer... for now, let's just skip those case
- */
- if (vlen <= buf.length) {
- value.getChars(0, vlen, buf, 0);
- if (prefix != null && prefix.length() > 0) {
- mWriter.writeAttribute(prefix, localName, buf, 0, vlen);
- } else {
- mWriter.writeAttribute(localName, buf, 0, vlen);
- }
- return;
- }
- }
- if (prefix != null && prefix.length() > 0) {
- mWriter.writeAttribute(prefix, localName, value);
- } else {
- mWriter.writeAttribute(localName, value);
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- protected final void doWriteAttr(String localName, String nsURI, String prefix,
- char[] buf, int start, int len)
- throws XMLStreamException
- {
- if (mCheckAttrs) { // still need to ensure no duplicate attrs?
- mCurrElem.checkAttrWrite(nsURI, localName);
- }
-
- if (mValidator != null) {
- /* No need to get it normalized... even if validator does normalize
- * it, we don't use that for anything
- */
- mValidator.validateAttribute(localName, nsURI, prefix, buf, start, len);
- }
- try {
- if (prefix != null && prefix.length() > 0) {
- mWriter.writeAttribute(prefix, localName, buf, start, len);
- } else {
- mWriter.writeAttribute(localName, buf, start, len);
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- protected void doWriteNamespace(String prefix, String nsURI)
- throws XMLStreamException
- {
- try {
- int vlen = nsURI.length();
- // Worthwhile to make a local copy?
- if (vlen >= ATTR_MIN_ARRAYCOPY) {
- char[] buf = mCopyBuffer;
- if (buf == null) {
- mCopyBuffer = buf = mConfig.allocMediumCBuffer(DEFAULT_COPYBUFFER_LEN);
- }
- // Let's not bother with too long, though
- if (vlen <= buf.length) {
- nsURI.getChars(0, vlen, buf, 0);
- mWriter.writeAttribute(XMLConstants.XMLNS_ATTRIBUTE, prefix, buf, 0, vlen);
- return;
- }
- }
- mWriter.writeAttribute(XMLConstants.XMLNS_ATTRIBUTE, prefix, nsURI);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- protected void doWriteDefaultNs(String nsURI)
- throws XMLStreamException
- {
- try {
- int vlen = (nsURI == null) ? 0 : nsURI.length();
- // Worthwhile to make a local copy?
- if (vlen >= ATTR_MIN_ARRAYCOPY) {
- char[] buf = mCopyBuffer;
- if (buf == null) {
- mCopyBuffer = buf = mConfig.allocMediumCBuffer(DEFAULT_COPYBUFFER_LEN);
- }
- // Let's not bother with too long, though
- if (vlen <= buf.length) {
- nsURI.getChars(0, vlen, buf, 0);
- mWriter.writeAttribute(XMLConstants.XMLNS_ATTRIBUTE, buf, 0, vlen);
- return;
- }
- }
- mWriter.writeAttribute(XMLConstants.XMLNS_ATTRIBUTE, nsURI);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- protected final void doWriteStartTag(String localName)
- throws XMLStreamException
- {
- mAnyOutput = true;
- mStartElementOpen = true;
- try {
- mWriter.writeStartTagStart(localName);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- protected final void doWriteStartTag(String prefix, String localName)
- throws XMLStreamException
- {
- mAnyOutput = true;
- mStartElementOpen = true;
- try {
- boolean hasPrefix = (prefix != null && prefix.length() > 0);
- if (hasPrefix) {
- mWriter.writeStartTagStart(prefix, localName);
- } else {
- mWriter.writeStartTagStart(localName);
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- /**
- *
- * @param expName Name that the closing element should have; null
- * if whatever is in stack should be used
- * @param allowEmpty If true, is allowed to create the empty element
- * if the closing element was truly empty; if false, has to write
- * the full empty element no matter what
- */
- protected void doWriteEndTag(QName expName, boolean allowEmpty)
- throws XMLStreamException
- {
- /* First of all, do we need to close up an earlier empty element?
- * (open start element that was not created via call to
- * writeEmptyElement gets handled later on)
- */
- if (mStartElementOpen && mEmptyElement) {
- mEmptyElement = false;
- closeStartElement(true);
- }
-
- // Better have something to close... (to figure out what to close)
- if (mState != STATE_TREE) {
- // Have to always throw exception... don't necessarily know the name
- reportNwfStructure("No open start element, when trying to write end element");
- }
-
- SimpleOutputElement thisElem = mCurrElem;
- String prefix = thisElem.getPrefix();
- String localName = thisElem.getLocalName();
- String nsURI = thisElem.getNamespaceURI();
-
- // Ok, and then let's pop that element from the stack
- mCurrElem = thisElem.getParent();
- // Need to return the instance to pool?
- if (mPoolSize < MAX_POOL_SIZE) {
- thisElem.addToPool(mOutputElemPool);
- mOutputElemPool = thisElem;
- ++mPoolSize;
- }
-
- if (mCheckStructure) {
- if (expName != null) {
- // Let's only check the local name, for now...
- if (!localName.equals(expName.getLocalPart())) {
- /* Only gets called when trying to output an XMLEvent... in
- * which case names can actually be compared
- */
- reportNwfStructure("Mismatching close element local name, '"+localName+"'; expected '"+expName.getLocalPart()+"'.");
- }
- }
- }
-
- /* Now, do we have an unfinished start element (created via
- * writeStartElement() earlier)?
- */
- if (mStartElementOpen) {
- /* Can't/shouldn't call closeStartElement, but need to do same
- * processing. Thus, this is almost identical to closeStartElement:
- */
- if (mValidator != null) {
- /* Note: return value is not of much use, since the
- * element will be closed right away...
- */
- mVldContent = mValidator.validateElementAndAttributes();
- }
- mStartElementOpen = false;
- try {
- //If an EmptyElementHandler is provided use it to determine if allowEmpty is set
- if (mEmptyElementHandler != null) {
- allowEmpty = mEmptyElementHandler.allowEmptyElement(prefix, localName, nsURI, allowEmpty);
- }
- // We could write an empty element, implicitly?
- if (allowEmpty) {
- mWriter.writeStartTagEmptyEnd();
- if (mCurrElem.isRoot()) {
- mState = STATE_EPILOG;
- }
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd(localName, nsURI, prefix);
- }
- return;
- }
- // Nah, need to close open elem, and then output close elem
- mWriter.writeStartTagEnd();
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- try {
- mWriter.writeEndTag(prefix, localName);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- if (mCurrElem.isRoot()) {
- mState = STATE_EPILOG;
- }
-
- // Ok, time to validate...
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd(localName, nsURI, prefix);
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // More abstract methods for sub-classes to implement
- ////////////////////////////////////////////////////
- */
-
- public abstract void doSetPrefix(String prefix, String uri)
- throws XMLStreamException;
-
- public abstract void writeDefaultNamespace(String nsURI)
- throws XMLStreamException;
-
- public abstract void writeNamespace(String prefix, String nsURI)
- throws XMLStreamException;
-
- public abstract void writeStartElement(StartElement elem)
- throws XMLStreamException;
-
- protected abstract void writeStartOrEmpty(String localName, String nsURI)
- throws XMLStreamException;
-
- protected abstract void writeStartOrEmpty(String prefix, String localName, String nsURI)
- throws XMLStreamException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BaseStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BaseStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BaseStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BaseStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1746 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.text.MessageFormat;
-
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLReporter;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-
-// unfortunate dependencies to StAX events:
-import javax.xml.stream.events.Characters;
-import javax.xml.stream.events.StartElement;
-
-import org.codehaus.stax2.DTDInfo;
-import org.codehaus.stax2.XMLStreamLocation2;
-import org.codehaus.stax2.XMLStreamReader2;
-import org.codehaus.stax2.ri.Stax2WriterImpl;
-import org.codehaus.stax2.validation.*;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.api.WstxInputProperties;
-import com.ctc.wstx.api.WstxOutputProperties;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.OutputConfigFlags;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.exc.*;
-import com.ctc.wstx.io.WstxInputLocation;
-import com.ctc.wstx.sr.StreamReaderImpl;
-import com.ctc.wstx.sr.AttributeCollector;
-import com.ctc.wstx.sr.InputElementStack;
-import com.ctc.wstx.util.DataUtil;
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Base class for {@link XMLStreamWriter} implementations Woodstox has.
- * Contains partial stream writer implementation, plus utility methods
- * shared by concrete implementation classes. Main reason for such
- * abstract base class is to allow other parts of Woodstox core to refer
- * to any of stream writer implementations in general way.
- */
-public abstract class BaseStreamWriter
- extends Stax2WriterImpl
- implements ValidationContext, OutputConfigFlags
-{
- protected final static int STATE_PROLOG = 1;
- protected final static int STATE_TREE = 2;
- protected final static int STATE_EPILOG = 3;
-
- protected final static char CHAR_SPACE = ' ';
-
- /**
- * This constant defines minimum length of a String, for which it
- * is beneficial to do an intermediate copy (using String.getChars()),
- * and iterate over intermediate array, instead of iterating using
- * String.charAt(). Former is generally faster for longer Strings, but
- * has some overhead for shorter Strings. Tests indicate that the
- * threshold is somewhere between 8 and 16 characters, at least on
- * x86 platform.
- */
- protected final static int MIN_ARRAYCOPY = 12;
-
- protected final static int ATTR_MIN_ARRAYCOPY = 12;
-
- protected final static int DEFAULT_COPYBUFFER_LEN = 512;
-
- /*
- ///////////////////////////////////////////////////////////
- // Output objects
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Actual physical writer to output serialized XML content to
- */
- protected final XmlWriter mWriter;
-
- /**
- * Intermediate buffer into which characters of a String can be
- * copied, in cases where such a copy followed by array access
- * is faster than calling
- * Note: current implementation of this method is not perfect; it
- * may be possible it can return true even if we are only using a DTD
- * to get some limited info, without validating?
- */
- protected boolean isValidating() {
- return (mValidator != null);
- }
-
- /**
- * Convenience method needed by {@link javax.xml.stream.XMLEventWriter}
- * implementation, to use when
- * writing a start element, and possibly its attributes and namespace
- * declarations.
- */
- public abstract void writeStartElement(StartElement elem)
- throws XMLStreamException;
-
- /**
- * Method called by {@link javax.xml.stream.XMLEventWriter}
- * (instead of the version
- * that takes no argument), so that we can verify it does match the
- * start element if necessary.
- */
- public abstract void writeEndElement(QName name)
- throws XMLStreamException;
-
- /**
- * Method called by {@link javax.xml.stream.XMLEventWriter}
- * (instead of more generic
- * text output methods), so that we can verify (if necessary) that
- * this character output type is legal in this context. Specifically,
- * it's not acceptable to add non-whitespace content outside root
- * element (in prolog/epilog).
- *
- * Note: cut'n pasted from the main
- * One design goal for this class is to avoid unnecessary buffering: since
- * there will be another Writer doing the actual encoding, amount of
- * buffering needed should still be limited. To this end, a threshold is
- * used to define what's the threshold of writes that we do want to
- * coalesce, ie. buffer. Writes bigger than this should in general proceed
- * without buffering.
- */
-public final class BufferingXmlWriter
- extends XmlWriter
- implements XMLStreamConstants
-{
- /**
- * Let's use a typical default to have a compromise between large
- * enough chunks to output, and minimizing memory overhead.
- * Compared to encoding writers, buffer size can be bit smaller
- * since there's one more level of processing (at encoding), which
- * may use bigger buffering.
- */
- final static int DEFAULT_BUFFER_SIZE = 1000;
-
- /**
- * Choosing threshold for 'small size' is a compromise between
- * excessive buffering (high small size), and too many fragmented
- * calls to the underlying writer (low small size). Let's just
- * use about 1/4 of the full buffer size.
- */
- final static int DEFAULT_SMALL_SIZE = 256;
-
- /**
- * Highest valued character that may need to be encoded (minus charset
- * encoding requirements) when writing attribute values.
- */
- protected final static int HIGHEST_ENCODABLE_ATTR_CHAR = (int)'<';
-
- /**
- * Highest valued character that may need to be encoded (minus charset
- * encoding requirements) when writing attribute values.
- */
- protected final static int HIGHEST_ENCODABLE_TEXT_CHAR = (int)'>';
-
- /*
- ////////////////////////////////////////////////
- // Output state, buffering
- ////////////////////////////////////////////////
- */
-
- /**
- * Actual Writer to use for outputting buffered data as appropriate.
- */
- protected final Writer mOut;
-
- protected char[] mOutputBuffer;
-
- /**
- * This is the threshold used to check what is considered a "small"
- * write; small writes will be buffered until resulting size will
- * be above the threshold.
- */
- protected final int mSmallWriteSize;
-
- protected int mOutputPtr;
-
- protected int mOutputBufLen;
-
- /**
- * Actual physical stream that the writer is using, if known.
- * Not used for actual output, only needed so that calling
- * application may (try to) figure out the original
- * source.
- */
- protected final OutputStream mUnderlyingStream;
-
- /*
- ////////////////////////////////////////////////
- // Encoding/escaping configuration
- ////////////////////////////////////////////////
- */
-
- /**
- * First Unicode character (one with lowest value) after (and including)
- * which character entities have to be used. For
- */
- private final int mEncHighChar;
-
- /**
- * Character that is considered to be the enclosing quote character;
- * for XML either single or double quote.
- */
- final char mEncQuoteChar;
-
- /**
- * Entity String to use for escaping the quote character.
- */
- final String mEncQuoteEntity;
-
- /*
- ////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////
- */
-
- /**
- * @param outs Underlying OutputStream that the writer
- * (
- * Implementation notes:
- *
- * Parts of surrogate handling are implemented here in the base class:
- * storage for the first part of a split surrogate (only possible when
- * character content is output split in multiple calls) is within
- * base class. Also, simple checks for unmatched surrogate pairs are
- * in
- * Regarding surrogate pair handling: most of the checks are in the base
- * class, and here we only need to worry about
- * TreeSet is used mostly because clearing it up is faster than
- * clearing up HashSet, and the only access is done by
- * adding entries and see if an value was already set.
- */
- TreeSet mAttrNames;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle (ctors)
- ////////////////////////////////////////////////////
- */
-
- public NonNsStreamWriter(XmlWriter xw, String enc, WriterConfig cfg)
- {
- super(xw, enc, cfg);
- mElements = new StringVector(32);
- }
-
- /*
- ////////////////////////////////////////////////////
- // XMLStreamWriter API
- ////////////////////////////////////////////////////
- */
-
- public NamespaceContext getNamespaceContext() {
- return EmptyNamespaceContext.getInstance();
- }
-
- public String getPrefix(String uri) {
- return null;
- }
-
- public void setDefaultNamespace(String uri)
- throws XMLStreamException
- {
- reportIllegalArg("Can not set default namespace for non-namespace writer.");
- }
-
- public void setNamespaceContext(NamespaceContext context)
- {
- reportIllegalArg("Can not set NamespaceContext for non-namespace writer.");
- }
-
- public void setPrefix(String prefix, String uri)
- throws XMLStreamException
- {
- reportIllegalArg("Can not set namespace prefix for non-namespace writer.");
- }
-
- public void writeAttribute(String localName, String value)
- throws XMLStreamException
- {
- // No need to set mAnyOutput, nor close the element
- if (!mStartElementOpen && mCheckStructure) {
- reportNwfStructure(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- if (mCheckAttrs) {
- /* 11-Dec-2005, TSa: Should use a more efficient Set/Map value
- * for this in future.
- */
- if (mAttrNames == null) {
- mAttrNames = new TreeSet();
- }
- if (!mAttrNames.add(localName)) {
- reportNwfAttr("Trying to write attribute '"+localName+"' twice");
- }
- }
- if (mValidator != null) {
- /* No need to get it normalized... even if validator does normalize
- * it, we don't use that for anything
- */
- mValidator.validateAttribute(localName, XmlConsts.ATTR_NO_NS_URI, XmlConsts.ATTR_NO_PREFIX, value);
- }
-
- try {
- mWriter.writeAttribute(localName, value);
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
- }
-
- public void writeAttribute(String nsURI, String localName, String value)
- throws XMLStreamException
- {
- writeAttribute(localName, value);
- }
-
- public void writeAttribute(String prefix, String nsURI,
- String localName, String value)
- throws XMLStreamException
- {
- writeAttribute(localName, value);
- }
-
- public void writeDefaultNamespace(String nsURI)
- throws XMLStreamException
- {
- reportIllegalMethod("Can not call writeDefaultNamespace namespaces with non-namespace writer.");
- }
-
- public void writeEmptyElement(String localName)
- throws XMLStreamException
- {
- doWriteStartElement(localName);
- mEmptyElement = true;
- }
-
- public void writeEmptyElement(String nsURI, String localName)
- throws XMLStreamException
- {
- writeEmptyElement(localName);
- }
-
- public void writeEmptyElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- writeEmptyElement(localName);
- }
-
- public void writeEndElement()
- throws XMLStreamException
- {
- doWriteEndTag(null, mCfgAutomaticEmptyElems);
- }
-
- public void writeNamespace(String prefix, String nsURI)
- throws XMLStreamException
- {
- reportIllegalMethod("Can not set write namespaces with non-namespace writer.");
- }
-
- public void writeStartElement(String localName)
- throws XMLStreamException
- {
- doWriteStartElement(localName);
- mEmptyElement = false;
- }
-
- public void writeStartElement(String nsURI, String localName)
- throws XMLStreamException
- {
- writeStartElement(localName);
- }
-
- public void writeStartElement(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- writeStartElement(localName);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Remaining XMLStreamWriter2 methods (StAX2)
- ////////////////////////////////////////////////////
- */
-
- /**
- * Similar to {@link #writeEndElement}, but never allows implicit
- * creation of empty elements.
- */
- public void writeFullEndElement()
- throws XMLStreamException
- {
- doWriteEndTag(null, false);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Remaining ValidationContext methods (StAX2)
- ////////////////////////////////////////////////////
- */
-
- public QName getCurrentElementName() {
- if (mElements.isEmpty()) {
- return null;
- }
- return new QName(mElements.getLastString());
- }
-
- public String getNamespaceURI(String prefix) {
- return null;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Package methods:
- ////////////////////////////////////////////////////
- */
-
- public void writeStartElement(StartElement elem)
- throws XMLStreamException
- {
- QName name = elem.getName();
- writeStartElement(name.getLocalPart());
- Iterator it = elem.getAttributes();
- while (it.hasNext()) {
- Attribute attr = (Attribute) it.next();
- name = attr.getName();
- writeAttribute(name.getLocalPart(), attr.getValue());
- }
- }
-
- /**
- * Method called by {@link javax.xml.stream.XMLEventWriter} implementation
- * (instead of the version
- * that takes no argument), so that we can verify it does match the
- * start element, if necessary
- */
- public void writeEndElement(QName name)
- throws XMLStreamException
- {
- doWriteEndTag(mCheckStructure ? name.getLocalPart() : null,
- mCfgAutomaticEmptyElems);
- }
-
- protected void writeTypedAttribute(String prefix, String nsURI, String localName,
- AsciiValueEncoder enc)
- throws XMLStreamException
- {
- // note: mostly copied from the other writeAttribute() method..
- if (!mStartElementOpen && mCheckStructure) {
- reportNwfStructure(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- if (mCheckAttrs) { // doh. Not good, need to construct non-transient value...
- if (mAttrNames == null) {
- mAttrNames = new TreeSet();
- }
- if (!mAttrNames.add(localName)) {
- reportNwfAttr("Trying to write attribute '"+localName+"' twice");
- }
- }
-
- try {
- if (mValidator == null) {
- mWriter.writeTypedAttribute(localName, enc);
- } else {
- mWriter.writeTypedAttribute(null, localName, null, enc, mValidator, getCopyBuffer());
- }
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
- }
-
- /**
- * Method called to close an open start element, when another
- * main-level element (not namespace declaration or
- * attribute) is being output; except for end element which is
- * handled differently.
- */
- protected void closeStartElement(boolean emptyElem)
- throws XMLStreamException
- {
- mStartElementOpen = false;
- if (mAttrNames != null) {
- mAttrNames.clear();
- }
-
- try {
- if (emptyElem) {
- mWriter.writeStartTagEmptyEnd();
- } else {
- mWriter.writeStartTagEnd();
- }
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
-
- if (mValidator != null) {
- mVldContent = mValidator.validateElementAndAttributes();
- }
-
- // Need bit more special handling for empty elements...
- if (emptyElem) {
- String localName = mElements.removeLast();
- if (mElements.isEmpty()) {
- mState = STATE_EPILOG;
- }
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
- }
- }
-
- /**
- * Element copier method implementation suitable to be used with
- * non-namespace-aware writers. The only special thing here is that
- * the copier can convert namespace declarations to equivalent
- * attribute writes.
- */
- public void copyStartElement(InputElementStack elemStack,
- AttributeCollector attrCollector)
- throws IOException, XMLStreamException
- {
- String ln = elemStack.getLocalName();
- boolean nsAware = elemStack.isNamespaceAware();
-
- /* First, since we are not to output namespace stuff as is,
- * we just need to copy the element:
- */
- if (nsAware) { // but reader is ns-aware? Need to add prefix?
- String prefix = elemStack.getPrefix();
- if (prefix != null && prefix.length() > 0) { // yup
- ln = prefix + ":" + ln;
- }
- }
- writeStartElement(ln);
-
- /* However, if there are any namespace declarations, we probably
- * better output them just as 'normal' attributes:
- */
- if (nsAware) {
- int nsCount = elemStack.getCurrentNsCount();
- if (nsCount > 0) {
- for (int i = 0; i < nsCount; ++i) {
- String prefix = elemStack.getLocalNsPrefix(i);
- if (prefix == null || prefix.length() == 0) { // default NS decl
- prefix = XMLConstants.XML_NS_PREFIX;
- } else {
- prefix = "xmlns:"+prefix;
- }
- writeAttribute(prefix, elemStack.getLocalNsURI(i));
- }
- }
- }
-
- /* And then let's just output attributes, if any (whether to copy
- * implicit, aka "default" attributes, is configurable)
- */
- int attrCount = mCfgCopyDefaultAttrs ?
- attrCollector.getCount() :
- attrCollector.getSpecifiedCount();
-
- if (attrCount > 0) {
- for (int i = 0; i < attrCount; ++i) {
- attrCollector.writeAttribute(i, mWriter);
- }
- }
- }
-
- protected String getTopElementDesc()
- {
- return mElements.isEmpty() ? "#root" : mElements.getLastString();
- }
-
- public String validateQNamePrefix(QName name)
- {
- // Can either strip prefix out, or return as is
- return name.getPrefix();
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////
- */
-
- private void doWriteStartElement(String localName)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- } else if (mState == STATE_PROLOG) {
- // 20-Dec-2005, TSa: Does this match DOCTYPE declaration?
- verifyRootElement(localName, null);
- } else if (mState == STATE_EPILOG) {
- if (mCheckStructure) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_SECOND_ROOT, localName);
- }
- // Outputting fragment? Better reset to tree, then...
- mState = STATE_TREE;
- }
-
- /* Note: need not check for CONTENT_ALLOW_NONE here, since the
- * validator should handle this particular case...
- */
- /*if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) { // EMPTY content
- reportInvalidContent(START_ELEMENT);
- }*/
- if (mValidator != null) {
- mValidator.validateElementStart(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
-
- mStartElementOpen = true;
- mElements.addString(localName);
- try {
- mWriter.writeStartTagStart(localName);
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
- }
-
- /**
- *
- * Note: Caller has to do actual removal of the element from element
- * stack, before calling this method.
- *
- * @param expName Name that the closing element should have; null
- * if whatever is in stack should be used
- * @param allowEmpty If true, is allowed to create the empty element
- * if the closing element was truly empty; if false, has to write
- * the full empty element no matter what
- */
- private void doWriteEndTag(String expName, boolean allowEmpty)
- throws XMLStreamException
- {
- /* First of all, do we need to close up an earlier empty element?
- * (open start element that was not created via call to
- * writeEmptyElement gets handled later on)
- */
- if (mStartElementOpen && mEmptyElement) {
- mEmptyElement = false;
- // note: this method guarantees proper updates to validation
- closeStartElement(true);
- }
-
- // Better have something to close... (to figure out what to close)
- if (mState != STATE_TREE) {
- // Have to throw an exception always, don't know elem name
- reportNwfStructure("No open start element, when trying to write end element");
- }
-
- /* Now, do we have an unfinished start element (created via
- * writeStartElement() earlier)?
- */
- String localName = mElements.removeLast();
- if (mCheckStructure) {
- if (expName != null && !localName.equals(expName)) {
- /* Only gets called when trying to output an XMLEvent... in
- * which case names can actually be compared
- */
- reportNwfStructure("Mismatching close element name, '"+localName+"'; expected '"+expName+"'.");
- }
- }
-
- /* Can't yet validate, since we have two paths; one for empty
- * elements, another for non-empty...
- */
-
- // Got a half output start element to close?
- if (mStartElementOpen) {
- /* Can't/shouldn't call closeStartElement, but need to do same
- * processing. Thus, this is almost identical to closeStartElement:
- */
- if (mValidator != null) {
- /* Note: return value is not of much use, since the
- * element will be closed right away...
- */
- mVldContent = mValidator.validateElementAndAttributes();
- }
- mStartElementOpen = false;
- if (mAttrNames != null) {
- mAttrNames.clear();
- }
- try {
- // We could write an empty element, implicitly?
- if (allowEmpty) {
- mWriter.writeStartTagEmptyEnd();
- if (mElements.isEmpty()) {
- mState = STATE_EPILOG;
- }
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
- return;
- }
- // Nah, need to close open elem, and then output close elem
- mWriter.writeStartTagEnd();
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
- }
-
- try {
- mWriter.writeEndTag(localName);
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
-
- if (mElements.isEmpty()) {
- mState = STATE_EPILOG;
- }
-
- // Ok, time to validate...
- if (mValidator != null) {
- mVldContent = mValidator.validateElementEnd(localName, XmlConsts.ELEM_NO_NS_URI, XmlConsts.ELEM_NO_PREFIX);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/OutputElementBase.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/OutputElementBase.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/OutputElementBase.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/OutputElementBase.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,373 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2005 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.util.*;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.ri.EmptyIterator;
-
-import com.ctc.wstx.util.BijectiveNsMap;
-
-/**
- * Class that encapsulates information about a specific element in virtual
- * output stack for namespace-aware writers.
- * It provides support for URI-to-prefix mappings as well as namespace
- * mapping generation.
- *
- * One noteworthy feature of the class is that it is designed to allow
- * "short-term recycling", ie. instances can be reused within context
- * of a simple document output. While reuse/recycling of such lightweight
- * object is often useless or even counter productive, here it may
- * be worth using, due to simplicity of the scheme (basically using
- * a very simple free-elements linked list).
- */
-public abstract class OutputElementBase
- implements NamespaceContext
-{
- public final static int PREFIX_UNBOUND = 0;
- public final static int PREFIX_OK = 1;
- public final static int PREFIX_MISBOUND = 2;
-
- final static String sXmlNsPrefix = XMLConstants.XML_NS_PREFIX;
- final static String sXmlNsURI = XMLConstants.XML_NS_URI;
-
- /*
- ////////////////////////////////////////////
- // Namespace binding/mapping information
- ////////////////////////////////////////////
- */
-
- /**
- * Namespace context end application may have supplied, and that
- * (if given) should be used to augment explicitly defined bindings.
- */
- protected NamespaceContext mRootNsContext;
-
- protected String mDefaultNsURI;
-
- /**
- * Mapping of namespace prefixes to URIs and back.
- */
- protected BijectiveNsMap mNsMapping;
-
- /**
- * True, if {@link #mNsMapping} is a shared copy from the parent;
- * false if a local copy was created (which happens when namespaces
- * get bound etc).
- */
- protected boolean mNsMapShared;
-
- /*
- ////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////
- */
-
- /**
- * Constructor for the virtual root element
- */
- protected OutputElementBase()
- {
- mNsMapping = null;
- mNsMapShared = false;
- mDefaultNsURI = "";
- mRootNsContext = null;
- }
-
- protected OutputElementBase(OutputElementBase parent, BijectiveNsMap ns)
- {
- mNsMapping = ns;
- mNsMapShared = (ns != null);
- mDefaultNsURI = parent.mDefaultNsURI;
- mRootNsContext = parent.mRootNsContext;
- }
-
- /**
- * Method called to reuse a pooled instance.
- */
- protected void relink(OutputElementBase parent)
- {
- mNsMapping = parent.mNsMapping;
- mNsMapShared = (mNsMapping != null);
- mDefaultNsURI = parent.mDefaultNsURI;
- mRootNsContext = parent.mRootNsContext;
- }
-
- protected abstract void setRootNsContext(NamespaceContext ctxt);
-
- /*
- ////////////////////////////////////////////
- // Public API, accessors
- ////////////////////////////////////////////
- */
-
- public abstract boolean isRoot();
-
- /**
- * @return String presentation of the fully-qualified name, in
- * "prefix:localName" format (no URI). Useful for error and
- * debugging messages.
- */
- public abstract String getNameDesc();
-
- public final String getDefaultNsUri() {
- return mDefaultNsURI;
- }
-
- /*
- ////////////////////////////////////////////
- // Public API, ns binding, checking
- ////////////////////////////////////////////
- */
-
- /**
- * Method similar to {@link #getPrefix}, but one that will not accept
- * the default namespace, only an explicit one. Usually used when
- * trying to find a prefix for attributes.
- */
- public final String getExplicitPrefix(String uri)
- {
- if (mNsMapping != null) {
- String prefix = mNsMapping.findPrefixByUri(uri);
- if (prefix != null) {
- return prefix;
- }
- }
- if (mRootNsContext != null) {
- String prefix = mRootNsContext.getPrefix(uri);
- if (prefix != null) {
- // Hmmh... still can't use the default NS:
- if (prefix.length() > 0) {
- return prefix;
- }
- // ... should we try to find an explicit one?
- }
- }
- return null;
- }
-
- /**
- * Method that verifies that passed-in prefix indeed maps to the specified
- * namespace URI; and depending on how it goes returns a status for
- * caller.
- *
- * @param isElement If true, rules for the default NS are those of elements
- * (ie. empty prefix can map to non-default namespace); if false,
- * rules are those of attributes (only non-default prefix can map to
- * a non-default namespace).
- *
- * @return PREFIX_OK, if passed-in prefix matches matched-in namespace URI
- * in current scope; PREFIX_UNBOUND if it's not bound to anything,
- * and PREFIX_MISBOUND if it's bound to another URI.
- *
- * @throws XMLStreamException True if default (no) prefix is allowed to
- * match a non-default URI (elements); false if not (attributes)
- */
- public final int isPrefixValid(String prefix, String nsURI,
- boolean isElement)
- throws XMLStreamException
- {
- // Hmmm.... caller shouldn't really pass null.
- if (nsURI == null) {
- nsURI = "";
- }
-
- /* First thing is to see if specified prefix is bound to a namespace;
- * and if so, verify it matches with data passed in:
- */
-
- // Checking default namespace?
- if (prefix == null || prefix.length() == 0) {
- if (isElement) {
- // It's fine for elements only if the URI actually matches:
- if (nsURI == mDefaultNsURI || nsURI.equals(mDefaultNsURI)) {
- return PREFIX_OK;
- }
- } else {
- /* Attributes never use the default namespace: "no prefix"
- * can only mean "no namespace"
- */
- if (nsURI.length() == 0) {
- return PREFIX_OK;
- }
- }
- return PREFIX_MISBOUND;
- }
-
- /* Need to handle 'xml' prefix and its associated
- * URI; they are always declared by default
- */
- if (prefix.equals(sXmlNsPrefix)) {
- // Should we thoroughly verify its namespace matches...?
- // 01-Apr-2005, TSa: Yes, let's always check this
- if (!nsURI.equals(sXmlNsURI)) {
- throwOutputError("Namespace prefix '"+sXmlNsPrefix
- +"' can not be bound to non-default namespace ('"+nsURI+"'); has to be the default '"
- +sXmlNsURI+"'");
- }
- return PREFIX_OK;
- }
-
- // Nope checking some other namespace
- String act;
-
- if (mNsMapping != null) {
- act = mNsMapping.findUriByPrefix(prefix);
- } else {
- act = null;
- }
-
- if (act == null && mRootNsContext != null) {
- act = mRootNsContext.getNamespaceURI(prefix);
- }
-
- // Not (yet) bound...
- if (act == null) {
- return PREFIX_UNBOUND;
- }
-
- return (act == nsURI || act.equals(nsURI)) ?
- PREFIX_OK : PREFIX_MISBOUND;
- }
-
- /*
- ////////////////////////////////////////////
- // Public API, mutators
- ////////////////////////////////////////////
- */
-
- public abstract void setDefaultNsUri(String uri);
-
- public final String generateMapping(String prefixBase, String uri, int[] seqArr)
- {
- // This is mostly cut'n pasted from addPrefix()...
- if (mNsMapping == null) {
- // Didn't have a mapping yet? Need to create one...
- mNsMapping = BijectiveNsMap.createEmpty();
- } else if (mNsMapShared) {
- /* Was shared with parent(s)? Need to create a derivative, to
- * allow for nesting/scoping of new prefix
- */
- mNsMapping = mNsMapping.createChild();
- mNsMapShared = false;
- }
- return mNsMapping.addGeneratedMapping(prefixBase, mRootNsContext,
- uri, seqArr);
- }
-
- public final void addPrefix(String prefix, String uri)
- {
- if (mNsMapping == null) {
- // Didn't have a mapping yet? Need to create one...
- mNsMapping = BijectiveNsMap.createEmpty();
- } else if (mNsMapShared) {
- /* Was shared with parent(s)? Need to create a derivative, to
- * allow for nesting/scoping of new prefix
- */
- mNsMapping = mNsMapping.createChild();
- mNsMapShared = false;
- }
- mNsMapping.addMapping(prefix, uri);
- }
-
- /*
- //////////////////////////////////////////////////
- // NamespaceContext implementation
- //////////////////////////////////////////////////
- */
-
- public final String getNamespaceURI(String prefix)
- {
- if (prefix.length() == 0) { //default NS
- return mDefaultNsURI;
- }
- if (mNsMapping != null) {
- String uri = mNsMapping.findUriByPrefix(prefix);
- if (uri != null) {
- return uri;
- }
- }
- return (mRootNsContext != null) ?
- mRootNsContext.getNamespaceURI(prefix) : null;
- }
-
- public final String getPrefix(String uri)
- {
- if (mDefaultNsURI.equals(uri)) {
- return "";
- }
- if (mNsMapping != null) {
- String prefix = mNsMapping.findPrefixByUri(uri);
- if (prefix != null) {
- return prefix;
- }
- }
- return (mRootNsContext != null) ?
- mRootNsContext.getPrefix(uri) : null;
- }
-
- public final Iterator getPrefixes(String uri)
- {
- List l = null;
-
- if (mDefaultNsURI.equals(uri)) {
- l = new ArrayList();
- l.add("");
- }
- if (mNsMapping != null) {
- l = mNsMapping.getPrefixesBoundToUri(uri, l);
- }
- // How about the root namespace context? (if any)
- /* Note: it's quite difficult to properly resolve masking, when
- * combining these things (not impossible, just tricky); for now
- * let's do best effort without worrying about masking:
- */
- if (mRootNsContext != null) {
- Iterator it = mRootNsContext.getPrefixes(uri);
- while (it.hasNext()) {
- String prefix = (String) it.next();
- if (prefix.length() == 0) { // default NS already checked
- continue;
- }
- // slow check... but what the heck
- if (l == null) {
- l = new ArrayList();
- } else if (l.contains(prefix)) { // double-defined...
- continue;
- }
- l.add(prefix);
- }
- }
- return (l == null) ? EmptyIterator.getInstance() :
- l.iterator();
- }
-
- /*
- ////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////
- */
-
- protected final void throwOutputError(String msg)
- throws XMLStreamException
- {
- throw new XMLStreamException(msg);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/RepairingNsStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/RepairingNsStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/RepairingNsStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/RepairingNsStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,646 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE,
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, softwar
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.util.HashMap;
-import java.util.Iterator;
-import java.util.Map;
-
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.events.Attribute;
-import javax.xml.stream.events.StartElement;
-
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.sr.AttributeCollector;
-import com.ctc.wstx.sr.InputElementStack;
-
-/**
- * Namespace-aware implementation of {@link XMLStreamWriter}, that does
- * namespace repairing, ie resolves possible conflicts between prefixes
- * (add new bindings as necessary), as well as automatically creates
- * namespace declarations as necessary.
- */
-public final class RepairingNsStreamWriter
- extends BaseNsStreamWriter
-{
- /*
- ///////////////////////////////////////////////////////////
- // Configuration (options, features)
- ///////////////////////////////////////////////////////////
- */
-
- // // // Additional specific config flags base class doesn't have
-
- protected final String mAutomaticNsPrefix;
-
- /*
- ///////////////////////////////////////////////////////////
- // Additional state
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Sequence number used for generating dynamic namespace prefixes.
- * Array used as a wrapper to allow for easy sharing of the sequence
- * number.
- */
- protected int[] mAutoNsSeq = null;
-
- protected String mSuggestedDefNs = null;
-
- /**
- * Map that contains URI-to-prefix entries that point out suggested
- * prefixes for URIs. These are populated by calls to
- * {@link #setPrefix}, and they are only used as hints for binding;
- * if there are conflicts, repairing writer can just use some other
- * prefix.
- */
- protected HashMap mSuggestedPrefixes = null;
-
- /*
- ///////////////////////////////////////////////////////////
- // Life-cycle (ctors)
- ///////////////////////////////////////////////////////////
- */
-
- public RepairingNsStreamWriter(XmlWriter xw, String enc, WriterConfig cfg)
- {
- super(xw, enc, cfg, true);
- mAutomaticNsPrefix = cfg.getAutomaticNsPrefix();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter API
- ///////////////////////////////////////////////////////////
- */
-
- //public NamespaceContext getNamespaceContext()
- //public void setNamespaceContext(NamespaceContext context)
- //public String getPrefix(String uri)
- //public void setPrefix(String prefix, String uri)
- //public void setDefaultNamespace(String uri)
-
- //public void writeAttribute(String localName, String value)
-
- public void writeAttribute(String nsURI, String localName, String value)
- throws XMLStreamException
- {
- // No need to set mAnyOutput, nor close the element
- if (!mStartElementOpen) {
- throwOutputError(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- doWriteAttr(localName, nsURI,
- findOrCreateAttrPrefix(null, nsURI, mCurrElem),
- value);
- }
-
- public void writeAttribute(String prefix, String nsURI,
- String localName, String value)
- throws XMLStreamException
- {
- if (!mStartElementOpen) {
- throwOutputError(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
-
- doWriteAttr(localName, nsURI, findOrCreateAttrPrefix(prefix, nsURI, mCurrElem),
- value);
- }
-
- public void writeDefaultNamespace(String nsURI)
- throws XMLStreamException
- {
- /* 01-Sep-2006, TSa: The use case for calling this method is that
- * of caller may wanting to 'suggest' that
- * such a namespace should indeed be bound at this level. This
- * may be necessary for canonicalization, or for minimizing number
- * of binding declarations (all children need the ns, but root
- * itself not).
- */
- if (!mStartElementOpen) {
- throwOutputError(ERR_NSDECL_WRONG_STATE);
- }
- /* ... We have one complication though: if the current element
- * uses default namespace, can not change it (attributes don't
- * matter -- they never use the default namespace, but either don't
- * belong to a namespace, or belong to one using explicit prefix)
- */
- String prefix = mCurrElem.getPrefix();
- if (prefix != null && prefix.length() > 0) { // ok, can change it
- mCurrElem.setDefaultNsUri(nsURI);
- doWriteDefaultNs(nsURI);
- }
- }
-
- //public void writeEmptyElement(String localName) throws XMLStreamException
-
- public void writeNamespace(String prefix, String nsURI)
- throws XMLStreamException
- {
- /* (see discussion in 'writeDefaultNamespace()' for details on
- * if and how this method may get called in repairing mode)
- */
- if (prefix == null || prefix.length() == 0) {
- writeDefaultNamespace(nsURI);
- return;
- }
- if (!mStartElementOpen) {
- throwOutputError(ERR_NSDECL_WRONG_STATE);
- }
- /* 01-Sep-2006, TSa: Let's only add the declaration if the prefix
- * is as of yet unbound. If we have to re-bind things in future,
- * so be it -- for now, this should suffice (and if we have to
- * add re-binding, must verify that no attribute, nor element
- * itself, is using overridden prefix)
- */
- int value = mCurrElem.isPrefixValid(prefix, nsURI, true);
- if (value == SimpleOutputElement.PREFIX_UNBOUND) {
- mCurrElem.addPrefix(prefix, nsURI);
- doWriteNamespace(prefix, nsURI);
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods:
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * With repairing writer, this is only taken as a suggestion as to how
- * the caller would prefer prefixes to be mapped.
- */
- public void setDefaultNamespace(String uri)
- throws XMLStreamException
- {
- mSuggestedDefNs = (uri == null || uri.length() == 0) ? null : uri;
- }
-
- public void doSetPrefix(String prefix, String uri)
- throws XMLStreamException
- {
- /* Ok; let's assume that passing in a null or empty String as
- * the URI means that we don't want passed prefix to be preferred
- * for any URI.
- */
- if (uri == null || uri.length() == 0) {
- if (mSuggestedPrefixes != null) {
- for (Iterator it = mSuggestedPrefixes.entrySet().iterator();
- it.hasNext(); ) {
- Map.Entry en = (Map.Entry) it.next();
- String thisP = (String) en.getValue();
- if (thisP.equals(prefix)) {
- it.remove();
- }
- }
- }
- } else {
- if (mSuggestedPrefixes == null) {
- mSuggestedPrefixes = new HashMap(16);
- }
- mSuggestedPrefixes.put(uri, prefix);
- }
- }
-
- public void writeStartElement(StartElement elem)
- throws XMLStreamException
- {
- /* In repairing mode this is simple: let's just pass info
- * we have, and things should work... a-may-zing!
- */
- QName name = elem.getName();
- writeStartElement(name.getPrefix(), name.getLocalPart(),
- name.getNamespaceURI());
- Iterator it = elem.getAttributes();
- while (it.hasNext()) {
- Attribute attr = (Attribute) it.next();
- name = attr.getName();
- writeAttribute(name.getPrefix(), name.getNamespaceURI(),
- name.getLocalPart(), attr.getValue());
- }
- }
-
- //public void writeEndElement(QName name) throws XMLStreamException
-
- protected void writeTypedAttribute(String prefix, String nsURI, String localName,
- AsciiValueEncoder enc)
- throws XMLStreamException
- {
- super.writeTypedAttribute(findOrCreateAttrPrefix(prefix, nsURI, mCurrElem),
- nsURI, localName, enc);
- }
-
- protected void writeStartOrEmpty(String localName, String nsURI)
- throws XMLStreamException
- {
- checkStartElement(localName, "");
-
- // First, need to find prefix matching URI, if any:
- String prefix = findElemPrefix(nsURI, mCurrElem);
- /* Then need to create the element, since it'll have to
- * contain the new namespace binding, if one needed
- * (changed to resolve [WSTX-135] as reported by Y-J Choi,
- * who also proposed the solution)
- */
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, prefix, localName, nsURI);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(prefix, localName, nsURI);
- }
-
- if (prefix != null) { // prefix ok, easy, no need to overwrite
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, prefix);
- }
- doWriteStartTag(prefix, localName);
- } else { // no prefix, more work
- prefix = generateElemPrefix(null, nsURI, mCurrElem);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, prefix);
- }
- mCurrElem.setPrefix(prefix);
- doWriteStartTag(prefix, localName);
- if (prefix == null || prefix.length() == 0) { // def NS
- mCurrElem.setDefaultNsUri(nsURI);
- doWriteDefaultNs(nsURI);
- } else { // explicit NS
- mCurrElem.addPrefix(prefix, nsURI);
- doWriteNamespace(prefix, nsURI);
- }
- }
- }
-
- protected void writeStartOrEmpty(String suggPrefix, String localName, String nsURI)
- throws XMLStreamException
- {
- checkStartElement(localName, suggPrefix);
-
- // In repairing mode, better ensure validity:
- String actPrefix = validateElemPrefix(suggPrefix, nsURI, mCurrElem);
- if (actPrefix != null) { // fine, an existing binding we can use:
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, actPrefix);
- }
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, actPrefix, localName, nsURI);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(actPrefix, localName, nsURI);
- }
- doWriteStartTag(actPrefix, localName);
- } else { // nah, need to create a new binding...
- /* Need to ensure that we'll pass "" as prefix, not null, so
- * that it is understood as "I want to use the default NS", not
- * as "whatever prefix, I don't care"
- */
- if (suggPrefix == null) {
- suggPrefix = "";
- }
- actPrefix = generateElemPrefix(suggPrefix, nsURI, mCurrElem);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, actPrefix);
- }
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, actPrefix, localName, nsURI);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(actPrefix, localName, nsURI);
- }
- mCurrElem.setPrefix(actPrefix);
- doWriteStartTag(actPrefix, localName);
- if (actPrefix == null || actPrefix.length() == 0) { // def NS
- mCurrElem.setDefaultNsUri(nsURI);
- doWriteDefaultNs(nsURI);
- } else { // explicit NS
- mCurrElem.addPrefix(actPrefix, nsURI);
- doWriteNamespace(actPrefix, nsURI);
- }
- }
- }
-
- /**
- * Element copier method implementation suitable for use with
- * namespace-aware writers in repairing mode.
- * The trickiest thing is having to properly
- * order calls to
- * One noteworthy feature of the class is that it is designed to allow
- * "short-term recycling", ie. instances can be reused within context
- * of a simple document output. While reuse/recycling of such lightweight
- * object is often useless or even counter productive, here it may
- * be worth using, due to simplicity of the scheme (basically using
- * a very simple free-elements linked list).
- */
-public final class SimpleOutputElement
- extends OutputElementBase
-{
- /*
- ////////////////////////////////////////////
- // Information about element itself:
- ////////////////////////////////////////////
- */
-
- /**
- * Reference to the parent element, element enclosing this element.
- * Null for root element.
- * Non-final only to allow temporary pooling
- * (on per-writer basis, to keep these short-lived).
- */
- SimpleOutputElement mParent;
-
- /**
- * Prefix that is used for the element. Can not be final, since sometimes
- * it needs to be dynamically generated and bound after creating the
- * element instance.
- */
- String mPrefix;
-
- /**
- * Local name of the element.
- * Non-final only to allow reuse.
- */
- String mLocalName;
-
- /**
- * Namespace of the element, whatever {@link #mPrefix} maps to.
- * Non-final only to allow reuse.
- */
- String mURI;
-
- /*
- ////////////////////////////////////////////
- // Attribute information
- ////////////////////////////////////////////
- */
-
- /**
- * Map used to check for duplicate attribute declarations, if
- * feature is enabled.
- */
- protected HashSet mAttrSet = null;
-
- /*
- ////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////
- */
-
- /**
- * Constructor for the virtual root element
- */
- private SimpleOutputElement()
- {
- super();
- mParent = null;
- mPrefix = null;
- mLocalName = "";
- mURI = null;
- }
-
- private SimpleOutputElement(SimpleOutputElement parent,
- String prefix, String localName, String uri,
- BijectiveNsMap ns)
- {
- super(parent, ns);
- mParent = parent;
- mPrefix = prefix;
- mLocalName = localName;
- mURI = uri;
- }
-
- /**
- * Method called to reuse a pooled instance.
- *
- * @return Chained pooled instance that should now be head of the
- * reuse chain
- */
- private void relink(SimpleOutputElement parent,
- String prefix, String localName, String uri)
- {
- super.relink(parent);
- mParent = parent;
- mPrefix = prefix;
- mLocalName = localName;
- mURI = uri;
- mNsMapping = parent.mNsMapping;
- mNsMapShared = (mNsMapping != null);
- mDefaultNsURI = parent.mDefaultNsURI;
- mRootNsContext = parent.mRootNsContext;
- }
-
- public static SimpleOutputElement createRoot()
- {
- return new SimpleOutputElement();
- }
-
- /**
- * Simplest factory method, which gets called when a 1-argument
- * element output method is called. It is, then, assumed to
- * use the default namespce.
- */
- protected SimpleOutputElement createChild(String localName)
- {
- /* At this point we can also discard attribute Map; it is assumed
- * that when a child element has been opened, no more attributes
- * can be output.
- */
- mAttrSet = null;
- return new SimpleOutputElement(this, null, localName,
- mDefaultNsURI, mNsMapping);
- }
-
- /**
- * @return New head of the recycle pool
- */
- protected SimpleOutputElement reuseAsChild(SimpleOutputElement parent,
- String localName)
- {
- mAttrSet = null;
- SimpleOutputElement poolHead = mParent;
- relink(parent, null, localName, mDefaultNsURI);
- return poolHead;
- }
-
- protected SimpleOutputElement reuseAsChild(SimpleOutputElement parent,
- String prefix, String localName,
- String uri)
- {
- mAttrSet = null;
- SimpleOutputElement poolHead = mParent;
- relink(parent, prefix, localName, uri);
- return poolHead;
- }
-
- /**
- * Full factory method, used for 'normal' namespace qualified output
- * methods.
- */
- protected SimpleOutputElement createChild(String prefix, String localName,
- String uri)
- {
- /* At this point we can also discard attribute Map; it is assumed
- * that when a child element has been opened, no more attributes
- * can be output.
- */
- mAttrSet = null;
- return new SimpleOutputElement(this, prefix, localName, uri, mNsMapping);
- }
-
- /**
- * Method called to temporarily link this instance to a pool, to
- * allow reusing of instances with the same reader.
- */
- protected void addToPool(SimpleOutputElement poolHead)
- {
- mParent = poolHead;
- }
-
- /*
- ////////////////////////////////////////////
- // Public API, accessors
- ////////////////////////////////////////////
- */
-
- public SimpleOutputElement getParent() {
- return mParent;
- }
-
- public boolean isRoot() {
- // (Virtual) Root element has no parent...
- return (mParent == null);
- }
-
- /**
- * @return String presentation of the fully-qualified name, in
- * "prefix:localName" format (no URI). Useful for error and
- * debugging messages.
- */
- public String getNameDesc() {
- if (mPrefix != null && mPrefix.length() > 0) {
- return mPrefix + ":" +mLocalName;
- }
- if (mLocalName != null && mLocalName.length() > 0) {
- return mLocalName;
- }
- return "#error"; // unexpected case
- }
-
- public String getPrefix() {
- return mPrefix;
- }
-
- public String getLocalName() {
- return mLocalName;
- }
-
- public String getNamespaceURI() {
- return mURI;
- }
-
- public QName getName() {
- return QNameCreator.create(mURI, mLocalName, mPrefix);
- }
-
- /*
- ////////////////////////////////////////////
- // Public API, ns binding, checking
- ////////////////////////////////////////////
- */
-
- public void checkAttrWrite(String nsURI, String localName)
- throws XMLStreamException
- {
- AttrName an = new AttrName(nsURI, localName);
- if (mAttrSet == null) {
- /* 13-Dec-2005, TSa: Should use a more efficient Set/Map value
- * for this in future -- specifically one that could use
- * ns/local-name pairs without intermediate objects
- */
- mAttrSet = new HashSet();
- }
- if (!mAttrSet.add(an)) {
- throw new XMLStreamException("Duplicate attribute write for attribute '"+an+"'");
- }
- }
-
- /*
- ////////////////////////////////////////////
- // Public API, mutators
- ////////////////////////////////////////////
- */
-
- public void setPrefix(String prefix) {
- mPrefix = prefix;
- }
-
- public void setDefaultNsUri(String uri) {
- mDefaultNsURI = uri;
- }
-
- /**
- * Note: this method can and will only be called before outputting
- * the root element.
- */
- protected final void setRootNsContext(NamespaceContext ctxt)
- {
- mRootNsContext = ctxt;
- // Let's also figure out the default ns binding, if any:
- String defURI = ctxt.getNamespaceURI("");
- if (defURI != null && defURI.length() > 0) {
- mDefaultNsURI = defURI;
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Helper classes:
- //////////////////////////////////////////////////
- */
-
- /**
- * Simple key class used to represent two-piece (attribute) names;
- * first part being optional (URI), and second non-optional (local name).
- */
- final static class AttrName
- implements Comparable
- {
- final String mNsURI;
- final String mLocalName;
-
- /**
- * Let's cache the hash code, since although hash calculation is
- * fast, hash code is needed a lot as this is always used as a
- * HashSet/TreeMap key.
- */
- final int mHashCode;
-
- public AttrName(String nsURI, String localName) {
- mNsURI = (nsURI == null) ? "" : nsURI;
- mLocalName = localName;
- mHashCode = mNsURI.hashCode() * 31 ^ mLocalName.hashCode();
- }
-
- public boolean equals(Object o) {
- if (o == this) {
- return true;
- }
- if (!(o instanceof AttrName)) {
- return false;
- }
- AttrName other = (AttrName) o;
- String otherLN = other.mLocalName;
- // Local names are shorter, more varying:
- if (otherLN != mLocalName && !otherLN.equals(mLocalName)) {
- return false;
- }
- String otherURI = other.mNsURI;
- return (otherURI == mNsURI || otherURI.equals(mNsURI));
- }
-
- public String toString() {
- if (mNsURI.length() > 0) {
- return "{"+mNsURI + "} " +mLocalName;
- }
- return mLocalName;
- }
-
- public int hashCode() {
- return mHashCode;
- }
-
- public int compareTo(Object o) {
- AttrName other = (AttrName) o;
- // Let's first order by namespace:
- int result = mNsURI.compareTo(other.mNsURI);
- if (result == 0) {
- result = mLocalName.compareTo(other.mLocalName);
- }
- return result;
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/TypedStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/TypedStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/TypedStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/TypedStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,330 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.math.BigDecimal;
-import java.math.BigInteger;
-
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.typed.Base64Variant;
-import org.codehaus.stax2.typed.Base64Variants;
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-import org.codehaus.stax2.ri.typed.ValueEncoderFactory;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.exc.WstxIOException;
-
-/**
- * Intermediate base class that implements Typed Access API (Stax2 v3)
- * for all (repairing, non-repairing, non-namespace) native stream
- * writer implementations.
- */
-public abstract class TypedStreamWriter
- extends BaseStreamWriter
-{
- /**
- * When outputting using Typed Access API, we will need
- * encoders. If so, they will created by lazily-constructed
- * factory
- */
- protected ValueEncoderFactory mValueEncoderFactory;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////////
- */
-
- protected TypedStreamWriter(XmlWriter xw, String enc, WriterConfig cfg)
- {
- super(xw, enc, cfg);
- }
-
- protected final ValueEncoderFactory valueEncoderFactory()
- {
- if (mValueEncoderFactory == null) {
- mValueEncoderFactory = new ValueEncoderFactory();
- }
- return mValueEncoderFactory;
- }
-
- /*
- /////////////////////////////////////////////////
- // TypedXMLStreamWriter2 implementation
- // (Typed Access API, Stax v3.0)
- /////////////////////////////////////////////////
- */
-
- // // // Typed element content write methods
-
- public void writeBoolean(boolean value)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value));
- }
-
- public void writeInt(int value)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value));
- }
-
- public void writeLong(long value)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value));
- }
-
- public void writeFloat(float value)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value));
- }
-
- public void writeDouble(double value)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value));
-
- }
-
- public void writeInteger(BigInteger value)
- throws XMLStreamException
- {
- /* No really efficient method exposed by JDK, keep it simple
- * (esp. considering that length is actually not bound)
- */
- writeTypedElement(valueEncoderFactory().getScalarEncoder(value.toString()));
- }
-
- public void writeDecimal(BigDecimal value)
- throws XMLStreamException
- {
- /* No really efficient method exposed by JDK, keep it simple
- * (esp. considering that length is actually not bound)
- */
- writeTypedElement(valueEncoderFactory().getScalarEncoder(value.toString()));
- }
-
- public void writeQName(QName name)
- throws XMLStreamException
- {
- /* Can't use AsciiValueEncoder, since QNames can contain
- * non-ascii characters
- */
- writeCharacters(serializeQName(name));
- }
-
- public final void writeIntArray(int[] value, int from, int length)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value, from, length));
- }
-
- public void writeLongArray(long[] value, int from, int length)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value, from, length));
- }
-
- public void writeFloatArray(float[] value, int from, int length)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value, from, length));
- }
-
- public void writeDoubleArray(double[] value, int from, int length)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(value, from, length));
- }
-
- public void writeBinary(byte[] value, int from, int length)
- throws XMLStreamException
- {
- Base64Variant v = Base64Variants.getDefaultVariant();
- writeTypedElement(valueEncoderFactory().getEncoder(v, value, from, length));
- }
-
- public void writeBinary(Base64Variant v, byte[] value, int from, int length)
- throws XMLStreamException
- {
- writeTypedElement(valueEncoderFactory().getEncoder(v, value, from, length));
- }
-
- protected final void writeTypedElement(AsciiValueEncoder enc)
- throws XMLStreamException
- {
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- // How about well-formedness?
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_NONWS_TEXT);
- }
- }
- // Or validity?
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- reportInvalidContent(CHARACTERS);
- }
-
- // So far so good: let's serialize
- try {
- XMLValidator vld = (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT) ?
- mValidator : null;
- if (vld == null) {
- mWriter.writeTypedElement(enc);
- } else {
- mWriter.writeTypedElement(enc, vld, getCopyBuffer());
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- // // // Typed attribute value write methods
-
- public void writeBooleanAttribute(String prefix, String nsURI, String localName, boolean value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value));
- }
-
- public void writeIntAttribute(String prefix, String nsURI, String localName, int value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value));
- }
-
- public void writeLongAttribute(String prefix, String nsURI, String localName, long value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value));
- }
-
- public void writeFloatAttribute(String prefix, String nsURI, String localName, float value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value));
- }
-
- public void writeDoubleAttribute(String prefix, String nsURI, String localName, double value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value));
- }
-
- public void writeIntegerAttribute(String prefix, String nsURI, String localName, BigInteger value)
- throws XMLStreamException
- {
- // not optimal, but has to do:
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getScalarEncoder(value.toString()));
- }
-
- public void writeDecimalAttribute(String prefix, String nsURI, String localName, BigDecimal value)
- throws XMLStreamException
- {
- // not optimal, but has to do:
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getScalarEncoder(value.toString()));
- }
-
- public void writeQNameAttribute(String prefix, String nsURI, String localName, QName name)
- throws XMLStreamException
- {
- /* Can't use AsciiValueEncoder, since QNames can contain
- * non-ascii characters
- */
- writeAttribute(prefix, nsURI, localName, serializeQName(name));
- }
-
- public void writeIntArrayAttribute(String prefix, String nsURI, String localName, int[] value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value, 0, value.length));
- }
-
- public void writeLongArrayAttribute(String prefix, String nsURI, String localName, long[] value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value, 0, value.length));
- }
-
- public void writeFloatArrayAttribute(String prefix, String nsURI, String localName, float[] value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value, 0, value.length));
- }
-
- public void writeDoubleArrayAttribute(String prefix, String nsURI, String localName, double[] value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(value, 0, value.length));
- }
-
- public void writeBinaryAttribute(String prefix, String nsURI, String localName, byte[] value)
- throws XMLStreamException
- {
- Base64Variant v = Base64Variants.getDefaultVariant();
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(v, value, 0, value.length));
- }
-
- public void writeBinaryAttribute(Base64Variant v, String prefix, String nsURI, String localName, byte[] value)
- throws XMLStreamException
- {
- writeTypedAttribute(prefix, nsURI, localName,
- valueEncoderFactory().getEncoder(v, value, 0, value.length));
- }
-
- /**
- * Method that will write attribute with value that is known not to
- * require additional escaping.
- */
- protected abstract void writeTypedAttribute(String prefix, String nsURI,
- String localName,
- AsciiValueEncoder enc)
- throws XMLStreamException;
-
- private String serializeQName(QName name)
- throws XMLStreamException
- {
- String vp = validateQNamePrefix(name);
- String local = name.getLocalPart();
- if (vp == null || vp.length() == 0) {
- return local;
- }
-
- // Not efficient... but should be ok
- return vp + ":" + local;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/XmlWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/XmlWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/XmlWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/XmlWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,625 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.io.OutputStream;
-import java.io.Writer;
-import java.text.MessageFormat;
-
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.io.EscapingWriterFactory;
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.api.InvalidCharHandler;
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.api.WstxOutputProperties;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.OutputConfigFlags;
-import com.ctc.wstx.exc.WstxIOException;
-import com.ctc.wstx.io.WstxInputData;
-
-/**
- * This is the base class for actual physical xml outputters. These
- * instances will only handle actual writing (possibly including
- * encoding) of the serialized textual xml, and will in general
- * not verify content being output. The exception are the
- * character-by-character checks that are most efficiently done
- * at encoding level (such as character escaping, and checks for
- * illegal character combinations), which are handled at this
- * level.
- *
- * Note that implementations can have different operating modes:
- * specifically, when dealing with illegal content (such as "--"
- * in a comment, "?>" in processing instruction, or "]]>" within
- * CDATA section), implementations can do one of 3 things:
- *
- * Note: can throw XMLStreamException, if name checking is enabled,
- * and name is invalid (name check has to be in this writer, not
- * caller, since it depends not only on xml limitations, but also
- * on encoding limitations)
- */
- public abstract void writeStartTagStart(String localName)
- throws IOException, XMLStreamException;
-
- /**
- *
- * Note: can throw XMLStreamException, if name checking is enabled,
- * and name is invalid (name check has to be in this writer, not
- * caller, since it depends not only on xml limitations, but also
- * on encoding limitations)
- */
- public abstract void writeStartTagStart(String prefix, String localName)
- throws IOException, XMLStreamException;
-
- public abstract void writeStartTagEnd()
- throws IOException;
-
- public abstract void writeStartTagEmptyEnd()
- throws IOException;
-
- public abstract void writeEndTag(String localName)
- throws IOException;
-
- public abstract void writeEndTag(String prefix, String localName)
- throws IOException;
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, attributes/ns
- ////////////////////////////////////////////////////
- */
-
- /**
- *
- * Note: can throw XMLStreamException, if name checking is enabled,
- * and name is invalid (name check has to be in this writer, not
- * caller, since it depends not only on xml limitations, but also
- * on encoding limitations)
- */
- public abstract void writeAttribute(String localName, String value)
- throws IOException, XMLStreamException;
-
- public abstract void writeAttribute(String localName, char[] value, int offset, int len)
- throws IOException, XMLStreamException;
-
- /**
- *
- * Note: can throw XMLStreamException, if name checking is enabled,
- * and name is invalid (name check has to be in this writer, not
- * caller, since it depends not only on xml limitations, but also
- * on encoding limitations)
- */
- public abstract void writeAttribute(String prefix, String localName, String value)
- throws IOException, XMLStreamException;
-
- public abstract void writeAttribute(String prefix, String localName, char[] value, int offset, int len)
- throws IOException, XMLStreamException;
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, Typed Access API support
- ////////////////////////////////////////////////////
- */
-
- /**
- * Like {@link #writeRaw}, but caller guarantees that the contents
- * additionally are known to be in 7-bit ascii range, and also
- * passes an encoder object that will encode values only when
- * being handed a buffer to append to.
- *
- * @param enc Encoder that will produce content
- */
- public abstract void writeTypedElement(AsciiValueEncoder enc)
- throws IOException;
-
- /**
- * Like {@link #writeRaw}, but caller guarantees that the contents
- * additionally are known to be in 7-bit ascii range, and also
- * passes an encoder object that will encode values only when
- * being handed a buffer to append to.
- *
- * @param enc Encoder that will produce content
- * @param validator Validator to use for validating serialized textual
- * content (can not be null)
- * @param copyBuffer Temporary buffer that writer can use for temporary
- * copies as necessary
- */
- public abstract void writeTypedElement(AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException;
-
- /**
- * Method similar to {@link #writeAttribute(String,String,char[],int,int)}
- * but where is known not to require escaping.
- * No validation needs to be performed.
- */
- public abstract void writeTypedAttribute(String localName, AsciiValueEncoder enc)
- throws IOException, XMLStreamException;
-
- /**
- * Method similar to {@link #writeAttribute(String,String,char[],int,int)}
- * but where is known not to require escaping.
- * No validation needs to be performed.
- */
- public abstract void writeTypedAttribute(String prefix, String localName, AsciiValueEncoder enc)
- throws IOException, XMLStreamException;
-
- /**
- * Method similar to {@link #writeAttribute(String,String,char[],int,int)}
- * but where is known not to require escaping.
- * Validation of the attribute value must be done by calling given
- * validator appropriately.
- */
- public abstract void writeTypedAttribute(String prefix, String localName, String nsURI,
- AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException;
-
- /*
- ////////////////////////////////////////////////////
- // Location information
- ////////////////////////////////////////////////////
- */
-
- protected abstract int getOutputPtr();
-
- public int getRow() {
- return mLocRowNr;
- }
-
- public int getColumn() {
- return (getOutputPtr() - mLocRowStartOffset) + 1;
- }
-
- public int getAbsOffset() {
- return mLocPastChars +getOutputPtr();
- }
-
- /*
- ////////////////////////////////////////////////////
- // Wrapper methods, semi-public
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method that can be called to get a wrapper instance that
- * can be used to essentially call the
- * Currently only used by stream writers, but could be more generally useful
- * too.
- */
-
-public final class BijectiveNsMap
-{
- /*
- ///////////////////////////////////////////////
- // Constants
- ///////////////////////////////////////////////
- */
-
- /**
- * Let's plan for having up to 14 explicit namespace declarations (2
- * defaults, for 'xml' and 'xmlns', are pre-populated)
- */
- final static int DEFAULT_ARRAY_SIZE = 2 * 16;
-
- /*
- ///////////////////////////////////////////////
- // Member vars
- ///////////////////////////////////////////////
- */
-
- final int mScopeStart;
-
- /**
- * Array that contains { prefix, ns-uri } pairs, up to (but not including)
- * index {@link #mScopeEnd}.
- */
- String[] mNsStrings;
-
- int mScopeEnd;
-
- /*
- ///////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////
- */
-
- private BijectiveNsMap(int scopeStart, String[] strs)
- {
- mScopeStart = mScopeEnd = scopeStart;
- mNsStrings = strs;
- }
-
- public static BijectiveNsMap createEmpty()
- {
- String[] strs = new String[DEFAULT_ARRAY_SIZE];
-
- strs[0] = XMLConstants.XML_NS_PREFIX;
- strs[1] = XMLConstants.XML_NS_URI;
- strs[2] = XMLConstants.XMLNS_ATTRIBUTE;
- strs[3] = XMLConstants.XMLNS_ATTRIBUTE_NS_URI;
-
- /* Let's consider pre-defined ones to be 'out of scope', i.e.
- * conceptually be part of (missing) parent's mappings.
- */
- return new BijectiveNsMap(4, strs);
- }
-
- public BijectiveNsMap createChild() {
- return new BijectiveNsMap(mScopeEnd, mNsStrings);
- }
-
- /*
- ///////////////////////////////////////////////
- // Public API, accessors
- ///////////////////////////////////////////////
- */
-
- public String findUriByPrefix(String prefix)
- {
- /* This is quite simple: just need to locate the last mapping
- * for the prefix, if any:
- */
- String[] strs = mNsStrings;
- int phash = prefix.hashCode();
-
- for (int ix = mScopeEnd - 2; ix >= 0; ix -= 2) {
- String thisP = strs[ix];
- if (thisP == prefix ||
- (thisP.hashCode() == phash && thisP.equals(prefix))) {
- return strs[ix+1];
- }
- }
- return null;
- }
-
- public String findPrefixByUri(String uri)
- {
-
- /* Finding a valid binding for the given URI is trickier, since
- * mappings can be masked by others... so, we need to first find
- * most recent binding, from the freshest one, and then verify
- * it's still unmasked; if not, continue with the first loop,
- * and so on.
- */
-
- String[] strs = mNsStrings;
- int uhash = uri.hashCode();
-
- main_loop:
- for (int ix = mScopeEnd - 1; ix > 0; ix -= 2) {
- String thisU = strs[ix];
- if (thisU == uri ||
- (thisU.hashCode() == uhash && thisU.equals(uri))) {
- // match, but has it been masked?
- String prefix = strs[ix-1];
- /* only need to check, if it wasn't within current scope
- * (no masking allowed within scopes)
- */
- if (ix < mScopeStart) {
- int phash = prefix.hashCode();
- for (int j = ix+1, end = mScopeEnd; j < end; j += 2) {
- String thisP = strs[j];
- if (thisP == prefix ||
- (thisP.hashCode() == phash && thisP.equals(prefix))) {
- // Masking... got to continue the main loop:
- continue main_loop;
- }
- }
- }
- // Ok, unmasked one, can return
- return prefix;
- }
- }
- return null;
- }
-
- public List getPrefixesBoundToUri(String uri, List l)
- {
- /* Same problems (masking) apply here, as well as with
- * findPrefixByUri...
- */
- String[] strs = mNsStrings;
- int uhash = uri.hashCode();
-
- main_loop:
- for (int ix = mScopeEnd - 1; ix > 0; ix -= 2) {
- String thisU = strs[ix];
- if (thisU == uri ||
- (thisU.hashCode() == uhash && thisU.equals(uri))) {
- // match, but has it been masked?
- String prefix = strs[ix-1];
- /* only need to check, if it wasn't within current scope
- * (no masking allowed within scopes)
- */
- if (ix < mScopeStart) {
- int phash = prefix.hashCode();
- for (int j = ix+1, end = mScopeEnd; j < end; j += 2) {
- String thisP = strs[j];
- if (thisP == prefix ||
- (thisP.hashCode() == phash && thisP.equals(prefix))) {
- // Masking... got to continue the main loop:
- continue main_loop;
- }
- }
- }
- // Ok, unmasked one, can add
- if (l == null) {
- l = new ArrayList();
- }
- l.add(prefix);
- }
- }
- return l;
- }
-
- public int size() {
- return (mScopeEnd >> 1);
- }
-
- public int localSize() {
- return ((mScopeEnd - mScopeStart) >> 1);
- }
-
- /*
- ///////////////////////////////////////////////
- // Public API, mutators
- ///////////////////////////////////////////////
- */
-
- /**
- * Method to add a new prefix-to-URI mapping for the current scope.
- * Note that it should NOT be used for the default namespace
- * declaration
- *
- * @param prefix Prefix to bind
- * @param uri URI to bind to the prefix
- *
- * @return If the prefix was already bound, the URI it was bound to:
- * null if it's a new binding for the current scope.
- */
- public String addMapping(String prefix, String uri)
- {
- String[] strs = mNsStrings;
- int phash = prefix.hashCode();
-
- for (int ix = mScopeStart, end = mScopeEnd; ix < end; ix += 2) {
- String thisP = strs[ix];
- if (thisP == prefix ||
- (thisP.hashCode() == phash && thisP.equals(prefix))) {
- // Overriding an existing mapping
- String old = strs[ix+1];
- strs[ix+1] = uri;
- return old;
- }
- }
- // no previous binding, let's just add it at the end
- if (mScopeEnd >= strs.length) {
- // let's just double the array sizes...
- strs = DataUtil.growArrayBy(strs, strs.length);
- mNsStrings = strs;
- }
- strs[mScopeEnd++] = prefix;
- strs[mScopeEnd++] = uri;
-
- return null;
- }
-
- /**
- * Method used to add a dynamic binding, and return the prefix
- * used to bind the specified namespace URI.
- */
- public String addGeneratedMapping(String prefixBase, NamespaceContext ctxt,
- String uri, int[] seqArr)
- {
- String[] strs = mNsStrings;
- int seqNr = seqArr[0];
- String prefix;
-
- main_loop:
- while (true) {
- /* We better intern the resulting prefix? Or not?
- * TODO: maybe soft cache these for other docs?
- */
- prefix = (prefixBase + seqNr).intern();
- ++seqNr;
-
- /* Ok, let's see if we have a mapping (masked or not) for
- * the prefix. If we do, let's just not use it: we could
- * of course mask it (unless it's in current scope), but
- * it's easier to just get a "virgin" prefix...
- */
- int phash = prefix.hashCode();
-
- for (int ix = mScopeEnd - 2; ix >= 0; ix -= 2) {
- String thisP = strs[ix];
- if (thisP == prefix ||
- (thisP.hashCode() == phash && thisP.equals(prefix))) {
- continue main_loop;
- }
- }
- /* So far so good... but do we have a root context that might
- * have something too?
- */
-
- if (ctxt != null && ctxt.getNamespaceURI(prefix) != null) {
- continue;
- }
- break;
- }
- seqArr[0] = seqNr;
-
- // Ok, good; then let's just add it in...
- if (mScopeEnd >= strs.length) {
- // let's just double the array sizes...
- strs = DataUtil.growArrayBy(strs, strs.length);
- mNsStrings = strs;
- }
- strs[mScopeEnd++] = prefix;
- strs[mScopeEnd++] = uri;
-
- return prefix;
- }
-
- /*
- ///////////////////////////////////////////////
- // Standard overridden methods
- ///////////////////////////////////////////////
- */
-
- public String toString() {
- return "["+getClass().toString()+"; "+size()+" entries; of which "
- +localSize()+" local]";
- }
-
- /*
- ///////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////
- */
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/DataUtil.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/DataUtil.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/DataUtil.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/DataUtil.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,135 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.lang.reflect.Array;
-import java.util.*;
-
-public final class DataUtil
-{
- final static char[] EMPTY_CHAR_ARRAY = new char[0];
-
- /**
- * If baseline requirement was JDK 1.5, we wouldn't need to
- * cache Integer instances like this (since it has
- * Integer.valueOf() which does it); but until then, we
- * alas need our known canonicalization.
- */
- final static Integer[] INTS = new Integer[100];
- static {
- for (int i = 0; i < INTS.length; ++i) {
- INTS[i] = new Integer(i);
- }
- }
-
- private DataUtil() { }
-
- /*
- ////////////////////////////////////////////////////////////
- // Pooling for immutable objects
- ////////////////////////////////////////////////////////////
- */
-
- public static char[] getEmptyCharArray() {
- return EMPTY_CHAR_ARRAY;
- }
-
- public static Integer Integer(int i)
- {
- /* !!! 13-Sep-2008, TSa: JDK 1.5 can use Integer.valueOf(int)
- * which does the same. When upgrading baseline, can get rid
- * of this method.
- */
- if (i < 0 || i >= INTS.length) {
- return new Integer(i);
- }
- return INTS[i];
- }
-
- /*
- ////////////////////////////////////////////////////////////
- // Methods for common operations on std data structs
- ////////////////////////////////////////////////////////////
- */
-
- /**
- * Method that can be used to efficiently check if 2 collections
- * share at least one common element.
- *
- * @return True if there is at least one element that's common
- * to both Collections, ie. that is contained in both of them.
- */
- public static boolean anyValuesInCommon(Collection c1, Collection c2)
- {
- // Let's always iterate over smaller collection:
- if (c1.size() > c2.size()) {
- Collection tmp = c1;
- c1 = c2;
- c2 = tmp;
- }
- Iterator it = c1.iterator();
- while (it.hasNext()) {
- if (c2.contains(it.next())) {
- return true;
- }
- }
- return false;
- }
-
- final static String NO_TYPE = "Illegal to pass null; can not determine component type";
-
- public static Object growArrayBy50Pct(Object arr)
- {
- if (arr == null) {
- throw new IllegalArgumentException(NO_TYPE);
- }
- Object old = arr;
- int len = Array.getLength(arr);
- arr = Array.newInstance(arr.getClass().getComponentType(), len + (len >> 1));
- System.arraycopy(old, 0, arr, 0, len);
- return arr;
- }
-
- /**
- * Method similar to {@link #growArrayBy50Pct}, but it also ensures that
- * the new size is at least as big as the specified minimum size.
- */
- public static Object growArrayToAtLeast(Object arr, int minLen)
- {
- if (arr == null) {
- throw new IllegalArgumentException(NO_TYPE);
- }
- Object old = arr;
- int oldLen = Array.getLength(arr);
- int newLen = oldLen + ((oldLen + 1) >> 1);
- if (newLen < minLen) {
- newLen = minLen;
- }
- arr = Array.newInstance(arr.getClass().getComponentType(), newLen);
- System.arraycopy(old, 0, arr, 0, oldLen);
- return arr;
- }
-
- public static String[] growArrayBy(String[] arr, int more)
- {
- if (arr == null) {
- return new String[more];
- }
- String[] old = arr;
- int len = arr.length;
- arr = new String[len + more];
- System.arraycopy(old, 0, arr, 0, len);
- return arr;
- }
-
- public static int[] growArrayBy(int[] arr, int more)
- {
- if (arr == null) {
- return new int[more];
- }
- int[] old = arr;
- int len = arr.length;
- arr = new int[len + more];
- System.arraycopy(old, 0, arr, 0, len);
- return arr;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/DefaultXmlSymbolTable.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/DefaultXmlSymbolTable.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/DefaultXmlSymbolTable.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/DefaultXmlSymbolTable.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,93 +0,0 @@
-package com.ctc.wstx.util;
-
-import com.ctc.wstx.util.SymbolTable;
-
-/**
- * Factory class used for instantiating pre-populated XML symbol
- * tables. Such tables already have basic String constants that
- * XML standard defines.
- */
-public final class DefaultXmlSymbolTable
-{
- /**
- * Root symbol table from which child instances are derived.
- */
- final static SymbolTable sInstance;
-
- final static String mNsPrefixXml;
- final static String mNsPrefixXmlns;
-
- /* Although theoretically there'd be no strict need to pre-populate
- * the default table, if all access was done using suggested usage
- * patterns (reuse input factories consistently, esp. for same types
- * of documents), it is possible some developers just use each factory
- * just once. As such, it does matter how tables are pre-populated.
- * Thus, let's use limited sensible set of predefined prefixes and
- * names.
- */
- static {
- /* 128 means it's ok without resize up to ~96 symbols; true that
- * default symbols added will be interned.
- */
- sInstance = new SymbolTable(true, 128);
-
- // Let's add default namespace binding prefixes
- mNsPrefixXml = sInstance.findSymbol("xml");
- mNsPrefixXmlns = sInstance.findSymbol("xmlns");
-
- /* No need to add keywords, as they are checked directly by
- * Reader, without constructing Strings.
- */
-
- // Ok, any common prefixes?
-
- // or local names (element, attribute)?
- sInstance.findSymbol("id");
- sInstance.findSymbol("name");
-
- // XML Schema?
- // prefixes:
- sInstance.findSymbol("xsd");
- sInstance.findSymbol("xsi");
- // local names:
- sInstance.findSymbol("type");
-
- // How about some common prefixes and names for Soap?
- // commonly used prefixes:
- sInstance.findSymbol("soap");
- sInstance.findSymbol("SOAP-ENC");
- sInstance.findSymbol("SOAP-ENV");
- // local names:
- sInstance.findSymbol("Body");
- sInstance.findSymbol("Envelope");
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API, factory method(s):
- ///////////////////////////////////////////////////
- */
-
- /**
- * Method that will return an instance of SymbolTable that has basic
- * XML 1.0 constants pre-populated.
- */
- public static SymbolTable getInstance() {
- return sInstance.makeChild();
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API, efficient access to (shared)
- // constants values:
- ///////////////////////////////////////////////////
- */
-
- public static String getXmlSymbol() {
- return mNsPrefixXml;
- }
-
- public static String getXmlnsSymbol() {
- return mNsPrefixXmlns;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ElementId.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ElementId.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ElementId.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ElementId.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,162 +0,0 @@
-package com.ctc.wstx.util;
-
-import javax.xml.stream.Location;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-
-/**
- * Simple container Object used to store information about id attribute
- * values, and references to such (as of yet undefined) values.
- *
- * Instances can be in one of 2 modes: either in fully defined mode,
- * in which case information refers to location where value was defined
- * (ie. we had id as a value of ID type attribute); or in undefined mode,
- * in which case information refers to the first reference.
- *
- * Note: this class is designed to be used with {@link ElementIdMap},
- * and as a result has some information specifically needed by the
- * map implementation (such as collision links).
- */
-public final class ElementId
-{
- /**
- * Flag that indicates whether this Object presents a defined id
- * value (value of an ID attribute) or just a reference to one.
- */
- private boolean mDefined;
-
- /*
- /////////////////////////////////////////////////
- // Information about id value or value reference,
- // depending on mDefined flag
- /////////////////////////////////////////////////
- */
-
- /**
- * Actual id value
- */
- private final String mIdValue;
-
- /**
- * Location of either definition (if {@link #mDefined} is true; or
- * first reference (otherwise). Used when reporting errors; either
- * a referenced id has not been defined, or there are multiple
- * definitions of same id.
- */
- private Location mLocation;
-
- /**
- * Name of element for which this id refers.
- */
- private PrefixedName mElemName;
-
- /**
- * Name of the attribute that contains this id value (often "id",
- * but need not be)
- */
- private PrefixedName mAttrName;
-
- /*
- ////////////////////////////////////////////////////
- // Linking information, needed by the map to keep
- // track of collided ids, as well as undefined ids
- ////////////////////////////////////////////////////
- */
-
- private ElementId mNextUndefined;
-
- /**
- * Pointer to the next element within collision chain.
- */
- private ElementId mNextColl;
-
- /*
- /////////////////////////////////////////////////
- // Life cycle
- /////////////////////////////////////////////////
- */
-
- ElementId(String id, Location loc, boolean defined,
- PrefixedName elemName, PrefixedName attrName)
- {
- mIdValue = id;
- mLocation = loc;
- mDefined = defined;
- mElemName = elemName;
- mAttrName = attrName;
- }
-
- protected void linkUndefined(ElementId undefined)
- {
- if (mNextUndefined != null) {
- throw new IllegalStateException("ElementId '"+this+"' already had net undefined set ('"+mNextUndefined+"')");
- }
- mNextUndefined = undefined;
- }
-
- protected void setNextColliding(ElementId nextColl)
- {
- // May add/remove link, no point in checking
- mNextColl = nextColl;
- }
-
- /*
- /////////////////////////////////////////////////
- // Public API
- /////////////////////////////////////////////////
- */
-
- public String getId() { return mIdValue; }
- public Location getLocation() { return mLocation; }
- public PrefixedName getElemName() { return mElemName; }
- public PrefixedName getAttrName() { return mAttrName; }
-
- public boolean isDefined() { return mDefined; }
-
- public boolean idMatches(char[] buf, int start, int len)
- {
- if (mIdValue.length() != len) {
- return false;
- }
- // Assumes it's always at least one char long
- if (buf[start] != mIdValue.charAt(0)) {
- return false;
- }
- int i = 1;
- len += start;
- while (++start < len) {
- if (buf[start] != mIdValue.charAt(i)) {
- return false;
- }
- ++i;
- }
- return true;
- }
-
- public boolean idMatches(String idStr)
- {
- return mIdValue.equals(idStr);
- }
-
- public ElementId nextUndefined() { return mNextUndefined; }
- public ElementId nextColliding() { return mNextColl; }
-
- public void markDefined(Location defLoc) {
- if (mDefined) { // sanity check
- throw new IllegalStateException(ErrorConsts.ERR_INTERNAL);
- }
- mDefined = true;
- mLocation = defLoc;
- }
-
- /*
- /////////////////////////////////////////////////
- // Other methods
- /////////////////////////////////////////////////
- */
-
- public String toString() {
- return mIdValue;
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ElementIdMap.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ElementIdMap.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ElementIdMap.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ElementIdMap.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,425 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.util;
-
-import javax.xml.stream.Location;
-
-
-/**
- * This class is a specialized type-safe linked hash map used for
- * storing {@link ElementId} instances. {@link ElementId} instances
- * represent both id definitions (values of element attributes that
- * have type ID in DTD), and references (values of element attributes
- * of type IDREF and IDREFS). These definitions and references are
- * stored for the purpose of verifying
- * that all referenced id values are defined, and that none are defined
- * more than once.
- *
- * Note: there are 2 somewhat distinct usage modes, by DTDValidator and
- * by MSV-based validators.
- * DTDs pass raw character arrays, whereas
- * MSV-based validators operate on Strings. This is the main reason
- * for 2 distinct sets of methods.
- */
-
-public final class ElementIdMap
-{
- /**
- * Default initial table size; set so that usually it need not
- * be expanded.
- */
- protected static final int DEFAULT_SIZE = 128;
-
- protected static final int MIN_SIZE = 16;
-
- /**
- * Let's use 80% fill factor...
- */
- protected static final int FILL_PCT = 80;
-
- /*
- ////////////////////////////////////////
- // Actual hash table structure
- ////////////////////////////////////////
- */
-
- /**
- * Actual hash table area
- */
- protected ElementId[] mTable;
-
- /**
- * Current size (number of entries); needed to know if and when
- * rehash.
- */
- protected int mSize;
-
- /**
- * Limit that indicates maximum size this instance can hold before
- * it needs to be expanded and rehashed. Calculated using fill
- * factor passed in to constructor.
- */
- protected int mSizeThreshold;
-
- /**
- * Mask used to get index from hash values; equal to
- *
- * Note: identical to {@link com.ctc.wstx.util.SymbolTable#calcHash},
- * although not required to be.
- *
- * @param len Length of String; has to be at least 1 (caller guarantees
- * this pre-condition)
- */
- public static int calcHash(char[] buffer, int start, int len)
- {
- int hash = (int) buffer[0];
- for (int i = 1; i < len; ++i) {
- hash = (hash * 31) + (int) buffer[i];
- }
- return hash;
- }
-
- public static int calcHash(String key)
- {
- int hash = (int) key.charAt(0);
- for (int i = 1, len = key.length(); i < len; ++i) {
- hash = (hash * 31) + (int) key.charAt(i);
-
- }
- return hash;
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Internal methods
- //////////////////////////////////////////////////////////
- */
-
- /**
- * Method called when size (number of entries) of symbol table grows
- * so big that load factor is exceeded. Since size has to remain
- * power of two, arrays will then always be doubled. Main work
- * is really redistributing old entries into new String/Bucket
- * entries.
- */
- private void rehash()
- {
- int size = mTable.length;
- /* Let's grow aggressively; this should minimize number of
- * resizes, while adding to mem usage. But since these Maps
- * are never long-lived (only during parsing and validation of
- * a single doc), that shouldn't greatly matter.
- */
- int newSize = (size << 2);
- ElementId[] oldSyms = mTable;
- mTable = new ElementId[newSize];
-
- // Let's update index mask, threshold, now (needed for rehashing)
- mIndexMask = newSize - 1;
- mSizeThreshold <<= 2;
-
- int count = 0; // let's do sanity check
-
- for (int i = 0; i < size; ++i) {
- for (ElementId id = oldSyms[i]; id != null; ) {
- ++count;
- int index = calcHash(id.getId()) & mIndexMask;
- ElementId nextIn = id.nextColliding();
- id.setNextColliding(mTable[index]);
- mTable[index] = id;
- id = nextIn;
- }
- }
-
- if (count != mSize) {
- ExceptionUtil.throwInternal("on rehash(): had "+mSize+" entries; now have "+count+".");
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/EmptyNamespaceContext.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/EmptyNamespaceContext.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/EmptyNamespaceContext.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/EmptyNamespaceContext.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,71 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.io.Writer;
-import java.util.Iterator;
-
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.stream.XMLStreamWriter;
-
-import org.codehaus.stax2.ri.EmptyIterator;
-
-/**
- * Dummy {@link NamespaceContext} (and {@link BaseNsContext})
- * implementation that is usually used in
- * non-namespace-aware mode.
- *
- * Note: differs from Stax2 reference implementation's version
- * slightly, since it needs to support Woodstox specific extensions
- * for efficient namespace declaration serialization.
- */
-public final class EmptyNamespaceContext
- extends BaseNsContext
-{
- final static EmptyNamespaceContext sInstance = new EmptyNamespaceContext();
-
- private EmptyNamespaceContext() { }
-
- public static EmptyNamespaceContext getInstance() { return sInstance; }
-
- /*
- /////////////////////////////////////////////
- // Extended API
- /////////////////////////////////////////////
- */
-
- public Iterator getNamespaces() {
- return EmptyIterator.getInstance();
- }
-
- /**
- * Method called by the matching start element class to
- * output all namespace declarations active in current namespace
- * scope, if any.
- */
- public void outputNamespaceDeclarations(Writer w)
- {
- ; // nothing to output
- }
-
- public void outputNamespaceDeclarations(XMLStreamWriter w)
- {
- ; // nothing to output
- }
-
- /*
- /////////////////////////////////////////////////
- // Template methods sub-classes need to implement
- /////////////////////////////////////////////////
- */
-
- public String doGetNamespaceURI(String prefix) {
- return null;
- }
-
- public String doGetPrefix(String nsURI) {
- return null;
- }
-
- public Iterator doGetPrefixes(String nsURI) {
- return EmptyIterator.getInstance();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ExceptionUtil.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ExceptionUtil.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ExceptionUtil.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ExceptionUtil.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,72 +0,0 @@
-package com.ctc.wstx.util;
-
-public final class ExceptionUtil
-{
- private ExceptionUtil() { }
-
- /**
- * Method that can be used to convert any Throwable to a RuntimeException;
- * conversion is only done for checked exceptions.
- */
- public static void throwRuntimeException(Throwable t)
- {
- // Unchecked? Can re-throw as is
- throwIfUnchecked(t);
- // Otherwise, let's just change its type:
- RuntimeException rex = new RuntimeException("[was "+t.getClass()+"] "+t.getMessage());
- // And indicate the root cause
- setInitCause(rex, t);
- throw rex;
- }
-
- public static void throwAsIllegalArgument(Throwable t)
- {
- // Unchecked? Can re-throw as is
- throwIfUnchecked(t);
- // Otherwise, let's just change its type:
- IllegalArgumentException rex = new IllegalArgumentException("[was "+t.getClass()+"] "+t.getMessage());
- // And indicate the root cause
- setInitCause(rex, t);
- throw rex;
- }
-
- public static void throwIfUnchecked(Throwable t)
- {
- // If it's not checked, let's throw it as is
- if (t instanceof RuntimeException) {
- throw (RuntimeException) t;
- }
- if (t instanceof Error) {
- throw (Error) t;
- }
- }
-
- /**
- * This method is just added for convenience, and only to be used for
- * assertion style of exceptions. For errors that actually occur, method
- * with the string arg should be called instead.
- */
- public static void throwGenericInternal()
- {
- throwInternal(null);
- }
-
- public static void throwInternal(String msg)
- {
- if (msg == null) {
- msg = "[no description]";
- }
- throw new RuntimeException("Internal error: "+msg);
- }
-
- public static void setInitCause(Throwable newT, Throwable rootT)
- {
- /* [WSTX-110]: Better make sure we do not already have
- * a chained exception...
- */
- if (newT.getCause() == null) {
- newT.initCause(rootT);
- }
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/InternCache.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/InternCache.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/InternCache.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/InternCache.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,72 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.util.LinkedHashMap;
-import java.util.Map;
-
-/**
- * Singleton class that implements "fast intern" functionality, essentially
- * adding a layer that caches Strings that have been previously intern()ed,
- * but that probably shouldn't be added to symbol tables.
- * This is usually used by improving intern()ing of things like namespace
- * URIs.
- *
- * Note: that this class extends {@link LinkedHashMap} is an implementation
- * detail -- no code should ever directly call Map methods.
- */
-public final class InternCache extends LinkedHashMap //
-Note that some of more generic classes may eventually be moved to more
-generic packages under
- * One important note about usage is that two of the name components (prefix
- * and local name) HAVE to have been interned some way, as all comparisons
- * are done using identity comparison; whereas URI is NOT necessarily
- * interned.
- *
- * Note that the main reason this class is mutable -- unlike most key classes
- * -- is that this allows reusing key objects for access, as long as the code
- * using it knows ramifications of trying to modify a key that's used
- * in a data structure.
- *
- * Note, too, that the hash code is cached as this class is mostly used as
- * a Map key, and hash code is used a lot.
- */
-public final class PrefixedName
- implements Comparable // to allow alphabetic ordering
-{
- private String mPrefix, mLocalName;
-
- volatile int mHash = 0;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public PrefixedName(String prefix, String localName)
- {
- mLocalName = localName;
- mPrefix = (prefix != null && prefix.length() == 0) ?
- null : prefix;
- }
-
- public PrefixedName reset(String prefix, String localName)
- {
- mLocalName = localName;
- mPrefix = (prefix != null && prefix.length() == 0) ?
- null : prefix;
- mHash = 0;
- return this;
- }
-
- public static PrefixedName valueOf(QName n)
- {
- return new PrefixedName(n.getPrefix(), n.getLocalPart());
- }
-
- /*
- ///////////////////////////////////////////////////
- // Accessors:
- ///////////////////////////////////////////////////
- */
-
- public String getPrefix() { return mPrefix; }
-
- public String getLocalName() { return mLocalName; }
-
- /**
- * @return True, if this attribute name would result in a namespace
- * binding (ie. it's "xmlns" or starts with "xmlns:").
- */
- public boolean isaNsDeclaration()
- {
- if (mPrefix == null) {
- return mLocalName == "xmlns";
- }
- return mPrefix == "xmlns";
- }
-
- /**
- * Method used to check for xml reserved attribute names, like
- * "xml:space" and "xml:id".
- *
- * Note: it is assumed that the passed-in localName is also
- * interned.
- */
- public boolean isXmlReservedAttr(boolean nsAware, String localName)
- {
- if (nsAware) {
- if ("xml" == mPrefix) {
- return mLocalName == localName;
- }
- } else {
- if (mLocalName.length() == (4 + localName.length())) {
- return (mLocalName.startsWith("xml:")
- && mLocalName.endsWith(localName));
- }
- }
- return false;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Overridden standard methods:
- ///////////////////////////////////////////////////
- */
-
- public String toString()
- {
- if (mPrefix == null || mPrefix.length() == 0) {
- return mLocalName;
- }
- StringBuffer sb = new StringBuffer(mPrefix.length() + 1 + mLocalName.length());
- sb.append(mPrefix);
- sb.append(':');
- sb.append(mLocalName);
- return sb.toString();
- }
-
- public boolean equals(Object o)
- {
- if (o == this) {
- return true;
- }
- if (!(o instanceof PrefixedName)) { // also filters out nulls
- return false;
- }
- PrefixedName other = (PrefixedName) o;
-
- if (mLocalName != other.mLocalName) { // assumes equality
- return false;
- }
- return (mPrefix == other.mPrefix);
- }
-
- public int hashCode() {
- int hash = mHash;
-
- if (hash == 0) {
- hash = mLocalName.hashCode();
- if (mPrefix != null) {
- hash ^= mPrefix.hashCode();
- }
- mHash = hash;
- }
- return hash;
- }
-
- public int compareTo(Object o)
- {
- PrefixedName other = (PrefixedName) o;
-
- // First, by prefix, then by local name:
- String op = other.mPrefix;
-
- // Missing prefix is ordered before existing prefix
- if (op == null || op.length() == 0) {
- if (mPrefix != null && mPrefix.length() > 0) {
- return 1;
- }
- } else if (mPrefix == null || mPrefix.length() == 0) {
- return -1;
- } else {
- int result = mPrefix.compareTo(op);
- if (result != 0) {
- return result;
- }
- }
-
- return mLocalName.compareTo(other.mLocalName);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/SimpleCache.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/SimpleCache.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/SimpleCache.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/SimpleCache.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,59 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.util.*;
-
-/**
- * Simple Map implementation usable for caches where contents do not
- * expire, but where size needs to remain bounded.
- *
- * Note: we probably should use weak references, or something similar
- * to limit maximum memory usage. This could be implemented in many
- * ways, perhaps by using two areas: first, smaller one, with strong
- * refs, and secondary bigger one that uses soft references.
- */
-public final class SimpleCache
-{
- protected final LimitMap mItems;
-
- protected final int mMaxSize;
-
- public SimpleCache(int maxSize)
- {
- mItems = new LimitMap(maxSize);
- mMaxSize = maxSize;
- }
-
- public Object find(Object key) {
- return mItems.get(key);
- }
-
- public void add(Object key, Object value)
- {
- mItems.put(key, value);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Helper classes
- ///////////////////////////////////////////////////////////////////////
- */
-
- final static class LimitMap
- extends LinkedHashMap
- {
- private static final long serialVersionUID = 1L;
-
- protected final int mMaxSize;
-
- public LimitMap(int size)
- {
- super(size, 0.8f, true);
- // Let's not allow silly low values...
- mMaxSize = size;
- }
-
- public boolean removeEldestEntry(Map.Entry eldest) {
- return (size() >= mMaxSize);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/StringUtil.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/StringUtil.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/StringUtil.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/StringUtil.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,339 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.util.Collection;
-import java.util.Iterator;
-
-public final class StringUtil
-{
- final static char CHAR_SPACE = ' '; // 0x0020
- private final static char INT_SPACE = 0x0020;
-
- static String sLF = null;
-
- public static String getLF()
- {
- String lf = sLF;
- if (lf == null) {
- try {
- lf = (String) System.getProperty("line.separator");
- sLF = (lf == null) ? "\n" : lf;
- } catch (Throwable t) {
- // Doh.... whatever; most likely SecurityException
- sLF = lf = "\n";
- }
- }
- return lf;
- }
-
- public static void appendLF(StringBuffer sb) {
- sb.append(getLF());
- }
-
- public static String concatEntries(Collection coll, String sep, String lastSep) {
- if (lastSep == null) {
- lastSep = sep;
- }
- int len = coll.size();
- StringBuffer sb = new StringBuffer(16 + (len << 3));
- Iterator it = coll.iterator();
- int i = 0;
- while (it.hasNext()) {
- if (i == 0) {
- ;
- } else if (i == (len - 1)) {
- sb.append(lastSep);
- } else {
- sb.append(sep);
- }
- ++i;
- sb.append(it.next());
- }
- return sb.toString();
- }
-
- /**
- * Method that will check character array passed, and remove all
- * "extra" spaces (leading and trailing space), and normalize
- * other white space (more than one consequtive space character
- * replaced with a single space).
- *
- * NOTE: we only remove explicit space characters (char code 0x0020);
- * the reason being that other white space must have come from
- * non-normalizable sources, ie. via entity expansion, and is thus
- * not to be normalized
- *
- * @param buf Buffer that contains the String to check
- * @param origStart Offset of the first character of the text to check
- * in the buffer
- * @param origEnd Offset of the character following the last character
- * of the text (as per usual Java API convention)
- *
- * @return Normalized String, if any white space was removed or
- * normalized; null if no changes were necessary.
- */
- public static String normalizeSpaces(char[] buf, int origStart, int origEnd)
- {
- --origEnd;
-
- int start = origStart;
- int end = origEnd;
-
- // First let's trim start...
- while (start <= end && buf[start] == CHAR_SPACE) {
- ++start;
- }
- // Was it all empty?
- if (start > end) {
- return "";
- }
-
- /* Nope, need to trim from the end then (note: it's known that char
- * at index 'start' is not a space, at this point)
- */
- while (end > start && buf[end] == CHAR_SPACE) {
- --end;
- }
-
- /* Ok, may have changes or not: now need to normalize
- * intermediate duplicate spaces. We also now that the
- * first and last characters can not be spaces.
- */
- int i = start+1;
-
- while (i < end) {
- if (buf[i] == CHAR_SPACE) {
- if (buf[i+1] == CHAR_SPACE) {
- break;
- }
- // Nah; no hole for these 2 chars!
- i += 2;
- } else {
- ++i;
- }
- }
-
- // Hit the end?
- if (i >= end) {
- // Any changes?
- if (start == origStart && end == origEnd) {
- return null; // none
- }
- return new String(buf, start, (end-start)+1);
- }
-
- /* Nope, got a hole, need to constuct the damn thing. Shouldn't
- * happen too often... so let's just use StringBuffer()
- */
- StringBuffer sb = new StringBuffer(end-start); // can't be longer
- sb.append(buf, start, i-start); // won't add the starting space
-
- while (i <= end) {
- char c = buf[i++];
- if (c == CHAR_SPACE) {
- sb.append(CHAR_SPACE);
- // Need to skip dups
- while (true) {
- c = buf[i++];
- if (c != CHAR_SPACE) {
- sb.append(c);
- break;
- }
- }
- } else {
- sb.append(c);
- }
- }
-
- return sb.toString();
- }
-
- public static boolean isAllWhitespace(String str)
- {
- for (int i = 0, len = str.length(); i < len; ++i) {
- if (str.charAt(i) > CHAR_SPACE) {
- return false;
- }
- }
- return true;
- }
-
- public static boolean isAllWhitespace(char[] ch, int start, int len)
- {
- len += start;
- for (; start < len; ++start) {
- if (ch[start] > CHAR_SPACE) {
- return false;
- }
- }
- return true;
- }
-
- /**
- * Internal constant used to denote END-OF-STRING
- */
- private final static int EOS = 0x10000;
-
- /**
- * Method that implements a loose String compairon for encoding
- * Strings. It will work like {@link String#equalsIgnoreCase},
- * except that it will also ignore all hyphen, underscore and
- * space characters.
- */
- public static boolean equalEncodings(String str1, String str2)
- {
- final int len1 = str1.length();
- final int len2 = str2.length();
-
- // Need to loop completely over both Strings
- for (int i1 = 0, i2 = 0; i1 < len1 || i2 < len2; ) {
- int c1 = (i1 >= len1) ? EOS : str1.charAt(i1++);
- int c2 = (i2 >= len2) ? EOS : str2.charAt(i2++);
-
- // Can first do a quick comparison (usually they are equal)
- if (c1 == c2) {
- continue;
- }
-
- // if not equal, maybe there are WS/hyphen/underscores to skip
- while (c1 <= INT_SPACE || c1 == '_' || c1 == '-') {
- c1 = (i1 >= len1) ? EOS : str1.charAt(i1++);
- }
- while (c2 <= INT_SPACE || c2 == '_' || c2 == '-') {
- c2 = (i2 >= len2) ? EOS : str2.charAt(i2++);
- }
- // Ok, how about case differences, then?
- if (c1 != c2) {
- // If one is EOF, can't match (one is substring of the other)
- if (c1 == EOS || c2 == EOS) {
- return false;
- }
- if (c1 < 127) { // ascii is easy...
- if (c1 <= 'Z' && c1 >= 'A') {
- c1 = c1 + ('a' - 'A');
- }
- } else {
- c1 = Character.toLowerCase((char)c1);
- }
- if (c2 < 127) { // ascii is easy...
- if (c2 <= 'Z' && c2 >= 'A') {
- c2 = c2 + ('a' - 'A');
- }
- } else {
- c2 = Character.toLowerCase((char)c2);
- }
- if (c1 != c2) {
- return false;
- }
- }
- }
-
- // If we got this far, we are ok as long as we got through it all
- return true;
- }
-
- public static boolean encodingStartsWith(String enc, String prefix)
- {
- int len1 = enc.length();
- int len2 = prefix.length();
-
- int i1 = 0, i2 = 0;
-
- // Need to loop completely over both Strings
- while (i1 < len1 || i2 < len2) {
- int c1 = (i1 >= len1) ? EOS : enc.charAt(i1++);
- int c2 = (i2 >= len2) ? EOS : prefix.charAt(i2++);
-
- // Can first do a quick comparison (usually they are equal)
- if (c1 == c2) {
- continue;
- }
-
- // if not equal, maybe there are WS/hyphen/underscores to skip
- while (c1 <= CHAR_SPACE || c1 == '_' || c1 == '-') {
- c1 = (i1 >= len1) ? EOS : enc.charAt(i1++);
- }
- while (c2 <= CHAR_SPACE || c2 == '_' || c2 == '-') {
- c2 = (i2 >= len2) ? EOS : prefix.charAt(i2++);
- }
- // Ok, how about case differences, then?
- if (c1 != c2) {
- if (c2 == EOS) { // Prefix done, good!
- return true;
- }
- if (c1 == EOS) { // Encoding done, not good
- return false;
- }
- if (Character.toLowerCase((char)c1) != Character.toLowerCase((char)c2)) {
- return false;
- }
- }
- }
-
- // Ok, prefix was exactly the same as encoding... that's fine
- return true;
- }
-
- /**
- * Method that will remove all non-alphanumeric characters, and optionally
- * upper-case included letters, from the given String.
- */
- public static String trimEncoding(String str, boolean upperCase)
- {
- int i = 0;
- int len = str.length();
-
- // Let's first check if String is fine as is:
- for (; i < len; ++i) {
- char c = str.charAt(i);
- if (c <= CHAR_SPACE || !Character.isLetterOrDigit(c)) {
- break;
- }
- }
-
- if (i == len) {
- return str;
- }
-
- // Nope: have to trim it
- StringBuffer sb = new StringBuffer();
- if (i > 0) {
- sb.append(str.substring(0, i));
- }
- for (; i < len; ++i) {
- char c = str.charAt(i);
- if (c > CHAR_SPACE && Character.isLetterOrDigit(c)) {
- if (upperCase) {
- c = Character.toUpperCase(c);
- }
- sb.append(c);
- }
- }
-
- return sb.toString();
- }
-
- public static boolean matches(String str, char[] cbuf, int offset, int len)
- {
- if (str.length() != len) {
- return false;
- }
- for (int i = 0; i < len; ++i) {
- if (str.charAt(i) != cbuf[offset+i]) {
- return false;
- }
- }
- return true;
- }
-
- /**
- *
- * Note that it is assumed that any "weird" white space
- * (xml 1.1 LSEP and NEL) have been replaced by canonical
- * alternatives (linefeed for element content, regular space
- * for attributes)
- */
- public final static boolean isSpace(char c)
- {
- return ((int) c) <= 0x0020;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/StringVector.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/StringVector.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/StringVector.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/StringVector.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,244 +0,0 @@
-package com.ctc.wstx.util;
-
-/**
- * Data container similar {@link java.util.List} (from storage perspective),
- * but that can be used in multiple ways. For some uses it acts more like
- * type-safe String list/vector; for others as order associative list of
- * String-to-String mappings.
- */
-public final class StringVector
-{
- private String[] mStrings;
-
- private int mSize;
-
- /*
- ///////////////////////////////////////////////////////
- // Life-cycle:
- ///////////////////////////////////////////////////////
- */
-
- public StringVector(int initialCount) {
- mStrings = new String[initialCount];
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Basic accessors
- ///////////////////////////////////////////////////////
- */
-
- public int size() { return mSize; }
-
- public boolean isEmpty() { return mSize == 0; }
-
- public String getString(int index) {
- if (index < 0 || index >= mSize) {
- throw new IllegalArgumentException("Index "+index+" out of valid range; current size: "+mSize+".");
- }
- return mStrings[index];
- }
-
- public String getLastString() {
- if (mSize < 1) {
- throw new IllegalStateException("getLastString() called on empty StringVector.");
- }
- return mStrings[mSize-1];
- }
-
- public String[] getInternalArray() {
- return mStrings;
- }
-
- public String[] asArray() {
- String[] strs = new String[mSize];
- System.arraycopy(mStrings, 0, strs, 0, mSize);
- return strs;
- }
-
- public boolean containsInterned(String value) {
- String[] str = mStrings;
- for (int i = 0, len = mSize; i < len; ++i) {
- if (str[i] == value) {
- return true;
- }
- }
- return false;
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Mutators:
- ///////////////////////////////////////////////////////
- */
-
- public void addString(String str) {
- if (mSize == mStrings.length) {
- String[] old = mStrings;
- int oldSize = old.length;
- mStrings = new String[oldSize + (oldSize << 1)];
- System.arraycopy(old, 0, mStrings, 0, oldSize);
- }
- mStrings[mSize++] = str;
- }
-
- public void addStrings(String str1, String str2) {
- if ((mSize + 2) > mStrings.length) {
- String[] old = mStrings;
- int oldSize = old.length;
- mStrings = new String[oldSize + (oldSize << 1)];
- System.arraycopy(old, 0, mStrings, 0, oldSize);
- }
- mStrings[mSize] = str1;
- mStrings[mSize+1] = str2;
- mSize += 2;
- }
-
- public void setString(int index, String str) {
- mStrings[index] = str;
- }
-
- public void clear(boolean removeRefs) {
- if (removeRefs) {
- for (int i = 0, len = mSize; i < len; ++i) {
- mStrings[i] = null;
- }
- }
- mSize = 0;
- }
-
- public String removeLast() {
- String result = mStrings[--mSize];
- mStrings[mSize] = null;
- return result;
- }
-
- public void removeLast(int count) {
- while (--count >= 0) {
- mStrings[--mSize] = null;
- }
- }
-
- /*
- ///////////////////////////////////////////////////////
- // Specialized "map accessors":
- ///////////////////////////////////////////////////////
- */
-
- /**
- * Specialized access method; treats vector as a Map, with 2 Strings
- * per entry; first one being key, second value. Further, keys are
- * assumed to be canonicalized with passed in key (ie. either intern()ed,
- * or resolved from symbol table).
- * Starting from the
- * end (assuming even number of entries), tries to find an entry with
- * matching key, and if so, returns value.
- */
- public String findLastFromMap(String key) {
- int index = mSize;
- while ((index -= 2) >= 0) {
- if (mStrings[index] == key) {
- return mStrings[index+1];
- }
- }
- return null;
- }
-
- public String findLastNonInterned(String key)
- {
- int index = mSize;
- while ((index -= 2) >= 0) {
- String curr = mStrings[index];
- if (curr == key || (curr != null && curr.equals(key))) {
- return mStrings[index+1];
- }
- }
- return null;
- }
-
- public int findLastIndexNonInterned(String key) {
- int index = mSize;
- while ((index -= 2) >= 0) {
- String curr = mStrings[index];
- if (curr == key || (curr != null && curr.equals(key))) {
- return index;
- }
- }
- return -1;
- }
-
- public String findLastByValueNonInterned(String value) {
- for (int index = mSize-1; index > 0; index -= 2) {
- String currVal = mStrings[index];
- if (currVal == value || (currVal != null && currVal.equals(value))) {
- return mStrings[index-1];
- }
- }
- return null;
- }
-
- public int findLastIndexByValueNonInterned(String value) {
- for (int index = mSize-1; index > 0; index -= 2) {
- String currVal = mStrings[index];
- if (currVal == value || (currVal != null && currVal.equals(value))) {
- return index-1;
- }
- }
- return -1;
- }
-
- /*
- // Not needed any more
- public Iterator findAllByValueNonInterned(String value) {
- String first = null;
- ArrayList all = null;
- for (int index = mSize-1; index > 0; index -= 2) {
- String currVal = mStrings[index];
- if (currVal == value || (currVal != null && currVal.equals(value))) {
- if (first == null) {
- first = mStrings[index-1];
- } else {
- if (all == null) {
- all = new ArrayList();
- all.add(first);
- }
- all.add(mStrings[index-1]);
- }
- }
- }
- if (all != null) {
- return all.iterator();
- }
- if (first != null) {
- return new SingletonIterator(first);
- }
- return EmptyIterator.getInstance();
- }
- */
-
- /*
- ///////////////////////////////////////////////////////
- // Other methods
- ///////////////////////////////////////////////////////
- */
-
- public String toString() {
- StringBuffer sb = new StringBuffer(mSize * 16);
- sb.append("[(size = ");
- sb.append(mSize);
- sb.append(" ) ");
- for (int i = 0; i < mSize; ++i) {
- if (i > 0) {
- sb.append(", ");
- }
- sb.append('"');
- sb.append(mStrings[i]);
- sb.append('"');
-
- sb.append(" == ");
- sb.append(Integer.toHexString(System.identityHashCode(mStrings[i])));
- }
- sb.append(']');
- return sb.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/SymbolTable.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/SymbolTable.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/SymbolTable.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/SymbolTable.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,731 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.util;
-
-/**
- * This class is a kind of specialized type-safe Map, from char array to
- * String value. Specialization means that in addition to type-safety
- * and specific access patterns (key char array, Value optionally interned
- * String; values added on access if necessary), and that instances are
- * meant to be used concurrently, but by using well-defined mechanisms
- * to obtain such concurrently usable instances. Main use for the class
- * is to store symbol table information for things like compilers and
- * parsers; especially when number of symbols (keywords) is limited.
- *
- * For optimal performance, usage pattern should be one where matches
- * should be very common (esp. after "warm-up"), and as with most hash-based
- * maps/sets, that hash codes are uniformly distributed. Also, collisions
- * are slightly more expensive than with HashMap or HashSet, since hash codes
- * are not used in resolving collisions; that is, equals() comparison is
- * done with all symbols in same bucket index.
- * Usual usage pattern is to create a single "master" instance, and either
- * use that instance in sequential fashion, or to create derived "child"
- * instances, which after use, are asked to return possible symbol additions
- * to master instance. In either case benefit is that symbol table gets
- * initialized so that further uses are more efficient, as eventually all
- * symbols needed will already be in symbol table. At that point no more
- * Symbol String allocations are needed, nor changes to symbol table itself.
- *
- * Note that while individual SymbolTable instances are NOT thread-safe
- * (much like generic collection classes), concurrently used "child"
- * instances can be freely used without synchronization. However, using
- * master table concurrently with child instances can only be done if
- * access to master instance is read-only (ie. no modifications done).
- */
-
-public class SymbolTable {
-
- /**
- * Default initial table size; no need to make it miniscule, due
- * to couple of things: first, overhead of array reallocation
- * is significant,
- * and second, overhead of rehashing is also non-negligible.
- *
- * Let's use 128 as the default; it allows for up to 96 symbols,
- * and uses about 512 bytes on 32-bit machines.
- */
- protected static final int DEFAULT_TABLE_SIZE = 128;
-
- protected static final float DEFAULT_FILL_FACTOR = 0.75f;
-
- protected static final String EMPTY_STRING = "";
-
- /*
- ////////////////////////////////////////
- // Configuration:
- ////////////////////////////////////////
- */
-
- /**
- * Flag that determines whether Strings to be added need to be
- * interned before being added or not. Forcing intern()ing will add
- * some overhead when adding new Strings, but may be beneficial if such
- * Strings are generally used by other parts of system. Note that even
- * without interning, all returned String instances are guaranteed
- * to be comparable with equality (==) operator; it's just that such
- * guarantees are not made for Strings other classes return.
- */
- protected boolean mInternStrings;
-
- /*
- ////////////////////////////////////////
- // Actual symbol table data:
- ////////////////////////////////////////
- */
-
- /**
- * Primary matching symbols; it's expected most match occur from
- * here.
- */
- protected String[] mSymbols;
-
- /**
- * Overflow buckets; if primary doesn't match, lookup is done
- * from here.
- *
- * Note: Number of buckets is half of number of symbol entries, on
- * assumption there's less need for buckets.
- */
- protected Bucket[] mBuckets;
-
- /**
- * Current size (number of entries); needed to know if and when
- * rehash.
- */
- protected int mSize;
-
- /**
- * Limit that indicates maximum size this instance can hold before
- * it needs to be expanded and rehashed. Calculated using fill
- * factor passed in to constructor.
- */
- protected int mSizeThreshold;
-
- /**
- * Mask used to get index from hash values; equal to
- *
- * Note: while this method is synchronized, it is generally not
- * safe to both use makeChild/mergeChild, AND to use instance
- * actively. Instead, a separate 'root' instance should be used
- * on which only makeChild/mergeChild are called, but instance itself
- * is not used as a symbol table.
- */
- public synchronized SymbolTable makeChild() {
- return new SymbolTable(mInternStrings, mSymbols, mBuckets,
- mSize, mSizeThreshold, mIndexMask,
- mThisVersion+1);
- }
-
- /**
- * Method that allows contents of child table to potentially be
- * "merged in" with contents of this symbol table.
- *
- * Note that caller has to make sure symbol table passed in is
- * really a child or sibling of this symbol table.
- */
- public synchronized void mergeChild(SymbolTable child)
- {
- // Let's do a basic sanity check first:
- if (child.size() <= size()) { // nothing to add
- return;
- }
-
- // Okie dokie, let's get the data in!
- mSymbols = child.mSymbols;
- mBuckets = child.mBuckets;
- mSize = child.mSize;
- mSizeThreshold = child.mSizeThreshold;
- mIndexMask = child.mIndexMask;
- mThisVersion++; // to prevent other children from overriding
-
- // Dirty flag... well, let's just clear it, to force copying just
- // in case. Shouldn't really matter, for master tables.
- mDirty = false;
-
- /* However, we have to mark child as dirty, so that it will not
- * be modifying arrays we "took over" (since child may have
- * returned an updated table before it stopped fully using
- * the SymbolTable: for example, it may still use it for
- * parsing PI targets in epilog)
- */
- child.mDirty = false;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public API, configuration
- ////////////////////////////////////////////////////
- */
-
- public void setInternStrings(boolean state) {
- mInternStrings = state;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public API, generic accessors:
- ////////////////////////////////////////////////////
- */
-
- public int size() { return mSize; }
-
- public int version() { return mThisVersion; }
-
- public boolean isDirty() { return mDirty; }
-
- public boolean isDirectChildOf(SymbolTable t)
- {
- /* Actually, this doesn't really prove it is a child (would have to
- * use sequence number, or identityHash to really prove it), but
- * it's good enough if relationship is known to exist.
- */
- /* (for real check, one would need to child/descendant stuff; or
- * at least an identity hash... or maybe even just a _static_ global
- * counter for instances... maybe that would actually be worth
- * doing?)
- */
- if (mThisVersion == (t.mThisVersion + 1)) {
- return true;
- }
- return false;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public API, accessing symbols:
- ////////////////////////////////////////////////////
- */
-
- /**
- * Main access method; will check if actual symbol String exists;
- * if so, returns it; if not, will create, add and return it.
- *
- * @return The symbol matching String in input array
- */
- /*
- public String findSymbol(char[] buffer, int start, int len)
- {
- return findSymbol(buffer, start, len, calcHash(buffer, start, len));
- }
- */
-
- public String findSymbol(char[] buffer, int start, int len, int hash)
- {
- // Sanity check:
- if (len < 1) {
- return EMPTY_STRING;
- }
-
- hash &= mIndexMask;
-
- String sym = mSymbols[hash];
-
- // Optimal case; checking existing primary symbol for hash index:
- if (sym != null) {
- // Let's inline primary String equality checking:
- if (sym.length() == len) {
- int i = 0;
- do {
- if (sym.charAt(i) != buffer[start+i]) {
- break;
- }
- } while (++i < len);
- // Optimal case; primary match found
- if (i == len) {
- return sym;
- }
- }
- // How about collision bucket?
- Bucket b = mBuckets[hash >> 1];
- if (b != null) {
- sym = b.find(buffer, start, len);
- if (sym != null) {
- return sym;
- }
- }
- }
-
- // Need to expand?
- if (mSize >= mSizeThreshold) {
- rehash();
- /* Need to recalc hash; rare occurence (index mask has been
- * recalculated as part of rehash)
- */
- hash = calcHash(buffer, start, len) & mIndexMask;
- } else if (!mDirty) {
- // Or perhaps we need to do copy-on-write?
- copyArrays();
- mDirty = true;
- }
- ++mSize;
-
- String newSymbol = new String(buffer, start, len);
- if (mInternStrings) {
- newSymbol = newSymbol.intern();
- }
- // Ok; do we need to add primary entry, or a bucket?
- if (mSymbols[hash] == null) {
- mSymbols[hash] = newSymbol;
- } else {
- int bix = hash >> 1;
- mBuckets[bix] = new Bucket(newSymbol, mBuckets[bix]);
- }
-
- return newSymbol;
- }
-
- /**
- * Similar to {link #findSymbol}, but will not add passed in symbol
- * if it is not in symbol table yet.
- */
- public String findSymbolIfExists(char[] buffer, int start, int len, int hash)
- {
- // Sanity check:
- if (len < 1) {
- return EMPTY_STRING;
- }
- hash &= mIndexMask;
-
- String sym = mSymbols[hash];
- // Optimal case; checking existing primary symbol for hash index:
- if (sym != null) {
- // Let's inline primary String equality checking:
- if (sym.length() == len) {
- int i = 0;
- do {
- if (sym.charAt(i) != buffer[start+i]) {
- break;
- }
- } while (++i < len);
- // Optimal case; primary match found
- if (i == len) {
- return sym;
- }
- }
- // How about collision bucket?
- Bucket b = mBuckets[hash >> 1];
- if (b != null) {
- sym = b.find(buffer, start, len);
- if (sym != null) {
- return sym;
- }
- }
- }
- return null;
- }
-
- /**
- * Similar to to {@link #findSymbol(char[],int,int,int)}; used to either
- * do potentially cheap intern() (if table already has intern()ed version),
- * or to pre-populate symbol table with known values.
- */
- public String findSymbol(String str)
- {
- int len = str.length();
- // Sanity check:
- if (len < 1) {
- return EMPTY_STRING;
- }
-
- int index = calcHash(str) & mIndexMask;
- String sym = mSymbols[index];
-
- // Optimal case; checking existing primary symbol for hash index:
- if (sym != null) {
- // Let's inline primary String equality checking:
- if (sym.length() == len) {
- int i = 0;
- for (; i < len; ++i) {
- if (sym.charAt(i) != str.charAt(i)) {
- break;
- }
- }
- // Optimal case; primary match found
- if (i == len) {
- return sym;
- }
- }
- // How about collision bucket?
- Bucket b = mBuckets[index >> 1];
- if (b != null) {
- sym = b.find(str);
- if (sym != null) {
- return sym;
- }
- }
- }
-
- // Need to expand?
- if (mSize >= mSizeThreshold) {
- rehash();
- /* Need to recalc hash; rare occurence (index mask has been
- * recalculated as part of rehash)
- */
- index = calcHash(str) & mIndexMask;
- } else if (!mDirty) {
- // Or perhaps we need to do copy-on-write?
- copyArrays();
- mDirty = true;
- }
- ++mSize;
-
- if (mInternStrings) {
- str = str.intern();
- }
- // Ok; do we need to add primary entry, or a bucket?
- if (mSymbols[index] == null) {
- mSymbols[index] = str;
- } else {
- int bix = index >> 1;
- mBuckets[bix] = new Bucket(str, mBuckets[bix]);
- }
-
- return str;
- }
-
- /**
- * Implementation of a hashing method for variable length
- * Strings. Most of the time intention is that this calculation
- * is done by caller during parsing, not here; however, sometimes
- * it needs to be done for parsed "String" too.
- *
- * @param len Length of String; has to be at least 1 (caller guarantees
- * this pre-condition)
- */
- public static int calcHash(char[] buffer, int start, int len) {
- int hash = (int) buffer[0];
- for (int i = 1; i < len; ++i) {
- hash = (hash * 31) + (int) buffer[i];
- }
- return hash;
- }
-
- public static int calcHash(String key) {
- int hash = (int) key.charAt(0);
- for (int i = 1, len = key.length(); i < len; ++i) {
- hash = (hash * 31) + (int) key.charAt(i);
-
- }
- return hash;
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Internal methods
- //////////////////////////////////////////////////////////
- */
-
- /**
- * Method called when copy-on-write is needed; generally when first
- * change is made to a derived symbol table.
- */
- private void copyArrays() {
- String[] oldSyms = mSymbols;
- int size = oldSyms.length;
- mSymbols = new String[size];
- System.arraycopy(oldSyms, 0, mSymbols, 0, size);
- Bucket[] oldBuckets = mBuckets;
- size = oldBuckets.length;
- mBuckets = new Bucket[size];
- System.arraycopy(oldBuckets, 0, mBuckets, 0, size);
- }
-
- /**
- * Method called when size (number of entries) of symbol table grows
- * so big that load factor is exceeded. Since size has to remain
- * power of two, arrays will then always be doubled. Main work
- * is really redistributing old entries into new String/Bucket
- * entries.
- */
- private void rehash()
- {
- int size = mSymbols.length;
- int newSize = size + size;
- String[] oldSyms = mSymbols;
- Bucket[] oldBuckets = mBuckets;
- mSymbols = new String[newSize];
- mBuckets = new Bucket[newSize >> 1];
- // Let's update index mask, threshold, now (needed for rehashing)
- mIndexMask = newSize - 1;
- mSizeThreshold += mSizeThreshold;
-
- int count = 0; // let's do sanity check
-
- /* Need to do two loops, unfortunately, since spillover area is
- * only half the size:
- */
- for (int i = 0; i < size; ++i) {
- String symbol = oldSyms[i];
- if (symbol != null) {
- ++count;
- int index = calcHash(symbol) & mIndexMask;
- if (mSymbols[index] == null) {
- mSymbols[index] = symbol;
- } else {
- int bix = index >> 1;
- mBuckets[bix] = new Bucket(symbol, mBuckets[bix]);
- }
- }
- }
-
- size >>= 1;
- for (int i = 0; i < size; ++i) {
- Bucket b = oldBuckets[i];
- while (b != null) {
- ++count;
- String symbol = b.getSymbol();
- int index = calcHash(symbol) & mIndexMask;
- if (mSymbols[index] == null) {
- mSymbols[index] = symbol;
- } else {
- int bix = index >> 1;
- mBuckets[bix] = new Bucket(symbol, mBuckets[bix]);
- }
- b = b.getNext();
- }
- }
-
- if (count != mSize) {
- throw new IllegalStateException("Internal error on SymbolTable.rehash(): had "+mSize+" entries; now have "+count+".");
- }
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Test/debug support:
- //////////////////////////////////////////////////////////
- */
-
- public double calcAvgSeek() {
- int count = 0;
-
- for (int i = 0, len = mSymbols.length; i < len; ++i) {
- if (mSymbols[i] != null) {
- ++count;
- }
- }
-
- for (int i = 0, len = mBuckets.length; i < len; ++i) {
- Bucket b = mBuckets[i];
- int cost = 2;
- while (b != null) {
- count += cost;
- ++cost;
- b = b.getNext();
- }
- }
-
- return ((double) count) / ((double) mSize);
- }
-
- /*
- //////////////////////////////////////////////////////////
- // Bucket class
- //////////////////////////////////////////////////////////
- */
-
- /**
- * This class is a symbol table entry. Each entry acts as a node
- * in a linked list.
- */
- static final class Bucket {
- private final String mSymbol;
- private final Bucket mNext;
-
- public Bucket(String symbol, Bucket next) {
- mSymbol = symbol;
- mNext = next;
- }
-
- public String getSymbol() { return mSymbol; }
- public Bucket getNext() { return mNext; }
-
- public String find(char[] buf, int start, int len) {
- String sym = mSymbol;
- Bucket b = mNext;
-
- while (true) { // Inlined equality comparison:
- if (sym.length() == len) {
- int i = 0;
- do {
- if (sym.charAt(i) != buf[start+i]) {
- break;
- }
- } while (++i < len);
- if (i == len) {
- return sym;
- }
- }
- if (b == null) {
- break;
- }
- sym = b.getSymbol();
- b = b.getNext();
- }
- return null;
- }
-
- public String find(String str) {
- String sym = mSymbol;
- Bucket b = mNext;
-
- while (true) {
- if (sym.equals(str)) {
- return sym;
- }
- if (b == null) {
- break;
- }
- sym = b.getSymbol();
- b = b.getNext();
- }
- return null;
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextAccumulator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextAccumulator.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextAccumulator.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextAccumulator.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,75 +0,0 @@
-package com.ctc.wstx.util;
-
-/**
- * Simple utility class used to efficiently accumulate and concatenate
- * text passed in various forms
- */
-public final class TextAccumulator
-{
- private String mText = null;
-
- /* !!! JDK 1.5: when we can upgrade to Java 5, can convert
- * to using
- * Over time more and more cruft has accumulated here, mostly to
- * support efficient access to collected text. Since access is
- * easiest to do efficiently using callbacks, this class now needs
- * to known interfaces of SAX classes and validators.
- *
- * Notes about usage: for debugging purposes, it's suggested to use
- * {@link #toString} method, as opposed to
- * {@link #contentsAsArray} or {@link #contentsAsString}. Internally
- * resulting code paths may or may not be different, WRT caching.
- *
- * @author Tatu Saloranta
- */
-public final class TextBuffer
-{
- /* 23-Mar-2006, TSa: Memory buffer clearing is a significant overhead
- * for small documents, no need to use huge buffer -- it will expand
- * as necessary for larger docs, but commonly text segments just
- * aren't that long.
- */
- /**
- * Size of the first text segment buffer to allocate; need not contain
- * the biggest segment, since new ones will get allocated as needed.
- * However, it's sensible to use something that often is big enough
- * to contain segments.
- */
- final static int DEF_INITIAL_BUFFER_SIZE = 500; // 1k
-
- /**
- * We will also restrict maximum length of individual segments
- * to allocate (not including cases where we must return a single
- * segment). Value is somewhat arbitrary, let's use it so that
- * memory used is no more than 1/2 megabytes.
- */
- final static int MAX_SEGMENT_LENGTH = 256 * 1024;
-
- final static int INT_SPACE = 0x0020;
-
- // // // Configuration:
-
- private final ReaderConfig mConfig;
-
- // // // Shared read-only input buffer:
-
- /**
- * Shared input buffer; stored here in case some input can be returned
- * as is, without being copied to collector's own buffers. Note that
- * this is read-only for this Objet.
- */
- private char[] mInputBuffer;
-
- /**
- * Character offset of first char in input buffer; -1 to indicate
- * that input buffer currently does not contain any useful char data
- */
- private int mInputStart;
-
- /**
- * When using shared buffer, offset after the last character in
- * shared buffer
- */
- private int mInputLen;
-
- // // // Internal non-shared collector buffers:
-
- private boolean mHasSegments = false;
-
- /**
- * List of segments prior to currently active segment.
- */
- private ArrayList mSegments;
-
-
- // // // Currently used segment; not (yet) contained in mSegments
-
- /**
- * Amount of characters in segments in {@link mSegments}
- */
- private int mSegmentSize;
-
- private char[] mCurrentSegment;
-
- /**
- * Number of characters in currently active (last) segment
- */
- private int mCurrentSize;
-
- // // // Temporary caching for Objects to return
-
- /**
- * String that will be constructed when the whole contents are
- * needed; will be temporarily stored in case asked for again.
- */
- private String mResultString;
-
- private char[] mResultArray;
-
- // // // Canonical indentation objects (up to 32 spaces, 8 tabs)
-
- public final static int MAX_INDENT_SPACES = 32;
- public final static int MAX_INDENT_TABS = 8;
-
- // Let's add one more space at the end, for safety...
- private final static String sIndSpaces =
- // 123456789012345678901234567890123
- "\n ";
- private final static char[] sIndSpacesArray = sIndSpaces.toCharArray();
- private final static String[] sIndSpacesStrings = new String[sIndSpacesArray.length];
-
- private final static String sIndTabs =
- // 1 2 3 4 5 6 7 8 9
- "\n\t\t\t\t\t\t\t\t\t";
- private final static char[] sIndTabsArray = sIndTabs.toCharArray();
- private final static String[] sIndTabsStrings = new String[sIndTabsArray.length];
-
- /*
- //////////////////////////////////////////////
- // Life-cycle
- //////////////////////////////////////////////
- */
-
- private TextBuffer(ReaderConfig cfg)
- {
- mConfig = cfg;
- }
-
- public static TextBuffer createRecyclableBuffer(ReaderConfig cfg)
- {
- return new TextBuffer(cfg);
- }
-
- public static TextBuffer createTemporaryBuffer()
- {
- return new TextBuffer(null);
- }
-
- /**
- * Method called to indicate that the underlying buffers should now
- * be recycled if they haven't yet been recycled. Although caller
- * can still use this text buffer, it is not advisable to call this
- * method if that is likely, since next time a buffer is needed,
- * buffers need to reallocated.
- * Note: calling this method automatically also clears contents
- * of the buffer.
- */
- public void recycle(boolean force)
- {
- if (mConfig != null && mCurrentSegment != null) {
- if (force) {
- /* If we are allowed to wipe out all existing data, it's
- * quite easy; we'll just wipe out contents, and return
- * biggest buffer:
- */
- resetWithEmpty();
- } else {
- /* But if there's non-shared data (ie. buffer is still
- * in use), can't return it yet:
- */
- if (mInputStart < 0 && (mSegmentSize + mCurrentSize) > 0) {
- return;
- }
- // If no data (or only shared data), can continue
- if (mSegments != null && mSegments.size() > 0) {
- // No need to use anything from list, curr segment not null
- mSegments.clear();
- mSegmentSize = 0;
- }
- }
-
- char[] buf = mCurrentSegment;
- mCurrentSegment = null;
- mConfig.freeMediumCBuffer(buf);
- }
- }
-
- /**
- * Method called to clear out any content text buffer may have, and
- * initializes buffer to use non-shared data.
- */
- public void resetWithEmpty()
- {
- mInputBuffer = null;
- mInputStart = -1; // indicates shared buffer not used
- mInputLen = 0;
-
- mResultString = null;
- mResultArray = null;
-
- // And then reset internal input buffers, if necessary:
- if (mHasSegments) {
- clearSegments();
- }
- mCurrentSize = 0;
- }
-
- /**
- * Similar to {@link #resetWithEmpty}, but actively marks current
- * text content to be empty string (whereas former method leaves
- * content as undefined).
- */
- public void resetWithEmptyString()
- {
- mInputBuffer = null;
- mInputStart = -1; // indicates shared buffer not used
- mInputLen = 0;
- mResultString = "";
- mResultArray = null;
- if (mHasSegments) {
- clearSegments();
- }
- mCurrentSize = 0;
- }
-
- /**
- * Method called to initialize the buffer with a shared copy of data;
- * this means that buffer will just have pointers to actual data. It
- * also means that if anything is to be appended to the buffer, it
- * will first have to unshare it (make a local copy).
- */
- public void resetWithShared(char[] buf, int start, int len)
- {
- // Let's first mark things we need about input buffer
- mInputBuffer = buf;
- mInputStart = start;
- mInputLen = len;
-
- // Then clear intermediate values, if any:
- mResultString = null;
- mResultArray = null;
-
- // And then reset internal input buffers, if necessary:
- if (mHasSegments) {
- clearSegments();
- }
- }
-
- public void resetWithCopy(char[] buf, int start, int len)
- {
- mInputBuffer = null;
- mInputStart = -1; // indicates shared buffer not used
- mInputLen = 0;
-
- mResultString = null;
- mResultArray = null;
-
- // And then reset internal input buffers, if necessary:
- if (mHasSegments) {
- clearSegments();
- } else {
- if (mCurrentSegment == null) {
- mCurrentSegment = allocBuffer(len);
- }
- mCurrentSize = mSegmentSize = 0;
- }
- append(buf, start, len);
- }
-
- /**
- * Method called to make sure there is a non-shared segment to use, without
- * appending any content yet.
- */
- public void resetInitialized()
- {
- resetWithEmpty();
- if (mCurrentSegment == null) {
- mCurrentSegment = allocBuffer(0);
- }
- }
-
- private final char[] allocBuffer(int needed)
- {
- int size = Math.max(needed, DEF_INITIAL_BUFFER_SIZE);
- char[] buf = null;
- if (mConfig != null) {
- buf = mConfig.allocMediumCBuffer(size);
- if (buf != null) {
- return buf;
- }
- }
- return new char[size];
- }
-
- private final void clearSegments()
- {
- mHasSegments = false;
- /* Since the current segment should be the biggest one
- * (as we allocate 50% bigger each time), let's retain it,
- * and clear others
- */
- mSegments.clear();
- mCurrentSize = mSegmentSize = 0;
- }
-
- public void resetWithIndentation(int indCharCount, char indChar)
- {
- mInputStart = 0;
- mInputLen = indCharCount+1;
- String text;
- if (indChar == '\t') { // tabs?
- mInputBuffer = sIndTabsArray;
- text = sIndTabsStrings[indCharCount];
- if (text == null) {
- sIndTabsStrings[indCharCount] = text = sIndTabs.substring(0, mInputLen);
- }
- } else { // nope, spaces (should assert indChar?)
- mInputBuffer = sIndSpacesArray;
- text = sIndSpacesStrings[indCharCount];
- if (text == null) {
- sIndSpacesStrings[indCharCount] = text = sIndSpaces.substring(0, mInputLen);
- }
- }
- mResultString = text;
-
- /* Should not need the explicit non-shared array; no point in
- * pre-populating it (can be changed if this is not true)
- */
- mResultArray = null;
-
- // And then reset internal input buffers, if necessary:
- if (mSegments != null && mSegments.size() > 0) {
- mSegments.clear();
- mCurrentSize = mSegmentSize = 0;
- }
- }
-
- /*
- //////////////////////////////////////////////
- // Accessors for implementing StAX interface:
- //////////////////////////////////////////////
- */
-
- /**
- * @return Number of characters currently stored by this collector
- */
- public int size() {
- if (mInputStart >= 0) { // shared copy from input buf
- return mInputLen;
- }
- // local segmented buffers
- return mSegmentSize + mCurrentSize;
- }
-
- public int getTextStart()
- {
- /* Only shared input buffer can have non-zero offset; buffer
- * segments start at 0, and if we have to create a combo buffer,
- * that too will start from beginning of the buffer
- */
- return (mInputStart >= 0) ? mInputStart : 0;
- }
-
- public char[] getTextBuffer()
- {
- // Are we just using shared input buffer?
- if (mInputStart >= 0) {
- return mInputBuffer;
- }
- // Nope; but does it fit in just one segment?
- if (mSegments == null || mSegments.size() == 0) {
- return mCurrentSegment;
- }
- // Nope, need to have/create a non-segmented array and return it
- return contentsAsArray();
- }
-
- /*
- /////////////////////////////////////////////////
- // Accessors for implementing StAX2 Typed access
- /////////////////////////////////////////////////
- */
-
- /**
- * Generic pass-through method which call given decoder
- * with accumulated data
- */
- public void decode(TypedValueDecoder tvd)
- throws IllegalArgumentException
- {
- char[] buf;
- int start, end;
-
- if (mInputStart >= 0) { // shared buffer, common case
- buf = mInputBuffer;
- start = mInputStart;
- end = start + mInputLen;
- } else {
- buf = getTextBuffer();
- start = 0;
- end = mSegmentSize + mCurrentSize;
- }
-
- // Need to trim first
- while (true) {
- if (start >= end) {
- tvd.handleEmptyValue();
- return;
- }
- if (!StringUtil.isSpace(buf[start])) {
- break;
- }
- ++start;
- }
- // Trailing space?
- while (--end > start && StringUtil.isSpace(buf[end])) { }
- tvd.decode(buf, start, end+1);
- }
-
- /**
- * Pass-through decode method called to find find the next token,
- * decode it, and repeat the process as long as there are more
- * tokens and the array decoder accepts more entries.
- * All tokens processed will be "consumed", such that they will
- * not be visible via buffer.
- *
- * @return Number of tokens decoded; 0 means that no (more) tokens
- * were found from this buffer.
- */
- public int decodeElements(TypedArrayDecoder tad, InputProblemReporter rep)
- throws TypedXMLStreamException
- {
- int count = 0;
-
- /* First: for simplicity, we require a single flat buffer to
- * decode from. Second: to be able to update start location
- * (to keep track of what's available), we need to fake that
- * we are using a shared buffer (since that has offset)
- */
- if (mInputStart < 0) {
- if (mHasSegments) {
- mInputBuffer = buildResultArray();
- mInputLen = mInputBuffer.length;
- // let's also clear segments since they are not needed any more
- clearSegments();
- } else {
- // just current buffer, easier to fake
- mInputBuffer = mCurrentSegment;
- mInputLen = mCurrentSize;
- }
- mInputStart = 0;
- }
-
- // And then let's decode
- int ptr = mInputStart;
- final int end = ptr + mInputLen;
- final char[] buf = mInputBuffer;
- int start = ptr;
-
- try {
- decode_loop:
- while (ptr < end) {
- // First, any space to skip?
- while (buf[ptr] <= INT_SPACE) {
- if (++ptr >= end) {
- break decode_loop;
- }
- }
- // Then let's figure out non-space char (token)
- start = ptr;
- ++ptr;
- while (ptr < end && buf[ptr] > INT_SPACE) {
- ++ptr;
- }
- ++count;
- int tokenEnd = ptr;
- ++ptr; // to skip trailing space (or, beyond end)
- // And there we have it
- if (tad.decodeValue(buf, start, tokenEnd)) {
- break;
- }
- }
- } catch (IllegalArgumentException iae) {
- // Need to convert to a checked stream exception
-
- /* Hmmh. This is probably not an accurate location... but
- * we can't do much better as content we have has been
- * normalized already.
- */
- Location loc = rep.getLocation();
- // -1 to move it back after being advanced earlier (to skip trailing space)
- String lexical = new String(buf, start, (ptr-start-1));
- throw new TypedXMLStreamException(lexical, iae.getMessage(), loc, iae);
- } finally {
- mInputStart = ptr;
- mInputLen = end-ptr;
- }
- return count;
- }
-
- /**
- * Method that needs to be called to configure given base64 decoder
- * with textual contents collected by this buffer.
- *
- * @param dec Decoder that will need data
- * @param firstChunk Whether this is the first segment fed or not;
- * if it is, state needs to be fullt reset; if not, only partially.
- */
- public void initBinaryChunks(Base64Variant v, CharArrayBase64Decoder dec, boolean firstChunk)
- {
- if (mInputStart < 0) { // non-shared
- dec.init(v, firstChunk, mCurrentSegment, 0, mCurrentSize, mSegments);
- } else { // shared
- dec.init(v, firstChunk, mInputBuffer, mInputStart, mInputLen, null);
- }
- }
-
- /*
- //////////////////////////////////////////////
- // Accessors:
- //////////////////////////////////////////////
- */
-
- public String contentsAsString()
- {
- if (mResultString == null) {
- // Has array been requested? Can make a shortcut, if so:
- if (mResultArray != null) {
- mResultString = new String(mResultArray);
- } else {
- // Do we use shared array?
- if (mInputStart >= 0) {
- if (mInputLen < 1) {
- return (mResultString = "");
- }
- mResultString = new String(mInputBuffer, mInputStart, mInputLen);
- } else { // nope... need to copy
- // But first, let's see if we have just one buffer
- int segLen = mSegmentSize;
- int currLen = mCurrentSize;
-
- if (segLen == 0) { // yup
- mResultString = (currLen == 0) ? "" : new String(mCurrentSegment, 0, currLen);
- } else { // no, need to combine
- StringBuffer sb = new StringBuffer(segLen + currLen);
- // First stored segments
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] curr = (char[]) mSegments.get(i);
- sb.append(curr, 0, curr.length);
- }
- }
- // And finally, current segment:
- sb.append(mCurrentSegment, 0, mCurrentSize);
- mResultString = sb.toString();
- }
- }
- }
- }
- return mResultString;
- }
-
- /**
- * Similar to {@link #contentsAsString}, but constructs a StringBuffer
- * for further appends.
- *
- * @param extraSpace Number of extra characters to preserve in StringBuffer
- * beyond space immediately needed to hold the contents
- */
- public StringBuffer contentsAsStringBuffer(int extraSpace)
- {
- if (mResultString != null) {
- return new StringBuffer(mResultString);
- }
- if (mResultArray != null) {
- StringBuffer sb = new StringBuffer(mResultArray.length + extraSpace);
- sb.append(mResultArray, 0, mResultArray.length);
- return sb;
- }
- if (mInputStart >= 0) { // shared array
- if (mInputLen < 1) {
- return new StringBuffer();
- }
- StringBuffer sb = new StringBuffer(mInputLen + extraSpace);
- sb.append(mInputBuffer, mInputStart, mInputLen);
- return sb;
- }
- int segLen = mSegmentSize;
- int currLen = mCurrentSize;
-
- StringBuffer sb = new StringBuffer(segLen + currLen + extraSpace);
- // First stored segments
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] curr = (char[]) mSegments.get(i);
- sb.append(curr, 0, curr.length);
- }
- }
- // And finally, current segment:
- sb.append(mCurrentSegment, 0, currLen);
- return sb;
- }
-
- public void contentsToStringBuffer(StringBuffer sb)
- {
- if (mResultString != null) {
- sb.append(mResultString);
- } else if (mResultArray != null) {
- sb.append(mResultArray);
- } else if (mInputStart >= 0) { // shared array
- if (mInputLen > 0) {
- sb.append(mInputBuffer, mInputStart, mInputLen);
- }
- } else {
- // First stored segments
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] curr = (char[]) mSegments.get(i);
- sb.append(curr, 0, curr.length);
- }
- }
- // And finally, current segment:
- sb.append(mCurrentSegment, 0, mCurrentSize);
- }
- }
-
- public char[] contentsAsArray()
- {
- char[] result = mResultArray;
- if (result == null) {
- mResultArray = result = buildResultArray();
- }
- return result;
- }
-
- public int contentsToArray(int srcStart, char[] dst, int dstStart, int len) {
- // Easy to copy from shared buffer:
- if (mInputStart >= 0) {
- int amount = mInputLen - srcStart;
- if (amount > len) {
- amount = len;
- } else if (amount < 0) {
- amount = 0;
- }
- if (amount > 0) {
- System.arraycopy(mInputBuffer, mInputStart+srcStart,
- dst, dstStart, amount);
- }
- return amount;
- }
-
- /* Could also check if we have array, but that'd only help with
- * braindead clients that get full array first, then segments...
- * which hopefully aren't that common
- */
- // Copying from segmented array is bit more involved:
- int totalAmount = 0;
- if (mSegments != null) {
- for (int i = 0, segc = mSegments.size(); i < segc; ++i) {
- char[] segment = (char[]) mSegments.get(i);
- int segLen = segment.length;
- int amount = segLen - srcStart;
- if (amount < 1) { // nothing from this segment?
- srcStart -= segLen;
- continue;
- }
- if (amount >= len) { // can get rest from this segment?
- System.arraycopy(segment, srcStart, dst, dstStart, len);
- return (totalAmount + len);
- }
- // Can get some from this segment, offset becomes zero:
- System.arraycopy(segment, srcStart, dst, dstStart, amount);
- totalAmount += amount;
- dstStart += amount;
- len -= amount;
- srcStart = 0;
- }
- }
-
- // Need to copy anything from last segment?
- if (len > 0) {
- int maxAmount = mCurrentSize - srcStart;
- if (len > maxAmount) {
- len = maxAmount;
- }
- if (len > 0) { // should always be true
- System.arraycopy(mCurrentSegment, srcStart, dst, dstStart, len);
- totalAmount += len;
- }
- }
-
- return totalAmount;
- }
-
- /**
- * Method that will stream contents of this buffer into specified
- * Writer.
- */
- public int rawContentsTo(Writer w)
- throws IOException
- {
- // Let's first see if we have created helper objects:
- if (mResultArray != null) {
- w.write(mResultArray);
- return mResultArray.length;
- }
- if (mResultString != null) {
- w.write(mResultString);
- return mResultString.length();
- }
-
- // Do we use shared array?
- if (mInputStart >= 0) {
- if (mInputLen > 0) {
- w.write(mInputBuffer, mInputStart, mInputLen);
- }
- return mInputLen;
- }
- // Nope, need to do full segmented output
- int rlen = 0;
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] ch = (char[]) mSegments.get(i);
- w.write(ch);
- rlen += ch.length;
- }
- }
- if (mCurrentSize > 0) {
- w.write(mCurrentSegment, 0, mCurrentSize);
- rlen += mCurrentSize;
- }
- return rlen;
- }
-
- public Reader rawContentsViaReader()
- throws IOException
- {
- // Let's first see if we have created helper objects:
- if (mResultArray != null) {
- return new CharArrayReader(mResultArray);
- }
- if (mResultString != null) {
- return new StringReader(mResultString);
- }
-
- // Do we use shared array?
- if (mInputStart >= 0) {
- if (mInputLen > 0) {
- return new CharArrayReader(mInputBuffer, mInputStart, mInputLen);
- }
- return new StringReader("");
- }
- // or maybe it's all in the current segment
- if (mSegments == null || mSegments.size() == 0) {
- return new CharArrayReader(mCurrentSegment, 0, mCurrentSize);
- }
- // Nope, need to do full segmented output
- return new BufferReader(mSegments, mCurrentSegment, mCurrentSize);
- }
-
- public boolean isAllWhitespace()
- {
- if (mInputStart >= 0) { // using single shared buffer?
- char[] buf = mInputBuffer;
- int i = mInputStart;
- int last = i + mInputLen;
- for (; i < last; ++i) {
- if (buf[i] > INT_SPACE) {
- return false;
- }
- }
- return true;
- }
-
- // Nope, need to do full segmented output
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] buf = (char[]) mSegments.get(i);
- for (int j = 0, len2 = buf.length; j < len2; ++j) {
- if (buf[j] > INT_SPACE) {
- return false;
- }
- }
- }
- }
-
- char[] buf = mCurrentSegment;
- for (int i = 0, len = mCurrentSize; i < len; ++i) {
- if (buf[i] > INT_SPACE) {
- return false;
- }
- }
- return true;
- }
-
- /**
- * Method that can be used to check if the contents of the buffer end
- * in specified String.
- *
- * @return True if the textual content buffer contains ends with the
- * specified String; false otherwise
- */
- public boolean endsWith(String str)
- {
- /* Let's just play this safe; should seldom if ever happen...
- * and because of that, can be sub-optimal, performancewise, to
- * alternatives.
- */
- if (mInputStart >= 0) {
- unshare(16);
- }
-
- int segIndex = (mSegments == null) ? 0 : mSegments.size();
- int inIndex = str.length() - 1;
- char[] buf = mCurrentSegment;
- int bufIndex = mCurrentSize-1;
-
- while (inIndex >= 0) {
- if (str.charAt(inIndex) != buf[bufIndex]) {
- return false;
- }
- if (--inIndex == 0) {
- break;
- }
- if (--bufIndex < 0) {
- if (--segIndex < 0) { // no more data?
- return false;
- }
- buf = (char[]) mSegments.get(segIndex);
- bufIndex = buf.length-1;
- }
- }
-
- return true;
- }
-
- /**
- * Note: it is assumed that this method is not used often enough to
- * be a bottleneck, or for long segments. Based on this, it is optimized
- * for common simple cases where there is only one single character
- * segment to use; fallback for other cases is to create such segment.
- */
- public boolean equalsString(String str)
- {
- int expLen = str.length();
-
- // First the easy check; if we have a shared buf:
- if (mInputStart >= 0) {
- if (mInputLen != expLen) {
- return false;
- }
- for (int i = 0; i < expLen; ++i) {
- if (str.charAt(i) != mInputBuffer[mInputStart+i]) {
- return false;
- }
- }
- return true;
- }
-
- // Otherwise, segments:
- if (expLen != size()) {
- return false;
- }
- char[] seg;
- if (mSegments == null || mSegments.size() == 0) {
- // just one segment, still easy
- seg = mCurrentSegment;
- } else {
- /* Ok; this is the sub-optimal case. Could obviously juggle through
- * segments, but probably not worth the hassle, we seldom if ever
- * get here...
- */
- seg = contentsAsArray();
- }
-
- for (int i = 0; i < expLen; ++i) {
- if (seg[i] != str.charAt(i)) {
- return false;
- }
- }
- return true;
- }
-
- /*
- //////////////////////////////////////////////
- // Access using SAX handlers:
- //////////////////////////////////////////////
- */
-
- public void fireSaxCharacterEvents(ContentHandler h)
- throws SAXException
- {
- if (mResultArray != null) { // already have single array?
- h.characters(mResultArray, 0, mResultArray.length);
- } else if (mInputStart >= 0) { // sharing input buffer?
- h.characters(mInputBuffer, mInputStart, mInputLen);
- } else {
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] ch = (char[]) mSegments.get(i);
- h.characters(ch, 0, ch.length);
- }
- }
- if (mCurrentSize > 0) {
- h.characters(mCurrentSegment, 0, mCurrentSize);
- }
- }
- }
-
- public void fireSaxSpaceEvents(ContentHandler h)
- throws SAXException
- {
- if (mResultArray != null) { // only happens for indentation
- h.ignorableWhitespace(mResultArray, 0, mResultArray.length);
- } else if (mInputStart >= 0) { // sharing input buffer?
- h.ignorableWhitespace(mInputBuffer, mInputStart, mInputLen);
- } else {
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] ch = (char[]) mSegments.get(i);
- h.ignorableWhitespace(ch, 0, ch.length);
- }
- }
- if (mCurrentSize > 0) {
- h.ignorableWhitespace(mCurrentSegment, 0, mCurrentSize);
- }
- }
- }
-
- public void fireSaxCommentEvent(LexicalHandler h)
- throws SAXException
- {
- // Comment can not be split, so may need to combine the array
- if (mResultArray != null) { // only happens for indentation
- h.comment(mResultArray, 0, mResultArray.length);
- } else if (mInputStart >= 0) { // sharing input buffer?
- h.comment(mInputBuffer, mInputStart, mInputLen);
- } else if (mSegments != null && mSegments.size() > 0) {
- char[] ch = contentsAsArray();
- h.comment(ch, 0, ch.length);
- } else {
- h.comment(mCurrentSegment, 0, mCurrentSize);
- }
- }
-
- public void fireDtdCommentEvent(DTDEventListener l)
- {
- // Comment can not be split, so may need to combine the array
- if (mResultArray != null) { // only happens for indentation
- l.dtdComment(mResultArray, 0, mResultArray.length);
- } else if (mInputStart >= 0) { // sharing input buffer?
- l.dtdComment(mInputBuffer, mInputStart, mInputLen);
- } else if (mSegments != null && mSegments.size() > 0) {
- char[] ch = contentsAsArray();
- l.dtdComment(ch, 0, ch.length);
- } else {
- l.dtdComment(mCurrentSegment, 0, mCurrentSize);
- }
- }
-
- /*
- //////////////////////////////////////////////
- // Support for validation
- //////////////////////////////////////////////
- */
-
- public void validateText(XMLValidator vld, boolean lastSegment)
- throws XMLStreamException
- {
- // Shared buffer? Let's just pass that
- if (mInputStart >= 0) {
- vld.validateText(mInputBuffer, mInputStart, mInputStart + mInputLen, lastSegment);
- } else {
- /* Otherwise, can either create a combine buffer, or construct
- * a String. While former could be more efficient, let's do latter
- * for now since current validator implementations work better
- * with Strings.
- */
- vld.validateText(contentsAsString(), lastSegment);
- }
- }
-
- /*
- //////////////////////////////////////////////
- // Public mutators:
- //////////////////////////////////////////////
- */
-
- /**
- * Method called to make sure that buffer is not using shared input
- * buffer; if it is, it will copy such contents to private buffer.
- */
- public void ensureNotShared() {
- if (mInputStart >= 0) {
- unshare(16);
- }
- }
-
- public void append(char c) {
- // Using shared buffer so far?
- if (mInputStart >= 0) {
- unshare(16);
- }
- mResultString = null;
- mResultArray = null;
- // Room in current segment?
- char[] curr = mCurrentSegment;
- if (mCurrentSize >= curr.length) {
- expand(1);
- curr = mCurrentSegment;
- }
- curr[mCurrentSize++] = c;
- }
-
- public void append(char[] c, int start, int len)
- {
- // Can't append to shared buf (sanity check)
- if (mInputStart >= 0) {
- unshare(len);
- }
- mResultString = null;
- mResultArray = null;
-
- // Room in current segment?
- char[] curr = mCurrentSegment;
- int max = curr.length - mCurrentSize;
-
- if (max >= len) {
- System.arraycopy(c, start, curr, mCurrentSize, len);
- mCurrentSize += len;
- } else {
- // No room for all, need to copy part(s):
- if (max > 0) {
- System.arraycopy(c, start, curr, mCurrentSize, max);
- start += max;
- len -= max;
- }
- /* And then allocate new segment; we are guaranteed to now
- * have enough room in segment.
- */
- expand(len); // note: curr != mCurrentSegment after this
- System.arraycopy(c, start, mCurrentSegment, 0, len);
- mCurrentSize = len;
- }
- }
-
- public void append(String str)
- {
- // Can't append to shared buf (sanity check)
- int len = str.length();
- if (mInputStart >= 0) {
- unshare(len);
- }
- mResultString = null;
- mResultArray = null;
-
- // Room in current segment?
- char[] curr = mCurrentSegment;
- int max = curr.length - mCurrentSize;
- if (max >= len) {
- str.getChars(0, len, curr, mCurrentSize);
- mCurrentSize += len;
- } else {
- // No room for all, need to copy part(s):
- if (max > 0) {
- str.getChars(0, max, curr, mCurrentSize);
- len -= max;
- }
- /* And then allocate new segment; we are guaranteed to now
- * have enough room in segment.
- */
- expand(len);
- str.getChars(max, max+len, mCurrentSegment, 0);
- mCurrentSize = len;
- }
- }
-
- /*
- //////////////////////////////////////////////
- // Raw access, for high-performance use:
- //////////////////////////////////////////////
- */
-
- public char[] getCurrentSegment()
- {
- /* Since the intention of the caller is to directly add stuff into
- * buffers, we should NOT have anything in shared buffer... ie. may
- * need to unshare contents.
- */
- if (mInputStart >= 0) {
- unshare(1);
- } else {
- char[] curr = mCurrentSegment;
- if (curr == null) {
- mCurrentSegment = allocBuffer(0);
- } else if (mCurrentSize >= curr.length) {
- // Plus, we better have room for at least one more char
- expand(1);
- }
- }
- return mCurrentSegment;
- }
-
- public int getCurrentSegmentSize() {
- return mCurrentSize;
- }
-
- public void setCurrentLength(int len) {
- mCurrentSize = len;
- }
-
- public char[] finishCurrentSegment()
- {
- if (mSegments == null) {
- mSegments = new ArrayList();
- }
- mHasSegments = true;
- mSegments.add(mCurrentSegment);
- int oldLen = mCurrentSegment.length;
- mSegmentSize += oldLen;
- char[] curr = new char[calcNewSize(oldLen)];
- mCurrentSize = 0;
- mCurrentSegment = curr;
- return curr;
- }
-
- /**
- * Method used to determine size of the next segment to
- * allocate to contain textual content.
- */
- private int calcNewSize(int latestSize)
- {
- // Let's grow segments by 50%, when over 8k
- int incr = (latestSize < 8000) ? latestSize : (latestSize >> 1);
- int size = latestSize + incr;
- // but let's not create too big chunks
- return Math.min(size, MAX_SEGMENT_LENGTH);
- }
-
- /*
- //////////////////////////////////////////////
- // Standard methods:
- //////////////////////////////////////////////
- */
-
- /**
- * Note: calling this method may not be as efficient as calling
- * {@link #contentsAsString}, since it's not guaranteed that resulting
- * String is cached.
- */
- public String toString() {
- return contentsAsString();
- }
-
- /*
- //////////////////////////////////////////////
- // Internal methods:
- //////////////////////////////////////////////
- */
-
- /**
- * Method called if/when we need to append content when we have been
- * initialized to use shared buffer.
- */
- public void unshare(int needExtra)
- {
- int len = mInputLen;
- mInputLen = 0;
- char[] inputBuf = mInputBuffer;
- mInputBuffer = null;
- int start = mInputStart;
- mInputStart = -1;
-
- // Is buffer big enough, or do we need to reallocate?
- int needed = len+needExtra;
- if (mCurrentSegment == null || needed > mCurrentSegment.length) {
- mCurrentSegment = allocBuffer(needed);
- }
- if (len > 0) {
- System.arraycopy(inputBuf, start, mCurrentSegment, 0, len);
- }
- mSegmentSize = 0;
- mCurrentSize = len;
- }
-
- /**
- * Method called when current segment is full, to allocate new
- * segment.
- *
- * @param roomNeeded Number of characters that the resulting
- * new buffer must have
- */
- private void expand(int roomNeeded)
- {
- // First, let's move current segment to segment list:
- if (mSegments == null) {
- mSegments = new ArrayList();
- }
- char[] curr = mCurrentSegment;
- mHasSegments = true;
- mSegments.add(curr);
- int oldLen = curr.length;
- mSegmentSize += oldLen;
- int newSize = Math.max(roomNeeded, calcNewSize(oldLen));
- curr = new char[newSize];
- mCurrentSize = 0;
- mCurrentSegment = curr;
- }
-
- private char[] buildResultArray()
- {
- if (mResultString != null) { // Can take a shortcut...
- return mResultString.toCharArray();
- }
- char[] result;
-
- // Do we use shared array?
- if (mInputStart >= 0) {
- if (mInputLen < 1) {
- return DataUtil.getEmptyCharArray();
- }
- result = new char[mInputLen];
- System.arraycopy(mInputBuffer, mInputStart, result, 0,
- mInputLen);
- } else { // nope
- int size = size();
- if (size < 1) {
- return DataUtil.getEmptyCharArray();
- }
- int offset = 0;
- result = new char[size];
- if (mSegments != null) {
- for (int i = 0, len = mSegments.size(); i < len; ++i) {
- char[] curr = (char[]) mSegments.get(i);
- int currLen = curr.length;
- System.arraycopy(curr, 0, result, offset, currLen);
- offset += currLen;
- }
- }
- System.arraycopy(mCurrentSegment, 0, result, offset, mCurrentSize);
- }
- return result;
- }
-
- private final static class BufferReader
- extends Reader
- {
- ArrayList _Segments;
- char[] _CurrentSegment;
- final int _CurrentLength;
-
- int _SegmentIndex;
- int _SegmentOffset;
- int _CurrentOffset;
-
- public BufferReader(ArrayList segs, char[] currSeg, int currSegLen)
- {
- _Segments = segs;
- _CurrentSegment = currSeg;
- _CurrentLength = currSegLen;
-
- _SegmentIndex = 0;
- _SegmentOffset = _CurrentOffset = 0;
- }
-
- public void close() {
- _Segments = null;
- _CurrentSegment = null;
- }
-
- public void mark(int x)
- throws IOException
- {
- throw new IOException("mark() not supported");
- }
-
- public boolean markSupported() {
- return false;
- }
-
- public int read(char[] cbuf, int offset, int len)
- {
- if (len < 1) {
- return 0;
- }
-
- int origOffset = offset;
- // First need to copy stuff from previous segments
- while (_Segments != null) {
- char[] curr = (char[]) _Segments.get(_SegmentIndex);
- int max = curr.length - _SegmentOffset;
- if (len <= max) { // this is enough
- System.arraycopy(curr, _SegmentOffset, cbuf, offset, len);
- _SegmentOffset += len;
- offset += len;
- return (offset - origOffset);
- }
- // Not enough, but helps...
- if (max > 0) {
- System.arraycopy(curr, _SegmentOffset, cbuf, offset, max);
- offset += max;
- }
- if (++_SegmentIndex >= _Segments.size()) { // last one
- _Segments = null;
- } else {
- _SegmentOffset = 0;
- }
- }
-
- // ok, anything to copy from the active segment?
- if (len > 0 && _CurrentSegment != null) {
- int max = _CurrentLength - _CurrentOffset;
- if (len >= max) { // reading it all
- len = max;
- System.arraycopy(_CurrentSegment, _CurrentOffset,
- cbuf, offset, len);
- _CurrentSegment = null;
- } else {
- System.arraycopy(_CurrentSegment, _CurrentOffset,
- cbuf, offset, len);
- _CurrentOffset += len;
- }
- offset += len;
- }
-
- return (origOffset == offset) ? -1 : (offset - origOffset);
- }
-
- public boolean ready() {
- return true;
- }
-
- public void reset()
- throws IOException
- {
- throw new IOException("reset() not supported");
- }
-
- public long skip(long amount)
- {
- /* Note: implementation is almost identical to that of read();
- * difference being that no data is copied.
- */
- if (amount < 0) {
- return 0L;
- }
-
- long origAmount= amount;
-
- while (_Segments != null) {
- char[] curr = (char[]) _Segments.get(_SegmentIndex);
- int max = curr.length - _SegmentOffset;
- if (max >= amount) { // this is enough
- _SegmentOffset += (int) amount;
- return origAmount;
- }
- // Not enough, but helps...
- amount -= max;
- if (++_SegmentIndex >= _Segments.size()) { // last one
- _Segments = null;
- } else {
- _SegmentOffset = 0;
- }
- }
-
- // ok, anything left in the active segment?
- if (amount > 0 && _CurrentSegment != null) {
- int max = _CurrentLength - _CurrentOffset;
- if (amount >= max) { // reading it all
- amount -= max;
- _CurrentSegment = null;
- } else {
- amount = 0L;
- _CurrentOffset += (int) amount;
- }
- }
-
- return (amount == origAmount) ? -1L : (origAmount - amount);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextBuilder.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextBuilder.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextBuilder.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextBuilder.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,133 +0,0 @@
-package com.ctc.wstx.util;
-
-/**
- * Class similar to {@link StringBuffer}, except that it can be used to
- * construct multiple Strings, that will share same underlying character
- * buffer. This is generally useful for closely related value Strings,
- * such as attribute values of a single XML start element.
- */
-public final class TextBuilder
-{
- private final static int MIN_LEN = 60;
- private final static int MAX_LEN = 120;
-
- private char[] mBuffer;
-
- private int mBufferLen;
-
- private String mResultString;
-
- /*
- ///////////////////////////////////////////////
- // Life-cycle:
- ///////////////////////////////////////////////
- */
-
- public TextBuilder(int initialSize)
- {
- int charSize = (initialSize << 4); // multiply by 16 (-> def. 192 chars)
- if (charSize < MIN_LEN) {
- charSize = MIN_LEN;
- } else if (charSize > MAX_LEN) {
- charSize = MAX_LEN;
- }
- mBuffer = new char[charSize];
- }
-
- /**
- * Method called before starting to (re)use the buffer, will discard
- * any existing content, and start collecting new set of values.
- */
- public void reset() {
- mBufferLen = 0;
- mResultString = null;
- }
-
- /*
- ///////////////////////////////////////////////
- // Accesors:
- ///////////////////////////////////////////////
- */
-
- public boolean isEmpty() {
- return mBufferLen == 0;
- }
-
- public String getAllValues()
- {
- if (mResultString == null) {
- mResultString = new String(mBuffer, 0, mBufferLen);
- }
- return mResultString;
- }
-
- /**
- * Method that gives access to underlying character buffer
- */
- public char[] getCharBuffer() {
- return mBuffer;
- }
-
- public int getCharSize() {
- return mBufferLen;
- }
-
- /*
- ///////////////////////////////////////////////
- // Mutators:
- ///////////////////////////////////////////////
- */
-
- public void append(char c) {
- if (mBuffer.length == mBufferLen) {
- resize(1);
- }
- mBuffer[mBufferLen++] = c;
- }
-
- public void append(char[] src, int start, int len) {
- if (len > (mBuffer.length - mBufferLen)) {
- resize(len);
- }
- System.arraycopy(src, start, mBuffer, mBufferLen, len);
- mBufferLen += len;
- }
-
- public void setBufferSize(int newSize) {
- mBufferLen = newSize;
- }
-
- public char[] bufferFull(int needSpaceFor) {
- mBufferLen = mBuffer.length;
- resize(1);
- return mBuffer;
- }
-
- /*
- ///////////////////////////////////////////////
- // Debugging:
- ///////////////////////////////////////////////
- */
-
- public String toString() {
- return new String(mBuffer, 0, mBufferLen);
- }
-
- /*
- ///////////////////////////////////////////////
- // Internal methods:
- ///////////////////////////////////////////////
- */
-
- private void resize(int needSpaceFor) {
- char[] old = mBuffer;
- int oldLen = old.length;
- int addition = oldLen >> 1; // Grow by 50%
- needSpaceFor -= (oldLen - mBufferLen);
- if (addition < needSpaceFor) {
- addition = needSpaceFor;
- }
- mBuffer = new char[oldLen+addition];
- System.arraycopy(old, 0, mBuffer, 0, mBufferLen);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/URLUtil.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/URLUtil.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/URLUtil.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/URLUtil.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,190 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.io.*;
-import java.net.URI;
-import java.net.URL;
-import java.net.URLDecoder;
-import java.net.MalformedURLException;
-import java.net.URISyntaxException;
-
-public final class URLUtil
-{
- private URLUtil() { }
-
- /**
- * Method that tries to figure out how to create valid URL from a system
- * id, without additional contextual information.
- * If we could use URIs this might be easier to do, but they are part
- * of JDK 1.4, and preferably code should only require 1.2 (or maybe 1.3)
- */
- public static URL urlFromSystemId(String sysId)
- throws IOException
- {
- try {
- /* Ok, does it look like a full URL? For one, you need a colon. Also,
- * to reduce likelihood of collision with Windows paths, let's only
- * accept it if there are 3 preceding other chars...
- * Not sure if Mac might be a problem? (it uses ':' as file path
- * separator, alas, at least prior to MacOS X)
- */
- int ix = sysId.indexOf(':', 0);
- /* Also, protocols are generally fairly short, usually 3 or 4
- * chars (http, ftp, urn); so let's put upper limit of 8 chars too
- */
- if (ix >= 3 && ix <= 8) {
- return new URL(sysId);
- }
- // Ok, let's just assume it's local file reference...
- /* 24-May-2006, TSa: Amazingly, this single call does show in
- * profiling, for small docs. The problem is that deep down it
- * tries to check physical file system, to check if the File
- * pointed to is a directory: and that is (relatively speaking)
- * a very expensive call. Since in this particular case it
- * should never be a dir (and/or doesn't matter), let's just
- * implement conversion locally
- */
- String absPath = new java.io.File(sysId).getAbsolutePath();
- // Need to convert colons/backslashes to regular slashes?
- {
- char sep = File.separatorChar;
- if (sep != '/') {
- absPath = absPath.replace(sep, '/');
- }
- }
- if (absPath.length() > 0 && absPath.charAt(0) != '/') {
- absPath = "/" + absPath;
- }
- return new URL("file", "", absPath);
- } catch (MalformedURLException e) {
- throwIOException(e, sysId);
- return null; // never gets here
- }
- }
-
- /**
- * @since 4.1
- */
- public static URI uriFromSystemId(String sysId) throws IOException
- {
- // note: mostly a copy of matching method above, but with URI instead of URL
- try {
- int ix = sysId.indexOf(':', 0);
- if (ix >= 3 && ix <= 8) {
- return new URI(sysId);
- }
- String absPath = new java.io.File(sysId).getAbsolutePath();
- char sep = File.separatorChar;
- if (sep != '/') {
- absPath = absPath.replace(sep, '/');
- }
- if (absPath.length() > 0 && absPath.charAt(0) != '/') {
- absPath = "/" + absPath;
- }
- return new URI("file", absPath, null);
- } catch (URISyntaxException e) {
- throwIOException(e, sysId);
- return null; // never gets here
- }
- }
-
- public static URL urlFromSystemId(String sysId, URL ctxt) throws IOException
- {
- if (ctxt == null) {
- return urlFromSystemId(sysId);
- }
- try {
- return new URL(ctxt, sysId);
- } catch (MalformedURLException e) {
- throwIOException(e, sysId);
- return null; // never gets here
- }
- }
-
- /**
- * Method that tries to create and return URL that denotes current
- * working directory. Usually used to create a context, when one is
- * not explicitly passed.
- */
- public static URL urlFromCurrentDir()
- throws java.net.MalformedURLException /* an IOException */
- {
- /* This seems to work; independent of whether there happens to
- * be such/file dir or not.
- */
- return new File("a").getAbsoluteFile().getParentFile().toURL();
- }
-
- /**
- * Method that tries to get a stream (ideally, optimal one) to read from
- * the specified URL.
- * Currently it just means creating a simple file input stream if the
- * URL points to a (local) file, and otherwise relying on URL classes
- * input stream creation method.
- */
- public static InputStream inputStreamFromURL(URL url)
- throws IOException
- {
- if ("file".equals(url.getProtocol())) {
- /* As per [WSTX-82], can not do this if the path refers
- * to a network drive on windows. This fixes the problem;
- * might not be needed on all platforms (NFS?), but should not
- * matter a lot: performance penalty of extra wrapping is more
- * relevant when accessing local file system.
- */
- String host = url.getHost();
- if (host == null || host.length() == 0) {
- /* One more test: if there are quoted characters, need
- * to decoded [WSTX-207]:
- */
- String path = url.getPath();
- if (path.indexOf('%') >= 0) {
- path = URLDecoder.decode(path, "UTF-8");
- }
- return new FileInputStream(path);
- }
- }
- return url.openStream();
- }
-
- /**
- * Method that tries to get a stream (ideally, optimal one) to write to
- * the resource specified by given URL.
- * Currently it just means creating a simple file output stream if the
- * URL points to a (local) file, and otherwise relying on URL classes
- * input stream creation method.
- */
- public static OutputStream outputStreamFromURL(URL url)
- throws IOException
- {
- if ("file".equals(url.getProtocol())) {
- /* As per [WSTX-82], can not do this if the path refers
- * to a network drive on windows.
- */
- String host = url.getHost();
- if (host == null || host.length() == 0) {
- return new FileOutputStream(url.getPath());
- }
- }
- return url.openConnection().getOutputStream();
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Private helper methods
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Helper method that tries to fully convert strange URL-specific exception
- * to more general IO exception. Also, to try to use JDK 1.4 feature without
- * creating requirement, uses reflection to try to set the root cause, if
- * we are running on JDK1.4
- */
- private static void throwIOException(Exception mex, String sysId)
- throws IOException
- {
- IOException ie = new IOException("[resolving systemId '"+sysId+"']: "+mex.toString());
- ExceptionUtil.setInitCause(ie, mex);
- throw ie;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/WordResolver.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/WordResolver.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/WordResolver.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/WordResolver.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,580 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.util.*;
-
-/**
- * A specialized Map/Symbol table - like data structure that can be used
- * for both checking whether a word (passed in as a char array) exists
- * in certain set of words AND getting that word as a String.
- * It is reasonably efficient both time and speed-wise, at least for
- * certain use cases; specifically, if there is no existing key to use,
- * it is more efficient way to get to a shared copy of that String
- * The general usage pattern is expected
- * to be such that most checks are positive, ie. that the word indeed
- * is contained in the structure.
- *
- * Although this is an efficient data struct for specific set of usage
- * patterns, one restriction is that the full set of words to include has to
- * be known before constructing the instnace. Also, the size of the set is
- * limited to total word content of about 20k characters.
- *
- * TODO: Should document the internal data structure...
- */
-public final class WordResolver
-{
- /**
- * Maximum number of words (Strings) an instance can contain
- */
- public final static int MAX_WORDS = 0x2000;
-
- final static char CHAR_NULL = (char) 0;
-
- /**
- * Offset added to numbers to mark 'negative' numbers. Asymmetric,
- * since range of negative markers needed is smaller than positive
- * numbers...
- */
- final static int NEGATIVE_OFFSET = 0x10000 - MAX_WORDS;
-
- /**
- * This is actually just a guess; but in general linear search should
- * be faster for short sequences (definitely for 4 or less; maybe up
- * to 8 or less?)
- */
- final static int MIN_BINARY_SEARCH = 7;
-
- /**
- * Compressed presentation of the word set.
- */
- final char[] mData;
-
- /**
- * Array of actual words returned resolved for matches.
- */
- final String[] mWords;
-
- /*
- ////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////
- */
-
- private WordResolver(String[] words, char[] index) {
- mWords = words;
- mData = index;
- }
-
- /**
- * Tries to construct an instance given ordered set of words.
- *
- * Note: currently maximum number of words that can be contained
- * is limited to {@link #MAX_WORDS}; additionally, maximum length
- * of all such words can not exceed roughly 28000 characters.
- *
- * @return WordResolver constructed for given set of words, if
- * the word set size is not too big; null to indicate "too big"
- * instance.
- */
- public static WordResolver constructInstance(TreeSet wordSet)
- {
- if (wordSet.size() > MAX_WORDS) {
- return null;
- }
- return new Builder(wordSet).construct();
- }
-
- /*
- ////////////////////////////////////////////////
- // Public API
- ////////////////////////////////////////////////
- */
-
- /**
- * @return Number of words contained
- */
- public int size() {
- return mWords.length;
- }
-
- /*
- public int indexSize() {
- return mData.length;
- }
- */
-
- /**
- * @param str Character array that contains the word to find
- * @param start Index of the first character of the word
- * @param end Index following the last character of the word,
- * so that
- * Performance of the set is comparable to that of {@link java.util.TreeSet}
- * for Strings, ie. 2-3x slower than {@link java.util.HashSet} when
- * using pre-constructed Strings. This is generally result of algorithmic
- * complexity of structures; Word and Tree sets are roughly logarithmic
- * to the whole data, whereas Hash set is linear to the length of key.
- * However:
- *
- * Although this is an efficient set for specific set of usage patterns,
- * one restriction is that the full set of words to include has to be
- * known before constructing the set. Also, the size of the set is
- * limited to total word content of about 20k characters; factory method
- * does verify the limit and indicates if an instance can not be created.
- */
-public final class WordSet
-{
- final static char CHAR_NULL = (char) 0;
-
- /**
- * Offset added to numbers to mark 'negative' numbers. Asymmetric,
- * since range of negative markers needed is smaller than positive
- * numbers...
- */
- final static int NEGATIVE_OFFSET = 0xC000;
-
- /**
- * This is actually just a guess; but in general linear search should
- * be faster for short sequences (definitely for 4 or less; maybe up
- * to 8 or less?)
- */
- final static int MIN_BINARY_SEARCH = 7;
-
- /**
- * Compressed presentation of the word set.
- */
- final char[] mData;
-
- /*
- ////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////
- */
-
- private WordSet(char[] data) {
- mData = data;
- }
-
- public static WordSet constructSet(TreeSet wordSet)
- {
- return new WordSet(new Builder(wordSet).construct());
- }
-
- public static char[] constructRaw(TreeSet wordSet)
- {
- return new Builder(wordSet).construct();
- }
-
- /*
- ////////////////////////////////////////////////
- // Public API
- ////////////////////////////////////////////////
- */
-
- public boolean contains(char[] buf, int start, int end) {
- return contains(mData, buf, start, end);
- }
-
- public static boolean contains(char[] data, char[] str, int start, int end)
- {
- int ptr = 0; // pointer to compressed set data
-
- main_loop:
- do {
- int left = end-start;
-
- // End of input String? Need to have the run entry:
- if (left == 0) {
- return (data[ptr+1] == CHAR_NULL);
- }
-
- int count = data[ptr++];
-
- // Nope, but do we have an end marker?
- if (count >= NEGATIVE_OFFSET) {
- // How many chars do we need to have left to match?
- int expCount = count - NEGATIVE_OFFSET;
- if (left != expCount) {
- return false;
- }
- while (start < end) {
- if (data[ptr] != str[start]) {
- return false;
- }
- ++ptr;
- ++start;
- }
- return true;
- }
-
- // No, need to find the branch to follow, if any
- char c = str[start++];
-
- // Linear or binary search?
- if (count < MIN_BINARY_SEARCH) {
- // always at least two branches; never less
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- continue main_loop;
- }
- if (data[ptr+2] == c) {
- ptr = (int) data[ptr+3];
- continue main_loop;
- }
- int branchEnd = ptr + (count << 1);
- // Starts from entry #3, if such exists
- for (ptr += 4; ptr < branchEnd; ptr += 2) {
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- continue main_loop;
- }
- }
- return false; // No match!
- }
-
- { // Ok, binary search:
- int low = 0;
- int high = count-1;
- int mid;
-
- while (low <= high) {
- mid = (low + high) >> 1;
- int ix = ptr + (mid << 1);
- int diff = data[ix] - c;
- if (diff > 0) { // char was 'higher', need to go down
- high = mid-1;
- } else if (diff < 0) { // lower, need to go up
- low = mid+1;
- } else { // match
- ptr = (int) data[ix+1];
- continue main_loop;
- }
- }
- }
-
- // If we fall here, no match!
- return false;
-
- } while (ptr != 0);
-
- // If we reached an end state, must match the length
- return (start == end);
- }
-
- public boolean contains(String str) {
- return contains(mData, str);
- }
-
- public static boolean contains(char[] data, String str)
- {
- // Let's use same vars as array-based code, to allow cut'n pasting
- int ptr = 0; // pointer to compressed set data
- int start = 0;
- int end = str.length();
-
- main_loop:
- do {
- int left = end-start;
-
- // End of input String? Need to have the run entry:
- if (left == 0) {
- return (data[ptr+1] == CHAR_NULL);
- }
-
- int count = data[ptr++];
-
- // Nope, but do we have an end marker?
- if (count >= NEGATIVE_OFFSET) {
- // How many chars do we need to have left to match?
- int expCount = count - NEGATIVE_OFFSET;
- if (left != expCount) {
- return false;
- }
- while (start < end) {
- if (data[ptr] != str.charAt(start)) {
- return false;
- }
- ++ptr;
- ++start;
- }
- return true;
- }
-
- // No, need to find the branch to follow, if any
- char c = str.charAt(start++);
-
- // Linear or binary search?
- if (count < MIN_BINARY_SEARCH) {
- // always at least two branches; never less
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- continue main_loop;
- }
- if (data[ptr+2] == c) {
- ptr = (int) data[ptr+3];
- continue main_loop;
- }
- int branchEnd = ptr + (count << 1);
- // Starts from entry #3, if such exists
- for (ptr += 4; ptr < branchEnd; ptr += 2) {
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- continue main_loop;
- }
- }
- return false; // No match!
- }
-
- { // Ok, binary search:
- int low = 0;
- int high = count-1;
- int mid;
-
- while (low <= high) {
- mid = (low + high) >> 1;
- int ix = ptr + (mid << 1);
- int diff = data[ix] - c;
- if (diff > 0) { // char was 'higher', need to go down
- high = mid-1;
- } else if (diff < 0) { // lower, need to go up
- low = mid+1;
- } else { // match
- ptr = (int) data[ix+1];
- continue main_loop;
- }
- }
- }
-
- // If we fall here, no match!
- return false;
-
- } while (ptr != 0);
-
- // If we reached an end state, must match the length
- return (start == end);
- }
-
- /*
- ////////////////////////////////////////////////
- // Private methods
- ////////////////////////////////////////////////
- */
-
- /*
- ////////////////////////////////////////////////
- // Helper classes
- ////////////////////////////////////////////////
- */
-
- private final static class Builder
- {
- final String[] mWords;
-
- char[] mData;
-
- /**
- * Number of characters currently used from mData
- */
- int mSize;
-
- public Builder(TreeSet wordSet) {
- int wordCount = wordSet.size();
- mWords = new String[wordCount];
- wordSet.toArray(mWords);
-
- /* Let's guess approximate size we should need, assuming
- * average word length of 6 characters, and 100% overhead
- * in structure:
- */
- int size = wordCount * 12;
- if (size < 256) {
- size = 256;
- }
- mData = new char[size];
- }
-
- /**
- * @return Raw character data that contains compressed structure
- * of the word set
- */
- public char[] construct()
- {
-// Uncomment if you need to debug array-out-of-bound probs
-//try {
- // Let's check degenerate case of 1 word:
- if (mWords.length == 1) {
- constructLeaf(0, 0);
- } else {
- constructBranch(0, 0, mWords.length);
- }
-//} catch (Throwable t) { System.err.println("Error: "+t); }
-
- char[] result = new char[mSize];
- System.arraycopy(mData, 0, result, 0, mSize);
- return result;
- }
-
- /**
- * Method that is called recursively to build the data
- * representation for a branch, ie. part of word set tree
- * that still has more than one ending
- *
- * @param charIndex Index of the character in words to consider
- * for this round
- * @param start Index of the first word to be processed
- * @param end Index of the word after last word to be processed
- * (so that number of words is
+ * Users of this class are encouraged to use a {@link TreeSet} with the {@link String#CASE_INSENSITIVE_ORDER}
+ * comparator if case-insensitive comparison is needed (like when dealing with HTML tags).
+ */
+ public static class SetEmptyElementHandler
+ implements EmptyElementHandler
+ {
+ final protected Set
+ * Note that element name comparison is case-insensitive as required
+ * by HTML specification.
+ */
+ public static class HtmlEmptyElementHandler
+ extends SetEmptyElementHandler
+ {
+ private final static HtmlEmptyElementHandler sInstance = new HtmlEmptyElementHandler();
+
+ public static HtmlEmptyElementHandler getInstance() { return sInstance; }
+
+ protected HtmlEmptyElementHandler()
+ {
+ super(new TreeSet
+ * Note about exceptions: choice of only allowing throwing of
+ * {@link IOException}s is due to the way Woodstox stream writer
+ * backend works;
+ * In addition to its main task as a configuration container, this class
+ * also acts as a wrapper around simple buffer recycling functionality.
+ * The reason is that while conceptually this is a separate concern,
+ * there are enough commonalities with the life-cycle of this object to
+ * make this a very convenience place to add that functionality...
+ * (that is: conceptually this is not right, but from pragmatic viewpoint
+ * it just makes sense)
+ */
+public final class ReaderConfig
+ extends CommonConfig
+ implements InputConfigFlags
+{
+ // Default limit values
+
+ public final static int DEFAULT_MAX_ATTRIBUTES_PER_ELEMENT = 1000;
+ public final static int DEFAULT_MAX_ATTRIBUTE_LENGTH = 65536 * 8;
+
+ public final static int DEFAULT_MAX_ELEMENT_DEPTH = 1000;
+
+ public final static int DEFAULT_MAX_ENTITY_DEPTH = 500;
+ public final static int DEFAULT_MAX_ENTITY_COUNT = 100 * 1000;
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Constants for reader properties:
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ // // First, standard StAX properties:
+
+ // Simple flags:
+ final static int PROP_COALESCE_TEXT = 1;
+ final static int PROP_NAMESPACE_AWARE = 2;
+ final static int PROP_REPLACE_ENTITY_REFS = 3;
+ final static int PROP_SUPPORT_EXTERNAL_ENTITIES = 4;
+ final static int PROP_VALIDATE_AGAINST_DTD = 5;
+ final static int PROP_SUPPORT_DTD = 6;
+
+ // Object type properties
+ public final static int PROP_EVENT_ALLOCATOR = 7;
+ final static int PROP_WARNING_REPORTER = 8;
+ final static int PROP_XML_RESOLVER = 9;
+
+ // // Then StAX2 standard properties:
+
+ // Simple flags:
+ final static int PROP_INTERN_NS_URIS = 20;
+ final static int PROP_INTERN_NAMES = 21;
+ final static int PROP_REPORT_CDATA = 22;
+ final static int PROP_REPORT_PROLOG_WS = 23;
+ final static int PROP_PRESERVE_LOCATION = 24;
+ final static int PROP_AUTO_CLOSE_INPUT = 25;
+
+ // Enum / Object type properties:
+ final static int PROP_SUPPORT_XMLID = 26; // shared with WriterConfig
+ final static int PROP_DTD_OVERRIDE = 27;
+
+ // // // Constants for additional Wstx properties:
+
+ // Simple flags:
+
+ /**
+ * Note: this entry was deprecated for 4.0 versions up until
+ * and including 4.0.7; was brought back for 4.0.8 (and will
+ * be retained for 4.1)
+ */
+ final static int PROP_NORMALIZE_LFS = 40;
+
+ /* This entry was deprecated for 3.2 and removed in 4.0
+ * version. There are no plans to bring it back.
+ */
+ //final static int PROP_NORMALIZE_ATTR_VALUES = 41;
+
+ final static int PROP_CACHE_DTDS = 42;
+ final static int PROP_CACHE_DTDS_BY_PUBLIC_ID = 43;
+ final static int PROP_LAZY_PARSING = 44;
+ final static int PROP_SUPPORT_DTDPP = 45;
+ final static int PROP_TREAT_CHAR_REFS_AS_ENTS = 46;
+
+ // Object type properties:
+
+ final static int PROP_INPUT_BUFFER_LENGTH = 50;
+ //final static int PROP_TEXT_BUFFER_LENGTH = 51;
+ final static int PROP_MIN_TEXT_SEGMENT = 52;
+ final static int PROP_CUSTOM_INTERNAL_ENTITIES = 53;
+ final static int PROP_DTD_RESOLVER = 54;
+ final static int PROP_ENTITY_RESOLVER = 55;
+ final static int PROP_UNDECLARED_ENTITY_RESOLVER = 56;
+ final static int PROP_BASE_URL = 57;
+ final static int PROP_INPUT_PARSING_MODE = 58;
+
+ // Size limitation to prevent various DOS attacks
+ final static int PROP_MAX_ATTRIBUTES_PER_ELEMENT = 60;
+ final static int PROP_MAX_CHILDREN_PER_ELEMENT = 61;
+ final static int PROP_MAX_ELEMENT_COUNT = 62;
+ final static int PROP_MAX_ELEMENT_DEPTH = 63;
+ final static int PROP_MAX_CHARACTERS = 64;
+ final static int PROP_MAX_ATTRIBUTE_SIZE = 65;
+ final static int PROP_MAX_TEXT_LENGTH = 66;
+ final static int PROP_MAX_ENTITY_COUNT = 67;
+ final static int PROP_MAX_ENTITY_DEPTH = 68;
+
+ /*
+ ////////////////////////////////////////////////
+ // Limits for numeric properties
+ ////////////////////////////////////////////////
+ */
+
+ /**
+ * Need to set a minimum size, since there are some limitations to
+ * smallest consequtive block that can be used.
+ */
+ final static int MIN_INPUT_BUFFER_LENGTH = 8; // 16 bytes
+
+ /**
+ * Let's allow caching of just a dozen DTDs... shouldn't really
+ * matter, how many DTDs does one really use?
+ */
+ final static int DTD_CACHE_SIZE_J2SE = 12;
+
+ final static int DTD_CACHE_SIZE_J2ME = 5;
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Default values for custom properties:
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ /**
+ * By default, let's require minimum of 64 chars to be delivered
+ * as shortest partial (piece of) text (CDATA, text) segment;
+ * same for both J2ME subset and full readers. Prevents tiniest
+ * runts from getting passed
+ */
+ final static int DEFAULT_SHORTEST_TEXT_SEGMENT = 64;
+
+ /**
+ * Default config flags are converted from individual settings,
+ * to conform to StAX 1.0 specifications.
+ */
+ final static int DEFAULT_FLAGS_FULL =
+ 0
+ // First, default settings StAX specs dictate:
+
+ | CFG_NAMESPACE_AWARE
+ // Coalescing to be disabled
+ //| CFG_COALESCE_TEXT
+ | CFG_REPLACE_ENTITY_REFS
+ | CFG_SUPPORT_EXTERNAL_ENTITIES
+ | CFG_SUPPORT_DTD
+
+ // and then custom setting defaults:
+
+ // and namespace URI interning
+ | CFG_INTERN_NAMES
+ | CFG_INTERN_NS_URIS
+
+ // we will also accurately report CDATA, by default
+ | CFG_REPORT_CDATA
+
+ /* 20-Jan-2006, TSa: As per discussions on stax-builders list
+ * (and input from xml experts), 4.0 will revert to "do not
+ * report SPACE events outside root element by default"
+ * settings. Conceptually this is what xml specification
+ * implies should be done: there is no content outside of
+ * the element tree, including any ignorable content, just
+ * processing instructions and comments.
+ */
+ //| CFG_REPORT_PROLOG_WS
+
+ /* but enable DTD caching (if they are handled):
+ * (... maybe J2ME subset shouldn't do it?)
+ */
+ | CFG_CACHE_DTDS
+ /* 29-Mar-2006, TSa: But note, no caching by public-id, due
+ * to problems with cases where public-id/system-id were
+ * inconsistently used, leading to problems.
+ */
+
+ /* by default, let's also allow lazy parsing, since it tends
+ * to improve performance
+ */
+ | CFG_LAZY_PARSING
+
+ /* and also make Event objects preserve location info...
+ * can be turned off for maximum performance
+ */
+ | CFG_PRESERVE_LOCATION
+
+ // As per Stax 1.0 specs, we can not enable this by default:
+ //| CFG_AUTO_CLOSE_INPUT);
+
+ /* Also, let's enable dtd++ support (shouldn't hurt with non-dtd++
+ * dtds)
+ */
+
+ | CFG_SUPPORT_DTDPP
+
+ /*
+ * Set this as a default, as this is required in xml;
+ */
+ | CFG_NORMALIZE_LFS
+
+ /* Regarding Xml:id, let's enabled typing by default, but not
+ * uniqueness validity checks: latter will be taken care of
+ * by DTD validation if enabled, otherwise needs to be explicitly
+ * enabled
+ */
+ | CFG_XMLID_TYPING
+ // | CFG_XMLID_UNIQ_CHECKS
+ ;
+
+ /**
+ * For now defaults for J2ME flags can be identical to 'full' set;
+ * differences are in buffer sizes.
+ */
+ final static int DEFAULT_FLAGS_J2ME = DEFAULT_FLAGS_FULL;
+
+ // // //
+
+ /**
+ * Map to use for converting from String property ids to ints
+ * described above; useful to allow use of switch later on.
+ */
+ final static HashMap
+ * In addition to the standard settings, following Woodstox-specific
+ * settings are also done:
+ *
+ * Notes: Does NOT change 'performance' settings (buffer sizes,
+ * DTD caching, coalescing, interning, accurate location info).
+ */
+ public void configureForXmlConformance()
+ {
+ // // StAX 1.0 settings
+ doSupportNamespaces(true);
+ doSupportDTDs(true);
+ doSupportExternalEntities(true);
+ doReplaceEntityRefs(true);
+
+ // // Stax2 additional settings
+
+ // Better enable full xml:id checks:
+ doXmlIdTyping(true);
+ doXmlIdUniqChecks(true);
+
+ // Woodstox-specific ones:
+ }
+
+ /**
+ * Method to call to make Reader created be as "convenient" to use
+ * as possible; ie try to avoid having to deal with some of things
+ * like segmented text chunks. This may incur some slight performance
+ * penalties, but should not affect XML conformance.
+ * See {@link XMLInputFactory2#configureForConvenience} for
+ * required settings for standard StAX/StAX properties.
+ *
+ * In addition to the standard settings, following Woodstox-specific
+ * settings are also done:
+ *
+ * In addition to the standard settings, following Woodstox-specific
+ * settings are also done:
+ *
+ * See {@link XMLInputFactory2#configureForLowMemUsage} for
+ * required settings for standard StAX/StAX properties.
+ *
+ * In addition to the standard settings, following Woodstox-specific
+ * settings are also done:
+ *
+ * See {@link XMLInputFactory2#configureForLowMemUsage} for
+ * required settings for standard StAX/StAX properties.
+ *
+ * In addition to the standard settings, following Woodstox-specific
+ * settings are also done:
+ *
+ * TODO:
+ *
+ * - CHECK_CHAR_VALIDITY (separate for white spaces?)
+ * - CATALOG_RESOLVER? (or at least, ENABLE_CATALOGS)
+ */
+public final class WstxInputProperties
+{
+ /**
+ * Constants used when no DTD handling is done, and we do not know the
+ * 'real' type of an attribute. Seems like CDATA is the safe choice.
+ */
+ public final static String UNKNOWN_ATTR_TYPE = "CDATA";
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Simple on/off settings:
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ // // // Normalization:
+
+ /**
+ * Feature that controls whether linefeeds are normalized into
+ * canonical linefeed as mandated by xml specification.
+ *
+ * Note that disabling this property (from its default enabled
+ * state) will result in non-conforming XML processing. It may
+ * be useful for use cases where changes to input content should
+ * be minimized.
+ *
+ * Note: this property was initially removed from Woodstox 4.0,
+ * but was reintroduced in 4.0.8 due to user request.
+ */
+ public final static String P_NORMALIZE_LFS = "com.ctc.wstx.normalizeLFs";
+
+ //public final static String P_NORMALIZE_ATTR_VALUES = "com.ctc.wstx.normalizeAttrValues";
+
+ // // // XML character validation:
+
+ /**
+ * Whether readers will verify that characters in text content are fully
+ * valid XML characters (not just Unicode). If true, will check
+ * that they are valid (including white space); if false, will not
+ * check.
+ *
+ * Note that this property will NOT have effect on all encoding problems,
+ * specifically:
+ *
+ * Turning this option off may improve parsing performance; leaving
+ * it on guarantees compatibility with XML 1.0 specs regarding character
+ * validity rules.
+ */
+ public final static String P_VALIDATE_TEXT_CHARS = "com.ctc.wstx.validateTextChars";
+
+ // // // Caching:
+
+ /**
+ * Whether readers will try to cache parsed external DTD subsets or not.
+ */
+
+ public final static String P_CACHE_DTDS = "com.ctc.wstx.cacheDTDs";
+
+ /**
+ * Whether reader is to cache DTDs (when caching enabled) based on public id
+ * or not: if not, system id will be primarily used. Although theoretically
+ * public IDs should be unique, and should be good caching keys, sometimes
+ * broken documents use 'wrong' public IDs, and such by default caching keys
+ * are based on system id only.
+ */
+ public final static String P_CACHE_DTDS_BY_PUBLIC_ID = "com.ctc.wstx.cacheDTDsByPublicId";
+
+
+ // // // Enabling/disabling lazy/incomplete parsing
+
+ /**
+ * Whether stream readers are allowed to do lazy parsing, meaning
+ * to parse minimal part of the event when
+ * {@link javax.xml.stream.XMLStreamReader#next} is called, and only parse the rest
+ * as needed (or skip remainder of no extra information is needed).
+ * Alternative to lazy parsing is called "eager parsing", and is
+ * what most xml parsers use by default.
+ *
+ * Enabling lazy parsing can improve performance for tasks where
+ * number of textual events are skipped. The downside is that
+ * not all well-formedness problems are reported when
+ * {@link javax.xml.stream.XMLStreamReader#next} is called, but only when the
+ * rest of event are read or skipped.
+ *
+ * Default value for Woodstox is such that lazy parsing is
+ * enabled.
+ *
+ * @deprecated As of Woodstox 4.0 use
+ * {@link XMLInputFactory2#P_LAZY_PARSING} instead (from
+ * Stax2 extension API, v3.0)
+ */
+ @Deprecated
+ public final static String P_LAZY_PARSING = XMLInputFactory2.P_LAZY_PARSING;
+
+ // // // API behavior (for backwards compatibility)
+
+ /**
+ * This read-only property indicates whether null is returned for default name space prefix;
+ * Boolean.TRUE indicates it does, Boolean.FALSE that it does not.
+ *
+ * Default value for 4.1 is 'false'; this will most likely change for 5.0 since
+ * Stax API actually specifies null to be used.
+ *
+ * @since 4.1.2
+ */
+ public final static String P_RETURN_NULL_FOR_DEFAULT_NAMESPACE = "com.ctc.wstx.returnNullForDefaultNamespace";
+
+ // // // Enabling/disabling support for dtd++
+
+ /**
+ * Whether the Reader will recognized DTD++ extensions when parsing
+ * DTD subsets.
+ *
+ * Note: not implemented by Woodstox.
+ *
+ * @deprecated Never implement, let's phase this out (deprecated in 4.2)
+ */
+ @Deprecated
+ public final static String P_SUPPORT_DTDPP = "com.ctc.wstx.supportDTDPP";
+
+ /**
+ * Whether the Reader will treat character references as entities while parsing
+ * XML documents.
+ */
+ public static final String P_TREAT_CHAR_REFS_AS_ENTS = "com.ctc.wstx.treatCharRefsAsEnts";
+
+ // // // Enabling alternate mode for parsing XML fragments instead
+ // // // of full documents
+
+ // Automatic W3C Schema support?
+ /*
+ * Whether W3C Schema hint attributes are recognized within document,
+ * and used to locate Schema to use for validation.
+ */
+ //public final static String P_AUTOMATIC_W3C_SCHEMA = 0x00100000;
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // More complex settings
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ // // // Buffer sizes;
+
+ /**
+ * Size of input buffer (in chars), to use for reading XML content
+ * from input stream/reader.
+ */
+ public final static String P_INPUT_BUFFER_LENGTH = "com.ctc.wstx.inputBufferLength";
+
+ // // // Constraints on sizes of text segments parsed:
+
+
+ /**
+ * Property to specify shortest non-complete text segment (part of
+ * CDATA section or text content) that parser is allowed to return,
+ * if not required to coalesce text.
+ */
+ public final static String P_MIN_TEXT_SEGMENT = "com.ctc.wstx.minTextSegment";
+
+ // // // Other size constraints (4.2+)
+
+ /**
+ * Maximum number of attributes allowed for single XML element.
+ * @since 4.2
+ */
+ public final static String P_MAX_ATTRIBUTES_PER_ELEMENT = "com.ctc.wstx.maxAttributesPerElement";
+
+ /**
+ * Maximum length of of individual attribute values (in characters)
+ * @since 4.2
+ */
+ public final static String P_MAX_ATTRIBUTE_SIZE = "com.ctc.wstx.maxAttributeSize";
+
+ /**
+ * Maximum number of child elements for any given element.
+ * @since 4.2
+ */
+ public final static String P_MAX_CHILDREN_PER_ELEMENT = "com.ctc.wstx.maxChildrenPerElement";
+
+ /**
+ * Maximum number of all elements in a single document.
+ * @since 4.2
+ */
+ public final static String P_MAX_ELEMENT_COUNT = "com.ctc.wstx.maxElementCount";
+
+ /**
+ * Maximum level of nesting of XML elements, starting with root element.
+ * @since 4.2
+ */
+ public final static String P_MAX_ELEMENT_DEPTH = "com.ctc.wstx.maxElementDepth";
+
+ /**
+ * Maximum length of input document, in characters.
+ * @since 4.2
+ */
+ public final static String P_MAX_CHARACTERS = "com.ctc.wstx.maxCharacters";
+
+ /**
+ * Maximum length of individual text (cdata) segments in input, in characters.
+ * @since 4.2
+ */
+ public final static String P_MAX_TEXT_LENGTH = "com.ctc.wstx.maxTextLength";
+
+ // and more size constraints (4.3+)
+
+ /**
+ * Maximum number of total (general parsed) entity expansions within input.
+ *
+ * @since 4.3
+ */
+ public final static String P_MAX_ENTITY_COUNT = "com.ctc.wstx.maxEntityCount";
+
+ /**
+ * Maximum depth of nested (general parsed) entity expansions.
+ *
+ * @since 4.3
+ */
+ public final static String P_MAX_ENTITY_DEPTH = "com.ctc.wstx.maxEntityDepth";
+
+ // // // Entity handling
+
+ /**
+ * Property of type {@link java.util.Map}, that defines explicit set of
+ * internal (generic) entities that will define of override any entities
+ * defined in internal or external subsets; except for the 5 pre-defined
+ * entities (lt, gt, amp, apos, quot). Can be used to explicitly define
+ * entites that would normally come from a DTD.
+ *
+ * @deprecated This feature may be removed from future versions of
+ * Woodstox, since the same functionality can be achieved by using
+ * custom entity resolvers.
+ */
+ @Deprecated
+ public final static String P_CUSTOM_INTERNAL_ENTITIES = "com.ctc.wstx.customInternalEntities";
+
+ /**
+ * Property of type {@link XMLResolver}, that
+ * will allow overriding of default DTD and external parameter entity
+ * resolution.
+ */
+ public final static String P_DTD_RESOLVER = "com.ctc.wstx.dtdResolver";
+
+ /**
+ * Property of type {@link XMLResolver}, that
+ * will allow overriding of default external general entity
+ * resolution. Note that using this property overrides settings done
+ * using {@link javax.xml.stream.XMLInputFactory#RESOLVER} (and vice versa).
+ */
+ public final static String P_ENTITY_RESOLVER = "com.ctc.wstx.entityResolver";
+
+ /**
+ * Property of type {@link XMLResolver}, that
+ * will allow graceful handling of references to undeclared (general)
+ * entities.
+ */
+ public final static String P_UNDECLARED_ENTITY_RESOLVER = "com.ctc.wstx.undeclaredEntityResolver";
+
+ /**
+ * Property of type {@link java.net.URL}, that will allow specifying
+ * context URL to use when resolving relative references, for the
+ * main-level entities (external DTD subset, references from the internal
+ * DTD subset).
+ */
+ public final static String P_BASE_URL = "com.ctc.wstx.baseURL";
+
+ // // // Alternate parsing modes
+
+ /**
+ * Three-valued property (one of
+ * {@link #PARSING_MODE_DOCUMENT},
+ * {@link #PARSING_MODE_FRAGMENT} or
+ * {@link #PARSING_MODE_DOCUMENTS}; default being the document mode)
+ * that can be used to handle "non-standard" XML content. The default
+ * mode (
+ * The main difference from the API perspective is that in first two
+ * modes, START_DOCUMENT and END_DOCUMENT are used as usual (as the first
+ * and last events returned), whereas the multi-document mode can return
+ * multiple pairs of these events: although it is still true that the
+ * first event (one cursor points to when reader is instantiated or
+ * returned by the event reader), there may be intervening pairs that
+ * signal boundary between two adjacent enclosed documents.
+ */
+ public final static String P_INPUT_PARSING_MODE = "com.ctc.wstx.fragmentMode";
+
+ // // // DTD defaulting, overriding
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Helper classes, values enumerations
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ public final static ParsingMode PARSING_MODE_DOCUMENT = new ParsingMode();
+ public final static ParsingMode PARSING_MODE_FRAGMENT = new ParsingMode();
+ public final static ParsingMode PARSING_MODE_DOCUMENTS = new ParsingMode();
+
+ /**
+ * Inner class used for creating type-safe enumerations (prior to JDK 1.5).
+ */
+ public final static class ParsingMode
+ {
+ ParsingMode() { }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/api/WstxOutputProperties.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/api/WstxOutputProperties.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/api/WstxOutputProperties.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/api/WstxOutputProperties.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,186 @@
+package com.ctc.wstx.api;
+
+/**
+ * Class that contains constant for property names used to configure
+ * cursor and event writers produced by Wstx implementation of
+ * {@link javax.xml.stream.XMLOutputFactory}.
+ *
+ */
+public final class WstxOutputProperties
+{
+ /**
+ * Default xml version number output, if none was specified by
+ * application. Version 1.0 is used
+ * to try to maximize compatibility (some older parsers
+ * may barf on 1.1 and later...)
+ */
+ public final static String DEFAULT_XML_VERSION = "1.0";
+
+ /**
+ * If no encoding is passed, we should just default to what xml
+ * in general expects (and can determine), UTF-8.
+ *
+ * Note: you can check out bug entry [WSTX-18] for more details
+ */
+ public final static String DEFAULT_OUTPUT_ENCODING = "UTF-8";
+
+ // // // Output options, simple on/off settings:
+
+ /**
+ * Whether writer should use double quotes in the XML declaration.
+ * The default is to use single quotes.
+ *
+ * @since 4.2.2
+ */
+ public final static String P_USE_DOUBLE_QUOTES_IN_XML_DECL = "com.ctc.wstx.useDoubleQuotesInXmlDecl";
+
+ /**
+ * Whether writer should just automatically convert all calls that
+ * would normally produce CDATA to produce (quoted) text.
+ */
+ public final static String P_OUTPUT_CDATA_AS_TEXT = "com.ctc.wstx.outputCDataAsText";
+
+ /**
+ * Whether writer should copy attributes that were initially expanded
+ * using default settings ("implicit" attributes) or not.
+ */
+ public final static String P_COPY_DEFAULT_ATTRS = "com.ctc.wstx.copyDefaultAttrs";
+
+ /**
+ * Whether writer is to add a single white space before closing "/>"
+ * of the empty element or not. It is sometimes useful to add to
+ * increase compatibility with HTML browsers, or to increase
+ * readability.
+ *
+ * The default value is 'false', up to Woodstox 4.x.
+ *
+ * NOTE: JavaDocs for versions 4.0.0 - 4.0.7 incorrectly state that
+ * default is 'true': this is NOT the case.
+ *
+ * Note: added to resolve Jira entry
+ * WSTX-125.
+ */
+ public final static String P_ADD_SPACE_AFTER_EMPTY_ELEM = "com.ctc.wstx.addSpaceAfterEmptyElem";
+
+ /**
+ * Whether stream writer is to automatically add end elements that are
+ * needed to properly close the output tree, when the stream is closed
+ * (either explicitly by a call to
+ * The default value is 'true' as of Woodstox 4.x.
+ * Prior to 4.0, this feature was always enabled and there was no
+ * way to disable it)
+ *
+ * @since 3.2.8
+ */
+ public final static String P_AUTOMATIC_END_ELEMENTS = "com.ctc.wstx.automaticEndElements";
+
+ // // // Validation options:
+
+ /**
+ * Whether output classes should do basic verification that the output
+ * structure is well-formed (start and end elements match); that
+ * there is one and only one root, and that there is no textual content
+ * in prolog/epilog. If false, won't do any checking regarding structure.
+ */
+ public final static String P_OUTPUT_VALIDATE_STRUCTURE = "com.ctc.wstx.outputValidateStructure";
+
+ /**
+ * Whether output classes should do basic verification that the textual
+ * content output as part of nodes should be checked for validity,
+ * if there's a possibility of invalid content. Nodes that include
+ * such constraints are: comment/'--', cdata/']]>',
+ * proc. instr/'?>'.
+ */
+ public final static String P_OUTPUT_VALIDATE_CONTENT = "com.ctc.wstx.outputValidateContent";
+
+ /**
+ * Whether output classes should check uniqueness of attribute names,
+ * to prevent accidental output of duplicate attributes.
+ */
+ public final static String P_OUTPUT_VALIDATE_ATTR = "com.ctc.wstx.outputValidateAttr";
+
+ /**
+ * Whether output classes should check validity of names, ie that they
+ * only contain legal XML identifier characters.
+ */
+ public final static String P_OUTPUT_VALIDATE_NAMES = "com.ctc.wstx.outputValidateNames";
+
+ /**
+ * Property that further modifies handling of invalid content so
+ * that if {@link #P_OUTPUT_VALIDATE_CONTENT} is enabled, instead of
+ * reporting an error, writer will try to fix the problem.
+ * Invalid content in this context refers to comment
+ * content with "--", CDATA with "]]>" and proc. instr data with "?>".
+ * This can
+ * be done for some content (CDATA, possibly comment), by splitting
+ * content into separate
+ * segments; but not for others (proc. instr, since that might
+ * change the semantics in unintended ways).
+ */
+ public final static String P_OUTPUT_FIX_CONTENT = "com.ctc.wstx.outputFixContent";
+
+ /**
+ * Property that determines whether Carriage Return (\r) characters are
+ * to be escaped when output or not. If enabled, all instances of
+ * of character \r are escaped using a character entity (where possible,
+ * that is, within CHARACTERS events, and attribute values). Otherwise
+ * they are output as is. The main reason to enable this property is
+ * to ensure that carriage returns are preserved as is through parsing,
+ * since otherwise they will be converted to canonical xml linefeeds
+ * (\n), when occuring along or as part of \r\n pair.
+ */
+ public final static String P_OUTPUT_ESCAPE_CR = "com.ctc.wstx.outputEscapeCr";
+
+ /**
+ * Property that defines a {@link InvalidCharHandler} used to determine
+ * what to do with a Java character that app tries to output but which
+ * is not a valid xml character. Alternatives are converting it to
+ * another character or throw an exception: default implementations
+ * exist for both behaviors.
+ */
+ public final static String P_OUTPUT_INVALID_CHAR_HANDLER = "com.ctc.wstx.outputInvalidCharHandler";
+
+ /**
+ * Property that defines an {@link EmptyElementHandler} used to determine
+ * if the end tag for an empty element should be written or not.
+ *
+ * If specified {@link org.codehaus.stax2.XMLOutputFactory2#P_AUTOMATIC_EMPTY_ELEMENTS} is ignored.
+ */
+ public final static String P_OUTPUT_EMPTY_ELEMENT_HANDLER = "com.ctc.wstx.outputEmptyElementHandler";
+
+ // // // Per-instance access to underlying output objects
+
+ /**
+ * Property that can be used to find out the underlying
+ * {@link java.io.OutputStream} that an
+ * {@link javax.xml.stream.XMLStreamWriter} instance is using,
+ * if known (not known if constructed with a {@link java.io.Writer},
+ * or other non-stream destination). Null is returned, if not
+ * known.
+ *
+ * Note: in general it is dangerous to operate on returned stream
+ * (if any), due to buffering stream writer can do. As such, caller
+ * has to take care to know what he is doing, including properly
+ * flushing output.
+ */
+ public final static String P_OUTPUT_UNDERLYING_STREAM = "com.ctc.wstx.outputUnderlyingStream";
+
+ /**
+ * Property that can be used to find out the underlying
+ * {@link java.io.Writer} that an
+ * {@link javax.xml.stream.XMLStreamWriter} instance is using,
+ * if known (may not be known if constructed with a {@link java.io.OutputStream},
+ * or other non-Writer destination). Null is returned, if not
+ * known. Note that the Writer may be an internal wrapper over
+ * an output stream.
+ *
+ * Note: in general it is dangerous to operate on returned Writer
+ * (if any), due to buffering stream writer can do. As such, caller
+ * has to take care to know what he is doing, including properly
+ * flushing output.
+ */
+ public final static String P_OUTPUT_UNDERLYING_WRITER = "com.ctc.wstx.outputUnderlyingWriter";
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/ErrorConsts.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/ErrorConsts.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/ErrorConsts.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/ErrorConsts.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,193 @@
+package com.ctc.wstx.cfg;
+
+import javax.xml.XMLConstants;
+import javax.xml.stream.XMLStreamConstants;
+
+/**
+ * "Static" class that contains error message constants. Note that the
+ * error message constants are NOT made final; reason is that doing so
+ * would make compiler inline them in other classes. Doing so would increase
+ * class size (although not mem usage -- Strings do get interned), with
+ * minimal performance impact.
+ */
+public class ErrorConsts
+ implements XMLStreamConstants
+{
+ // // // Types of warnings we issue via XMLReporter
+
+ public static String WT_ENT_DECL = "entity declaration";
+ public static String WT_ELEM_DECL = "element declaration";
+ public static String WT_ATTR_DECL = "attribute declaration";
+ public static String WT_XML_DECL = "xml declaration";
+ public static String WT_DT_DECL = "doctype declaration";
+ public static String WT_NS_DECL = "namespace declaration";
+
+ /**
+ * This is the generic type for warnings based on XMLValidationProblem
+ * objects.
+ */
+ public static String WT_VALIDATION = "schema validation";
+
+ // // And then warning strings
+
+ public static String W_UNDEFINED_ELEM = "Undefined element \"{0}\"; referred to by attribute(s)";
+ public static String W_MIXED_ENCODINGS = "Inconsistent text encoding; declared as \"{0}\" in xml declaration, application had passed \"{1}\"";
+ public static String W_MISSING_DTD = "Missing DOCTYPE declaration in validating mode; can not validate elements or attributes";
+ public static String W_DTD_DUP_ATTR = "Attribute \"{0}\" (for element <{1}>) declared multiple times";
+ public static String W_DTD_ATTR_REDECL = "Attribute \"{0}\" already declared for element <{1}>; ignoring re-declaration";
+
+ // // // Generic errors:
+
+ public static String ERR_INTERNAL = "Internal error";
+ public static String ERR_NULL_ARG = "Illegal to pass null as argument";
+ public static String ERR_UNKNOWN_FEATURE = "Unrecognized feature \"{0}\"";
+
+ // // // Wrong reader state:
+
+ public static String ERR_STATE_NOT_STELEM = "Current event not START_ELEMENT";
+ public static String ERR_STATE_NOT_ELEM = "Current event not START_ELEMENT or END_ELEMENT";
+ public static String ERR_STATE_NOT_PI = "Current event not PROCESSING_INSTRUCTION";
+ public static String ERR_STATE_NOT_ELEM_OR_TEXT = "Current event ({0}) not START_ELEMENT, END_ELEMENT, CHARACTERS or CDATA";
+
+ // // // XML declaration related problems
+
+ public static String ERR_XML_10_VS_11 = "XML 1.0 document can not refer to XML 1.1 parsed external entities";
+
+ // // // Structural problems, prolog/epilog:
+
+ public static String ERR_DTD_IN_EPILOG = "Can not have DOCTYPE declaration in epilog";
+ public static String ERR_DTD_DUP = "Duplicate DOCTYPE declaration";
+ public static String ERR_CDATA_IN_EPILOG = " (CDATA not allowed in prolog/epilog)";
+
+ // // // Illegal input:
+
+ public static String ERR_HYPHENS_IN_COMMENT = "String '--' not allowed in comment (missing '>'?)";
+ public static String ERR_BRACKET_IN_TEXT = "String ']]>' not allowed in textual content, except as the end marker of CDATA section";
+
+ // // // Generic parsing errors:
+
+ public static String ERR_UNEXP_KEYWORD = "Unexpected keyword \"{0}\"; expected \"{1}\"";
+
+ public static String ERR_WF_PI_MISSING_TARGET = "Missing processing instruction target";
+ public static String ERR_WF_PI_XML_TARGET = "Illegal processing instruction target (\"{0}\"); 'xml' (case insensitive) is reserved by the specs.";
+ public static String ERR_WF_PI_XML_MISSING_SPACE = "excepted either space or \"?>\" after PI target";
+
+ // // // Entity problems:
+
+ public static String ERR_WF_ENTITY_EXT_DECLARED = "Entity \"{0}\" declared externally, but referenced from a document declared 'standalone=\"yes\"'";
+
+ public static String ERR_WF_GE_UNDECLARED = "Undeclared general entity \"{0}\"";
+
+ public static String ERR_WF_GE_UNDECLARED_SA = "Undeclared general entity \"{0}\" (document in stand-alone mode; perhaps declared externally?)";
+
+ // // // Namespace problems:
+
+ public static String ERR_NS_UNDECLARED = "Undeclared namespace prefix \"{0}\"";
+ public static String ERR_NS_UNDECLARED_FOR_ATTR = "Undeclared namespace prefix \"{0}\" (for attribute \"{1}\")";
+
+ public static String ERR_NS_REDECL_XML = "Trying to redeclare prefix 'xml' from its default URI '"
+ +XMLConstants.XML_NS_URI
+ +"' to \"{0}\"";
+
+ public static String ERR_NS_REDECL_XMLNS = "Trying to declare prefix 'xmlns' (illegal as per NS 1.1 #4)";
+
+ public static String ERR_NS_REDECL_XML_URI = "Trying to bind URI '"
+ +XMLConstants.XML_NS_URI+" to prefix \"{0}\" (can only bind to 'xml')";
+
+ public static String ERR_NS_REDECL_XMLNS_URI = "Trying to bind URI '"
+ +XMLConstants.XMLNS_ATTRIBUTE_NS_URI+" to prefix \"{0}\" (can not be explicitly bound)";
+
+ public static String ERR_NS_EMPTY =
+"Non-default namespace can not map to empty URI (as per Namespace 1.0 # 2) in XML 1.0 documents";
+
+
+ // // // DTD-specific:
+
+ public static String ERR_DTD_MAINLEVEL_KEYWORD = "; expected a keyword (ATTLIST, ELEMENT, ENTITY, NOTATION), comment, or conditional section";
+
+ public static String ERR_DTD_ATTR_TYPE = "; expected one of type (CDATA, ID, IDREF, IDREFS, ENTITY, ENTITIES NOTATION, NMTOKEN or NMTOKENS)";
+
+ public static String ERR_DTD_DEFAULT_TYPE = "; expected #REQUIRED, #IMPLIED or #FIXED";
+
+ public static String ERR_DTD_ELEM_REDEFD =
+ "Trying to redefine element \"{0}\" (originally defined at {1})";
+ public static String ERR_DTD_NOTATION_REDEFD =
+ "Trying to redefine notation \"{0}\" (originally defined at {1})";
+
+ public static String ERR_DTD_UNDECLARED_ENTITY =
+ "Undeclared {0} entity \"{1}\"";
+
+ public static String ERR_DTD_XML_SPACE = "Attribute xml:space has to be defined of type enumerated, and have 1 or 2 values, 'default' and/or 'preserve'";
+
+ public static String ERR_DTD_XML_ID = "Attribute xml:id has to have attribute type of ID, as per Xml:id specification";
+
+
+ // // // DTD-validation:
+
+ public static String ERR_VLD_UNKNOWN_ELEM = "Undefined element <{0}> encountered";
+
+ public static String ERR_VLD_EMPTY = "Element <{0}> has EMPTY content specification; can not contain {1}";
+ public static String ERR_VLD_NON_MIXED = "Element <{0}> has non-mixed content specification; can not contain non-white space text, or any CDATA sections";
+ public static String ERR_VLD_ANY = "Element <{0}> has ANY content specification; can not contain {1}";
+ public static String ERR_VLD_UNKNOWN_ATTR = "Element <{0}> has no attribute \"{1}\"";
+ public static String ERR_VLD_WRONG_ROOT = "Unexpected root element <{0}>; expected <{0}> as per DOCTYPE declaration";
+
+ // // // Output problems:
+
+ public static String WERR_PROLOG_CDATA =
+ "Trying to output a CDATA block outside main element tree (in prolog or epilog)";
+ public static String WERR_PROLOG_NONWS_TEXT =
+ "Trying to output non-whitespace characters outside main element tree (in prolog or epilog)";
+ public static String WERR_PROLOG_SECOND_ROOT =
+ "Trying to output second root, <{0}>";
+
+ public static String WERR_CDATA_CONTENT =
+ "Illegal input: CDATA block has embedded ']]>' in it (index {0})";
+ public static String WERR_COMMENT_CONTENT =
+ "Illegal input: comment content has embedded '--' in it (index {0})";
+
+ public static String WERR_ATTR_NO_ELEM =
+ "Trying to write an attribute when there is no open start element.";
+
+ public static String WERR_NAME_EMPTY = "Illegal to pass empty name";
+
+ public static String WERR_NAME_ILLEGAL_FIRST_CHAR = "Illegal first name character {0}";
+ public static String WERR_NAME_ILLEGAL_CHAR = "Illegal name character {0}";
+
+ /*
+ ////////////////////////////////////////////////////
+ // Utility methods
+ ////////////////////////////////////////////////////
+ */
+
+ public static String tokenTypeDesc(int type)
+ {
+ switch (type) {
+ case START_ELEMENT:
+ return "START_ELEMENT";
+ case END_ELEMENT:
+ return "END_ELEMENT";
+ case START_DOCUMENT:
+ return "START_DOCUMENT";
+ case END_DOCUMENT:
+ return "END_DOCUMENT";
+
+ case CHARACTERS:
+ return "CHARACTERS";
+ case CDATA:
+ return "CDATA";
+ case SPACE:
+ return "SPACE";
+
+ case COMMENT:
+ return "COMMENT";
+ case PROCESSING_INSTRUCTION:
+ return "PROCESSING_INSTRUCTION";
+ case DTD:
+ return "DTD";
+ case ENTITY_REFERENCE:
+ return "ENTITY_REFERENCE";
+ }
+ return "["+type+"]";
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/InputConfigFlags.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/InputConfigFlags.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/InputConfigFlags.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/InputConfigFlags.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,204 @@
+package com.ctc.wstx.cfg;
+
+/**
+ * Constant interface that contains configuration flag used by parser
+ * and parser factory, as well as some other input constants.
+ */
+public interface InputConfigFlags
+{
+ /*
+ //////////////////////////////////////////////////////
+ // Flags for standard StAX features:
+ //////////////////////////////////////////////////////
+ */
+
+ // // // Namespace handling:
+
+ /**
+ * If true, parser will handle namespaces according to XML specs; if
+ * false, will only pass them as part of element/attribute name value
+ * information.
+ */
+ final static int CFG_NAMESPACE_AWARE = 0x0001;
+
+
+ // // // Text normalization
+
+
+ /// Flag that indicates iterator should coalesce all text segments.
+ final static int CFG_COALESCE_TEXT = 0x0002;
+
+
+ // // // Entity handling
+
+ /**
+ * Flag that enables automatic replacement of internal entities
+ */
+ final static int CFG_REPLACE_ENTITY_REFS = 0x0004;
+
+ /**
+ * Flag that enables support for expanding external entities. Woodstox
+ * pretty much ignores the setting, since effectively it is irrelevant,
+ * as {@link #CFG_REPLACE_ENTITY_REFS} and {@link #CFG_SUPPORT_DTD}
+ * both need to be enabled for external entities to be supported.
+ */
+ final static int CFG_SUPPORT_EXTERNAL_ENTITIES = 0x0008;
+
+ // // // DTD handling
+
+ /**
+ * Whether DTD handling is enabled or disabled; disabling means both
+ * internal and external subsets will just be skipped unprocessed.
+ */
+ final static int CFG_SUPPORT_DTD = 0x0010;
+
+ /**
+ * Not yet (fully) supported; added as the placeholder
+ */
+ final static int CFG_VALIDATE_AGAINST_DTD = 0x0020;
+
+ // // Note: can add 2 more 'standard' flags here...
+
+ /*
+ //////////////////////////////////////////////////////
+ // Flags for StAX2 features
+ //////////////////////////////////////////////////////
+ */
+
+ /**
+ * If true, parser will report (ignorable) white space events in prolog
+ * and epilog; if false, it will silently ignore them.
+ */
+ final static int CFG_REPORT_PROLOG_WS = 0x0100;
+
+ // // // Type conversions:
+
+
+ /**
+ * If true, parser will accurately report CDATA event as such (unless
+ * coalescing); otherwise will always report them as CHARACTERS
+ * independent of coalescing settings.
+ */
+ final static int CFG_REPORT_CDATA = 0x0200;
+
+ // // // String interning:
+
+ /**
+ * If true, will guarantee that all names (attribute/element local names
+ * have been intern()ed. If false, this is not guaranteed although
+ * implementation may still choose to do it.
+ */
+ final static int CFG_INTERN_NAMES = 0x0400;
+
+ /**
+ * It true, will call intern() on all namespace URIs parsed; otherwise
+ * will just use 'regular' Strings created from parsed contents. Interning
+ * makes namespace-based access faster, but has initial overhead of
+ * intern() call.
+ */
+ final static int CFG_INTERN_NS_URIS = 0x0800;
+
+ // // // Lazy/incomplete parsing
+
+ /**
+ * Property that determines whether Event objects created will
+ * contain (accurate) {@link javax.xml.stream.Location} information or not. If not,
+ * Location may be null or a fixed location (beginning of main
+ * XML file).
+ *
+ * Note, however, that the underlying parser will still keep track
+ * of location information for error reporting purposes; it's only
+ * Event objects that are affected.
+ */
+ final static int CFG_PRESERVE_LOCATION = 0x1000;
+
+ // // // Input source handling
+
+ /**
+ * Property that enables/disables stream reader to close the underlying
+ * input source, either when it is asked to (.close() is called), or
+ * when it doesn't need it any more (reaching EOF, hitting an
+ * unrecoverable exception).
+ * As per Stax 1.0 specification, automatic closing is NOT enabled by
+ * default; except if the caller has no access to the target (i.e.
+ * when factory created it)
+ */
+ final static int CFG_AUTO_CLOSE_INPUT = 0x2000;
+
+ /*
+ //////////////////////////////////////////////////////
+ // Flags for Woodstox-specific features
+ //////////////////////////////////////////////////////
+ */
+
+ // // // Content normalization
+
+ // 20-Jan-2007, TSa: These properties removed from 4.0, deprecated:
+ final static int CFG_NORMALIZE_LFS = 0x4000;
+ //final static int CFG_NORMALIZE_ATTR_VALUES = 0x8000;
+
+ // // // Caching
+
+ /**
+ * If true, input factory is allowed cache parsed external DTD subsets,
+ * potentially speeding up things for which DTDs are needed for: entity
+ * substitution, attribute defaulting, and of course DTD-based validation.
+ */
+ final static int CFG_CACHE_DTDS = 0x00010000;
+
+ /**
+ * If true, key used for matching DTD subsets can be the public id,
+ * if false, only system id can be used.
+ */
+ final static int CFG_CACHE_DTDS_BY_PUBLIC_ID = 0x00020000;
+
+ // // // Lazy/incomplete parsing
+
+ /**
+ * If true, input factory can defer parsing of nodes until data is
+ * actually needed; if false, it has to read all the data in right
+ * away when next type is requested. Setting it to true is good for
+ * performance, in the cases where some of the nodes (like comments,
+ * processing instructions, or whole subtrees) are ignored. Otherwise
+ * setting will not make much of a difference. Downside is that error
+ * reporting is also done 'lazily'; not right away when getting the next
+ * even type but when either accessing data, or skipping it.
+ */
+ final static int CFG_LAZY_PARSING = 0x00040000;
+
+ // // // Validation support
+
+ // DTD++ support
+
+ /**
+ * If true, DTD-parser will recognize DTD++ features, and the validator
+ * will also use any such information found from DTD when DTD validation
+ * is enabled.
+ */
+ final static int CFG_SUPPORT_DTDPP = 0x00080000;
+
+ // Automatic W3C Schema support?
+ //final static int CFG_AUTOMATIC_W3C_SCHEMA = 0x00100000;
+
+ // // // Xml:id support
+
+ /**
+ * If true, xml:id attribute type assignment and matching checks will
+ * be done as per Xml:id specification. Needs to be enabled for xml:id
+ * uniqueness checks to function properly
+ */
+ final static int CFG_XMLID_TYPING = 0x00200000;
+
+ /**
+ * If true, xml:id attribute uniqueness constraints are enforced, even
+ * if not validating against DTD otherwise.
+ */
+ final static int CFG_XMLID_UNIQ_CHECKS = 0x00400000;
+
+ /**
+ * If true, the XML parser will treat character references as entities.
+ *
+ */
+ final static int CFG_TREAT_CHAR_REFS_AS_ENTS = 0x00800000;
+}
+
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/OutputConfigFlags.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/OutputConfigFlags.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/OutputConfigFlags.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/OutputConfigFlags.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,112 @@
+package com.ctc.wstx.cfg;
+
+/**
+ * Constant interface that contains configuration flag used by output
+ * classes internally, for presenting on/off configuration options.
+ */
+public interface OutputConfigFlags
+{
+ /**
+ * Flag that indicates whether writer is namespace-aware or not; if not,
+ * only local part is ever used.
+ */
+ final static int CFG_ENABLE_NS = 0x0001;
+
+ /// Flag that indicates that output class should auto-generate namespace prefixes as necessary.
+ final static int CFG_AUTOMATIC_NS = 0x0002;
+
+ /// Flag that indicates we can output 'automatic' empty elements.
+ final static int CFG_AUTOMATIC_EMPTY_ELEMENTS = 0x0004;
+
+ /**
+ * Whether writer should just automatically convert all calls that
+ * would normally produce CDATA to produce (quoted) text.
+ */
+ final static int CFG_OUTPUT_CDATA_AS_TEXT = 0x0008;
+
+ /**
+ * Flag that indicates whether attributes expanded from default attribute
+ * values should be copied to output, when using copy methods.
+ */
+ final static int CFG_COPY_DEFAULT_ATTRS = 0x0010;
+
+ /**
+ * Flag that indicates whether CR (\r, ascii 13) characters occuring
+ * in text (CHARACTERS) and attribute values should be escaped using
+ * character entities or not. Escaping is needed to enable seamless
+ * round-tripping (preserving CR characters).
+ */
+ final static int CFG_ESCAPE_CR = 0x0020;
+
+ /**
+ * Flag that indicates
+ * whether writer is to add a single white space before closing "/>"
+ * of the empty element or not. It is sometimes useful to add to
+ * increase compatibility with HTML browsers, or to increase
+ * readability.
+ */
+ final static int CFG_ADD_SPACE_AFTER_EMPTY_ELEM = 0x0040;
+
+ /**
+ * Flag that indicates we can output 'automatic' empty elements;
+ * end elements needed to close the logical output tree when
+ * stream writer is closed (by closing it explicitly, or by writing
+ * end-document event)
+ *
+ * @since 3.2.8
+ */
+ final static int CFG_AUTOMATIC_END_ELEMENTS = 0x0080;
+
+ /// Flag that indicates we should check validity of output XML structure.
+ final static int CFG_VALIDATE_STRUCTURE = 0x0100;
+
+ /**
+ * Flag that indicates we should check validity of textual content of
+ * nodes that have constraints.
+ *
+ * Specifically: comments can not have '--', CDATA sections can not
+ * have ']]>' and processing instruction can not have '?<' character
+ * combinations in content passed in.
+ */
+ final static int CFG_VALIDATE_CONTENT = 0x0200;
+
+ /**
+ * Flag that indicates we should check validity of names (element and
+ * attribute names and prefixes; processing instruction names), that they
+ * contain only legal identifier characters.
+ */
+ final static int CFG_VALIDATE_NAMES = 0x0400;
+
+ /**
+ * Flag that indicates we should check uniqueness of attribute names,
+ * to prevent accidental output of duplicate attributes.
+ */
+ final static int CFG_VALIDATE_ATTR = 0x0800;
+
+ /**
+ * Flag that will enable writer that checks for validity of content
+ * to try to fix the problem, by splitting output segments as
+ * necessary. If disabled, validation will throw an exception; and
+ * without validation no problem is noticed by writer (but instead
+ * invalid output is created).
+ */
+ final static int CFG_FIX_CONTENT = 0x1000;
+
+ /**
+ * Property that enables/disables stream write to close the underlying
+ * output target, either when it is asked to (.close() is called), or
+ * when it doesn't need it any more (reaching EOF, hitting an
+ * unrecoverable exception).
+ * As per Stax 1.0 specification, automatic closing is NOT enabled by
+ * default; except if the caller has no access to the target (i.e.
+ * when factory created it)
+ */
+ final static int CFG_AUTO_CLOSE_OUTPUT = 0x2000;
+
+ /**
+ * Property that indicates if singe quotes or double quotes should be
+ * used in the XML declaration.
+ * The default is to use single quotes.
+ */
+ final static int CFG_USE_DOUBLE_QUOTES_IN_XML_DECL = 0x4000;
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/package.html libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/package.html
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/cfg/package.html 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/cfg/package.html 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,3 @@
+
+ * As per Stax specs, related clarifying discussion on
+ * the mailing list, and especially JDK 1.6 definitions
+ * in {@link javax.xml.XMLConstants} constants, empty String
+ * should be used instead of null.
+ */
+ public final static String DEFAULT_NAMESPACE_URI = ELEM_NO_NS_URI;
+
+ // // // Well, these are not strictly xml constants, but for
+ // // // now can live here
+
+ /**
+ * This constant defines the highest Unicode character allowed
+ * in XML content.
+ */
+ final static int MAX_UNICODE_CHAR = 0x10FFFF;
+
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/compat/package.html libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/compat/package.html
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/compat/package.html 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/compat/package.html 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,7 @@
+
+ * Note: choice of java.util.logging logging is only based on the
+ * fact that it is guaranteed to be present (we have JDK 1.4 baseline
+ * requirement) so that we do not add external dependencies.
+ * It is not a recommendation for using JUL per se; most users would
+ * do well to just use slf4j or log4j directly instead.
+ *
+ * @author Tatu Saloranta
+ *
+ * @since 3.2.8
+ */
+public final class QNameCreator
+{
+ /**
+ * Creator object that creates QNames using proper 3-arg constructor.
+ * If dynamic class loading fails
+ */
+ private final static Helper _helper;
+ static {
+ Helper h = null;
+ try {
+ // Not sure where it'll fail, constructor or create...
+ Helper h0 = new Helper();
+ /*QName n =*/ h0.create("elem", "http://dummy", "ns");
+ h = h0;
+ } catch (Throwable t) {
+ String msg = "Could not construct QNameCreator.Helper; assume 3-arg QName constructor not available and use 2-arg method instead. Problem: "+t.getMessage();
+ try {
+ Logger.getLogger("com.ctc.wstx.compat.QNameCreator").warning(msg);
+ } catch (Throwable t2) { // just in case JUL craps out...
+ System.err.println("ERROR: failed to log error using Logger (problem "+t.getMessage()+"), original problem: "+msg);
+ }
+ }
+ _helper = h;
+ }
+
+ public static QName create(String uri, String localName, String prefix)
+ {
+ if (_helper == null) { // can't use 3-arg constructor; but 2-arg will be there
+ return new QName(uri, localName);
+ }
+ return _helper.create(uri, localName, prefix);
+ }
+
+ /**
+ * Helper class used to encapsulate calls to the missing method.
+ */
+ private final static class Helper
+ {
+ public Helper() { }
+
+ public QName create(String localName, String nsURI, String prefix)
+ {
+ return new QName(localName, nsURI, prefix);
+ }
+ }
+}
+
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/DOMOutputElement.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/DOMOutputElement.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/DOMOutputElement.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/DOMOutputElement.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,216 @@
+package com.ctc.wstx.dom;
+
+import javax.xml.namespace.NamespaceContext;
+
+import org.w3c.dom.Element;
+import org.w3c.dom.Node;
+
+import com.ctc.wstx.sw.OutputElementBase;
+import com.ctc.wstx.util.BijectiveNsMap;
+
+/**
+ * Context object that holds information about an open element
+ * (one for which START_ELEMENT has been sent, but no END_ELEMENT)
+ */
+public final class DOMOutputElement
+ extends OutputElementBase
+{
+ /**
+ * Reference to the parent element, element enclosing this element.
+ * Null for root element.
+ * Non-final to allow temporary pooling
+ * (on per-writer basis, to keep these short-lived).
+ */
+ private DOMOutputElement mParent;
+
+ /**
+ * DOM node that is the root under which content is written, in case
+ * where there is no parent (mParent == null). If mParent is not null,
+ * this will be null.
+ * Value is of type
+ * {@link Document}, {@link DocumentFragment} or {@link Element}
+ */
+ private final Node mRootNode;
+
+ /**
+ * Actual DOM element for which this element object acts as a proxy.
+ */
+ private Element mElement;
+
+ private boolean mDefaultNsSet;
+
+ /**
+ * Constructor for the virtual root element
+ */
+ private DOMOutputElement(Node rootNode)
+ {
+ super();
+ mRootNode = rootNode;
+ mParent = null;
+ mElement = null;
+ mNsMapping = null;
+ mNsMapShared = false;
+ mDefaultNsURI = "";
+ mRootNsContext = null;
+ mDefaultNsSet = false;
+ }
+
+ private DOMOutputElement(DOMOutputElement parent, Element element, BijectiveNsMap ns)
+ {
+ super(parent, ns);
+ mRootNode = null;
+ mParent = parent;
+ mElement = element;
+ mNsMapping = ns;
+ mNsMapShared = (ns != null);
+ mDefaultNsURI = parent.mDefaultNsURI;
+ mRootNsContext = parent.mRootNsContext;
+ mDefaultNsSet = false;
+ }
+
+ /**
+ * Method called to reuse a pooled instance.
+ *
+ * @returns Chained pooled instance that should now be head of the
+ * reuse chain
+ */
+ private void relink(DOMOutputElement parent, Element element)
+ {
+ super.relink(parent);
+ mParent = parent;
+ mElement = element;
+ parent.appendNode(element);
+ mDefaultNsSet = false;
+ }
+
+ public static DOMOutputElement createRoot(Node rootNode)
+ {
+ return new DOMOutputElement(rootNode);
+ }
+
+ /**
+ * Simplest factory method, which gets called when a 1-argument
+ * element output method is called. It is, then, assumed to
+ * use the default namespace.
+ * Will both create the child element and attach it to parent element,
+ * or lacking own owner document.
+ */
+ protected DOMOutputElement createAndAttachChild(Element element)
+ {
+ if (mRootNode != null) {
+ mRootNode.appendChild(element);
+ } else {
+ mElement.appendChild(element);
+ }
+ return createChild(element);
+ }
+
+ protected DOMOutputElement createChild(Element element)
+ {
+ return new DOMOutputElement(this, element, mNsMapping);
+ }
+
+ /**
+ * @return New head of the recycle pool
+ */
+ protected DOMOutputElement reuseAsChild(DOMOutputElement parent, Element element)
+ {
+ DOMOutputElement poolHead = mParent;
+ relink(parent, element);
+ return poolHead;
+ }
+
+ /**
+ * Method called to temporarily link this instance to a pool, to
+ * allow reusing of instances with the same reader.
+ */
+ protected void addToPool(DOMOutputElement poolHead)
+ {
+ mParent = poolHead;
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Public API, accessors
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ public DOMOutputElement getParent() {
+ return mParent;
+ }
+
+ @Override
+ public boolean isRoot() {
+ // (Virtual) Root element has no parent...
+ return (mParent == null);
+ }
+
+ /**
+ * @return String presentation of the fully-qualified name, in
+ * "prefix:localName" format (no URI). Useful for error and
+ * debugging messages.
+ */
+ @Override
+ public String getNameDesc() {
+ if(mElement != null) {
+ return mElement.getLocalName();
+ }
+ return "#error"; // unexpected case
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Public API, mutators
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ @Override
+ public void setDefaultNsUri(String uri) {
+ mDefaultNsURI = uri;
+ mDefaultNsSet = true;
+ }
+
+ @Override
+ protected void setRootNsContext(NamespaceContext ctxt)
+ {
+ mRootNsContext = ctxt;
+ /* Let's also see if we have an active default ns mapping:
+ * (provided it hasn't yet explicitly been set for this element)
+ */
+ if (!mDefaultNsSet) {
+ String defURI = ctxt.getNamespaceURI("");
+ if (defURI != null && defURI.length() > 0) {
+ mDefaultNsURI = defURI;
+ }
+ }
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Public API, DOM manipulation
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ protected void appendNode(Node n)
+ {
+ if (mRootNode != null) {
+ mRootNode.appendChild(n);
+ } else {
+ mElement.appendChild(n);
+ }
+ }
+
+ protected void addAttribute(String pname, String value)
+ {
+ mElement.setAttribute(pname, value);
+ }
+
+ protected void addAttribute(String uri, String qname, String value)
+ {
+ mElement.setAttributeNS(uri, qname, value);
+ }
+
+ public void appendChild(Node n) {
+ mElement.appendChild(n);
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingReader.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,92 @@
+package com.ctc.wstx.dom;
+
+import java.util.Collections;
+
+import javax.xml.stream.*;
+import javax.xml.transform.dom.DOMSource;
+
+import org.codehaus.stax2.ri.dom.DOMWrappingReader;
+
+import com.ctc.wstx.api.ReaderConfig;
+import com.ctc.wstx.exc.WstxParsingException;
+
+public class WstxDOMWrappingReader
+ extends DOMWrappingReader
+{
+ protected final ReaderConfig mConfig;
+
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ protected WstxDOMWrappingReader(DOMSource src, ReaderConfig cfg)
+ throws XMLStreamException
+ {
+ super(src, cfg.willSupportNamespaces(), cfg.willCoalesceText());
+ mConfig = cfg;
+ // [WSTX-162]: allow enabling/disabling name/ns intern()ing
+ if (cfg.hasInternNamesBeenEnabled()) {
+ setInternNames(true);
+ }
+ if (cfg.hasInternNsURIsBeenEnabled()) {
+ setInternNsURIs(true);
+ }
+ }
+
+ public static WstxDOMWrappingReader createFrom(DOMSource src, ReaderConfig cfg)
+ throws XMLStreamException
+ {
+ return new WstxDOMWrappingReader(src, cfg);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Defined/Overridden config methods
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public boolean isPropertySupported(String name) {
+ // !!! TBI: not all these properties are really supported
+ return mConfig.isPropertySupported(name);
+ }
+
+ @Override
+ public Object getProperty(String name)
+ {
+ if (name.equals("javax.xml.stream.entities")) {
+ // !!! TBI
+ return Collections.EMPTY_LIST;
+ }
+ if (name.equals("javax.xml.stream.notations")) {
+ // !!! TBI
+ return Collections.EMPTY_LIST;
+ }
+ return mConfig.getProperty(name);
+ }
+
+ @Override
+ public boolean setProperty(String name, Object value)
+ {
+ // Basic config accessor works just fine...
+ return mConfig.setProperty(name, value);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Defined/Overridden error reporting
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ protected void throwStreamException(String msg, Location loc)
+ throws XMLStreamException
+ {
+ if (loc == null) {
+ throw new WstxParsingException(msg);
+ }
+ throw new WstxParsingException(msg, loc);
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dom/WstxDOMWrappingWriter.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,761 @@
+package com.ctc.wstx.dom;
+
+import java.util.*;
+
+import javax.xml.XMLConstants;
+import javax.xml.namespace.*;
+import javax.xml.stream.*;
+import javax.xml.transform.dom.DOMResult;
+
+import org.w3c.dom.*;
+
+import org.codehaus.stax2.ri.EmptyNamespaceContext;
+import org.codehaus.stax2.ri.dom.DOMWrappingWriter;
+
+import com.ctc.wstx.api.WriterConfig;
+import com.ctc.wstx.cfg.ErrorConsts;
+import com.ctc.wstx.sw.OutputElementBase;
+
+/* TODO:
+ *
+ * - validator interface implementation
+ */
+
+/**
+ * This is an adapter class that allows building a DOM tree using
+ * {@link XMLStreamWriter} interface.
+ *
+ * Note that the implementation is only to be used for use with
+ *
+ * Some notes regarding missing/incomplete functionality:
+ *
+ * Note: explicit empty element (written using
+ *
+ * Note: while this is often the same as {@link #mCurrElem},
+ * it's not always. Specifically, an empty element (written
+ * explicitly using
+ * Note:
+ */
+ public int getValueType() {
+ return TYPE_CDATA;
+ }
+
+ public String getValueTypeString()
+ {
+ return sTypes[getValueType()];
+ }
+
+ public boolean typeIsId() {
+ return false;
+ }
+
+ public boolean typeIsNotation() {
+ return false;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ public abstract String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException;
+
+ /**
+ *
+ * Note: the default implementation is not optimized, as it does
+ * a potentially unnecessary copy of the contents. It is expected that
+ * this method is seldom called (Woodstox never directly calls it; it
+ * only gets called for chained validators when one validator normalizes
+ * the value, and then following validators are passed a String, not
+ * char array)
+ */
+ public String validate(DTDValidatorBase v, String value, boolean normalize)
+ throws XMLStreamException
+ {
+ int len = value.length();
+ /* Temporary buffer has to come from the validator itself, since
+ * attribute objects are stateless and shared...
+ */
+ char[] cbuf = v.getTempAttrValueBuffer(value.length());
+ if (len > 0) {
+ value.getChars(0, len, cbuf, 0);
+ }
+ return validate(v, cbuf, 0, len, normalize);
+ }
+
+ /**
+ * Method called by the {@link DTDValidator}
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ public abstract void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException;
+
+ /**
+ * Method called when no validation is to be done, but value is still
+ * to be normalized as much as it can. What this usually means is that
+ * all regular space (parser earlier on converts other white space to
+ * spaces, except for specific character entities; and these special
+ * cases are NOT to be normalized).
+ *
+ * The only exception is that CDATA will not do any normalization. But
+ * for now, let's implement basic functionality that CDTA instance will
+ * override
+ *
+ * @param v Validator that invoked normalization
+ *
+ * @return Normalized value as a String, if any changes were done;
+ * null if input was normalized
+ */
+ public String normalize(DTDValidatorBase v, char[] cbuf, int start, int end)
+ {
+ return StringUtil.normalizeSpaces(cbuf, start, end);
+ }
+
+ /**
+ * Method called to do initial normalization of the default attribute
+ * value, without trying to verify its validity. Thus, it's
+ * called independent of whether we are fully validating the document.
+ */
+ public void normalizeDefault()
+ {
+ String val = mDefValue.getValue();
+ if (val.length() > 0) {
+ char[] cbuf = val.toCharArray();
+ String str = StringUtil.normalizeSpaces(cbuf, 0, cbuf.length);
+ if (str != null) {
+ mDefValue.setValue(str);
+ }
+ }
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Package methods, validation helper methods
+ ///////////////////////////////////////////////////
+ */
+
+ protected String validateDefaultName(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String origDefValue = mDefValue.getValue();
+ String defValue = origDefValue.trim();
+
+ if (defValue.length() == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; empty String is not a valid name");
+ }
+
+ // Ok, needs to be a valid XML name:
+ int illegalIx = WstxInputData.findIllegalNameChar(defValue, mCfgNsAware, mCfgXml11);
+ if (illegalIx >= 0) {
+ if (illegalIx == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue+"'; character "
+ +WstxInputData.getCharDesc(defValue.charAt(0))
+ +") not valid first character of a name");
+ } else {
+ reportValidationProblem(rep, "Invalid default value '"+defValue+"'; character #"+illegalIx+" ("
+ +WstxInputData.getCharDesc(defValue.charAt(illegalIx))
+ +") not valid name character");
+ }
+ }
+
+ // Ok, cool it's ok...
+ return normalize ? defValue : origDefValue;
+ }
+
+ protected String validateDefaultNames(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String defValue = mDefValue.getValue().trim();
+ int len = defValue.length();
+
+ // Then code similar to actual value validation:
+ StringBuilder sb = null;
+ int count = 0;
+ int start = 0;
+
+ main_loop:
+ while (start < len) {
+ char c = defValue.charAt(start);
+
+ // Ok, any white space to skip?
+ while (true) {
+ if (!WstxInputData.isSpaceChar(c)) {
+ break;
+ }
+ if (++start >= len) {
+ break main_loop;
+ }
+ c = defValue.charAt(start);
+ }
+
+ // Then need to find the token itself:
+ int i = start+1;
+
+ for (; i < len; ++i) {
+ if (WstxInputData.isSpaceChar(defValue.charAt(i))) {
+ break;
+ }
+ }
+ String token = defValue.substring(start, i);
+ int illegalIx = WstxInputData.findIllegalNameChar(token, mCfgNsAware, mCfgXml11);
+ if (illegalIx >= 0) {
+ if (illegalIx == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; character "
+ +WstxInputData.getCharDesc(defValue.charAt(start))
+ +") not valid first character of a name token");
+ } else {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; character "
+ +WstxInputData.getCharDesc(c)
+ +") not a valid name character");
+ }
+ }
+ ++count;
+ if (normalize) {
+ if (sb == null) {
+ sb = new StringBuilder(i - start + 32);
+ } else {
+ sb.append(' ');
+ }
+ sb.append(token);
+ }
+ start = i+1;
+ }
+
+ if (count == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; empty String is not a valid name value");
+ }
+
+ return normalize ? sb.toString() : defValue;
+ }
+
+ protected String validateDefaultNmToken(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String origDefValue = mDefValue.getValue();
+ String defValue = origDefValue.trim();
+
+ if (defValue.length() == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue+"'; empty String is not a valid NMTOKEN");
+ }
+ int illegalIx = WstxInputData.findIllegalNmtokenChar(defValue, mCfgNsAware, mCfgXml11);
+ if (illegalIx >= 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; character #"+illegalIx+" ("
+ +WstxInputData.getCharDesc(defValue.charAt(illegalIx))
+ +") not valid NMTOKEN character");
+ }
+ // Ok, cool it's ok...
+ return normalize ? defValue : origDefValue;
+ }
+
+ /**
+ * Method called by validation/normalization code for enumeration-valued
+ * attributes, to trim
+ * specified attribute value (full normalization not needed -- called
+ * for values that CAN NOT have spaces inside; such values can not
+ * be legal), and then check whether it is included
+ * in set of words (tokens) passed in. If actual value was included,
+ * will return the normalized word (as well as store shared String
+ * locally); otherwise will return null.
+ */
+ public String validateEnumValue(char[] cbuf, int start, int end,
+ boolean normalize,
+ WordResolver res)
+ {
+ /* Better NOT to build temporary Strings quite yet; can resolve
+ * matches via resolver more efficiently.
+ */
+ // Note: at this point, should only have real spaces...
+ if (normalize) {
+ while (start < end && cbuf[start] <= CHAR_SPACE) {
+ ++start;
+ }
+ while (--end > start && cbuf[end] <= CHAR_SPACE) {
+ ;
+ }
+ ++end; // so it'll point to the first char (or beyond end of buffer)
+ }
+
+ // Empty String is never legal for enums:
+ if (start >= end) {
+ return null;
+ }
+ return res.find(cbuf, start, end);
+ }
+
+ protected EntityDecl findEntityDecl(DTDValidatorBase v,
+ char[] ch, int start, int len /*, int hash*/)
+ throws XMLStreamException
+ {
+ Map
+ * Notes about thread-safety: this class is not thread-safe, since it does
+ * not have to be, in general case. That is, the only instances that can
+ * be shared are external subset instances, and those are used in read-only
+ * manner (with the exception of temporary arrays constructed on-demand).
+ */
+public final class DTDElement
+{
+
+ /*
+ ///////////////////////////////////////////////////
+ // Information about the element itself
+ ///////////////////////////////////////////////////
+ */
+
+ final PrefixedName mName;
+
+ /**
+ * Location of the (real) definition of the element; may be null for
+ * placeholder elements created to hold ATTLIST definitions
+ */
+ final Location mLocation;
+
+ /**
+ * Base validator object for validating content model of this element;
+ * may be null for some simple content models (ANY, EMPTY).
+ */
+ StructValidator mValidator;
+
+ int mAllowedContent;
+
+ /**
+ * True if the DTD was parsed (and is to be used) in namespace-aware
+ * mode.
+ * Affects (name) validation amongst other things.
+ */
+ final boolean mNsAware;
+
+ /**
+ * True if the DTD was parsed in xml1.1 compliant mode (referenced to
+ * from an xml 1.1 document).
+ * Affects (name) validation amongst other things.
+ */
+ final boolean mXml11;
+
+ /*
+ ///////////////////////////////////////////////////
+ // Attribute info
+ ///////////////////////////////////////////////////
+ */
+
+ HashMap
+ * note: although ID attributes are not to have default value,
+ * this is 'only' a validity constraint, and in dtd-aware-but-
+ * not-validating mode it is apparently 'legal' to add default
+ * values.
+ */
+ public DTDIdAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex,
+ boolean nsAware, boolean xml11)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex) {
+ return new DTDIdAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_ID;
+ }
+
+ @Override
+ public boolean typeIsId() {
+ return true;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called by the validator
+ * to let the attribute do necessary normalization and/or validation
+ * for the value.
+ */
+ @SuppressWarnings("cast")
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ // Let's trim leading white space first...
+ while (start < end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+
+ // No id?
+ if (start >= end) {
+ return reportValidationProblem(v, "Empty ID value");
+ }
+ --end; // so that it now points to the last char
+ while (end > start && WstxInputData.isSpaceChar(cbuf[end])) {
+ --end;
+ }
+
+ // Ok, need to check char validity, and also calc hash code:
+ char c = cbuf[start];
+ if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as the first ID character");
+ }
+ int hash = (int) c;
+ for (int i = start+1; i <= end; ++i) {
+ c = cbuf[i];
+ if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as an ID character");
+ }
+ hash = (hash * 31) + (int) c;
+ }
+
+ // Either way, we do need to validate characters, and calculate hash
+ ElementIdMap m = v.getIdMap();
+ PrefixedName elemName = v.getElemName();
+ Location loc = v.getLocation();
+ ElementId id = m.addDefined(cbuf, start, (end - start + 1), hash,
+ loc, elemName, mName);
+
+ // We can detect dups by checking if Location is the one we passed:
+ if (id.getLocation() != loc) {
+ return reportValidationProblem(v, "Duplicate id '"+id.getId()+"', first declared at "
+ +id.getLocation());
+ }
+
+ if (normalize) {
+ return id.getId();
+ }
+ return null;
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ {
+ // Should never get called
+ throw new IllegalStateException(ErrorConsts.ERR_INTERNAL);
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDId.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDId.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDId.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDId.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,142 @@
+/* Woodstox XML processor
+ *
+ * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
+ *
+ * Licensed under the License specified in the file LICENSE which is
+ * included with the source code.
+ * You may not use this file except in compliance with the License.
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.ctc.wstx.dtd;
+
+import java.net.URI;
+
+/**
+ * Simple key object class, used for accessing (external) DTDs when stored for
+ * caching. Main idea is that the primary id of a DTD (public or system id;
+ * latter normalized if possible)
+ * has to match, as well as couple of on/off settings for parsing (namespace
+ * support, text normalization).
+ * Latter restriction is needed since although DTDs do not deal
+ * with (or understand) namespaces, some parsing is done to be able to validate
+ * namespace aware/non-aware documents, and handling differs between the two.
+ * As to primary key part, public id is used if one was defined; if so,
+ * comparison is String equality. If not, then system id is compared: system
+ * id has to be expressed as URL if so.
+ */
+public final class DTDId
+{
+ protected final String mPublicId;
+
+ protected final URI mSystemId;
+
+ protected final int mConfigFlags;
+
+ protected final boolean mXml11;
+
+ protected int mHashCode = 0;
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Life-cycle:
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ private DTDId(String publicId, URI systemId, int configFlags, boolean xml11)
+ {
+ mPublicId = publicId;
+ mSystemId = systemId;
+ mConfigFlags = configFlags;
+ mXml11 = xml11;
+ }
+
+ public static DTDId constructFromPublicId(String publicId, int configFlags,
+ boolean xml11)
+ {
+ if (publicId == null || publicId.length() == 0) {
+ throw new IllegalArgumentException("Empty/null public id.");
+ }
+ return new DTDId(publicId, null, configFlags, xml11);
+ }
+
+ public static DTDId constructFromSystemId(URI systemId, int configFlags,
+ boolean xml11)
+ {
+ if (systemId == null) {
+ throw new IllegalArgumentException("Null system id.");
+ }
+ return new DTDId(null, systemId, configFlags, xml11);
+ }
+
+ public static DTDId construct(String publicId, URI systemId, int configFlags, boolean xml11)
+ {
+ if (publicId != null && publicId.length() > 0) {
+ return new DTDId(publicId, null, configFlags, xml11);
+ }
+ if (systemId == null) {
+ throw new IllegalArgumentException("Illegal arguments; both public and system id null/empty.");
+ }
+ return new DTDId(null, systemId, configFlags, xml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Overridden standard methods
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ @Override
+ public int hashCode() {
+ int hash = mHashCode;
+ if (hash == 0) {
+ hash = mConfigFlags;
+ if (mPublicId != null) {
+ hash ^= mPublicId.hashCode();
+ } else {
+ hash ^= mSystemId.hashCode();
+ }
+ if (mXml11) {
+ hash ^= 1;
+ }
+ mHashCode = hash;
+ }
+ return hash;
+ }
+
+ @Override
+ public String toString() {
+ StringBuffer sb = new StringBuffer(60);
+ sb.append("Public-id: ");
+ sb.append(mPublicId);
+ sb.append(", system-id: ");
+ sb.append(mSystemId);
+ sb.append(" [config flags: 0x");
+ sb.append(Integer.toHexString(mConfigFlags));
+ sb.append("], xml11: ");
+ sb.append(mXml11);
+ return sb.toString();
+ }
+
+ @Override
+ public boolean equals(Object o)
+ {
+ if (o == this) return true;
+ if (o == null || o.getClass() != getClass()) return false;
+ DTDId other = (DTDId) o;
+ if (other.mConfigFlags != mConfigFlags
+ || other.mXml11 != mXml11) {
+ return false;
+ }
+ if (mPublicId != null) {
+ String op = other.mPublicId;
+ return (op != null) && op.equals(mPublicId);
+ }
+ return mSystemId.equals(other.mSystemId);
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDIdRefAttr.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDIdRefAttr.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDIdRefAttr.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDIdRefAttr.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,120 @@
+package com.ctc.wstx.dtd;
+
+import javax.xml.stream.Location;
+import javax.xml.stream.XMLStreamException;
+
+import com.ctc.wstx.io.WstxInputData;
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.ElementId;
+import com.ctc.wstx.util.ElementIdMap;
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Attribute class for attributes that contain references
+ * to elements that have matching identifier specified.
+ */
+public final class DTDIdRefAttr
+ extends DTDAttribute
+{
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Main constructor.
+ */
+ public DTDIdRefAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex,
+ boolean nsAware, boolean xml11)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex) {
+ return new DTDIdRefAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_IDREF;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called by the validator
+ * to let the attribute do necessary normalization and/or validation
+ * for the value.
+ */
+ @SuppressWarnings("cast")
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ /* Let's skip leading/trailing white space, even if we are not
+ * to normalize visible attribute value. This allows for better
+ * round-trip handling, but still allow validation.
+ */
+ while (start < end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+
+ if (start >= end) { // empty (all white space) value?
+ return reportValidationProblem(v, "Empty IDREF value");
+ }
+
+ --end; // so that it now points to the last char
+ while (end > start && WstxInputData.isSpaceChar(cbuf[end])) {
+ --end;
+ }
+
+ // Ok, need to check char validity, and also calc hash code:
+ char c = cbuf[start];
+ if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as the first IDREF character");
+ }
+ int hash = (int) c;
+ for (int i = start+1; i <= end; ++i) {
+ c = cbuf[i];
+ if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as an IDREF character");
+ }
+ hash = (hash * 31) + (int) c;
+ }
+
+ // Ok, let's check and update id ref list...
+ ElementIdMap m = v.getIdMap();
+ Location loc = v.getLocation();
+ ElementId id = m.addReferenced(cbuf, start, (end - start + 1), hash,
+ loc, v.getElemName(), mName);
+ // and that's all; no more checks needed here
+ return normalize ? id.getId() : null;
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String def = validateDefaultName(rep, normalize);
+ if (normalize) {
+ mDefValue.setValue(def);
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDIdRefsAttr.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,162 @@
+package com.ctc.wstx.dtd;
+
+import javax.xml.stream.Location;
+import javax.xml.stream.XMLStreamException;
+
+import com.ctc.wstx.io.WstxInputData;
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.ElementId;
+import com.ctc.wstx.util.ElementIdMap;
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Attribute class for attributes that contain multiple references
+ * to elements that have matching identifier specified.
+ */
+public final class DTDIdRefsAttr
+ extends DTDAttribute
+{
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Main constructor.
+ */
+ public DTDIdRefsAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex,
+ boolean nsAware, boolean xml11)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex) {
+ return new DTDIdRefsAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_IDREFS;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ @SuppressWarnings("cast")
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ /* Let's skip leading/trailing white space, even if we are not
+ * to normalize visible attribute value. This allows for better
+ * round-trip handling (no changes for physical value caller
+ * gets), but still allows succesful validation.
+ */
+ while (start < end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+
+ // No id?
+ if (start >= end) {
+ return reportValidationProblem(v, "Empty IDREFS value");
+ }
+
+ --end; // so that it now points to the last char
+ // We now the first char is not a space by now...
+ while (end > start && WstxInputData.isSpaceChar(cbuf[end])) {
+ --end;
+ }
+
+ // Ok; now start points to first, end to last char (both inclusive)
+ ElementIdMap m = v.getIdMap();
+ Location loc = v.getLocation();
+
+ String idStr = null;
+ StringBuilder sb = null;
+ while (start <= end) {
+ // Ok, need to check char validity, and also calc hash code:
+ char c = cbuf[start];
+ if (!WstxInputData.isNameStartChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as the first IDREFS character");
+ }
+ int hash = (int) c;
+ int i = start+1;
+ for (; i <= end; ++i) {
+ c = cbuf[i];
+ if (WstxInputData.isSpaceChar(c)) {
+ break;
+ }
+ if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as an IDREFS character");
+ }
+ hash = (hash * 31) + (int) c;
+ }
+
+ // Ok, got the next id ref...
+ ElementId id = m.addReferenced(cbuf, start, i - start, hash,
+ loc, v.getElemName(), mName);
+
+ // Can skip the trailing space char (if there was one)
+ start = i+1;
+
+ /* When normalizing, we can possibly share id String, or
+ * alternatively, compose normalized String if multiple
+ */
+ if (normalize) {
+ if (idStr == null) { // first idref
+ idStr = id.getId();
+ } else {
+ if (sb == null) {
+ sb = new StringBuilder(idStr);
+ }
+ idStr = id.getId();
+ sb.append(' ');
+ sb.append(idStr);
+ }
+ }
+
+ // Ok, any white space to skip?
+ while (start <= end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+ }
+
+ if (normalize) {
+ if (sb != null) {
+ idStr = sb.toString();
+ }
+ return idStr;
+ }
+
+ return null;
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ *
+ * It's unlikely there will be default values... but just in case,
+ * let's implement it properly.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String def = validateDefaultNames(rep, normalize);
+ if (normalize) {
+ mDefValue.setValue(def);
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNmTokenAttr.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,112 @@
+package com.ctc.wstx.dtd;
+
+import javax.xml.stream.XMLStreamException;
+
+import com.ctc.wstx.io.WstxInputData;
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Specific attribute class for attributes that contain (unique)
+ * identifiers.
+ */
+public final class DTDNmTokenAttr
+ extends DTDAttribute
+{
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Main constructor.
+ */
+ public DTDNmTokenAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex,
+ boolean nsAware, boolean xml11)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex)
+ {
+ return new DTDNmTokenAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_NMTOKEN;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called by the validator
+ * to let the attribute do necessary normalization and/or validation
+ * for the value.
+ */
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ int origLen = end-start;
+
+ // Let's trim leading white space first...
+ while (start < end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+
+ // Empty value?
+ if (start >= end) {
+ return reportValidationProblem(v, "Empty NMTOKEN value");
+ }
+
+ --end; // so that it now points to the last char
+ while (end > start && WstxInputData.isSpaceChar(cbuf[end])) {
+ --end;
+ }
+
+ // Ok, need to check char validity
+ for (int i = start; i <= end; ++i) {
+ char c = cbuf[i];
+ if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid NMTOKEN character");
+ }
+ }
+
+ if (normalize) {
+ // Let's only create the String if we trimmed something
+ int len = (end - start)+1;
+ if (len != origLen) {
+ return new String(cbuf, start, len);
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String def = validateDefaultNmToken(rep, normalize);
+ if (normalize) {
+ mDefValue.setValue(def);
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNmTokensAttr.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,208 @@
+package com.ctc.wstx.dtd;
+
+import javax.xml.stream.XMLStreamException;
+
+import com.ctc.wstx.io.WstxInputData;
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Specific attribute class for attributes that contain (unique)
+ * identifiers.
+ */
+public final class DTDNmTokensAttr
+ extends DTDAttribute
+{
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Main constructor.
+ */
+ public DTDNmTokensAttr(PrefixedName name, DefaultAttrValue defValue, int specIndex,
+ boolean nsAware, boolean xml11)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex) {
+ return new DTDNmTokensAttr(mName, mDefValue, specIndex, mCfgNsAware, mCfgXml11);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_NMTOKENS;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called by the validator
+ * to let the attribute do necessary normalization and/or validation
+ * for the value.
+ */
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ //int origStart = start;
+
+ /* First things first; let's ensure value is not empty (all
+ * white space)...
+ */
+ while (start < end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+ // Empty value?
+ if (start >= end) {
+ return reportValidationProblem(v, "Empty NMTOKENS value");
+ }
+
+ /* Then, let's have separate handling for normalizing and
+ * non-normalizing case, since latter is trivially easy case:
+ */
+ if (!normalize) {
+ for (; start < end; ++start) {
+ char c = cbuf[start];
+ if (!WstxInputData.isSpaceChar(c)
+ && !WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as NMTOKENS character");
+ }
+ }
+ return null; // ok, all good
+ }
+
+ //boolean trimmed = (origStart != start);
+ //origStart = start;
+
+ --end; // so that it now points to the last char
+ // Wouldn't absolutely have to trim trailing... but is easy to do
+ while (end > start && WstxInputData.isSpaceChar(cbuf[end])) {
+ --end;
+ //trimmed = true;
+ }
+
+ /* Ok, now, need to check we only have valid chars, and maybe
+ * also coalesce multiple spaces, if any.
+ */
+ StringBuilder sb = null;
+
+ while (start <= end) {
+ int i = start;
+ for (; i <= end; ++i) {
+ char c = cbuf[i];
+ if (WstxInputData.isSpaceChar(c)) {
+ break;
+ }
+ if (!WstxInputData.isNameChar(c, mCfgNsAware, mCfgXml11)) {
+ return reportInvalidChar(v, c, "not valid as an NMTOKENS character");
+ }
+ }
+
+ if (sb == null) {
+ sb = new StringBuilder(end - start + 1);
+ } else {
+ sb.append(' ');
+ }
+ sb.append(cbuf, start, (i - start));
+
+ start = i + 1;
+ // Ok, any white space to skip?
+ while (start <= end && WstxInputData.isSpaceChar(cbuf[start])) {
+ ++start;
+ }
+ }
+
+ /* 27-Nov-2005, TSa: Could actually optimize trimming, and often
+ * avoid using StringBuilder... but let's only do it if it turns
+ * out dealing with NMTOKENS normalization shows up on profiling...
+ */
+ return sb.toString();
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ String defValue = mDefValue.getValue();
+ int len = defValue.length();
+
+ // Then code similar to actual value validation:
+ StringBuilder sb = null;
+ int count = 0;
+ int start = 0;
+
+ main_loop:
+ while (start < len) {
+ char c = defValue.charAt(start);
+
+ // Ok, any white space to skip?
+ while (true) {
+ if (!WstxInputData.isSpaceChar(c)) {
+ break;
+ }
+ if (++start >= len) {
+ break main_loop;
+ }
+ c = defValue.charAt(start);
+ }
+
+ int i = start+1;
+
+ do {
+ if (++i >= len) {
+ break;
+ }
+ c = defValue.charAt(i);
+ } while (!WstxInputData.isSpaceChar(c));
+ ++count;
+ String token = defValue.substring(start, i);
+ int illegalIx = WstxInputData.findIllegalNmtokenChar(token, mCfgNsAware, mCfgXml11);
+ if (illegalIx >= 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; character #"+illegalIx+" ("
+ +WstxInputData.getCharDesc(defValue.charAt(illegalIx))
+ +") not a valid NMTOKENS character");
+ }
+
+ if (normalize) {
+ if (sb == null) {
+ sb = new StringBuilder(i - start + 32);
+ } else {
+ sb.append(' ');
+ }
+ sb.append(token);
+ }
+ start = i+1;
+ }
+
+ if (count == 0) {
+ reportValidationProblem(rep, "Invalid default value '"+defValue
+ +"'; empty String is not a valid NMTOKENS value");
+ return;
+ }
+
+ if (normalize) {
+ mDefValue.setValue(sb.toString());
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNotationAttr.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNotationAttr.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDNotationAttr.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDNotationAttr.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,105 @@
+package com.ctc.wstx.dtd;
+
+import javax.xml.stream.XMLStreamException;
+
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.PrefixedName;
+import com.ctc.wstx.util.WordResolver;
+
+/**
+ * Specific attribute class for attributes that are of NOTATION type,
+ * and also contain enumerated set of legal values.
+ */
+public final class DTDNotationAttr
+ extends DTDAttribute
+{
+ final WordResolver mEnumValues;
+
+ /*
+ ///////////////////////////////////////////////////
+ // Life-cycle
+ ///////////////////////////////////////////////////
+ */
+
+ public DTDNotationAttr(PrefixedName name, DefaultAttrValue defValue,
+ int specIndex, boolean nsAware, boolean xml11,
+ WordResolver enumValues)
+ {
+ super(name, defValue, specIndex, nsAware, xml11);
+ mEnumValues = enumValues;
+ }
+
+ @Override
+ public DTDAttribute cloneWith(int specIndex)
+ {
+ return new DTDNotationAttr(mName, mDefValue, specIndex,
+ mCfgNsAware, mCfgXml11, mEnumValues);
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API
+ ///////////////////////////////////////////////////
+ */
+
+ @Override
+ public int getValueType() {
+ return TYPE_NOTATION;
+ }
+
+ @Override
+ public boolean typeIsNotation() {
+ return true;
+ }
+
+ /*
+ ///////////////////////////////////////////////////
+ // Public API, validation
+ ///////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called by the validator
+ * to let the attribute do necessary normalization and/or validation
+ * for the value.
+ *
+ * Note: identical to the implementation in {@link DTDEnumAttr}
+ */
+ @Override
+ public String validate(DTDValidatorBase v, char[] cbuf, int start, int end, boolean normalize)
+ throws XMLStreamException
+ {
+ String ok = validateEnumValue(cbuf, start, end, normalize, mEnumValues);
+ if (ok == null) {
+ String val = new String(cbuf, start, (end-start));
+ return reportValidationProblem(v, "Invalid notation value '"+val+"': has to be one of ("
+ +mEnumValues+")");
+ }
+ return ok;
+ }
+
+ /**
+ * Method called by the validator
+ * to ask attribute to verify that the default it has (if any) is
+ * valid for such type.
+ */
+ @Override
+ public void validateDefault(InputProblemReporter rep, boolean normalize)
+ throws XMLStreamException
+ {
+ // First, basic checks that it's a valid non-empty name:
+ String def = validateDefaultName(rep, normalize);
+
+ // And then that it's one of listed values:
+ String shared = mEnumValues.find(def);
+ if (shared == null) {
+ reportValidationProblem(rep, "Invalid default value '"+def+"': has to be one of ("
+ +mEnumValues+")");
+ }
+
+ // Ok, cool it's ok...
+ if (normalize) {
+ mDefValue.setValue(shared);
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDSchemaFactory.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDSchemaFactory.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDSchemaFactory.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDSchemaFactory.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,204 @@
+/* Woodstox XML processor
+ *
+ * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
+ *
+ * Licensed under the License specified in the file LICENSE which is
+ * included with the source code.
+ * You may not use this file except in compliance with the License.
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.ctc.wstx.dtd;
+
+import java.io.*;
+import java.net.URL;
+
+import javax.xml.stream.*;
+
+import org.codehaus.stax2.validation.*;
+
+import com.ctc.wstx.api.ReaderConfig;
+import com.ctc.wstx.api.ValidatorConfig;
+import com.ctc.wstx.cfg.XmlConsts;
+import com.ctc.wstx.exc.WstxIOException;
+import com.ctc.wstx.io.*;
+import com.ctc.wstx.util.DefaultXmlSymbolTable;
+import com.ctc.wstx.util.SymbolTable;
+import com.ctc.wstx.util.URLUtil;
+
+/**
+ * Factory for creating DTD validator schema objects (shareable stateless
+ * "blueprints" for creating actual validators).
+ *
+ * Due to close coupling of XML and DTD, some of the functionality
+ * implemented (like that of reading internal subsets embedded in XML
+ * documents) is only accessible by core Woodstox. The externally
+ * accessible
+ */
+public class DTDSchemaFactory
+ extends XMLValidationSchemaFactory
+{
+ /*
+ /////////////////////////////////////////////////////
+ // Objects shared by actual parsers
+ /////////////////////////////////////////////////////
+ */
+
+ /**
+ * 'Root' symbol table, used for creating actual symbol table instances,
+ * but never as is.
+ */
+ final static SymbolTable mRootSymbols = DefaultXmlSymbolTable.getInstance();
+ static {
+ mRootSymbols.setInternStrings(true);
+ }
+
+ /**
+ * Current configurations for this factory
+ */
+ protected final ValidatorConfig mSchemaConfig;
+
+ /**
+ * This configuration object is used (instead of a more specific one)
+ * since the actual DTD reader uses such configuration object.
+ */
+ protected final ReaderConfig mReaderConfig;
+
+ public DTDSchemaFactory()
+ {
+ super(XMLValidationSchema.SCHEMA_ID_DTD);
+ mReaderConfig = ReaderConfig.createFullDefaults();
+ mSchemaConfig = ValidatorConfig.createDefaults();
+ }
+
+ /*
+ ////////////////////////////////////////////////////////////
+ // Stax2, Configuration methods
+ ////////////////////////////////////////////////////////////
+ */
+
+ @Override
+ public boolean isPropertySupported(String propName) {
+ return mSchemaConfig.isPropertySupported(propName);
+ }
+
+ @Override
+ public boolean setProperty(String propName, Object value) {
+ return mSchemaConfig.setProperty(propName, value);
+ }
+
+ @Override
+ public Object getProperty(String propName) {
+ return mSchemaConfig.getProperty(propName);
+ }
+
+ /*
+ ////////////////////////////////////////////////////////////
+ // Stax2, Factory methods
+ ////////////////////////////////////////////////////////////
+ */
+
+ @Override
+ public XMLValidationSchema createSchema(InputStream in, String encoding,
+ String publicId, String systemId)
+ throws XMLStreamException
+ {
+ ReaderConfig rcfg = createPrivateReaderConfig();
+ return doCreateSchema(rcfg, StreamBootstrapper.getInstance
+ (publicId, SystemId.construct(systemId), in), publicId, systemId, null);
+ }
+
+ @Override
+ public XMLValidationSchema createSchema(Reader r,
+ String publicId, String systemId)
+ throws XMLStreamException
+ {
+ ReaderConfig rcfg = createPrivateReaderConfig();
+ return doCreateSchema(rcfg, ReaderBootstrapper.getInstance
+ (publicId, SystemId.construct(systemId), r, null), publicId, systemId, null);
+ }
+
+ @SuppressWarnings("resource")
+ @Override
+ public XMLValidationSchema createSchema(URL url)
+ throws XMLStreamException
+ {
+ ReaderConfig rcfg = createPrivateReaderConfig();
+ try {
+ InputStream in = URLUtil.inputStreamFromURL(url);
+ return doCreateSchema(rcfg, StreamBootstrapper.getInstance
+ (null, null, in),
+ null, url.toExternalForm(), url);
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+
+ @SuppressWarnings("resource")
+ @Override
+ public XMLValidationSchema createSchema(File f)
+ throws XMLStreamException
+ {
+ ReaderConfig rcfg = createPrivateReaderConfig();
+ try {
+ URL url = URLUtil.toURL(f);
+ return doCreateSchema(rcfg, StreamBootstrapper.getInstance
+ (null, null, new FileInputStream(f)),
+ null, url.toExternalForm(), url);
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+
+ /*
+ ////////////////////////////////////////////////////////////
+ // Internal methods
+ ////////////////////////////////////////////////////////////
+ */
+
+ /**
+ * The main validator construction method, called by all externally
+ * visible methods.
+ */
+ @SuppressWarnings("resource")
+ protected XMLValidationSchema doCreateSchema
+ (ReaderConfig rcfg, InputBootstrapper bs, String publicId, String systemIdStr, URL ctxt)
+ throws XMLStreamException
+ {
+ try {
+ Reader r = bs.bootstrapInput(rcfg, false, XmlConsts.XML_V_UNKNOWN);
+ if (bs.declaredXml11()) {
+ rcfg.enableXml11(true);
+ }
+ if (ctxt == null) { // this is just needed as context for param entity expansion
+ ctxt = URLUtil.urlFromCurrentDir();
+ }
+ /* Note: need to pass unknown for 'xmlVersion' here (as well as
+ * above for bootstrapping), since this is assumed to be the main
+ * level parsed document and no xml version compatibility checks
+ * should be done.
+ */
+ SystemId systemId = SystemId.construct(systemIdStr, ctxt);
+ WstxInputSource src = InputSourceFactory.constructEntitySource
+ (rcfg, null, null, bs, publicId, systemId, XmlConsts.XML_V_UNKNOWN, r);
+
+ /* true -> yes, fully construct for validation
+ * (does not mean it has to be used for validation, but required
+ * if it is to be used for that purpose)
+ */
+ return FullDTDReader.readExternalSubset(src, rcfg, /*int.subset*/null, true, bs.getDeclaredVersion());
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+
+ private ReaderConfig createPrivateReaderConfig()
+ {
+ return mReaderConfig.createNonShared(mRootSymbols.makeChild());
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDSubsetImpl.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDSubsetImpl.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DTDSubsetImpl.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DTDSubsetImpl.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,531 @@
+/* Woodstox XML processor
+ *
+ * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
+ *
+ * Licensed under the License specified in file LICENSE, included with
+ * the source code.
+ * You may not use this file except in compliance with the License.
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.ctc.wstx.dtd;
+
+import java.text.MessageFormat;
+import java.util.*;
+
+import javax.xml.stream.Location;
+import javax.xml.stream.XMLStreamException;
+import javax.xml.stream.events.NotationDeclaration;
+
+import org.codehaus.stax2.validation.*;
+
+import com.ctc.wstx.cfg.ErrorConsts;
+import com.ctc.wstx.ent.EntityDecl;
+import com.ctc.wstx.exc.WstxParsingException;
+import com.ctc.wstx.sr.InputProblemReporter;
+import com.ctc.wstx.util.DataUtil;
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * The default implementation of {@link DTDSubset}
+ */
+public final class DTDSubsetImpl
+ extends DTDSubset
+{
+ /**
+ * Whether this subset is cachable. Only those external
+ * subsets that do not refer to PEs defined by internal subsets (or
+ * GEs via default attribute value expansion) are cachable.
+ */
+ final boolean mIsCachable;
+
+ /**
+ * Whether this subset has full validation information; and
+ * consequently whether it will do actual validation, or just allow
+ * access to type information, notations, entities, and add default
+ * attribute values.
+ */
+ final boolean mFullyValidating;
+
+ /**
+ * Flag that indicates whether any of the elements declarared
+ * has any attribute default values for namespace pseudo-attributes.
+ */
+ final boolean mHasNsDefaults;
+
+ /*
+ //////////////////////////////////////////////////////
+ // Entity information
+ //////////////////////////////////////////////////////
+ */
+
+ /**
+ * Map (name-to-EntityDecl) of general entity declarations (internal,
+ * external) for this DTD subset.
+ */
+ final HashMap
+ * Needed
+ * for determinining if external subset materially depends on definitions
+ * from internal subset; if so, such subset is not cachable.
+ * This also
+ * means that information is not stored for non-cachable instance.
+ */
+ final Set
+ * Needed
+ * for determinining if external subset materially depends on definitions
+ * from internal subset; if so, such subset is not cachable.
+ * This also
+ * means that information is not stored for non-cachable instance.
+ */
+ final Set
+ * Note: The first Map argument WILL be modified; second one
+ * not. Caller needs to ensure this is acceptable.
+ */
+ private static
+ * API is separated from its implementation to reduce coupling; for example,
+ * it is possible to have DTD subset implementations that do not implement
+ * validation logics, just entity expansion.
+ */
+public abstract class DTDSubset
+ implements DTDValidationSchema
+{
+ /*
+ //////////////////////////////////////////////////////
+ // Life-cycle
+ //////////////////////////////////////////////////////
+ */
+
+ protected DTDSubset() { }
+
+ /**
+ * Method that will combine definitions from this internal subset with
+ * definitions from passed-in external subset, producing a new combined
+ * DTDSubset instance.
+ */
+ public abstract DTDSubset combineWithExternalSubset(InputProblemReporter rep,
+ DTDSubset extSubset)
+ throws XMLStreamException;
+
+ /*
+ //////////////////////////////////////////////////////
+ // XMLValidationSchema implementation
+ //////////////////////////////////////////////////////
+ */
+
+ @Override
+ public abstract XMLValidator createValidator(ValidationContext ctxt)
+ throws XMLStreamException;
+
+ @Override
+ public String getSchemaType() {
+ return XMLValidationSchema.SCHEMA_ID_DTD;
+ }
+
+ /*
+ //////////////////////////////////////////////////////
+ // And extended DTDValidationSchema
+ //////////////////////////////////////////////////////
+ */
+
+ @Override
+ public abstract int getEntityCount();
+
+ @Override
+ public abstract int getNotationCount();
+
+ /*
+ //////////////////////////////////////////////////////
+ // Woodstox-specific API, caching support
+ //////////////////////////////////////////////////////
+ */
+
+ public abstract boolean isCachable();
+
+ /**
+ * Method used in determining whether cached external subset instance
+ * can be used with specified internal subset. If ext. subset references
+ * any parameter entities int subset (re-)defines, it can not; otherwise
+ * it can be used.
+ *
+ * @return True if this (external) subset refers to a parameter entity
+ * defined in passed-in internal subset.
+ */
+ public abstract boolean isReusableWith(DTDSubset intSubset);
+
+ /*
+ //////////////////////////////////////////////////////
+ // Woodstox-specific API, entity/notation handling
+ //////////////////////////////////////////////////////
+ */
+
+ public abstract HashMap
+ * Note, too, that underlying {@link IOException}s are generally wrapped
+ * as {@link XMLStreamException}s. This is needed to reduce amount of
+ * work caller has to do for wrapping. It will still be possible to
+ * unwrap these exceptions further up the call stack if need be.
+ */
+final class DTDWriter
+{
+ /*
+ //////////////////////////////////////////////////
+ // Configuration
+ //////////////////////////////////////////////////
+ */
+
+ final Writer mWriter;
+
+ final boolean mIncludeComments;
+
+ final boolean mIncludeConditionals;
+
+ final boolean mIncludePEs;
+
+ /*
+ //////////////////////////////////////////////////
+ // Output status
+ //////////////////////////////////////////////////
+ */
+
+ /**
+ * Counter that indicates whether flattened output should be written to
+ * (non-null) mWriter; values above zero indicate output is enabled,
+ * zero and below that output is disabled.
+ * Only enabled if mWriter is not
+ * null; will be temporarily disabled during processing of content
+ * that is not to be included (PE reference; or comments / conditional
+ * sections if comment/cs output is suppressed)
+ */
+ int mIsFlattening = 0;
+
+ /**
+ * Pointer to first character in the current input buffer that
+ * has not yet been written to flatten writer.
+ */
+ int mFlattenStart = 0;
+
+ /*
+ //////////////////////////////////////////////////
+ // Life-cycle
+ //////////////////////////////////////////////////
+ */
+
+ public DTDWriter(Writer out, boolean inclComments, boolean inclCond,
+ boolean inclPEs)
+ {
+ mWriter = out;
+ mIncludeComments = inclComments;
+ mIncludeConditionals = inclCond;
+ mIncludePEs = inclPEs;
+
+ mIsFlattening = 1; // starts enabled
+ }
+
+ /*
+ //////////////////////////////////////////////////
+ // Public API, accessors, state change
+ //////////////////////////////////////////////////
+ */
+
+ public boolean includeComments() {
+ return mIncludeComments;
+ }
+
+ public boolean includeConditionals() {
+ return mIncludeConditionals;
+ }
+
+ public boolean includeParamEntities() {
+ return mIncludePEs;
+ }
+
+ public void disableOutput()
+ {
+ --mIsFlattening;
+ }
+
+ public void enableOutput(int newStart)
+ {
+ ++mIsFlattening;
+ mFlattenStart = newStart;
+ }
+
+ public void setFlattenStart(int ptr) {
+ mFlattenStart = ptr;
+ }
+
+ public int getFlattenStart() {
+ return mFlattenStart;
+ }
+
+ /*
+ //////////////////////////////////////////////////
+ // Public API, output methods:
+ //////////////////////////////////////////////////
+ */
+
+ public void flush(char[] buf, int upUntil)
+ throws XMLStreamException
+ {
+ if (mFlattenStart < upUntil) {
+ if (mIsFlattening > 0) {
+ try {
+ mWriter.write(buf, mFlattenStart, upUntil - mFlattenStart);
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+ mFlattenStart = upUntil;
+ }
+ }
+
+ /**
+ * Method called when explicit output has to be done for flatten output:
+ * this is usually done when there's need to do speculative checks
+ * before it's known if some chars are output (when suppressing comments
+ * or conditional sections)
+ */
+ public void output(String output)
+ throws XMLStreamException
+ {
+ if (mIsFlattening > 0) {
+ try {
+ mWriter.write(output);
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+ }
+
+ public void output(char c)
+ throws XMLStreamException
+ {
+ if (mIsFlattening > 0) {
+ try {
+ mWriter.write(c);
+ } catch (IOException ioe) {
+ throw new WstxIOException(ioe);
+ }
+ }
+ }
+}
+
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/EmptyValidator.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/EmptyValidator.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/EmptyValidator.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/EmptyValidator.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,49 @@
+package com.ctc.wstx.dtd;
+
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Simple content model validator that accepts no elements, ever; this
+ * is true for pure #PCDATA content model as well as EMPTY content model.
+ * Can be used as a singleton, since all info needed for diagnostics
+ * is passed via methods.
+ */
+public class EmptyValidator
+ extends StructValidator
+{
+ final static EmptyValidator sPcdataInstance = new EmptyValidator("No elements allowed in pure #PCDATA content model");
+
+ final static EmptyValidator sEmptyInstance = new EmptyValidator("No elements allowed in EMPTY content model");
+
+ final String mErrorMsg;
+
+ private EmptyValidator(String errorMsg) {
+ mErrorMsg = errorMsg;
+ }
+
+ public static EmptyValidator getPcdataInstance() { return sPcdataInstance; }
+ public static EmptyValidator getEmptyInstance() { return sPcdataInstance; }
+
+ /**
+ * Simple; can always (re)use instance itself; no state information
+ * is kept.
+ */
+ @Override
+ public StructValidator newInstance() {
+ return this;
+ }
+
+ @Override
+ public String tryToValidate(PrefixedName elemName) {
+ return mErrorMsg;
+ }
+
+ /**
+ * If we ever get as far as element closing, things are all good;
+ * can just return null.
+ */
+ @Override
+ public String fullyValid() {
+ return null;
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/FullDTDReader.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/FullDTDReader.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/FullDTDReader.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/FullDTDReader.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,3520 @@
+/* Woodstox XML processor
+ *
+ * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
+ *
+ * Licensed under the License specified in file LICENSE, included with
+ * the source code.
+ * You may not use this file except in compliance with the License.
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.ctc.wstx.dtd;
+
+import java.io.IOException;
+import java.io.Writer;
+import java.net.URL;
+import java.text.MessageFormat;
+import java.util.*;
+
+import javax.xml.stream.Location;
+import javax.xml.stream.XMLReporter;
+import javax.xml.stream.XMLStreamException;
+import javax.xml.stream.events.NotationDeclaration;
+
+import org.codehaus.stax2.validation.XMLValidationProblem;
+import org.codehaus.stax2.validation.XMLValidator;
+
+import com.ctc.wstx.api.ReaderConfig;
+import com.ctc.wstx.cfg.ErrorConsts;
+import com.ctc.wstx.cfg.XmlConsts;
+import com.ctc.wstx.ent.*;
+import com.ctc.wstx.evt.WNotationDeclaration;
+import com.ctc.wstx.exc.WstxIOException;
+import com.ctc.wstx.io.WstxInputData;
+import com.ctc.wstx.io.WstxInputSource;
+import com.ctc.wstx.util.*;
+
+/**
+ * Reader that reads in DTD information from internal or external subset.
+ *
+ * There are 2 main modes for DTDReader, depending on whether it is parsing
+ * internal or external subset. Parsing of internal subset is somewhat
+ * simpler, since no dependency checking is needed. For external subset,
+ * handling of parameter entities is bit more complicated, as care has to
+ * be taken to distinguish between using PEs defined in int. subset, and
+ * ones defined in ext. subset itself. This determines cachability of
+ * external subsets.
+ *
+ * Reader also implements simple stand-alone functionality for flattening
+ * DTD files (expanding all references to their eventual textual form);
+ * this is sometimes useful when optimizing modularized DTDs
+ * (which are more maintainable) into single monolithic DTDs (which in
+ * general can be more performant).
+ *
+ * @author Tatu Saloranta
+ */
+public class FullDTDReader
+ extends MinimalDTDReader
+{
+ /**
+ * Flag that can be changed to enable or disable interning of shared
+ * names; shared names are used for enumerated values to reduce
+ * memory usage.
+ */
+ final static boolean INTERN_SHARED_NAMES = false;
+
+ // // // Entity expansion types:
+
+ final static Boolean ENTITY_EXP_GE = Boolean.FALSE;
+
+ final static Boolean ENTITY_EXP_PE = Boolean.TRUE;
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Configuration
+ ///////////////////////////////////////////////////////////
+ */
+
+ final int mConfigFlags;
+
+ // Extracted wstx-specific settings:
+
+ final boolean mCfgSupportDTDPP;
+
+ /**
+ * This flag indicates whether we should build a validating 'real'
+ * validator (true, the usual case),
+ * or a simpler pseudo-validator that can do all non-validation tasks
+ * that are based on DTD info (entity expansion, notation references,
+ * default attribute values). Latter is used in non-validating mode.
+ *
+ */
+ final boolean mCfgFullyValidating;
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Entity handling, parameter entities (PEs)
+ ///////////////////////////////////////////////////////////
+ */
+
+ /**
+ * Set of parameter entities defined so far in the currently parsed
+ * subset. Note: the first definition sticks, entities can not be
+ * redefined.
+ *
+ * Keys are entity name Strings; values are instances of EntityDecl
+ */
+ HashMap
+ * Keys are entity name Strings; values are instances of EntityDecl
+ *
+ * Note: this Map only contains entities declared and defined in the
+ * subset being parsed; no previously defined values are passed.
+ */
+ HashMap
+ * Keys are entity name Strings; values are instances of
+ * NotationDecl objects
+ */
+ HashMap
+ * Note: see base class for some additional remarks about this
+ * method.
+ */
+ @Override
+ public EntityDecl findEntity(String entName)
+ {
+ if (mPredefdGEs != null) {
+ EntityDecl decl = mPredefdGEs.get(entName);
+ if (decl != null) {
+ return decl;
+ }
+ }
+ return mGeneralEntities.get(entName);
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Main-level parsing methods
+ ///////////////////////////////////////////////////////////
+ */
+
+ protected DTDSubset parseDTD()
+ throws XMLStreamException
+ {
+ while (true) {
+ mCheckForbiddenPEs = false; // PEs are ok at this point
+ int i = getNextAfterWS();
+ if (i < 0) {
+ if (mIsExternal) { // ok for external DTDs
+ break;
+ }
+ // Error for internal subset
+ throwUnexpectedEOF(SUFFIX_IN_DTD_INTERNAL);
+ }
+ if (i == '%') { // parameter entity
+ expandPE();
+ continue;
+ }
+
+ /* First, let's keep track of start of the directive; needed for
+ * entity and notation declaration events.
+ */
+ mTokenInputTotal = mCurrInputProcessed + mInputPtr;
+ mTokenInputRow = mCurrInputRow;
+ mTokenInputCol = mInputPtr - mCurrInputRowStart;
+
+ if (i == '<') {
+ // PEs not allowed within declarations, in the internal subset proper
+ mCheckForbiddenPEs = !mIsExternal && (mInput == mRootInput);
+ if (mFlattenWriter == null) {
+ parseDirective();
+ } else {
+ parseDirectiveFlattened();
+ }
+ continue;
+ }
+
+ if (i == ']') {
+ if (mIncludeCount == 0 && !mIsExternal) { // End of internal subset
+ break;
+ }
+ if (mIncludeCount > 0) { // active INCLUDE block(s) open?
+ boolean suppress = (mFlattenWriter != null) && !mFlattenWriter.includeConditionals();
+
+ if (suppress) {
+ mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
+ mFlattenWriter.disableOutput();
+ }
+
+ try {
+ // ]]> needs to be a token, can not come from PE:
+ char c = dtdNextFromCurr();
+ if (c == ']') {
+ c = dtdNextFromCurr();
+ if (c == '>') {
+ // Ok, fine, conditional include section ended.
+ --mIncludeCount;
+ continue;
+ }
+ }
+ throwDTDUnexpectedChar(c, "; expected ']]>' to close conditional include section");
+ } finally {
+ if (suppress) {
+ mFlattenWriter.enableOutput(mInputPtr);
+ }
+ }
+ }
+ // otherwise will fall through, and give an error
+ }
+
+ if (mIsExternal) {
+ throwDTDUnexpectedChar(i, "; expected a '<' to start a directive");
+ }
+ throwDTDUnexpectedChar(i, "; expected a '<' to start a directive, or \"]>\" to end internal subset");
+ }
+
+ /* 05-Feb-2006, TSa: Not allowed to have unclosed INCLUDE/IGNORE
+ * blocks...
+ */
+ if (mIncludeCount > 0) { // active INCLUDE block(s) open?
+ String suffix = (mIncludeCount == 1) ? "an INCLUDE block" : (""+mIncludeCount+" INCLUDE blocks");
+ throwUnexpectedEOF(getErrorMsg()+"; expected closing marker for "+suffix);
+ }
+
+ /* First check: have all notation references been resolved?
+ * (related to [WSTX-121])
+ */
+ if (mNotationForwardRefs != null && mNotationForwardRefs.size() > 0) {
+ _reportUndefinedNotationRefs();
+ }
+
+ // Ok; time to construct and return DTD data object.
+ DTDSubset ss;
+
+ // There are more settings for ext. subsets:
+ if (mIsExternal) {
+ /* External subsets are cachable if they did not refer to any
+ * PEs or GEs defined in internal subset passed in (if any),
+ * nor to any notations.
+ * We don't care about PEs it defined itself, but need to pass
+ * in Set of PEs it refers to, to check if cached copy can be
+ * used with different int. subsets.
+ * We need not worry about notations referred, since they are
+ * not allowed to be re-defined.
+ */
+ boolean cachable = !mUsesPredefdEntities && !mUsesPredefdNotations;
+ ss = DTDSubsetImpl.constructInstance(cachable,
+ mGeneralEntities, mRefdGEs,
+ null, mRefdPEs,
+ mNotations, mElements,
+ mCfgFullyValidating);
+ } else {
+ /* Internal subsets are not cachable (no unique way to refer
+ * to unique internal subsets), and there can be no references
+ * to pre-defined PEs, as none were passed.
+ */
+ ss = DTDSubsetImpl.constructInstance(false, mGeneralEntities, null,
+ mParamEntities, null,
+ mNotations, mElements,
+ mCfgFullyValidating);
+ }
+ return ss;
+ }
+
+ protected void parseDirective()
+ throws XMLStreamException
+ {
+ /* Hmmh. Don't think PEs are allowed to contain starting
+ * '!' (or '?')... and it has to come from the same
+ * input source too (no splits)
+ */
+ char c = dtdNextFromCurr();
+ if (c == '?') { // xml decl?
+ readPI();
+ return;
+ }
+ if (c != '!') { // nothing valid
+ throwDTDUnexpectedChar(c, "; expected '!' to start a directive");
+ }
+
+ /* ignore/include, comment, or directive; we are still getting
+ * token from same section though
+ */
+ c = dtdNextFromCurr();
+ if (c == '-') { // plain comment
+ c = dtdNextFromCurr();
+ if (c != '-') {
+ throwDTDUnexpectedChar(c, "; expected '-' for a comment");
+ }
+ if (mEventListener != null && mEventListener.dtdReportComments()) {
+ readComment(mEventListener);
+ } else {
+ skipComment();
+ }
+ } else if (c == '[') {
+ checkInclusion();
+ } else if (c >= 'A' && c <= 'Z') {
+ handleDeclaration(c);
+ } else {
+ throwDTDUnexpectedChar(c, ErrorConsts.ERR_DTD_MAINLEVEL_KEYWORD);
+ }
+ }
+
+ /**
+ * Method similar to {@link #parseDirective}, but one that takes care
+ * to properly output dtd contents via {@link com.ctc.wstx.dtd.DTDWriter}
+ * as necessary.
+ * Separated to simplify both methods; otherwise would end up with
+ * 'if (... flatten...) ... else ...' spaghetti code.
+ */
+ protected void parseDirectiveFlattened()
+ throws XMLStreamException
+ {
+ /* First, need to flush any flattened output there may be, at
+ * this point (except for opening lt char): and then need to
+ * temporarily disable more output until we know the type and
+ * whether it should be output or not:
+ */
+ mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
+ mFlattenWriter.disableOutput();
+
+ /* Let's determine type here, and call appropriate skip/parse
+ * methods.
+ */
+ char c = dtdNextFromCurr();
+ if (c == '?') { // xml decl?
+ mFlattenWriter.enableOutput(mInputPtr);
+ mFlattenWriter.output("= 'A' && c <= 'Z') {
+ mFlattenWriter.enableOutput(mInputPtr);
+ mFlattenWriter.output(" 0) {
+ if (mFlattenWriter != null) {
+ mFlattenWriter.setFlattenStart(mInputPtr);
+ }
+ return true;
+ }
+ input.close();
+ } catch (IOException ioe) {
+ throw constructFromIOE(ioe);
+ }
+ if (input == mRootInput) {
+ return false;
+ }
+ WstxInputSource parent = input.getParent();
+ if (parent == null) { // sanity check!
+ throwNullParent(input);
+ }
+ /* 13-Feb-2006, TSa: Ok, do we violate a proper nesting constraints
+ * with this input block closure?
+ */
+ if (mCurrDepth != input.getScopeId()) {
+ handleIncompleteEntityProblem(input);
+ }
+
+ mInput = input = parent;
+ input.restoreContext(this);
+ if (mFlattenWriter != null) {
+ mFlattenWriter.setFlattenStart(mInputPtr);
+ }
+ mInputTopDepth = input.getScopeId();
+ /* 21-Feb-2006, TSa: Since linefeed normalization needs to be
+ * suppressed for internal entity expansion, we may need to
+ * change the state...
+ */
+ if (!mNormalizeLFs) {
+ mNormalizeLFs = !input.fromInternalEntity();
+ }
+ // Maybe there are leftovers from that input in buffer now?
+ } while (mInputPtr >= mInputEnd);
+
+ return true;
+ }
+
+ @Override
+ protected boolean loadMoreFromCurrent() throws XMLStreamException
+ {
+ // Any flattened not-yet-output input to flush?
+ if (mFlattenWriter != null) {
+ mFlattenWriter.flush(mInputBuffer, mInputEnd);
+ }
+
+ // Need to update offsets properly
+ mCurrInputProcessed += mInputEnd;
+ mCurrInputRowStart -= mInputEnd;
+ try {
+ int count = mInput.readInto(this);
+ if (count > 0) {
+ if (mFlattenWriter != null) {
+ mFlattenWriter.setFlattenStart(mInputPtr);
+ }
+ return true;
+ }
+ } catch (IOException ie) {
+ throwFromIOE(ie);
+ }
+ return false;
+ }
+
+ @Override
+ protected boolean ensureInput(int minAmount) throws XMLStreamException
+ {
+ int currAmount = mInputEnd - mInputPtr;
+ if (currAmount >= minAmount) {
+ return true;
+ }
+ // Any flattened not-yet-output input to flush?
+ if (mFlattenWriter != null) {
+ mFlattenWriter.flush(mInputBuffer, mInputEnd);
+ }
+ try {
+ if (mInput.readMore(this, minAmount)) {
+ if (mFlattenWriter != null) {
+ //mFlattenWriter.setFlattenStart(mInputPtr);
+ mFlattenWriter.setFlattenStart(currAmount);
+ }
+ return true;
+ }
+ } catch (IOException ie) {
+ throwFromIOE(ie);
+ }
+ return false;
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Internal methods, input access:
+ ///////////////////////////////////////////////////////////
+ */
+
+ private void loadMoreScoped(WstxInputSource currScope,
+ String entityName, Location loc)
+ throws XMLStreamException
+ {
+ boolean check = (mInput == currScope);
+ loadMore(getErrorMsg());
+ // Did we get out of the scope?
+ if (check && (mInput != currScope)) {
+ _reportWFCViolation("Unterminated entity value for entity '"
+ +entityName+"' (definition started at "
+ +loc+")");
+ }
+ }
+
+ /**
+ * @return Next character from the current input block, if any left;
+ * NULL if end of block (entity expansion)
+ */
+ private char dtdNextIfAvailable()
+ throws XMLStreamException
+ {
+ char c;
+ if (mInputPtr < mInputEnd) {
+ c = mInputBuffer[mInputPtr++];
+ } else {
+ int i = peekNext();
+ if (i < 0) {
+ return CHAR_NULL;
+ }
+ ++mInputPtr;
+ c = (char) i;
+ }
+ if (c == CHAR_NULL) {
+ throwNullChar();
+ }
+ return c;
+ }
+
+ /**
+ * Method that will get next character, and either return it as is (for
+ * normal chars), or expand parameter entity that starts with next
+ * character (which has to be '%').
+ */
+ private char getNextExpanded()
+ throws XMLStreamException
+ {
+ while (true) {
+ char c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
+ if (c != '%') {
+ return c;
+ }
+ expandPE();
+ }
+ }
+
+ private char skipDtdWs(boolean handlePEs)
+ throws XMLStreamException
+ {
+ while (true) {
+ char c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
+ if (c > CHAR_SPACE) {
+ if (c == '%' && handlePEs) {
+ expandPE();
+ continue;
+ }
+ return c;
+ }
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ } else if (c != CHAR_SPACE && c != '\t') {
+ throwInvalidSpace(c);
+ }
+ }
+ }
+
+ /**
+ * Note: Apparently a parameter entity expansion does also count
+ * as white space (that is, PEs outside of quoted text are considered
+ * to be separated by white spaces on both sides). Fortunately this
+ * can be handled by 2 little hacks: both a start of a PE, and an
+ * end of input block (== end of PE expansion) count as succesful
+ * spaces.
+ *
+ * @return Character following the obligatory boundary (white space
+ * or PE start/end)
+ */
+ private char skipObligatoryDtdWs()
+ throws XMLStreamException
+ {
+ /* Ok; since we need at least one space, or a PE, or end of input
+ * block, let's do this unique check first...
+ */
+ int i = peekNext();
+ char c;
+
+ if (i == -1) { // just means 'local' EOF (since peek only checks current)
+ c = getNextChar(getErrorMsg());
+ // Non-space, non PE is ok, due to end-of-block...
+ if (c > CHAR_SPACE && c != '%') {
+ return c;
+ }
+ } else {
+ c = mInputBuffer[mInputPtr++]; // was peek, need to read
+ if (c > CHAR_SPACE && c != '%') {
+ throwDTDUnexpectedChar(c, "; expected a separating white space");
+ }
+ }
+
+ // Ok, got it, now can loop...
+ while (true) {
+ if (c == '%') {
+ expandPE();
+ } else if (c > CHAR_SPACE) {
+ break;
+ } else {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ } else if (c != CHAR_SPACE && c != '\t') {
+ throwInvalidSpace(c);
+ }
+ }
+ /* Now we got one space (or end of input block) -- no need to
+ * restrict get next on current block (in case PE ends); happens
+ * with xmltest/valid/not-sa/003.xml, for eaxmple.
+ */
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
+ }
+ return c;
+ }
+
+ /**
+ * Method called to handle expansion of parameter entities. When called,
+ * '%' character has been encountered as a reference indicator, and
+ * now we should get parameter entity name.
+ */
+ private void expandPE()
+ throws XMLStreamException
+ {
+ String id;
+ char c;
+
+ if (mCheckForbiddenPEs) {
+ /* Ok; we hit a PE where we should not have (within the internal
+ * dtd subset proper, within a declaration). This is a WF error.
+ */
+ throwForbiddenPE();
+ }
+
+ // 01-Jul-2004, TSa: When flattening, need to flush previous output
+ if (mFlattenWriter != null) {
+ // Flush up to but not including ampersand...
+ mFlattenWriter.flush(mInputBuffer, mInputPtr-1);
+ mFlattenWriter.disableOutput();
+ c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ id = readDTDName(c);
+ try {
+ c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ } finally {
+ // will ignore name and colon (or whatever was parsed)
+ mFlattenWriter.enableOutput(mInputPtr);
+ }
+ } else {
+ c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ id = readDTDName(c);
+ c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ }
+
+ // Should now get semicolon...
+ if (c != ';') {
+ throwDTDUnexpectedChar(c, "; expected ';' to end parameter entity name");
+ }
+ mExpandingPE = true;
+ expandEntity(id, true, ENTITY_EXP_PE);
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Internal methods, low-level parsing:
+ ///////////////////////////////////////////////////////////
+ */
+
+ /**
+ * Method called to verify whether input has specified keyword; if it
+ * has, returns null and points to char after the keyword; if not,
+ * returns whatever constitutes a keyword matched, for error
+ * reporting purposes.
+ */
+ protected String checkDTDKeyword(String exp)
+ throws XMLStreamException
+ {
+ int i = 0;
+ int len = exp.length();
+ char c = ' ';
+
+ for (; i < len; ++i) {
+ if (mInputPtr < mInputEnd) {
+ c = mInputBuffer[mInputPtr++];
+ } else {
+ c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) { // end of block, fine
+ return exp.substring(0, i);
+ }
+ }
+ if (c != exp.charAt(i)) {
+ break;
+ }
+ }
+
+ if (i == len) {
+ // Got a match? Cool... except if identifier still continues...
+ c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) { // EOB, fine
+ return null;
+ }
+ if (!isNameChar(c)) {
+ --mInputPtr; // to push it back
+ return null;
+ }
+ }
+ StringBuilder sb = new StringBuilder(exp.substring(0, i));
+ sb.append(c);
+ while (true) {
+ c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) { // EOB, fine
+ break;
+ }
+ if (!isNameChar(c) && c != ':') {
+ --mInputPtr; // to push it back
+ break;
+ }
+ sb.append(c);
+ }
+ return sb.toString();
+ }
+
+ /**
+ * Method called usually to indicate an error condition; will read rest
+ * of specified keyword (including characters that can be part of XML
+ * identifiers), append that to passed prefix (which is optional), and
+ * return resulting String.
+ *
+ * @param prefix Part of keyword already read in.
+ */
+ protected String readDTDKeyword(String prefix)
+ throws XMLStreamException
+ {
+ StringBuilder sb = new StringBuilder(prefix);
+
+ while (true) {
+ char c;
+ if (mInputPtr < mInputEnd) {
+ c = mInputBuffer[mInputPtr++];
+ } else {
+ // Don't want to cross block boundary
+ c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) {
+ break; // end-of-block
+ }
+ }
+ if (!isNameChar(c) && c != ':') {
+ --mInputPtr;
+ break;
+ }
+ sb.append(c);
+ }
+ return sb.toString();
+ }
+
+ /**
+ * @return True, if input contains 'PUBLIC' keyword; false if it
+ * contains 'SYSTEM'; otherwise throws an exception.
+ */
+ private boolean checkPublicSystemKeyword(char c)
+ throws XMLStreamException
+ {
+ String errId;
+
+ if (c == 'P') {
+ errId = checkDTDKeyword("UBLIC");
+ if (errId == null) {
+ return true;
+ }
+ errId = "P" + errId;
+ } else if (c == 'S') {
+ errId = checkDTDKeyword("YSTEM");
+ if (errId == null) {
+ return false;
+ }
+ errId = "S" + errId;
+ } else {
+ if (!isNameStartChar(c)) {
+ throwDTDUnexpectedChar(c, "; expected 'PUBLIC' or 'SYSTEM' keyword");
+ }
+ errId = readDTDKeyword(String.valueOf(c));
+ }
+
+ _reportWFCViolation("Unrecognized keyword '"+errId+"'; expected 'PUBLIC' or 'SYSTEM'");
+ return false; // never gets here
+ }
+
+ private String readDTDName(char c)
+ throws XMLStreamException
+ {
+ // Let's just check this before trying to parse the id...
+ if (!isNameStartChar(c)) {
+ throwDTDUnexpectedChar(c, "; expected an identifier");
+ }
+ return parseFullName(c);
+ }
+
+ private String readDTDLocalName(char c, boolean checkChar)
+ throws XMLStreamException
+ {
+ /* Let's just check this first, to get better error msg
+ * (parseLocalName() will double-check it too)
+ */
+ if (checkChar && !isNameStartChar(c)) {
+ throwDTDUnexpectedChar(c, "; expected an identifier");
+ }
+ return parseLocalName(c);
+ }
+
+ /**
+ * Similar to {@link #readDTDName}, except that the rules are bit looser,
+ * ie. there are no additional restrictions for the first char
+ */
+ private String readDTDNmtoken(char c)
+ throws XMLStreamException
+ {
+ char[] outBuf = getNameBuffer(64);
+ int outLen = outBuf.length;
+ int outPtr = 0;
+
+ while (true) {
+ /* Note: colon not included in name char array, since it has
+ * special meaning WRT QNames, need to add into account here:
+ */
+ if (!isNameChar(c) && c != ':') {
+ // Need to get at least one char
+ if (outPtr == 0) {
+ throwDTDUnexpectedChar(c, "; expected a NMTOKEN character to start a NMTOKEN");
+ }
+ --mInputPtr;
+ break;
+ }
+ if (outPtr >= outLen) {
+ outBuf = expandBy50Pct(outBuf);
+ outLen = outBuf.length;
+ }
+ outBuf[outPtr++] = c;
+ if (mInputPtr < mInputEnd) {
+ c = mInputBuffer[mInputPtr++];
+ } else {
+ c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) { // end-of-block
+ break;
+ }
+ }
+ }
+
+ /* Nmtokens need not be canonicalized; they will be processed
+ * as necessary later on:
+ */
+ return new String(outBuf, 0, outPtr);
+ }
+
+ /**
+ * Method that will read an element or attribute name from DTD; depending
+ * on namespace mode, it can have prefix as well.
+ *
+ * Note: returned {@link PrefixedName} instances are canonicalized so that
+ * all instances read during parsing of a single DTD subset so that
+ * identity comparison can be used instead of calling
+ * Note: c is known to be a letter (from 'A' to 'Z') at this poit.
+ */
+ private void handleDeclaration(char c)
+ throws XMLStreamException
+ {
+ String keyw = null;
+
+ /* We need to ensure that PEs do not span declaration boundaries
+ * (similar to element nesting wrt. GE expansion for xml content).
+ * This VC is defined in xml 1.0, section 2.8 as
+ * "VC: Proper Declaration/PE Nesting"
+ */
+ /* We have binary depths within DTDs, for now: since the declaration
+ * just started, we should now have 1 as the depth:
+ */
+ mCurrDepth = 1;
+
+ try {
+ do { // dummy loop, for break
+ if (c == 'A') { // ATTLIST?
+ keyw = checkDTDKeyword("TTLIST");
+ if (keyw == null) {
+ mCurrDeclaration = "ATTLIST";
+ handleAttlistDecl();
+ break;
+ }
+ keyw = "A" + keyw;
+ } else if (c == 'E') { // ENTITY, ELEMENT?
+ c = dtdNextFromCurr();
+ if (c == 'N') {
+ keyw = checkDTDKeyword("TITY");
+ if (keyw == null) {
+ mCurrDeclaration = "ENTITY";
+ handleEntityDecl(false);
+ break;
+ }
+ keyw = "EN" + keyw;
+ } else if (c == 'L') {
+ keyw = checkDTDKeyword("EMENT");
+ if (keyw == null) {
+ mCurrDeclaration = "ELEMENT";
+ handleElementDecl();
+ break;
+ }
+ keyw = "EL" + keyw;
+ } else {
+ keyw = readDTDKeyword("E"+c);
+ }
+ } else if (c == 'N') { // NOTATION?
+ keyw = checkDTDKeyword("OTATION");
+ if (keyw == null) {
+ mCurrDeclaration = "NOTATION";
+ handleNotationDecl();
+ break;
+ }
+ keyw = "N" + keyw;
+ } else if (c == 'T' && mCfgSupportDTDPP) { // (dtd++ only) TARGETNS?
+ keyw = checkDTDKeyword("ARGETNS");
+ if (keyw == null) {
+ mCurrDeclaration = "TARGETNS";
+ handleTargetNsDecl();
+ break;
+ }
+ keyw = "T" + keyw;
+ } else {
+ keyw = readDTDKeyword(String.valueOf(c));
+ }
+ // If we got this far, we got a problem...
+ _reportBadDirective(keyw);
+ } while (false);
+ /* Ok: now, the current input can not have been started
+ * within the scope... so:
+ */
+ if (mInput.getScopeId() > 0) {
+ handleGreedyEntityProblem(mInput);
+ }
+
+ } finally {
+ // Either way, declaration has ended now...
+ mCurrDepth = 0;
+ mCurrDeclaration = null;
+ }
+ }
+
+ /**
+ * Specialized method that handles potentially suppressable entity
+ * declaration. Specifically: at this point it is known that first
+ * letter is 'E', that we are outputting flattened DTD info,
+ * and that parameter entity declarations are to be suppressed.
+ * Furthermore, flatten output is still being disabled, and needs
+ * to be enabled by the method at some point.
+ */
+ private void handleSuppressedDeclaration()
+ throws XMLStreamException
+ {
+ String keyw;
+ char c = dtdNextFromCurr();
+
+ if (c == 'N') {
+ keyw = checkDTDKeyword("TITY");
+ if (keyw == null) {
+ handleEntityDecl(true);
+ return;
+ }
+ keyw = "EN" + keyw;
+ mFlattenWriter.enableOutput(mInputPtr); // error condition...
+ } else {
+ mFlattenWriter.enableOutput(mInputPtr);
+ mFlattenWriter.output(" m = getElementMap();
+ DTDElement elem = m.get(elemName);
+
+ if (elem == null) { // ok, need a placeholder
+ // Let's add ATTLIST location as the temporary location too
+ elem = DTDElement.createPlaceholder(mConfig, loc, elemName);
+ m.put(elemName, elem);
+ }
+
+ // Ok, need to loop to get all attribute defs:
+ int index = 0;
+
+ while (true) {
+ /* White space is optional, if we get the closing '>' char;
+ * otherwise it's obligatory.
+ */
+ c = getNextExpanded();
+ if (isSpaceChar(c)) {
+ // Let's push it back in case it's LF, to be handled properly
+ --mInputPtr;
+ c = skipDtdWs(true);
+
+ /* 26-Jan-2006, TSa: actually there are edge cases where
+ * we may get the attribute name right away (esp.
+ * with PEs...); so let's defer possible error for
+ * later on. Should not allow missing spaces between
+ * attribute declarations... ?
+ */
+ /*
+ } else if (c != '>') {
+ throwDTDUnexpectedChar(c, "; excepted either '>' closing ATTLIST declaration, or a white space character separating individual attribute declarations");
+ */
+ }
+ if (c == '>') {
+ break;
+ }
+ handleAttrDecl(elem, c, index, loc);
+ ++index;
+ }
+ }
+
+ private void handleElementDecl()
+ throws XMLStreamException
+ {
+ char c = skipObligatoryDtdWs();
+ final PrefixedName elemName = readDTDQName(c);
+
+ /* Ok, event needs to know its exact starting point (opening '<'
+ * char), let's get that info now (note: data has been preserved
+ * earlier)
+ */
+ Location loc = getStartLocation();
+
+ // Ok; name got, need some white space next
+ c = skipObligatoryDtdWs();
+
+ /* Then the content spec: either a special case (ANY, EMPTY), or
+ * a parenthesis group for 'real' content spec
+ */
+ StructValidator val = null;
+ int vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT;
+
+ if (c == '(') { // real content model
+ c = skipDtdWs(true);
+ if (c == '#') {
+ val = readMixedSpec(elemName, mCfgFullyValidating);
+ vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT; // checked against DTD
+ } else {
+ --mInputPtr; // let's push it back...
+ ContentSpec spec = readContentSpec(elemName, true, mCfgFullyValidating);
+ val = spec.getSimpleValidator();
+ if (val == null) {
+ val = new DFAValidator(DFAState.constructDFA(spec));
+ }
+ vldContent = XMLValidator.CONTENT_ALLOW_WS; // checked against DTD
+ }
+ } else if (isNameStartChar(c)) {
+ do { // dummy loop to allow break:
+ String keyw = null;
+ if (c == 'A') {
+ keyw = checkDTDKeyword("NY");
+ if (keyw == null) {
+ vldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT; // no DTD checks
+ break;
+ }
+ keyw = "A"+keyw;
+ } else if (c == 'E') {
+ keyw = checkDTDKeyword("MPTY");
+ if (keyw == null) {
+ val = EmptyValidator.getPcdataInstance();
+ vldContent = XMLValidator.CONTENT_ALLOW_NONE; // needed to prevent non-elements too
+ break;
+ }
+ keyw = "E"+keyw;
+ } else {
+ --mInputPtr;
+ keyw = readDTDKeyword(String.valueOf(c));
+ }
+ _reportWFCViolation("Unrecognized DTD content spec keyword '"
+ +keyw+"' (for element <"+elemName+">); expected ANY or EMPTY");
+ } while (false);
+ } else {
+ throwDTDUnexpectedChar(c, ": excepted '(' to start content specification for element <"+elemName+">");
+ }
+
+ // Ok, still need the trailing gt-char to close the declaration:
+ c = skipDtdWs(true);
+ if (c != '>') {
+ throwDTDUnexpectedChar(c, "; expected '>' to finish the element declaration for <"+elemName+">");
+ }
+
+ LinkedHashMap
+ * Note: only valid for DTD++, in 'plain DTD' mode shouldn't get
+ * called.
+ */
+ private void handleTargetNsDecl()
+ throws XMLStreamException
+ {
+ mAnyDTDppFeatures = true;
+
+ char c = skipObligatoryDtdWs();
+ String name;
+
+ // Explicit namespace name?
+ if (isNameStartChar(c)) {
+ name = readDTDLocalName(c, false);
+ c = skipObligatoryDtdWs();
+ } else { // no, default namespace (or error)
+ name = null;
+ }
+
+ // Either way, should now get a quote:
+ if (c != '"' && c != '\'') {
+ if (c == '>') { // slightly more accurate error
+ _reportWFCViolation("Missing namespace URI for TARGETNS directive");
+ }
+ throwDTDUnexpectedChar(c, "; expected a single or double quote to enclose the namespace URI");
+ }
+
+ /* !!! 07-Nov-2004, TSa: what's the exact value we should get
+ * here? Ns declarations can have any attr value...
+ */
+ String uri = parseSystemId(c, false, "in namespace URI");
+
+ // Do we need to normalize the URI?
+ if ((mConfigFlags & CFG_INTERN_NS_URIS) != 0) {
+ uri = InternCache.getInstance().intern(uri);
+ }
+
+ // Ok, and then the closing '>':
+ c = skipDtdWs(true);
+ if (c != '>') {
+ throwDTDUnexpectedChar(c, "; expected '>' to end TARGETNS directive");
+ }
+
+ if (name == null) { // default NS URI
+ mDefaultNsURI = uri;
+ } else {
+ if (mNamespaces == null) {
+ mNamespaces = new HashMap
+ * Note: it is assumed at this point that access is only from a single
+ * thread, and non-recursive -- generally valid assumption as readers are
+ * not shared. Restriction is needed since the method is not re-entrant:
+ * it uses mAccessKey during the method call.
+ */
+ private PrefixedName findSharedName(String prefix, String localName)
+ {
+ HashMap
+ * Notes about usage:
+ * equals()
- * method (but only within a single subset!). This also reduces memory
- * usage to some extent.
- */
- private PrefixedName readDTDQName(char firstChar)
- throws XMLStreamException
- {
- String prefix, localName;
-
- if (!mCfgNsEnabled) {
- prefix = null;
- localName = parseFullName(firstChar);
- } else {
- localName = parseLocalName(firstChar);
- /* Hmmh. This is tricky; should only read from the current
- * scope, but it is ok to hit end-of-block if it was a PE
- * expansion...
- */
- char c = dtdNextIfAvailable();
- if (c == CHAR_NULL) { // end-of-block
- // ok, that's it...
- prefix = null;
- } else {
- if (c == ':') { // Ok, got namespace and local name
- prefix = localName;
- c = dtdNextFromCurr();
- localName = parseLocalName(c);
- } else {
- prefix = null;
- --mInputPtr;
- }
- }
- }
-
- return findSharedName(prefix, localName);
- }
-
- private char readArity()
- throws XMLStreamException
- {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
- if (c == '?' || c == '*' || c == '+') {
- return c;
- }
- // Hmmh, not recognized, let's put it back:
- --mInputPtr;
-
- // Default is 'just one'
- return ' ';
- }
-
- /**
- * Method that reads and pre-processes replacement text for an internal
- * entity (parameter or generic).
- */
- private char[] parseEntityValue(String id, Location loc, char quoteChar)
- throws XMLStreamException
- {
- /* 25-Jun-2004, TSa: Let's first mark current input source as the
- * scope, so we can both make sure it ends in this input
- * context (file), and that embedded single/double quotes
- * in potentially expanded entities do not end the value
- * definition (as per XML 1.0/3, 4.4.5)
- */
- WstxInputSource currScope = mInput;
-
- /* 18-Jul-2004, TSa: Also, let's see if parameter entities are
- * allowed; they are only legal outside of main internal subset
- * (ie. main XML input) file (or to be precise; they are legal
- * in the int. subset only as complete declarations)
- */
- //boolean allowPEs = mIsExternal || (mInput != mRootInput);
-
- TextBuffer tb = mValueBuffer;
- if (tb == null) {
- tb = TextBuffer.createTemporaryBuffer();
- }
- tb.resetInitialized();
-
- char[] outBuf = tb.getCurrentSegment();
- int outPtr = tb.getCurrentSegmentSize();
-
- while (true) {
- if (mInputPtr >= mInputEnd) {
- loadMoreScoped(currScope, id, loc);
- }
- char c = mInputBuffer[mInputPtr++];
-
- // Let's get most normal chars 'skipped' first
- if (c >= CHAR_FIRST_PURE_TEXT) {
- ;
- } else if (c == quoteChar) {
- // Only end if we are in correct scope:
- if (mInput == currScope) {
- break;
- }
- } else if (c == '&') { // char entity that needs to be replaced?
- /* 06-Sep-2004, TSa: We can NOT expand pre-defined entities, as
- * XML specs consider them 'real' (non-char) entities.
- * And expanding them would cause problems with entities
- * that have such entities.
- */
- int d = resolveCharOnlyEntity(false);
- // Did we get a real char entity?
- if (d != 0) {
- if (d <= 0xFFFF) {
- c = (char) d;
- } else {
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- d -= 0x10000;
- outBuf[outPtr++] = (char) ((d >> 10) + 0xD800);;
- c = (char) ((d & 0x3FF) + 0xDC00);
- }
- } else {
- /* 11-Feb-2006, TSa: Even so, must verify that the
- * entity reference is well-formed.
- */
- boolean first = true;
- while (true) {
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c; // starting with '&'
- if (mInputPtr >= mInputEnd) {
- loadMoreScoped(currScope, id, loc);
- }
- c = mInputBuffer[mInputPtr++];
- if (c == ';') {
- break;
- }
- if (first) {
- first = false;
- if (isNameStartChar(c)) {
- continue;
- }
- } else {
- if (isNameChar(c)) {
- continue;
- }
- }
- if (c == ':' && !mCfgNsEnabled) {
- continue; // fine in non-ns mode
- }
- if (first) { // missing name
- throwDTDUnexpectedChar(c, "; expected entity name after '&'");
- }
- throwDTDUnexpectedChar(c, "; expected semi-colon after entity name");
- }
- // we can just fall through to let semicolon be added
- }
- // Either '&' itself, or expanded char entity
- } else if (c == '%') { // param entity?
- expandPE();
- // Need to loop over, no char available yet
- continue;
- } else if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- if (skipCRLF(c)) {
- if (!mNormalizeLFs) {
- // Special handling, to output 2 chars at a time:
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- }
- c = '\n';
- } else {
- if (mNormalizeLFs) { // Mac LF
- c = '\n';
- }
- }
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
- }
- tb.setCurrentLength(outPtr);
-
- // Ok, now need the closing '>':
- char c = skipDtdWs(true);
- if (c != '>') {
- throwDTDUnexpectedChar(c, "; expected closing '>' after ENTITY declaration");
- }
- char[] result = tb.contentsAsArray();
- mValueBuffer = tb; // recycle, if needed later on
-
- return result;
- }
-
- /**
- * This method is similar to {@link #parseEntityValue} in some ways,
- * but has some notable differences, due to the way XML specs define
- * differences. Main differences are that parameter entities are not
- * allowed (or rather, recognized as entities), and that general
- * entities need to be verified, but NOT expanded right away.
- * Whether forward references are allowed or not is an open question
- * right now.
- */
- private void parseAttrDefaultValue(DefaultAttrValue defVal, char quoteChar, PrefixedName attrName,
- Location loc, boolean gotFixed)
- throws XMLStreamException
- {
- if (quoteChar != '"' && quoteChar != '\'') { // caller doesn't test it
- String msg = "; expected a single or double quote to enclose the default value";
- if (!gotFixed) {
- msg += ", or one of keywords (#REQUIRED, #IMPLIED, #FIXED)";
- }
- msg += " (for attribute '"+attrName+"')";
- throwDTDUnexpectedChar(quoteChar, msg);
- }
-
- /* Let's mark the current input source as the scope, so we can both
- * make sure it ends in this input context (DTD subset), and that
- * embedded single/double quotes in potentially expanded entities do
- * not end the value definition (as per XML 1.0/3, 4.4.5)
- */
- WstxInputSource currScope = mInput;
-
- TextBuffer tb = mValueBuffer;
- if (tb == null) {
- tb = TextBuffer.createTemporaryBuffer();
- }
- tb.resetInitialized();
-
- int outPtr = 0;
- char[] outBuf = tb.getCurrentSegment();
- int outLen = outBuf.length;
-
- /* One more note: this is mostly cut'n pasted from stream reader's
- * parseNormalizedAttrValue...
- */
- main_loop:
-
- while (true) {
- if (mInputPtr >= mInputEnd) {
- boolean check = (mInput == currScope);
- loadMore(getErrorMsg());
- // Did we get out of the scope?
- if (check && (mInput != currScope)) {
- _reportWFCViolation("Unterminated attribute default value for attribute '"
- +attrName+"' (definition started at "
- +loc+")");
- }
- }
- char c = mInputBuffer[mInputPtr++];
-
- // Let's do a quick for most attribute content chars:
- if (c < CHAR_FIRST_PURE_TEXT) {
- if (c <= CHAR_SPACE) {
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- c = getNextChar(SUFFIX_IN_DEF_ATTR_VALUE);
- if (c != '\n') { // nope, not 2-char lf (Mac?)
- --mInputPtr;
- c = mNormalizeLFs ? '\n' : '\r';
- } else {
- // Fine if we are to normalize lfs
- /* !!! 20-Jan-2007, TSa: Hmmh. Not sure if and
- * how to preserve: for now, let's assume there's
- * no need.
- */
- /*
- if (!mNormalizeLFs) {
- if (outPtr >= outLen) { // need more room?
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- outLen = outBuf.length;
- }
- outBuf[outPtr++] = '\r';
- // c is fine to continue
- }
- */
- }
- markLF();
- } else if (c != CHAR_SPACE && c != '\t') {
- throwInvalidSpace(c);
- }
- c = CHAR_SPACE;
- } else if (c == quoteChar) {
- /* It is possible to get these via expanded entities;
- * need to make sure this is the same input level as
- * the one that had starting quote
- */
- if (mInput == currScope) {
- break;
- }
- } else if (c == '&') { // an entity of some sort...
- /* Will need to expand char entities and pre-defd
- * int. entities (amp, lt, apos, gt): first method
- * is just more optimized than the second
- */
- int d;
- if (inputInBuffer() >= 3) {
- d = resolveSimpleEntity(true);
- } else {
- d = resolveCharOnlyEntity(true);
- }
- // Only get null if it's a 'real' general entity...
- if (d == 0) {
- c = getNextChar(SUFFIX_IN_ENTITY_REF);
- String id = parseEntityName(c);
- try {
- mCurrAttrDefault = defVal;
- mExpandingPE = false;
- expandEntity(id, false, ENTITY_EXP_GE);
- } finally {
- mCurrAttrDefault = null;
- }
- // Ok, should have updated the input source by now
- continue main_loop;
- }
-
- if (c <= 0xFFFF) {
-
- } else{
- if (d <= 0xFFFF) {
- c = (char) d;
- } else {
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- d -= 0x10000;
- outBuf[outPtr++] = (char) ((d >> 10) + 0xD800);;
- c = (char) ((d & 0x3FF) + 0xDC00);
- }
- }
- } else if (c == '<') {
- throwDTDUnexpectedChar(c, SUFFIX_IN_DEF_ATTR_VALUE);
- }
- } // if (c < CHAR_FIRST_PURE_TEXT)
-
- // Ok, let's just add char in, whatever it was
- if (outPtr >= outLen) { // need more room?
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- outLen = outBuf.length;
- }
- outBuf[outPtr++] = c;
- }
-
- tb.setCurrentLength(outPtr);
- defVal.setValue(tb.contentsAsString());
- mValueBuffer = tb;
- }
-
- /**
- * Method similar to {@link #skipPI}, but one that does basic
- * well-formedness checks.
- */
- protected void readPI()
- throws XMLStreamException
- {
- String target = parseFullName();
- if (target.length() == 0) {
- _reportWFCViolation(ErrorConsts.ERR_WF_PI_MISSING_TARGET);
- }
- if (target.equalsIgnoreCase("xml")) {
- _reportWFCViolation(ErrorConsts.ERR_WF_PI_XML_TARGET, target);
- }
-
- char c = dtdNextFromCurr();
- // Ok, need a space between target and data nonetheless
- if (!isSpaceChar(c)) { // except if it ends right away
- if (c != '?' || dtdNextFromCurr() != '>') {
- throwUnexpectedChar(c, ErrorConsts.ERR_WF_PI_XML_MISSING_SPACE);
- }
- if (mEventListener != null) {
- mEventListener.dtdProcessingInstruction(target, "");
- }
- } else if (mEventListener == null) {
- /* Otherwise, not that much to check since we don't care about
- * the contents.
- */
- while (true) {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c == '?') {
- do {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- } while (c == '?');
- if (c == '>') {
- break;
- }
- }
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
- }
- } else {
- // 24-Nov-2006, TSa: Actually, someone does care...
- // First, need to skip extra space (if any)
- while (c <= CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t' && c != ' ') {
- throwInvalidSpace(c);
- }
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- }
-
- TextBuffer tb = getTextBuffer();
- char[] outBuf = tb.getCurrentSegment();
- int outPtr = 0;
-
- while (true) {
- if (c == '?') {
- while (true) {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c != '?') {
- break;
- }
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- }
- if (c == '>') {
- break;
- }
- // Need to push back char that follows '?', output '?'
- --mInputPtr;
- c = '?';
- } else if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- c = '\n';
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- }
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- }
- tb.setCurrentLength(outPtr);
- String data = tb.contentsAsString();
- mEventListener.dtdProcessingInstruction(target, data);
- }
- }
-
- /**
- * Method similar to {@link #skipComment}, but that has to collect
- * contents, to be reported for a SAX handler.
- */
- protected void readComment(DTDEventListener l)
- throws XMLStreamException
- {
- TextBuffer tb = getTextBuffer();
- char[] outBuf = tb.getCurrentSegment();
- int outPtr = 0;
-
- while (true) {
- char c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- c = '\n';
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '-') {
- c = dtdNextFromCurr();
- if (c == '-') { // Ok, has to be end marker then:
- // Either get '>' or error:
- c = dtdNextFromCurr();
- if (c != '>') {
- throwParseError(ErrorConsts.ERR_HYPHENS_IN_COMMENT);
- }
- break;
- }
- c = '-';
- --mInputPtr; // need to push back the second char read
- }
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
- }
- tb.setCurrentLength(outPtr);
- tb.fireDtdCommentEvent(l);
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, conditional blocks:
- //////////////////////////////////////////////////
- */
-
- private void checkInclusion()
- throws XMLStreamException
- {
- String keyword;
-
- // INCLUDE/IGNORE not allowed in internal subset...
- /* 18-Jul-2004, TSa: Except if it's in an expanded parsed external
- * entity...
- */
- if (!mIsExternal && mInput == mRootInput) {
- _reportWFCViolation("Internal DTD subset can not use (INCLUDE/IGNORE) directives (except via external entities)");
- }
-
- char c = skipDtdWs(true);
- if (c != 'I') {
- // let's obtain the keyword for error reporting purposes:
- keyword = readDTDKeyword(String.valueOf(c));
- } else {
- c = dtdNextFromCurr();
- if (c == 'G') {
- keyword = checkDTDKeyword("NORE");
- if (keyword == null) {
- handleIgnored();
- return;
- }
- keyword = "IG"+keyword;
- } else if (c == 'N') {
- keyword = checkDTDKeyword("CLUDE");
- if (keyword == null) {
- handleIncluded();
- return;
- }
- keyword = "IN"+keyword;
- } else {
- --mInputPtr;
- keyword = readDTDKeyword("I");
- }
- }
-
- // If we get here, it was an error...
- _reportWFCViolation("Unrecognized directive '"+keyword+"'; expected either 'IGNORE' or 'INCLUDE'");
- }
-
- private void handleIncluded()
- throws XMLStreamException
- {
- char c = skipDtdWs(false);
- if (c != '[') {
- throwDTDUnexpectedChar(c, "; expected '[' to follow 'INCLUDE' directive");
- }
- ++mIncludeCount;
- }
-
- private void handleIgnored()
- throws XMLStreamException
- {
- char c = skipDtdWs(false);
- int count = 1; // Nesting of IGNORE/INCLUDE sections we have to match
-
- if (c != '[') {
- throwDTDUnexpectedChar(c, "; expected '[' to follow 'IGNORE' directive");
- }
-
- /* Ok; now, let's just skip until we get the closing ']]>'
- */
- String errorMsg = getErrorMsg();
- while (true) {
- c = (mInputPtr < mInputEnd)
- ? mInputBuffer[mInputPtr++] : getNextChar(errorMsg);
- if (c < CHAR_SPACE) {
- if (c == '\n' || c == '\r') {
- skipCRLF(c);
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == ']') { // closing?
- if (getNextChar(errorMsg) == ']'
- && getNextChar(errorMsg) == '>') {
- if (--count < 1) { // done!
- return;
- }
- // nested ignores, let's just continue
- } else {
- --mInputPtr; // need to push one char back, may be '<'
- }
- } else if (c == '<') {
- if (getNextChar(errorMsg) == '!'
- && getNextChar(errorMsg) == '[') {
- // Further nesting, sweet
- ++count;
- } else {
- --mInputPtr; // need to push one char back, may be '<'
- }
- }
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal methods, validation, exceptions:
- //////////////////////////////////////////////////
- */
-
- private void _reportUndefinedNotationRefs()
- throws XMLStreamException
- {
- int count = mNotationForwardRefs.size();
-
- String id = (String) mNotationForwardRefs.keySet().iterator().next();
- String msg = ""+count+" referenced notation"+((count == 1) ? "":"s")+" undefined: first one '"+id+"'";
- _reportVCViolation(msg);
- }
-
- private void _reportBadDirective(String dir)
- throws XMLStreamException
- {
- String msg = "Unrecognized DTD directive ''; expected ATTLIST, ELEMENT, ENTITY or NOTATION";
- if (mCfgSupportDTDPP) {
- msg += " (or, for DTD++, TARGETNS)";
- }
- _reportWFCViolation(msg);
- }
-
- private void _reportVCViolation(String msg)
- throws XMLStreamException
- {
- /* 01-Sep-2006, TSa: Not 100% sure what's the right way to do it --
- * they are errors (non-fatal, but not warnings), but the way
- * base class handles things, we probably better 'downgrade'
- * them to warnings in non-validating mode.
- */
- if (mCfgFullyValidating) {
- reportValidationProblem(msg, XMLValidationProblem.SEVERITY_ERROR);
- } else {
- reportValidationProblem(msg, XMLValidationProblem.SEVERITY_WARNING);
- }
- }
-
- private void _reportWFCViolation(String msg)
- throws XMLStreamException
- {
- throwParseError(msg);
- }
-
- private void _reportWFCViolation(String format, Object arg)
- throws XMLStreamException
- {
- throwParseError(format, arg, null);
- }
-
- private void throwDTDElemError(String msg, Object elem)
- throws XMLStreamException
- {
- _reportWFCViolation(elemDesc(elem) + ": " + msg);
- }
-
- private void throwDTDAttrError(String msg, DTDElement elem, PrefixedName attrName)
- throws XMLStreamException
- {
- _reportWFCViolation(attrDesc(elem, attrName) + ": " + msg);
- }
-
- private void throwDTDUnexpectedChar(int i, String extraMsg)
- throws XMLStreamException
- {
- if (extraMsg == null) {
- throwUnexpectedChar(i, getErrorMsg());
- }
- throwUnexpectedChar(i, getErrorMsg()+extraMsg);
- }
-
- private void throwForbiddenPE()
- throws XMLStreamException
- {
- _reportWFCViolation("Can not have parameter entities in the internal subset, except for defining complete declarations (XML 1.0, #2.8, WFC 'PEs In Internal Subset')");
- }
-
- private String elemDesc(Object elem) {
- return "Element <"+elem+">)";
- }
-
- private String attrDesc(Object elem, PrefixedName attrName) {
- return "Attribute '"+attrName+"' (of element <"+elem+">)";
- }
-
- private String entityDesc(WstxInputSource input) {
- return "Entity &"+input.getEntityId()+";";
- }
-
- /*
- /////////////////////////////////////////////////////
- // Internal methods, main-level declaration parsing:
- /////////////////////////////////////////////////////
- */
-
- /**
- *
- *
- */
-public final class LargePrefixedNameSet
- extends PrefixedNameSet
-{
- /**
- * Let's not bother creating tiny hash areas; should seldom be a problem
- * as smaller sets are usually created using different impl. class.
- */
- final static int MIN_HASH_AREA = 8;
-
- final boolean mNsAware;
-
- /**
- * Primary hash area in which NameKeys are added. Sized to be the smallest
- * power of two bigger than number of entries; but at least 4 (it doesn't
- * make sense to create smaller arrays)
- */
- final PrefixedName[] mNames;
-
- /**
- * Secondary (spill) area, in which keys whose hash values collide
- * with primary ones are added. Number of buckets is 1/4 of number
- * of primary entries,
- */
- final Bucket[] mBuckets;
-
- public LargePrefixedNameSet(boolean nsAware, PrefixedName[] names)
- {
- mNsAware = nsAware;
- int len = names.length;
-
- // Let's find the size first... let's except 1/8 slack (88% fill rate)
- int minSize = len + ((len + 7) >> 3);
- // Let's not create hash areas smaller than certain limit
- int tableSize = MIN_HASH_AREA;
-
- while (tableSize < minSize) {
- tableSize += tableSize;
- }
-
- mNames = new PrefixedName[tableSize];
- // and 1/4 of that for spill area... but let's do that lazily
-
- Bucket[] buckets = null;
- int mask = (tableSize - 1);
-
- for (int i = 0; i < len; ++i) {
- PrefixedName nk = names[i];
- int ix = (nk.hashCode() & mask);
- if (mNames[ix] == null) { // no collision
- mNames[ix] = nk;
- } else { // collision, need to add a bucket
- ix >>= 2;
-
- Bucket old;
- if (buckets == null) {
- buckets = new Bucket[tableSize >> 2];
- old = null;
- } else {
- old = buckets[ix];
- }
- buckets[ix] = new Bucket(nk, old);
- }
- }
-
- mBuckets = buckets;
- }
-
- public boolean hasMultiple() { return true; }
-
- /**
- * @return True if the set contains specified name; false if not.
- */
- public boolean contains(PrefixedName name)
- {
- PrefixedName[] hashArea = mNames;
- int index = name.hashCode() & (hashArea.length - 1);
- PrefixedName res = hashArea[index];
-
- if (res != null && res.equals(name)) {
- return true;
- }
-
- Bucket[] buckets = mBuckets;
- if (buckets != null) {
- for (Bucket bucket = buckets[index >> 2]; bucket != null;
- bucket = bucket.getNext()) {
- res = bucket.getName();
- if (res.equals(name)) {
- return true;
- }
- }
- }
- return false;
- }
-
- /**
- * Method called by debug/error handling code, to get a list of
- * all names contained.
- */
- public void appendNames(StringBuffer sb, String sep)
- {
- // Let's first get the alphabetized list of all names from main hash
- TreeSet ts = new TreeSet();
- for (int i = 0; i < mNames.length; ++i) {
- PrefixedName name = mNames[i];
- if (name != null) {
- ts.add(name);
- }
- }
-
- // then spill area
- if (mBuckets != null) {
- for (int i = 0; i < (mNames.length >> 2); ++i) {
- Bucket b = mBuckets[i];
- while (b != null) {
- ts.add(b.getName());
- b = b.getNext();
- }
- }
- }
-
- // And then append them:
- Iterator it = ts.iterator();
- boolean first = true;
- while (it.hasNext()) {
- if (first) {
- first = false;
- } else {
- sb.append(sep);
- }
- sb.append(it.next().toString());
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Helper class(es)
- ///////////////////////////////////////////////////////////
- */
-
- private final static class Bucket
- {
- final PrefixedName mName;
-
- final Bucket mNext;
-
- public Bucket(PrefixedName name, Bucket next) {
- mName = name;
- mNext = next;
- }
-
- public PrefixedName getName() { return mName; }
- public Bucket getNext() { return mNext; }
-
- public boolean contains(PrefixedName n) {
- return mName.equals(n);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/MinimalDTDReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/MinimalDTDReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/MinimalDTDReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/MinimalDTDReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,411 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.ent.EntityDecl;
-import com.ctc.wstx.io.WstxInputData;
-import com.ctc.wstx.io.WstxInputSource;
-import com.ctc.wstx.sr.StreamScanner;
-
-/**
- * Minimal DTD reader implementation that only knows how to skip
- * internal DTD subsets.
- */
-public class MinimalDTDReader
- extends StreamScanner
-{
- /*
- //////////////////////////////////////////////////
- // Configuration
- //////////////////////////////////////////////////
- */
-
- /**
- * True, when reading external subset, false when reading internal
- * subset.
- */
- final boolean mIsExternal;
-
- /*
- //////////////////////////////////////////////////
- // Life-cycle
- //////////////////////////////////////////////////
- */
-
- /**
- * Constructor used for reading/skipping internal subset.
- */
- private MinimalDTDReader(WstxInputSource input, ReaderConfig cfg)
- {
- this(input, cfg, false);
- }
-
- /**
- * Common initialization part of int/ext subset constructors.
- */
- protected MinimalDTDReader(WstxInputSource input, ReaderConfig cfg,
- boolean isExt)
- {
- super(input, cfg, cfg.getDtdResolver());
- mIsExternal = isExt;
- /* And let's force expansion (matters mostly/only for undefined
- * entities)
- */
- mCfgReplaceEntities = true;
- }
-
- /**
- * Method that just skims
- * through structure of internal subset, but without doing any sort
- * of validation, or parsing of contents. Method may still throw an
- * exception, if skipping causes EOF or there's an I/O problem.
- *
- * @param srcData Link back to the input buffer shared with the owning
- * stream reader.
- */
- public static void skipInternalSubset(WstxInputData srcData, WstxInputSource input,
- ReaderConfig cfg)
- throws XMLStreamException
- {
- MinimalDTDReader r = new MinimalDTDReader(input, cfg);
- // Need to read from same source as the master (owning stream reader)
- r.copyBufferStateFrom(srcData);
- try {
- r.skipInternalSubset();
- } finally {
- /* And then need to restore changes back to srcData (line nrs etc);
- * effectively means that we'll stop reading internal DTD subset,
- * if so.
- */
- srcData.copyBufferStateFrom(r);
- }
- }
-
- /*
- //////////////////////////////////////////////////
- // Abstract methods from StreamScanner
- //////////////////////////////////////////////////
- */
-
- /**
- * What DTD reader returns doesn't really matter, so let's just return
- * perceived start location (different from what stream readers actually
- * do)
- */
- public final Location getLocation()
- {
- return getStartLocation();
- }
-
- // @Override
- protected EntityDecl findEntity(String id, Object arg) {
- throwIllegalCall();
- return null; // never gets here but javac needs it
- }
-
- /**
- * This is a VC, not WFC, nothing to do when skipping through
- * DTD in non-supporting mode.
- */
- protected void handleUndeclaredEntity(String id)
- throws XMLStreamException
- {
- // nothing to do...
- }
-
- /**
- * Since improper entity/PE nesting is VC, not WFC, let's not
- * react to this failure at all when only skipping the DTD subset.
- */
- protected void handleIncompleteEntityProblem(WstxInputSource closing)
- throws XMLStreamException
- {
- // nothing to do...
- }
-
- protected char handleExpandedSurrogate(char first, char second)
- {
- // should we throw an exception?
- return first;
- }
-
- /*
- //////////////////////////////////////////////////
- // Internal API
- //////////////////////////////////////////////////
- */
-
- /**
- * Method that may need to be called by attribute default value
- * validation code, during parsing....
- *
- *
- */
-public final class SmallPrefixedNameSet
- extends PrefixedNameSet
-{
- final boolean mNsAware;
-
- final String[] mStrings;
-
- public SmallPrefixedNameSet(boolean nsAware, PrefixedName[] names)
- {
- mNsAware = nsAware;
- int len = names.length;
- if (len == 0) { // sanity check
- throw new IllegalStateException("Trying to construct empty PrefixedNameSet");
- }
- mStrings = new String[nsAware ? (len+len) : len];
- for (int out = 0, in = 0; in < len; ++in) {
- PrefixedName nk = names[in];
- if (nsAware) {
- mStrings[out++] = nk.getPrefix();
- }
- mStrings[out++] = nk.getLocalName();
- }
- }
-
- public boolean hasMultiple() {
- return mStrings.length > 1;
- }
-
- /**
- * @return True if the set contains specified name; false if not.
- */
- public boolean contains(PrefixedName name)
- {
- int len = mStrings.length;
- String ln = name.getLocalName();
- String[] strs = mStrings;
-
- if (mNsAware) {
- String prefix = name.getPrefix();
- if (strs[1] == ln && strs[0] == prefix) {
- return true;
- }
- for (int i = 2; i < len; i += 2) {
- if (strs[i+1] == ln && strs[i] == prefix) {
- return true;
- }
- }
- } else {
- if (strs[0] == ln) {
- return true;
- }
- for (int i = 1; i < len; ++i) {
- if (strs[i] == ln) {
- return true;
- }
- }
- }
-
- return false;
- }
-
- public void appendNames(StringBuffer sb, String sep)
- {
- for (int i = 0; i < mStrings.length; ) {
- if (i > 0) {
- sb.append(sep);
- }
- if (mNsAware) {
- String prefix = mStrings[i++];
- if (prefix != null) {
- sb.append(prefix);
- sb.append(':');
- }
- }
- sb.append(mStrings[i++]);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/StarModel.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/StarModel.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/StarModel.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/StarModel.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,83 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.*;
-
-/**
- * Model class that represents any number of repetitions of its submodel
- * (including no repetitions).
- */
-public class StarModel
- extends ModelNode
-{
- ModelNode mModel;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public StarModel(ModelNode model) {
- super();
- mModel = model;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public ModelNode cloneModel() {
- return new StarModel(mModel.cloneModel());
- }
-
- public boolean isNullable() {
- return true;
- }
-
- public void indexTokens(List tokens) {
- mModel.indexTokens(tokens);
- }
-
- public void addFirstPos(BitSet pos) {
- mModel.addFirstPos(pos);
- }
-
- public void addLastPos(BitSet pos) {
- mModel.addLastPos(pos);
- }
-
- public void calcFollowPos(BitSet[] followPosSets)
- {
- // First, let's let sub-model do its stuff
- mModel.calcFollowPos(followPosSets);
-
- /* And then add the closure for the model (since sub-model
- * can 'follow itself' as many times as it needs to)
- */
-
- BitSet foll = new BitSet();
- mModel.addFirstPos(foll);
-
- BitSet toAddTo = new BitSet();
- mModel.addLastPos(toAddTo);
-
- int ix = 0; // need to/can skip the null entry (index 0)
- while ((ix = toAddTo.nextSetBit(ix+1)) >= 0) {
- /* Ok; so token at this index needs to have follow positions
- * added...
- */
- followPosSets[ix].or(foll);
- }
- }
-
- public String toString() {
- return mModel.toString() + "*";
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/StructValidator.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/StructValidator.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/StructValidator.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/StructValidator.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,51 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.dtd;
-
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Base class for validator Objects used to validate tree structure of an
- * XML-document against DTD.
- */
-public abstract class StructValidator
-{
- /**
- * Method that should be called to get the actual usable validator
- * instance, from the 'template' validator.
- */
- public abstract StructValidator newInstance();
-
- /**
- * Method called when a new (start) element is encountered within the
- * scope of parent element this validator monitors.
- *
- * @return Null if element is valid in its current position; error
- * message if not.
- */
- public abstract String tryToValidate(PrefixedName elemName);
-
- /**
- * Method called when the end element of the scope this validator
- * validates is encountered. It should make sure that the content
- * model is valid, and if not, to construct an error message.
- *
- * @return Null if the content model for the element is valid; error
- * message if not.
- */
- public abstract String fullyValid();
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/TokenContentSpec.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/TokenContentSpec.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/TokenContentSpec.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/TokenContentSpec.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,138 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Content specification that defines content model consisting of just
- * one allowed element. In addition to the allowed element, spec can have
- * optional arity ("*", "+", "?") marker.
- */
-public class TokenContentSpec
- extends ContentSpec
-{
- final static TokenContentSpec sDummy = new TokenContentSpec
- (' ', new PrefixedName("*", "*"));
-
- final PrefixedName mElemName;
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public TokenContentSpec(char arity, PrefixedName elemName)
- {
- super(arity);
- mElemName = elemName;
- }
-
- public static TokenContentSpec construct(char arity, PrefixedName elemName)
- {
- return new TokenContentSpec(arity, elemName);
- }
-
- public static TokenContentSpec getDummySpec() {
- return sDummy;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- public boolean isLeaf() {
- return mArity == ' ';
- }
-
- public PrefixedName getName() {
- return mElemName;
- }
-
- public StructValidator getSimpleValidator() {
- return new Validator(mArity, mElemName);
- }
-
- public ModelNode rewrite() {
- TokenModel model = new TokenModel(mElemName);
- if (mArity == '*') {
- return new StarModel(model);
- }
- if (mArity == '?') {
- return new OptionalModel(model);
- }
- if (mArity == '+') {
- return new ConcatModel(model,
- new StarModel(new TokenModel(mElemName)));
- }
- return model;
- }
-
- public String toString() {
- return (mArity == ' ') ? mElemName.toString()
- : (mElemName.toString() + mArity);
- }
-
- /*
- ///////////////////////////////////////////////////
- // Validator class:
- ///////////////////////////////////////////////////
- */
-
- final static class Validator
- extends StructValidator
- {
- final char mArity;
- final PrefixedName mElemName;
-
- int mCount = 0;
-
- public Validator(char arity, PrefixedName elemName)
- {
- mArity = arity;
- mElemName = elemName;
- }
-
- /**
- * Rules for reuse are simple: if we can have any number of
- * repetitions, we can just use a shared root instance. Although
- * its count variable will get updated this doesn't really
- * matter as it won't be used. Otherwise a new instance has to
- * be created always, to keep track of instance counts.
- */
- public StructValidator newInstance() {
- return (mArity == '*') ? this : new Validator(mArity, mElemName);
- }
-
- public String tryToValidate(PrefixedName elemName)
- {
- if (!elemName.equals(mElemName)) {
- return "Expected element <"+mElemName+">";
- }
- if (++mCount > 1 && (mArity == '?' || mArity == ' ')) {
- return "More than one instance of element <"+mElemName+">";
- }
- return null;
- }
-
- public String fullyValid()
- {
- switch (mArity) {
- case '*':
- case '?':
- return null;
- case '+': // need at least one (and multiples checked earlier)
- case ' ':
- if (mCount > 0) {
- return null;
- }
- return "Expected "+(mArity == '+' ? "at least one" : "")
- +" element <"+mElemName+">";
- }
- // should never happen:
- throw new IllegalStateException(ErrorConsts.ERR_INTERNAL);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/TokenModel.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/TokenModel.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/dtd/TokenModel.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/dtd/TokenModel.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,90 +0,0 @@
-package com.ctc.wstx.dtd;
-
-import java.util.BitSet;
-import java.util.List;
-
-import com.ctc.wstx.util.PrefixedName;
-
-/**
- * Model class that encapsulates a single (obligatory) token instance.
- */
-public final class TokenModel
- extends ModelNode
-{
- final static TokenModel NULL_TOKEN = new TokenModel(null);
- static { // null token needs to have 0 as its index...
- NULL_TOKEN.mTokenIndex = 0;
- }
-
- final PrefixedName mElemName;
-
- int mTokenIndex = -1; // to catch errors...
-
- /*
- ///////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////
- */
-
- public TokenModel(PrefixedName elemName) {
- mElemName = elemName;
- }
-
- public static TokenModel getNullToken() {
- return NULL_TOKEN;
- }
-
- /*
- ///////////////////////////////////////////////////
- // Public API
- ///////////////////////////////////////////////////
- */
-
- public PrefixedName getName() { return mElemName; }
-
- /**
- * Method that has to create a deep copy of the model, without
- * sharing any of existing Objects.
- */
- public ModelNode cloneModel() {
- return new TokenModel(mElemName);
- }
-
- public boolean isNullable() {
- return false;
- }
-
- public void indexTokens(List tokens)
- {
- /* Doh. This is not clean... but need to make sure the null
- * token never gets reindexed or explicitly added:
- */
- if (this != NULL_TOKEN) {
- int index = tokens.size();
- mTokenIndex = index;
- tokens.add(this);
- }
- }
-
- public void addFirstPos(BitSet firstPos) {
- firstPos.set(mTokenIndex);
- }
-
- public void addLastPos(BitSet lastPos) {
- lastPos.set(mTokenIndex);
- }
-
- public void calcFollowPos(BitSet[] followPosSets) {
- // nothing to do, for tokens...
- }
-
- public String toString() {
- return (mElemName == null) ? "[null]" : mElemName.toString();
- }
-
- /*
- ///////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////
- */
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/EntityDecl.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/EntityDecl.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/EntityDecl.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/EntityDecl.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,135 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.ent;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.net.URL;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.evt.WEntityDeclaration;
-import com.ctc.wstx.io.WstxInputSource;
-
-/**
- * Abstract base class for various entity declarations DTD reader
- * has parsed from DTD subsets.
- */
-public abstract class EntityDecl
- extends WEntityDeclaration
-{
- /**
- * Name/id of the entity used to reference it.
- */
- final String mName;
-
- /**
- * Context that is to be used to resolve references encountered from
- * expanded contents of this entity.
- */
- final URL mContext;
-
- /**
- * Flag that can be set to indicate that the declaration was in the
- * external DTD subset. Default is false.
- */
- protected boolean mDeclaredExternally = false;
-
- public EntityDecl(Location loc, String name, URL ctxt)
- {
- super(loc);
- mName = name;
- mContext = ctxt;
- }
-
- public void markAsExternallyDeclared() {
- mDeclaredExternally = true;
- }
-
- public final String getBaseURI() {
- return mContext.toExternalForm();
- }
-
- public final String getName() {
- return mName;
- }
-
- public final Location getLocation() {
- return mLocation;
- }
-
- public abstract String getNotationName();
-
- public abstract String getPublicId();
-
- public abstract String getReplacementText();
-
- public abstract int getReplacementText(Writer w)
- throws IOException;
-
- public abstract String getSystemId();
-
- /**
- * @return True, if the declaration occured in the external DTD
- * subset; false if not (internal subset, custom declaration)
- */
- public boolean wasDeclaredExternally() {
- return mDeclaredExternally;
- }
-
- /*
- ///////////////////////////////////////////
- // Implementation of abstract base methods
- ///////////////////////////////////////////
- */
-
- public abstract void writeEnc(Writer w) throws IOException;
-
- /*
- ///////////////////////////////////////////
- // Extended API for Wstx core
- ///////////////////////////////////////////
- */
-
- // // // Access to data
-
- public abstract char[] getReplacementChars();
-
- public final int getReplacementTextLength() {
- String str = getReplacementText();
- return (str == null) ? 0 : str.length();
- }
-
- // // // Type information
-
- public abstract boolean isExternal();
-
- public abstract boolean isParsed();
-
- // // // Factory methods
-
- /**
- * Method called to create the new input source through which expansion
- * value of the entity can be read.
- */
- public abstract WstxInputSource expand(WstxInputSource parent,
- XMLResolver res, ReaderConfig cfg,
- int xmlVersion)
- throws IOException, XMLStreamException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/ExtEntity.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/ExtEntity.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/ExtEntity.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/ExtEntity.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,78 +0,0 @@
-package com.ctc.wstx.ent;
-
-import com.ctc.wstx.api.ReaderConfig;
-import java.io.IOException;
-import java.io.Writer;
-import java.net.URL;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.io.WstxInputSource;
-
-public abstract class ExtEntity
- extends EntityDecl
-{
- final String mPublicId;
- final String mSystemId;
-
- public ExtEntity(Location loc, String name, URL ctxt,
- String pubId, String sysId)
- {
- super(loc, name, ctxt);
- mPublicId = pubId;
- mSystemId = sysId;
- }
-
- public abstract String getNotationName();
-
- public String getPublicId() {
- return mPublicId;
- }
-
- public String getReplacementText() {
- return null;
- }
-
- public int getReplacementText(Writer w)
- //throws IOException
- {
- return 0;
- }
-
- public String getSystemId() {
- return mSystemId;
- }
-
- /*
- ///////////////////////////////////////////
- // Implementation of abstract base methods
- ///////////////////////////////////////////
- */
-
- public abstract void writeEnc(Writer w) throws IOException;
-
- /*
- ///////////////////////////////////////////
- // Extended API for Wstx core
- ///////////////////////////////////////////
- */
-
- // // // Access to data
-
- public char[] getReplacementChars() {
- return null;
- }
-
- // // // Type information
-
- public boolean isExternal() { return true; }
-
- public abstract boolean isParsed();
-
- public abstract WstxInputSource expand(WstxInputSource parent,
- XMLResolver res, ReaderConfig cfg,
- int xmlVersion)
- throws IOException, XMLStreamException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/IntEntity.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/IntEntity.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/ent/IntEntity.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/ent/IntEntity.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,134 +0,0 @@
-package com.ctc.wstx.ent;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.net.URL;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLResolver;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.io.InputSourceFactory;
-import com.ctc.wstx.io.TextEscaper;
-import com.ctc.wstx.io.WstxInputLocation;
-import com.ctc.wstx.io.WstxInputSource;
-
-public class IntEntity
- extends EntityDecl
-{
- /**
- * Location where entity content definition started;
- * points to the starting/opening quote for internal
- * entities.
- */
- protected final Location mContentLocation;
-
- /**
- * Replacement text of the entity; full array contents.
- */
- final char[] mRepl;
-
- String mReplText = null;
-
- public IntEntity(Location loc, String name, URL ctxt,
- char[] repl, Location defLoc)
- {
- super(loc, name, ctxt);
- mRepl = repl;
- mContentLocation = defLoc;
- }
-
- public static IntEntity create(String id, String repl)
- {
- return create(id, repl.toCharArray());
- }
-
- public static IntEntity create(String id, char[] val)
- {
- WstxInputLocation loc = WstxInputLocation.getEmptyLocation();
- return new IntEntity(loc, id, null, val, loc);
- }
-
- public String getNotationName() {
- return null;
- }
-
- public String getPublicId() {
- return null;
- }
-
- public String getReplacementText()
- {
- String repl = mReplText;
- if (repl == null) {
- repl = (mRepl.length == 0) ? "" : new String(mRepl);
- mReplText = repl;
- }
- return mReplText;
- }
-
- public int getReplacementText(Writer w)
- throws IOException
- {
- w.write(mRepl);
- return mRepl.length;
- }
-
- public String getSystemId() {
- return null;
- }
-
- /*
- ///////////////////////////////////////////
- // Implementation of abstract base methods
- ///////////////////////////////////////////
- */
-
- public void writeEnc(Writer w) throws IOException
- {
- w.write("");
- }
-
- /*
- ///////////////////////////////////////////
- // Extended API for Wstx core
- ///////////////////////////////////////////
- */
-
- // // // Access to data
-
- /**
- * Gives raw access to replacement text data...
- *
- *
- */
-public class MergedNsContext
- extends BaseNsContext
-{
- final NamespaceContext mParentCtxt;
-
- /**
- * List of {@link Namespace} instances.
- */
- final List mNamespaces;
-
- Map mNsByPrefix = null;
-
- Map mNsByURI = null;
-
- protected MergedNsContext(NamespaceContext parentCtxt, List localNs)
- {
- mParentCtxt = parentCtxt;
- mNamespaces = (localNs == null) ? Collections.EMPTY_LIST : localNs;
- }
-
- public static BaseNsContext construct(NamespaceContext parentCtxt,
- List localNs)
- {
- return new MergedNsContext(parentCtxt, localNs);
- }
-
- /*
- /////////////////////////////////////////////
- // NamespaceContext API
- /////////////////////////////////////////////
- */
-
- public String doGetNamespaceURI(String prefix)
- {
- // Note: base class checks for 'known' problems and prefixes:
- if (mNsByPrefix == null) {
- mNsByPrefix = buildByPrefixMap();
- }
- Namespace ns = (Namespace) mNsByPrefix.get(prefix);
- if (ns == null && mParentCtxt != null) {
- return mParentCtxt.getNamespaceURI(prefix);
- }
- return (ns == null) ? null : ns.getNamespaceURI();
- }
-
- public String doGetPrefix(String nsURI)
- {
- // Note: base class checks for 'known' problems and prefixes:
- if (mNsByURI == null) {
- mNsByURI = buildByNsURIMap();
- }
- Namespace ns = (Namespace) mNsByURI.get(nsURI);
- if (ns == null && mParentCtxt != null) {
- return mParentCtxt.getPrefix(nsURI);
- }
- return (ns == null) ? null : ns.getPrefix();
- }
-
- public Iterator doGetPrefixes(String nsURI)
- {
- // Note: base class checks for 'known' problems and prefixes:
- ArrayList l = null;
-
- for (int i = 0, len = mNamespaces.size(); i < len; ++i) {
- Namespace ns = (Namespace) mNamespaces.get(i);
- String uri = ns.getNamespaceURI();
- if (uri == null) {
- uri = "";
- }
- if (uri.equals(nsURI)) {
- if (l == null) {
- l = new ArrayList();
- }
- String prefix = ns.getPrefix();
- l.add((prefix == null) ? "" : prefix);
- }
- }
-
- if (mParentCtxt != null) {
- Iterator it = mParentCtxt.getPrefixes(nsURI);
- if (l == null) {
- return it;
- }
- while (it.hasNext()) {
- l.add(it.next());
- }
- }
-
- return (l == null) ? EmptyIterator.getInstance() : l.iterator();
- }
-
- /*
- /////////////////////////////////////////////
- // Extended API
- /////////////////////////////////////////////
- */
-
- /**
- * Method that returns information about namespace definition declared
- * in this scope; not including ones declared in outer scopes.
- */
- public Iterator getNamespaces()
- {
- return mNamespaces.iterator();
- }
-
- public void outputNamespaceDeclarations(Writer w) throws IOException
- {
- for (int i = 0, len = mNamespaces.size(); i < len; ++i) {
- Namespace ns = (Namespace) mNamespaces.get(i);
- w.write(' ');
- w.write(XMLConstants.XMLNS_ATTRIBUTE);
- if (!ns.isDefaultNamespaceDeclaration()) {
- w.write(':');
- w.write(ns.getPrefix());
- }
- w.write("=\"");
- w.write(ns.getNamespaceURI());
- w.write('"');
- }
- }
-
- /**
- * Method called by the matching start element class to
- * output all namespace declarations active in current namespace
- * scope, if any.
- */
- public void outputNamespaceDeclarations(XMLStreamWriter w) throws XMLStreamException
- {
- for (int i = 0, len = mNamespaces.size(); i < len; ++i) {
- Namespace ns = (Namespace) mNamespaces.get(i);
- if (ns.isDefaultNamespaceDeclaration()) {
- w.writeDefaultNamespace(ns.getNamespaceURI());
- } else {
- w.writeNamespace(ns.getPrefix(), ns.getNamespaceURI());
- }
- }
- }
-
- /*
- /////////////////////////////////////////////
- // Private methods:
- /////////////////////////////////////////////
- */
-
- private Map buildByPrefixMap()
- {
- int len = mNamespaces.size();
- if (len == 0) {
- return Collections.EMPTY_MAP;
- }
-
- LinkedHashMap m = new LinkedHashMap(1 + len + (len>>1));
- for (int i = 0; i < len; ++i) {
- Namespace ns = (Namespace) mNamespaces.get(i);
- String prefix = ns.getPrefix();
- if (prefix == null) { // shouldn't happen but...
- prefix = "";
- }
- m.put(prefix, ns);
- }
- return m;
- }
-
- private Map buildByNsURIMap()
- {
- int len = mNamespaces.size();
- if (len == 0) {
- return Collections.EMPTY_MAP;
- }
-
- LinkedHashMap m = new LinkedHashMap(1 + len + (len>>1));
- for (int i = 0; i < len; ++i) {
- Namespace ns = (Namespace) mNamespaces.get(i);
- String uri = ns.getNamespaceURI();
- if (uri == null) { // shouldn't happen but...
- uri = "";
- }
- m.put(uri, ns);
- }
- return m;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,4 +0,0 @@
-
-Package contains Woodstox implementation of the StAX event layer; functionality
-that is built on top of stream readers and writers.
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/SimpleStartElement.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/SimpleStartElement.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/SimpleStartElement.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/SimpleStartElement.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,172 +0,0 @@
-package com.ctc.wstx.evt;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.util.*;
-
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.stream.events.Attribute;
-import javax.xml.stream.events.Namespace;
-import javax.xml.stream.events.StartElement;
-
-import org.codehaus.stax2.ri.EmptyIterator;
-
-import com.ctc.wstx.io.TextEscaper;
-import com.ctc.wstx.util.BaseNsContext;
-
-/**
- * Wstx {@link StartElement} implementation used when event is constructed
- * from already objectified data, for example when constructed by the event
- * factory.
- */
-public class SimpleStartElement
- extends BaseStartElement
-{
- final Map mAttrs;
-
- /*
- /////////////////////////////////////////////
- // Life cycle
- /////////////////////////////////////////////
- */
-
- protected SimpleStartElement(Location loc, QName name, BaseNsContext nsCtxt,
- Map attr)
- {
- super(loc, name, nsCtxt);
- mAttrs = attr;
- }
-
- /**
- * Factory method called when a start element needs to be constructed
- * from an external source (most likely, non-woodstox stream reader).
- */
- public static SimpleStartElement construct(Location loc, QName name,
- Map attrs, List ns,
- NamespaceContext nsCtxt)
- {
- BaseNsContext myCtxt = MergedNsContext.construct(nsCtxt, ns);
- return new SimpleStartElement(loc, name, myCtxt, attrs);
- }
-
- public static SimpleStartElement construct(Location loc, QName name,
- Iterator attrs, Iterator ns,
- NamespaceContext nsCtxt)
- {
- Map attrMap;
- if (attrs == null || !attrs.hasNext()) {
- attrMap = null;
- } else {
- attrMap = new LinkedHashMap();
- do {
- Attribute attr = (Attribute) attrs.next();
- attrMap.put(attr.getName(), attr);
- } while (attrs.hasNext());
- }
-
- BaseNsContext myCtxt;
- if (ns != null && ns.hasNext()) {
- ArrayList l = new ArrayList();
- do {
- l.add((Namespace) ns.next()); // cast to catch type problems early
- } while (ns.hasNext());
- myCtxt = MergedNsContext.construct(nsCtxt, l);
- } else {
- /* Doh. Need specificially 'our' namespace context, to get them
- * output properly...
- */
- if (nsCtxt == null) {
- myCtxt = null;
- } else if (nsCtxt instanceof BaseNsContext) {
- myCtxt = (BaseNsContext) nsCtxt;
- } else {
- myCtxt = MergedNsContext.construct(nsCtxt, null);
- }
- }
- return new SimpleStartElement(loc, name, myCtxt, attrMap);
- }
-
- /*
- /////////////////////////////////////////////
- // Public API
- /////////////////////////////////////////////
- */
-
- public Attribute getAttributeByName(QName name)
- {
- if (mAttrs == null) {
- return null;
- }
- return (Attribute) mAttrs.get(name);
- }
-
- public Iterator getAttributes()
- {
- if (mAttrs == null) {
- return EmptyIterator.getInstance();
- }
- return mAttrs.values().iterator();
- }
-
- protected void outputNsAndAttr(Writer w) throws IOException
- {
- // First namespace declarations, if any:
- if (mNsCtxt != null) {
- mNsCtxt.outputNamespaceDeclarations(w);
- }
- // Then attributes, if any:
- if (mAttrs != null && mAttrs.size() > 0) {
- Iterator it = mAttrs.values().iterator();
- while (it.hasNext()) {
- Attribute attr = (Attribute) it.next();
- // Let's only output explicit attribute values:
- if (!attr.isSpecified()) {
- continue;
- }
-
- w.write(' ');
- QName name = attr.getName();
- String prefix = name.getPrefix();
- if (prefix != null && prefix.length() > 0) {
- w.write(prefix);
- w.write(':');
- }
- w.write(name.getLocalPart());
- w.write("=\"");
- String val = attr.getValue();
- if (val != null && val.length() > 0) {
- TextEscaper.writeEscapedAttrValue(w, val);
- }
- w.write('"');
- }
- }
- }
-
- protected void outputNsAndAttr(XMLStreamWriter w) throws XMLStreamException
- {
- // First namespace declarations, if any:
- if (mNsCtxt != null) {
- mNsCtxt.outputNamespaceDeclarations(w);
- }
- // Then attributes, if any:
- if (mAttrs != null && mAttrs.size() > 0) {
- Iterator it = mAttrs.values().iterator();
- while (it.hasNext()) {
- Attribute attr = (Attribute) it.next();
- // Let's only output explicit attribute values:
- if (!attr.isSpecified()) {
- continue;
- }
- QName name = attr.getName();
- String prefix = name.getPrefix();
- String ln = name.getLocalPart();
- String nsURI = name.getNamespaceURI();
- w.writeAttribute(prefix, nsURI, ln, attr.getValue());
- }
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WDTD.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WDTD.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WDTD.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WDTD.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,93 +0,0 @@
-package com.ctc.wstx.evt;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import javax.xml.stream.Location;
-
-import org.codehaus.stax2.ri.evt.DTDEventImpl;
-
-import com.ctc.wstx.dtd.DTDSubset;
-
-/**
- * Event that contains all StAX accessible information read from internal
- * and external DTD subsets.
- */
-public class WDTD
- extends DTDEventImpl
-{
- /**
- * Internal DTD Object that contains combined information from internal
- * and external subsets.
- */
- final DTDSubset mSubset;
-
- /*
- /////////////////////////////////////////////////////
- // Lazily constructed objects
- /////////////////////////////////////////////////////
- */
-
- List mEntities = null;
-
- List mNotations = null;
-
- /*
- /////////////////////////////////////////////////////
- // Constuctors
- /////////////////////////////////////////////////////
- */
-
- public WDTD(Location loc, String rootName,
- String sysId, String pubId, String intSubset,
- DTDSubset dtdSubset)
- {
- super(loc, rootName, sysId, pubId, intSubset, dtdSubset);
- mSubset = dtdSubset;
- }
-
- public WDTD(Location loc, String rootName,
- String sysId, String pubId, String intSubset)
- {
- this(loc, rootName, sysId, pubId, intSubset, null);
- }
-
- /**
- * Constructor used when only partial information is available...
- */
- public WDTD(Location loc, String rootName, String intSubset)
- {
- this(loc, rootName, null, null, intSubset, null);
- }
-
- public WDTD(Location loc, String fullText)
- {
- super(loc, fullText);
- mSubset = null;
- }
-
- /*
- /////////////////////////////////////////////////////
- // Accessors
- /////////////////////////////////////////////////////
- */
-
- public List getEntities()
- {
- if (mEntities == null && (mSubset != null)) {
- /* Better make a copy, so that caller can not modify list
- * DTD has, which may be shared (since DTD subset instances
- * are cached and reused)
- */
- mEntities = new ArrayList(mSubset.getGeneralEntityList());
- }
- return mEntities;
- }
-
- public List getNotations() {
- if (mNotations == null && (mSubset != null)) {
- mNotations = new ArrayList(mSubset.getNotationList());
- }
- return mNotations;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WEntityDeclaration.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WEntityDeclaration.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WEntityDeclaration.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WEntityDeclaration.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,109 +0,0 @@
-package com.ctc.wstx.evt;
-
-import java.io.IOException;
-import java.io.Writer;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.events.EntityDeclaration;
-
-import org.codehaus.stax2.XMLStreamWriter2;
-import org.codehaus.stax2.ri.evt.BaseEventImpl;
-
-import com.ctc.wstx.ent.EntityDecl;
-import com.ctc.wstx.exc.WstxIOException;
-
-/**
- * Simple implementation of StAX entity declaration events; for the
- * most just wraps a {@link EntityDecl} instance.
- */
-public abstract class WEntityDeclaration
- extends BaseEventImpl
- implements EntityDeclaration
-{
- public WEntityDeclaration(Location loc)
- {
- super(loc);
- }
-
- public abstract String getBaseURI();
-
- public abstract String getName();
-
- public abstract String getNotationName();
-
- public abstract String getPublicId();
-
- public abstract String getReplacementText();
-
- public abstract String getSystemId();
-
- /*
- ///////////////////////////////////////////
- // Implementation of abstract base methods
- ///////////////////////////////////////////
- */
-
- public int getEventType() {
- return ENTITY_DECLARATION;
- }
-
- public abstract void writeEnc(Writer w) throws IOException;
-
- public void writeAsEncodedUnicode(Writer w)
- throws XMLStreamException
- {
- try {
- writeEnc(w);
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
- }
-
- /**
- * This method does not make much sense for this event type -- the reason
- * being that the entity declarations can only be written as part of
- * a DTD (internal or external subset), not separately. Can basically
- * choose to either skip silently (output nothing), or throw an
- * exception.
- */
- public void writeUsing(XMLStreamWriter2 w) throws XMLStreamException
- {
- /* Fail silently, or throw an exception? Let's do latter; at least
- * then we'll get useful (?) bug reports!
- */
- throw new XMLStreamException("Can not write entity declarations using an XMLStreamWriter");
- }
-
- /*
- ///////////////////////////////////////////
- // Standard method impl: note, copied
- // from Stax2 RI "EntityDeclarationEventImpl"
- ///////////////////////////////////////////
- */
-
- public boolean equals(Object o)
- {
- if (o == this) return true;
- if (o == null) return false;
-
- if (!(o instanceof EntityDeclaration)) return false;
-
- EntityDeclaration other = (EntityDeclaration) o;
- return stringsWithNullsEqual(getName(), other.getName())
- && stringsWithNullsEqual(getBaseURI(), other.getBaseURI())
- && stringsWithNullsEqual(getNotationName(), other.getNotationName())
- && stringsWithNullsEqual(getPublicId(), other.getPublicId())
- && stringsWithNullsEqual(getReplacementText(), other.getReplacementText())
- && stringsWithNullsEqual(getSystemId(), other.getSystemId())
- ;
- }
-
- public int hashCode()
- {
- /* Hmmh. Could try using most of the data, but really, name
- * should be enough for most use cases
- */
- return getName().hashCode();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WEntityReference.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WEntityReference.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WEntityReference.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WEntityReference.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,44 +0,0 @@
-package com.ctc.wstx.evt;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.events.EntityReference;
-import javax.xml.stream.events.EntityDeclaration;
-
-import org.codehaus.stax2.ri.evt.EntityReferenceEventImpl;
-
-/**
- * We need a slightly specialized version to support concept of
- * undeclared entities, which can be used in (non-default, non-standard)
- * mode where undeclared entities are allowed to be handled.
- */
-public class WEntityReference
- extends EntityReferenceEventImpl
- implements EntityReference
-{
- final String mName;
-
- public WEntityReference(Location loc, EntityDeclaration decl)
- {
- super(loc, decl);
- mName = null;
- }
-
- /**
- * This constructor gets called for undeclared/defined entities: we will
- * still know the name (from the reference), but not how it's defined
- * (since it is not defined).
- */
- public WEntityReference(Location loc, String name)
- {
- super(loc, (EntityDeclaration) null);
- mName = name;
- }
-
- public String getName()
- {
- if (mName != null) {
- return mName;
- }
- return super.getName();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WNotationDeclaration.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WNotationDeclaration.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WNotationDeclaration.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WNotationDeclaration.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,43 +0,0 @@
-package com.ctc.wstx.evt;
-
-import java.net.URL;
-
-import javax.xml.stream.Location;
-
-import org.codehaus.stax2.ri.evt.NotationDeclarationEventImpl;
-
-/**
- * Woodstox implementation of {@link org.codehaus.stax2.evt.NotationDeclaration2}.
- * The only required addition is that of passing in the Base URI.
- *
- * @author Tatu Saloranta
- *
- * @since 4.0.0
- */
-public class WNotationDeclaration
- extends NotationDeclarationEventImpl
-{
- /**
- * Base URL that can be used to resolve the notation reference if
- * necessary.
- */
- final URL _baseURL;
-
- public WNotationDeclaration(Location loc,
- String name, String pubId, String sysId,
- URL baseURL)
- {
- super(loc, name, pubId, sysId);
- _baseURL = baseURL;
- }
-
- //@Override
- public String getBaseURI()
- {
- if (_baseURL == null) {
- return super.getBaseURI();
- }
- return _baseURL.toExternalForm();
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WstxEventReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WstxEventReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/evt/WstxEventReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/evt/WstxEventReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,84 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.evt;
-
-import javax.xml.stream.*;
-import javax.xml.stream.util.XMLEventAllocator;
-
-import org.codehaus.stax2.XMLStreamReader2;
-import org.codehaus.stax2.ri.Stax2EventReaderImpl;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.exc.WstxParsingException;
-
-/**
- * Woodstox version, based on generic Stax reference implementation
- * baseline of {@link Stax2EventReaderImpl}.
- */
-public class WstxEventReader
- extends Stax2EventReaderImpl
-{
- public WstxEventReader(XMLEventAllocator a, XMLStreamReader2 r)
- {
- super(a, r);
- }
-
- /*
- //////////////////////////////////////////////////////
- // Impl of abstract methods
- //////////////////////////////////////////////////////
- */
-
- protected String getErrorDesc(int errorType, int currEvent)
- {
- // Defaults are mostly fine, except we can easily add event type desc
- switch (errorType) {
- case ERR_GETELEMTEXT_NOT_START_ELEM:
- return ErrorConsts.ERR_STATE_NOT_STELEM+", got "+ErrorConsts.tokenTypeDesc(currEvent);
- case ERR_GETELEMTEXT_NON_TEXT_EVENT:
- return "Expected a text token, got "+ErrorConsts.tokenTypeDesc(currEvent);
- case ERR_NEXTTAG_NON_WS_TEXT:
- return "Only all-whitespace CHARACTERS/CDATA (or SPACE) allowed for nextTag(), got "+ErrorConsts.tokenTypeDesc(currEvent);
- case ERR_NEXTTAG_WRONG_TYPE:
- return "Got "+ErrorConsts.tokenTypeDesc(currEvent)+", instead of START_ELEMENT, END_ELEMENT or SPACE";
- }
- return null;
- }
-
- public boolean isPropertySupported(String name)
- {
- return ((XMLStreamReader2)getStreamReader()).isPropertySupported(name);
- }
-
- public boolean setProperty(String name, Object value)
- {
- return ((XMLStreamReader2)getStreamReader()).setProperty(name, value);
- }
-
-
- /*
- //////////////////////////////////////////////////////
- // Overrides
- //////////////////////////////////////////////////////
- */
-
- // @Override
- protected void reportProblem(String msg, Location loc)
- throws XMLStreamException
- {
- throw new WstxParsingException(msg, loc);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,4 +0,0 @@
-
-Package that contains all exceptions Woodstox readers and writers throw.
-Exceptions generally extend {@link javax.xml.stream.XMLStreamException}.
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxEOFException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxEOFException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxEOFException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxEOFException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,17 +0,0 @@
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.Location;
-
-/**
- * Exception thrown during parsing, if an unexpected EOF is encountered.
- * Location usually signals starting position of current Node.
- */
-public class WstxEOFException
- extends WstxParsingException
-{
- private static final long serialVersionUID = 1L;
-
- public WstxEOFException(String msg, Location loc) {
- super(msg, loc);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,104 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.util.ExceptionUtil;
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Base class for all implementatations of {@link XMLStreamException}
- * Wstx uses.
- */
-public class WstxException
- extends XMLStreamException
-{
- private static final long serialVersionUID = 1L;
-
- /**
- * D'oh. Super-class munges and hides the message, have to duplicate here
- */
- final protected String mMsg;
-
- public WstxException(String msg) {
- super(msg);
- mMsg = msg;
- }
-
- public WstxException(Throwable th) {
- super(th.getMessage(), th);
- mMsg = th.getMessage();
-
- // 13-Aug-2004, TSa: Better make sure root cause is set...
- ExceptionUtil.setInitCause(this, th);
- }
-
- public WstxException(String msg, Location loc) {
- super(msg, loc);
- mMsg = msg;
- }
-
- public WstxException(String msg, Location loc, Throwable th) {
- super(msg, loc, th);
- mMsg = msg;
-
- // 13-Aug-2004, TSa: Better make sure root cause is set...
- ExceptionUtil.setInitCause(this, th);
- }
-
- /**
- * Method is overridden for two main reasons: first, default method
- * does not display public/system id information, even if it exists, and
- * second, default implementation can not handle nested Location
- * information.
- */
- public String getMessage()
- {
- String locMsg = getLocationDesc();
- /* Better not use super's message if we do have location information,
- * since parent's message contains (part of) Location
- * info; something we can regenerate better...
- */
- if (locMsg == null) {
- return super.getMessage();
- }
- StringBuffer sb = new StringBuffer(mMsg.length() + locMsg.length() + 20);
- sb.append(mMsg);
- StringUtil.appendLF(sb);
- sb.append(" at ");
- sb.append(locMsg);
- return sb.toString();
- }
-
- public String toString()
- {
- return getClass().getName()+": "+getMessage();
- }
-
- /*
- ////////////////////////////////////////////////////////
- // Internal methods:
- ////////////////////////////////////////////////////////
- */
-
- protected String getLocationDesc()
- {
- Location loc = getLocation();
- return (loc == null) ? null : loc.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxIOException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxIOException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxIOException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxIOException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,21 +0,0 @@
-package com.ctc.wstx.exc;
-
-import java.io.IOException;
-
-/**
- * Simple wrapper for {@link IOException}s; needed when StAX does not expose
- * underlying I/O exceptions via its methods.
- */
-public class WstxIOException
- extends WstxException
-{
- private static final long serialVersionUID = 1L;
-
- public WstxIOException(IOException ie) {
- super(ie);
- }
-
- public WstxIOException(String msg) {
- super(msg);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxLazyException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxLazyException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxLazyException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxLazyException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,61 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.util.ExceptionUtil;
-
-/**
- * Alternative exception class Woodstox code uses when it is not allowed
- * to throw an instance of {@link XMLStreamException}; this generally
- * happens when doing lazy parsing.
- */
-public class WstxLazyException
- extends RuntimeException
-{
- private static final long serialVersionUID = 1L;
-
- final XMLStreamException mOrig;
-
- public WstxLazyException(XMLStreamException origEx)
- {
- super(origEx.getMessage());
- mOrig = origEx;
- // Let's additionally to set source message
- ExceptionUtil.setInitCause(this, origEx);
- }
-
- public static void throwLazily(XMLStreamException ex)
- throws WstxLazyException
- {
- throw new WstxLazyException(ex);
- }
-
- /**
- * Need to override this, to be able to dynamically construct and
- * display the location information...
- */
- public String getMessage()
- {
- return "["+getClass().getName()+"] "+mOrig.getMessage();
- }
-
- public String toString()
- {
- return "["+getClass().getName()+"] "+mOrig.toString();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxOutputException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxOutputException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxOutputException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxOutputException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,16 +0,0 @@
-package com.ctc.wstx.exc;
-
-/**
- * Exception class used for notifying about well-formedness errors that
- * writers would create. Such exceptions are thrown when strict output
- * validation is enabled.
- */
-public class WstxOutputException
- extends WstxException
-{
- private static final long serialVersionUID = 1L;
-
- public WstxOutputException(String msg) {
- super(msg);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxParsingException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxParsingException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxParsingException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxParsingException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,19 +0,0 @@
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.Location;
-
-/**
- * Intermediate base class for reporting actual Wstx parsing problems.
- */
-public class WstxParsingException
- extends WstxException
-{
- private static final long serialVersionUID = 1L;
-
- public WstxParsingException(String msg, Location loc) {
- super(msg, loc);
- }
-
- // !!! 13-Sep-2008, tatus: Only needed for DOMWrapping reader, for now
- public WstxParsingException(String msg) { super(msg); }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxUnexpectedCharException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxUnexpectedCharException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxUnexpectedCharException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxUnexpectedCharException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,26 +0,0 @@
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.Location;
-
-/**
- * Generic exception type that indicates that tokenizer/parser encountered
- * unexpected (but not necessarily invalid per se) character; character that
- * is not legal in current context. Could happen, for example, if white space
- * was missing between attribute value and name of next attribute.
- */
-public class WstxUnexpectedCharException
- extends WstxParsingException
-{
- private static final long serialVersionUID = 1L;
-
- final char mChar;
-
- public WstxUnexpectedCharException(String msg, Location loc, char c) {
- super(msg, loc);
- mChar = c;
- }
-
- public char getChar() {
- return mChar;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxValidationException.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxValidationException.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/exc/WstxValidationException.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/exc/WstxValidationException.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,101 +0,0 @@
-package com.ctc.wstx.exc;
-
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.validation.XMLValidationException;
-import org.codehaus.stax2.validation.XMLValidationProblem;
-
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * Specific exception thrown when document has validation (DTD, W3C Schema)
- * errors; things that are not well-formedness problems.
- *mByteBufferEnd
- */
- protected int mBytePtr;
-
- /**
- * Pointed to the end marker, that is, position one after the last
- * valid available byte.
- */
- protected int mByteBufferEnd;
-
- /**
- * Flag that indicates whether the read buffer is to be recycled
- * when Reader is closed or not.
- */
- private final boolean mRecycleBuffer;
-
- /*
- ////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////
- */
-
- protected BaseReader(ReaderConfig cfg, InputStream in, byte[] buf, int ptr, int len,
- boolean recycleBuffer)
- {
- mConfig = cfg;
- mIn = in;
- mByteBuffer = buf;
- mBytePtr = ptr;
- mByteBufferEnd = len;
- mRecycleBuffer = recycleBuffer;
- }
-
- /*
- ////////////////////////////////////////
- // Configuration
- ////////////////////////////////////////
- */
-
- /**
- * Method that can be called to indicate the xml conformance used
- * when reading content using this reader. Some of the character
- * validity checks need to be done at reader level, and sometimes
- * they depend on xml level (for example, xml 1.1 has new linefeeds
- * and both more and less restricted characters).
- */
- public abstract void setXmlCompliancy(int xmlVersion);
-
- /**
- * Method that can be used to see if we can actually modify the
- * underlying buffer. This is the case if we are managing the buffer,
- * but not if it was just given to us.
- */
- protected final boolean canModifyBuffer()
- {
- return mRecycleBuffer;
- }
-
- /*
- ////////////////////////////////////////
- // Reader API
- ////////////////////////////////////////
- */
-
- public void close()
- throws IOException
- {
- InputStream in = mIn;
-
- if (in != null) {
- mIn = null;
- freeBuffers();
- in.close();
- }
- }
-
- char[] mTmpBuf = null;
-
- /**
- * Although this method is implemented by the base class, AND it should
- * never be called by Woodstox code, let's still implement it bit more
- * efficiently just in case
- */
- public int read()
- throws IOException
- {
- if (mTmpBuf == null) {
- mTmpBuf = new char[1];
- }
- if (read(mTmpBuf, 0, 1) < 1) {
- return -1;
- }
- return mTmpBuf[0];
- }
-
- /*
- ////////////////////////////////////////
- // Internal/package methods:
- ////////////////////////////////////////
- */
-
- protected final InputStream getStream() { return mIn; }
-
- /**
- * Method for reading as many bytes from the underlying stream as possible
- * (that fit in the buffer), to the beginning of the buffer.
- */
- protected final int readBytes()
- throws IOException
- {
- mBytePtr = 0;
- mByteBufferEnd = 0;
- if (mIn != null) {
- int count = mIn.read(mByteBuffer, 0, mByteBuffer.length);
- if (count > 0) {
- mByteBufferEnd = count;
- }
- return count;
- }
- return -1;
- }
-
- /**
- * Method for reading as many bytes from the underlying stream as possible
- * (that fit in the buffer considering offset), to the specified offset.
- *
- * @return Number of bytes read, if any; -1 to indicate none available
- * (that is, end of input)
- */
- protected final int readBytesAt(int offset)
- throws IOException
- {
- // shouldn't modify mBytePtr, assumed to be 'offset'
- if (mIn != null) {
- int count = mIn.read(mByteBuffer, offset, mByteBuffer.length - offset);
- if (count > 0) {
- mByteBufferEnd += count;
- }
- return count;
- }
- return -1;
- }
-
- /**
- * This method should be called along with (or instead of) normal
- * close. After calling this method, no further reads should be tried.
- * Method will try to recycle read buffers (if any).
- */
- public final void freeBuffers()
- {
- /* 11-Apr-2005, TSa: Ok, we can release the buffer now, to be
- * recycled by the next stream reader instantiated by this
- * thread (if any).
- */
- if (mRecycleBuffer) {
- byte[] buf = mByteBuffer;
- if (buf != null) {
- mByteBuffer = null;
- if (mConfig != null) {
- mConfig.freeFullBBuffer(buf);
- }
- }
- }
- }
-
- protected void reportBounds(char[] cbuf, int start, int len)
- throws IOException
- {
- throw new ArrayIndexOutOfBoundsException("read(buf,"+start+","+len+"), cbuf["+cbuf.length+"]");
- }
-
- protected void reportStrangeStream()
- throws IOException
- {
- throw new IOException("Strange I/O stream, returned 0 bytes on read");
- }
-
- protected void reportInvalidXml11(int value, int bytePos, int charPos)
- throws IOException
- {
- throw new CharConversionException("Invalid character 0x"
- +Integer.toHexString(value)
- +", can only be included in xml 1.1 using character entities (at char #"+charPos+", byte #"+bytePos+")");
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BranchingReaderSource.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BranchingReaderSource.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/io/BranchingReaderSource.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/io/BranchingReaderSource.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,158 +0,0 @@
-package com.ctc.wstx.io;
-
-import java.io.IOException;
-import java.io.Reader;
-import java.net.URL;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.util.TextBuffer;
-
-/**
- * Specialized input source that can "branch" input it reads; essentially
- * both giving out read data AND also writing it out to a Writer.
- *ThreadLocal member of the owning class pointing to
- * instance of this class through a SoftReference. The
- * end result is a low-overhead GC-cleanable recycling: hopefully
- * ideal for use by stream readers.
- *fireStartTag, we must use
- * element stack, not attribute collector:
- */
- //String prefix = mAttrCollector.getNsPrefix(index);
- String prefix = mElemStack.getLocalNsPrefix(index);
- return (prefix == null || prefix.length() == 0) ?
- "xmlns" : prefix;
- }
- return null;
- }
-
- public String getQName(int index)
- {
- if (index < mAttrCount) {
- if (index < 0) {
- return null;
- }
- String prefix = mAttrCollector.getPrefix(index);
- String ln = mAttrCollector.getLocalName(index);
- return (prefix == null || prefix.length() == 0) ?
- ln : (prefix + ":" + ln);
- }
- index -= mAttrCount;
- if (index < mNsCount) {
- /* As discussed in fireStartTag, we must use
- * element stack, not attribute collector:
- */
- //String prefix = mAttrCollector.getNsPrefix(index);
- String prefix = mElemStack.getLocalNsPrefix(index);
- if (prefix == null || prefix.length() == 0) {
- return "xmlns";
- }
- return "xmlns:"+prefix;
- }
- return null;
- }
-
- public String getType(int index)
- {
- if (index < mAttrCount) {
- if (index < 0) {
- return null;
- }
- /* Note: Woodstox will have separate type for enumerated values;
- * SAX considers these NMTOKENs, so may need to convert (but
- * note: some SAX impls also use "ENUMERATED")
- */
- String type = mElemStack.getAttributeType(index);
- // Let's count on it being interned:
- if (type == "ENUMERATED") {
- type = "NMTOKEN";
- }
- return type;
- }
- // But how about namespace declarations... let's just call them CDATA?
- index -= mAttrCount;
- if (index < mNsCount) {
- return "CDATA";
- }
- return null;
- }
-
- public String getType(String qName)
- {
- return getType(getIndex(qName));
- }
-
- public String getType(String uri, String localName)
- {
- return getType(getIndex(uri, localName));
- }
-
- public String getURI(int index)
- {
- if (index < mAttrCount) {
- if (index < 0) {
- return null;
- }
- String uri = mAttrCollector.getURI(index);
- return (uri == null) ? "" : uri;
- }
- if ((index - mAttrCount) < mNsCount) {
- return XMLConstants.XMLNS_ATTRIBUTE_NS_URI;
- }
- return null;
- }
-
- public String getValue(int index)
- {
- if (index < mAttrCount) {
- return (index < 0) ? null : mAttrCollector.getValue(index);
- }
- index -= mAttrCount;
- if (index < mNsCount) {
- /* As discussed in fireStartTag, we must use
- * element stack, not attribute collector:
- */
- //String uri = mAttrCollector.getNsURI(index);
- String uri = mElemStack.getLocalNsURI(index);
- return (uri == null) ? "" : uri;
- }
- return null;
- }
-
- public String getValue(String qName)
- {
- return getValue(getIndex(qName));
- }
-
- public String getValue(String uri, String localName)
- {
- return getValue(getIndex(uri, localName));
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Attributes2 (SAX2) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- public boolean isDeclared(int index)
- {
- if (index < mAttrCount) {
- if (index >= 0) {
- // !!! TODO: implement properly
- return true;
- }
- } else {
- index -= mAttrCount;
- if (index < mNsCount) {
- /* DTD and namespaces don't really play nicely together;
- * and in general xmlns: pseudo-attributes are not declared...
- * so not quite sure what to return here. For now, let's
- * return true, to indicate they ought to be valid
- */
- return true;
- }
- }
- throwNoSuchAttribute(index);
- return false; // never gets here
- }
-
- public boolean isDeclared(String qName)
- {
- return false;
- }
-
- public boolean isDeclared(String uri, String localName)
- {
- return false;
- }
-
- public boolean isSpecified(int index)
- {
- if (index < mAttrCount) {
- if (index >= 0) {
- return mAttrCollector.isSpecified(index);
- }
- } else {
- index -= mAttrCount;
- if (index < mNsCount) {
- /* Determining default-attr - based namespace declarations
- * would need new accessors on Woodstox... but they are
- * extremely rare, too
- */
- return true;
- }
- }
- throwNoSuchAttribute(index);
- return false; // never gets here
- }
-
- public boolean isSpecified(String qName)
- {
- int ix = getIndex(qName);
- if (ix < 0) {
- throw new IllegalArgumentException("No attribute with qName '"+qName+"'");
- }
- return isSpecified(ix);
- }
-
- public boolean isSpecified(String uri, String localName)
- {
- int ix = getIndex(uri, localName);
- if (ix < 0) {
- throw new IllegalArgumentException("No attribute with uri "+uri+", local name '"+localName+"'");
- }
- return isSpecified(ix);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Locator (SAX1) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- public int getColumnNumber()
- {
- if (mScanner != null) {
- Location loc = mScanner.getLocation();
- return loc.getColumnNumber();
- }
- return -1;
- }
-
- public int getLineNumber()
- {
- if (mScanner != null) {
- Location loc = mScanner.getLocation();
- return loc.getLineNumber();
- }
- return -1;
- }
-
- public String getPublicId()
- {
- if (mScanner != null) {
- Location loc = mScanner.getLocation();
- return loc.getPublicId();
- }
- return null;
- }
-
- public String getSystemId()
- {
- if (mScanner != null) {
- Location loc = mScanner.getLocation();
- return loc.getSystemId();
- }
- return null;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Locator2 (SAX2) implementation
- ///////////////////////////////////////////////////////////////////////
- */
-
- public String getEncoding()
- {
- return mEncoding;
- }
-
- public String getXMLVersion()
- {
- return mXmlVersion;
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // DTDEventListener (woodstox internal API) impl
- ///////////////////////////////////////////////////////////////////////
- */
-
- public boolean dtdReportComments()
- {
- return (mLexicalHandler != null);
- }
-
- public void dtdComment(char[] data, int offset, int len)
- {
- if (mLexicalHandler != null) {
- try {
- mLexicalHandler.comment(data, offset, len);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdProcessingInstruction(String target, String data)
- {
- if (mContentHandler != null) {
- try {
- mContentHandler.processingInstruction(target, data);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdSkippedEntity(String name)
- {
- if (mContentHandler != null) {
- try {
- mContentHandler.skippedEntity(name);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- // DTD declarations that must be exposed
- public void dtdNotationDecl(String name, String publicId, String systemId, URL baseURL)
- throws XMLStreamException
- {
- if (mDTDHandler != null) {
- /* 24-Nov-2006, TSa: Note: SAX expects system identifiers to
- * be fully resolved when reported...
- */
- if (systemId != null && systemId.indexOf(':') < 0) {
- try {
- systemId = URLUtil.urlFromSystemId(systemId, baseURL).toExternalForm();
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
- try {
- mDTDHandler.notationDecl(name, publicId, systemId);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdUnparsedEntityDecl(String name, String publicId, String systemId, String notationName, URL baseURL)
- throws XMLStreamException
- {
- if (mDTDHandler != null) {
- // SAX expects system id to be fully resolved?
- if (systemId.indexOf(':') < 0) { // relative path...
- try {
- systemId = URLUtil.urlFromSystemId(systemId, baseURL).toExternalForm();
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
- try {
- mDTDHandler.unparsedEntityDecl(name, publicId, systemId, notationName);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- // DTD declarations that can be exposed
-
- public void attributeDecl(String eName, String aName, String type, String mode, String value)
- {
- if (mDeclHandler != null) {
- try {
- mDeclHandler.attributeDecl(eName, aName, type, mode, value);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdElementDecl(String name, String model)
- {
- if (mDeclHandler != null) {
- try {
- mDeclHandler.elementDecl(name, model);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdExternalEntityDecl(String name, String publicId, String systemId)
- {
- if (mDeclHandler != null) {
- try {
- mDeclHandler.externalEntityDecl(name, publicId, systemId);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- public void dtdInternalEntityDecl(String name, String value)
- {
- if (mDeclHandler != null) {
- try {
- mDeclHandler.internalEntityDecl(name, value);
- } catch (SAXException sex) {
- throw new WrappedSaxException(sex);
- }
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////////////////
- */
-
- private void throwSaxException(Exception src)
- throws SAXException
- {
- SAXParseException se = new SAXParseException(src.getMessage(), /*(Locator)*/ this, src);
- ExceptionUtil.setInitCause(se, src);
- if (mErrorHandler != null) {
- mErrorHandler.fatalError(se);
- }
- throw se;
- }
-
- private void throwSaxException(String msg)
- throws SAXException
- {
- SAXParseException se = new SAXParseException(msg, /*(Locator)*/ this);
- if (mErrorHandler != null) {
- mErrorHandler.fatalError(se);
- }
- throw se;
- }
-
- private void throwNoSuchAttribute(int index)
- {
- throw new IllegalArgumentException("No attribute with index "+index+" (have "+(mAttrCount+mNsCount)+" attributes)");
- }
-
- /*
- /////////////////////////////////////////////////
- // Helper class for dealing with entity resolution
- /////////////////////////////////////////////////
- */
-
- /**
- * Simple helper class that converts from Stax API into SAX
- * EntityResolver call(s)
- */
- final class ResolverProxy
- implements XMLResolver
- {
- public ResolverProxy() { }
-
- public Object resolveEntity(String publicID, String systemID, String baseURI, String namespace)
- throws XMLStreamException
- {
- if (mEntityResolver != null) {
- try {
- /* Hmmh. SAX expects system id to have been mangled prior
- * to call... this may work, depending on stax impl:
- */
- URL url = new URL(baseURI);
- String ref = new URL(url, systemID).toExternalForm();
- InputSource isrc = mEntityResolver.resolveEntity(publicID, ref);
- if (isrc != null) {
- //System.err.println("Debug: succesfully resolved '"+publicID+"', '"+systemID+"'");
- InputStream in = isrc.getByteStream();
- if (in != null) {
- return in;
- }
- Reader r = isrc.getCharacterStream();
- if (r != null) {
- return r;
- }
- }
-
- // Returning null should be fine, actually...
- return null;
- } catch (IOException ex) {
- throw new WstxIOException(ex);
- } catch (Exception ex) {
- throw new XMLStreamException(ex.getMessage(), ex);
- }
- }
- return null;
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Helper classes for SAX1 support
- ///////////////////////////////////////////////////////////////////////
- */
-
- final static class DocHandlerWrapper
- implements ContentHandler
- {
- final DocumentHandler mDocHandler;
-
- final AttributesWrapper mAttrWrapper = new AttributesWrapper();
-
- DocHandlerWrapper(DocumentHandler h)
- {
- mDocHandler = h;
- }
-
- public void characters(char[] ch, int start, int length)
- throws SAXException
- {
- mDocHandler.characters(ch, start, length);
- }
-
- public void endDocument() throws SAXException
- {
- mDocHandler.endDocument();
- }
-
- public void endElement(String uri, String localName, String qName)
- throws SAXException
- {
- if (qName == null) {
- qName = localName;
- }
- mDocHandler.endElement(qName);
- }
-
- public void endPrefixMapping(String prefix)
- {
- // no equivalent in SAX1, ignore
- }
-
- public void ignorableWhitespace(char[] ch, int start, int length)
- throws SAXException
- {
- mDocHandler.ignorableWhitespace(ch, start, length);
- }
-
- public void processingInstruction(String target, String data)
- throws SAXException
- {
- mDocHandler.processingInstruction(target, data);
- }
-
- public void setDocumentLocator(Locator locator)
- {
- mDocHandler.setDocumentLocator(locator);
- }
-
- public void skippedEntity(String name)
- {
- // no equivalent in SAX1, ignore
- }
-
- public void startDocument()
- throws SAXException
- {
- mDocHandler.startDocument();
- }
-
- public void startElement(String uri, String localName, String qName,
- Attributes attrs)
- throws SAXException
- {
- if (qName == null) {
- qName = localName;
- }
- // Also, need to wrap Attributes to look like AttributeLost
- mAttrWrapper.setAttributes(attrs);
- mDocHandler.startElement(qName, mAttrWrapper);
- }
-
- public void startPrefixMapping(String prefix, String uri)
- {
- // no equivalent in SAX1, ignore
- }
- }
-
- final static class AttributesWrapper
- implements AttributeList
- {
- Attributes mAttrs;
-
- public AttributesWrapper() { }
-
- public void setAttributes(Attributes a) {
- mAttrs = a;
- }
-
- public int getLength()
- {
- return mAttrs.getLength();
- }
-
- public String getName(int i)
- {
- String n = mAttrs.getQName(i);
- return (n == null) ? mAttrs.getLocalName(i) : n;
- }
-
- public String getType(int i)
- {
- return mAttrs.getType(i);
- }
-
- public String getType(String name)
- {
- return mAttrs.getType(name);
- }
-
- public String getValue(int i)
- {
- return mAttrs.getValue(i);
- }
-
- public String getValue(String name)
- {
- return mAttrs.getValue(name);
- }
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/AttributeCollector.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/AttributeCollector.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/AttributeCollector.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/AttributeCollector.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1145 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.io.IOException;
-import java.util.Arrays;
-
-import javax.xml.XMLConstants;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.namespace.QName;
-
-import org.codehaus.stax2.ri.typed.CharArrayBase64Decoder;
-import org.codehaus.stax2.ri.typed.ValueDecoderFactory;
-import org.codehaus.stax2.typed.Base64Variant;
-import org.codehaus.stax2.typed.TypedArrayDecoder;
-import org.codehaus.stax2.typed.TypedValueDecoder;
-import org.codehaus.stax2.typed.TypedXMLStreamException;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.sw.XmlWriter;
-import com.ctc.wstx.util.*;
-
-/**
- * Shared base class that defines API stream reader uses to communicate
- * with the attribute collector implementation, independent of whether it's
- * operating in namespace-aware or non-namespace modes.
- * Collector class is used to build up attribute lists; for the most part
- * will just hold references to few specialized {@link TextBuilder}s that
- * are used to create efficient semi-shared value Strings.
- */
-public final class AttributeCollector
-{
- final static int INT_SPACE = 0x0020;
-
- /**
- * Threshold value that indicates minimum length for lists instances
- * that need a Map structure, for fast attribute access by fully-qualified
- * name.
- */
- protected final static int LONG_ATTR_LIST_LEN = 4;
-
- /**
- * Expected typical maximum number of attributes for any element;
- * chosen to minimize need to resize, while trying not to waste space.
- * Dynamically grown; better not to set too high to avoid excessive
- * overhead for small attribute-less documents.
- */
- protected final static int EXP_ATTR_COUNT = 12;
-
- protected final static int EXP_NS_COUNT = 6;
-
- /**
- * This value is used to indicate that we shouldn't keep track
- * of index of xml:id attribute -- generally done when Xml:id
- * support is disabled
- */
- protected final static int XMLID_IX_DISABLED = -2;
-
- protected final static int XMLID_IX_NONE = -1;
-
- protected final static InternCache sInternCache = InternCache.getInstance();
-
- /*
- ///////////////////////////////////////////////////////////
- // Configuration
- ///////////////////////////////////////////////////////////
- */
-
- // // Settings for matching Xml:id attribute
-
- final String mXmlIdPrefix;
- final String mXmlIdLocalName;
-
- /*
- ///////////////////////////////////////////////////////////
- // Collected attribute (incl namespace attrs) information:
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Array of attributes collected for this element.
- */
- protected Attribute[] mAttributes;
-
- /**
- * Actual number of attributes collected, including attributes
- * added via default values.
- */
- protected int mAttrCount;
-
- /**
- * Number of attribute values actually parsed, not including
- * ones created via default value expansion. Equal to or less than
- * {@link #mAttrCount}.
- */
- protected int mNonDefCount;
-
- /**
- * Array of namespace declaration attributes collected for this element;
- * not used in non-namespace-aware mode
- */
- protected Attribute[] mNamespaces;
-
- /**
- * Number of valid namespace declarations in {@link #mNamespaces}.
- */
- protected int mNsCount;
-
- /**
- * Flag to indicate whether the default namespace has already been declared
- * for the current element.
- */
- protected boolean mDefaultNsDeclared = false;
-
- /**
- * Index of "xml:id" attribute, if one exists for the current
- * element; {@link #XMLID_IX_NONE} if none.
- */
- protected int mXmlIdAttrIndex;
-
- /*
- ///////////////////////////////////////////////////////////
- // Attribute (and ns) value builders
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * TextBuilder into which values of all attributes are appended
- * to, including default valued ones (defaults are added after
- * explicit ones).
- * Constructed lazily, if and when needed (not needed
- * for short attribute-less docs)
- */
- protected TextBuilder mValueBuilder = null;
-
- /**
- * TextBuilder into which values of namespace URIs are added (including
- * URI for the default namespace, if one defined).
- */
- private final TextBuilder mNamespaceBuilder = new TextBuilder(EXP_NS_COUNT);
-
- /*
- //////////////////////////////////////////////////////////////
- // Information that defines "Map-like" data structure used for
- // quick access to attribute values by fully-qualified name
- //////////////////////////////////////////////////////////////
- */
-
- /**
- * Encoding of a data structure that contains mapping from
- * attribute names to attribute index in main attribute name arrays.
- *mAttrHashSize), and remaining spillover area
- * that follows hash area up until (but not including)
- * mAttrSpillEnd index.
- * Main hash area only contains indexes (index+1; 0 signifying empty slot)
- * to actual attributes; spillover area has both hash and index for
- * any spilled entry. Spilled entries are simply stored in order
- * added, and need to be searched using linear search. In case of both
- * primary hash hits and spills, eventual comparison with the local
- * name needs to be done with actual name array.
- */
- protected int[] mAttrMap = null;
-
- /**
- * Size of hash area in mAttrMap; generally at least 20%
- * more than number of attributes (mAttrCount).
- */
- protected int mAttrHashSize;
-
- /**
- * Pointer to int slot right after last spill entr, in
- * mAttrMap array.
- */
- protected int mAttrSpillEnd;
-
- /*
- ///////////////////////////////////////////////
- // Life-cycle:
- ///////////////////////////////////////////////
- */
-
- protected AttributeCollector(ReaderConfig cfg, boolean nsAware)
- {
- mXmlIdAttrIndex = cfg.willDoXmlIdTyping() ? XMLID_IX_NONE : XMLID_IX_DISABLED;
- if (nsAware) {
- mXmlIdPrefix = "xml";
- mXmlIdLocalName = "id";
- } else {
- mXmlIdPrefix = null;
- mXmlIdLocalName = "xml:id";
- }
- }
-
- /**
- * Method called to allow reusing of collector, usually right before
- * starting collecting attributes for a new start tag.
- */
- /**
- * Method called to allow reusing of collector, usually right before
- * starting collecting attributes for a new start tag.
- *getTextXxx
- * methods can be called; which is less than those for which
- * {@link #getText} can be called. Specifically, DTD and
- * ENTITY_REFERENCE types do not support these extended
- */
- final protected static int MASK_GET_TEXT_XXX =
- (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE) | (1 << COMMENT);
-
- /**
- * This mask is used with Stax2 getText() method (one that takes
- * Writer as an argument): accepts even wider range of event types.
- */
- final protected static int MASK_GET_TEXT_WITH_WRITER =
- (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE)
- | (1 << COMMENT) | (1 << DTD) | (1 << ENTITY_REFERENCE)
- | (1 << PROCESSING_INSTRUCTION);
-
- final protected static int MASK_GET_ELEMENT_TEXT =
- (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE)
- | (1 << ENTITY_REFERENCE);
-
-
- // // // Indicator of type of text in text event (WRT white space)
-
- final static int ALL_WS_UNKNOWN = 0x0000;
- final static int ALL_WS_YES = 0x0001;
- final static int ALL_WS_NO = 0x0002;
-
- /* 2 magic constants used for enabling/disabling indentation checks:
- * (to minimize negative impact for both small docs, and large
- * docs with non-regular white space)
- */
-
- private final static int INDENT_CHECK_START = 16;
-
- private final static int INDENT_CHECK_MAX = 40;
-
- // // // Shared namespace symbols
-
- final protected static String sPrefixXml = DefaultXmlSymbolTable.getXmlSymbol();
-
- final protected static String sPrefixXmlns = DefaultXmlSymbolTable.getXmlnsSymbol();
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Configuration
- ///////////////////////////////////////////////////////////////////////
- */
-
- // note: mConfig defined in base class
-
- /**
- * Set of locally stored configuration flags
- */
- protected final int mConfigFlags;
-
- // // // Various extracted settings:
-
- protected final boolean mCfgCoalesceText;
-
- protected final boolean mCfgReportTextAsChars;
- protected final boolean mCfgLazyParsing;
-
- /**
- * Minimum number of characters parser can return as partial text
- * segment, IF it's not required to coalesce adjacent text
- * segments.
- */
- protected final int mShortestTextSegment;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Symbol handling:
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Object to notify about shared stuff, such as symbol tables, as well
- * as to query for additional config settings if necessary.
- */
- final protected ReaderCreator mOwner;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Additional XML document information, in addition
- // to what StreamScanner has
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Status about "stand-aloneness" of document; set to 'yes'/'no'/'unknown'
- * based on whether there was xml declaration, and if so, whether
- * it had standalone attribute.
- */
- protected int mDocStandalone = DOC_STANDALONE_UNKNOWN;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // DOCTYPE information from document type declaration
- // (if any found)
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Prefix of root element, as dictated by DOCTYPE declaration; null
- * if no DOCTYPE declaration, or no root prefix
- */
- String mRootPrefix;
-
- /**
- * Local name of root element, as dictated by DOCTYPE declaration; null
- * if no DOCTYPE declaration.
- */
- String mRootLName;
-
- /**
- * Public id of the DTD, if one exists and has been parsed.
- */
- protected String mDtdPublicId;
-
- /**
- * System id of the DTD, if one exists and has been parsed.
- */
- protected String mDtdSystemId;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Information about currently open subtree, content
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * TextBuffer mostly used to collect non-element textual content
- * (text, CDATA, comment content, pi data)
- */
- final protected TextBuffer mTextBuffer;
-
- /**
- * Currently open element tree
- */
- final protected InputElementStack mElementStack;
-
- /**
- * Object that stores information about currently accessible attributes.
- */
- final protected AttributeCollector mAttrCollector;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Tokenization state
- ////////////////////////////////////////////////////
- */
-
- /// Flag set when DOCTYPE declaration has been parsed
- protected boolean mStDoctypeFound = false;
-
- /**
- * State of the current token; one of M_ - constants from above.
- *getText and its variations), we may need
- * to delay throwing an exception until {@link #next} is called next
- * time. If so, this variable holds the pending stream exception.
- */
- protected XMLStreamException mPendingException = null;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // DTD information (entities, content spec stub)
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Entities parsed from internal/external DTD subsets. Although it
- * will remain null for this class, extended classes make use of it,
- * plus, to be able to share some of entity resolution code, instance
- * is left here even though it semantically belongs to the sub-class.
- */
- protected Map mGeneralEntities = null;
-
- /**
- * Mode information needed at this level; mostly to check what kind
- * of textual content (if any) is allowed in current element
- * context. Constants come from
- * {@link XMLValidator},
- * (like {@link XMLValidator#CONTENT_ALLOW_VALIDATABLE_TEXT}).
- * Only used inside tree; ignored for prolog/epilog (which
- * have straight-forward static rules).
- */
- protected int mVldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT;
-
- /**
- * Configuration from {@link XMLStreamProperties.RETURN_NULL_FOR_DEFAULT_NAMESPACE}
- *
- * @since 4.1.2
- */
- protected boolean mReturnNullForDefaultNamespace;
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Instance construction, initialization
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * @param elemStack Input element stack to use; if null, will create
- * instance locally.
- * @param forER Override indicator; if true, this stream reader will be
- * used by an event reader, and should modify some of the base config
- * settings appropriately. If false, configuration settings are to
- * be used as is.
- */
- protected BasicStreamReader(InputBootstrapper bs,
- BranchingReaderSource input, ReaderCreator owner,
- ReaderConfig cfg, InputElementStack elemStack,
- boolean forER)
- throws XMLStreamException
- {
- super(input, cfg, cfg.getEntityResolver());
-
- mOwner = owner;
-
- mTextBuffer = TextBuffer.createRecyclableBuffer(cfg);
-
- // // // First, configuration settings:
-
- mConfigFlags = cfg.getConfigFlags();
- mCfgCoalesceText = (mConfigFlags & CFG_COALESCE_TEXT) != 0;
- mCfgReportTextAsChars = (mConfigFlags & CFG_REPORT_CDATA) == 0;
- mXml11 = cfg.isXml11();
-
- // Can only use canonical white space if we are normalizing lfs
- mCheckIndentation = mNormalizeLFs ? 16 : 0;
-
- /* 30-Sep-2005, TSa: Let's not do lazy parsing when access is via
- * Event API. Reason is that there will be no performance benefit
- * (event objects always access full info right after traversal),
- * but the wrapping of stream exceptions within runtime exception
- * wrappers would happen, which is inconvenient (loss of stack trace,
- * not catching all exceptions as expected)
- */
- mCfgLazyParsing = !forER && ((mConfigFlags & CFG_LAZY_PARSING) != 0);
-
- /* There are a few derived settings used during tokenization that
- * need to be initialized now...
- */
- if (mCfgCoalesceText) {
- mStTextThreshold = TOKEN_FULL_COALESCED;
- mShortestTextSegment = Integer.MAX_VALUE;
- } else {
- mStTextThreshold = TOKEN_PARTIAL_SINGLE;
- if (forER) {
- /* 30-Sep-2005, TSa: No point in returning runt segments for
- * event readers (due to event object overhead, less
- * convenient); let's just force returning of full length
- * segments.
- */
- mShortestTextSegment = Integer.MAX_VALUE;
- } else {
- mShortestTextSegment = cfg.getShortestReportedTextSegment();
- }
- }
-
- // // // Then handling of xml declaration data:
-
- mDocXmlVersion = bs.getDeclaredVersion();
- mDocInputEncoding = bs.getInputEncoding();
- mDocXmlEncoding = bs.getDeclaredEncoding();
-
- String sa = bs.getStandalone();
- if (sa == null) {
- mDocStandalone = DOC_STANDALONE_UNKNOWN;
- } else {
- if (XmlConsts.XML_SA_YES.equals(sa)) {
- mDocStandalone = DOC_STANDALONE_YES;
- } else {
- mDocStandalone = DOC_STANDALONE_NO;
- }
- }
-
- /* Ok; either we got declaration or not, but in either case we can
- * now initialize prolog parsing settings, without having to really
- * parse anything more.
- */
- /* 07-Oct-2005, TSa: Except, if we are in fragment mode, in which
- * case we are kind of "in tree" mode...
- */
- mParseState = mConfig.inputParsingModeFragment() ?
- STATE_TREE : STATE_PROLOG;
-
- // // // And then connecting element stack and attribute collector
-
- mElementStack = elemStack;
- mAttrCollector = elemStack.getAttrCollector();
-
- // And finally, location information may have offsets:
- input.initInputLocation(this, mCurrDepth);
-
- elemStack.connectReporter(this);
-
- Object value = getProperty(WstxInputProperties.P_RETURN_NULL_FOR_DEFAULT_NAMESPACE);
- mReturnNullForDefaultNamespace = (value instanceof Boolean) && ((Boolean) value).booleanValue();
- }
-
- protected static InputElementStack createElementStack(ReaderConfig cfg)
- {
- return new InputElementStack(cfg, cfg.willSupportNamespaces());
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // XMLStreamReader, document info
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * As per Stax (1.0) specs, needs to return whatever xml declaration
- * claimed encoding is, if any; or null if no xml declaration found.
- *
- * Reads the content of a text-only element, an exception is thrown if
- * this is not a text-only element.
- * Regardless of value of javax.xml.stream.isCoalescing this method always
- * returns coalesced content.
- *
- */
- public String getElementText()
- throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throwParseError(ErrorConsts.ERR_STATE_NOT_STELEM, null, null);
- }
- /* Ok, now: with START_ELEMENT we know that it's not partially
- * processed; that we are in-tree (not prolog or epilog).
- * The only possible complication would be:
- */
- if (mStEmptyElem) {
- /* And if so, we'll then get 'virtual' close tag; things
- * are simple as location info was set when dealing with
- * empty start element; and likewise, validation (if any)
- * has been taken care of
- */
- mStEmptyElem = false;
- mCurrToken = END_ELEMENT;
- return "";
- }
- // First need to find a textual event
- while (true) {
- int type = next();
- if (type == END_ELEMENT) {
- return "";
- }
- if (type == COMMENT || type == PROCESSING_INSTRUCTION) {
- continue;
- }
- if (((1 << type) & MASK_GET_ELEMENT_TEXT) == 0) {
- throw _constructUnexpectedInTyped(type);
- }
- break;
- }
- if (mTokenState < TOKEN_FULL_SINGLE) {
- readCoalescedText(mCurrToken, false);
- }
- /* Ok: then a quick check; if it looks like we are directly
- * followed by the end tag, we need not construct String
- * quite yet.
- */
- if ((mInputPtr + 1) < mInputEnd &&
- mInputBuffer[mInputPtr] == '<' && mInputBuffer[mInputPtr+1] == '/') {
- // Note: next() has validated text, no need for more validation
- mInputPtr += 2;
- mCurrToken = END_ELEMENT;
- // must first get text, as call to readEndElem may break it:
- String result = mTextBuffer.contentsAsString();
- // Can by-pass next(), nextFromTree(), in this case:
- readEndElem();
- // and then return results
- return result;
- }
-
- // Otherwise, we'll need to do slower processing
- int extra = 1 + (mTextBuffer.size() >> 1); // let's add 50% space
- StringBuffer sb = mTextBuffer.contentsAsStringBuffer(extra);
- int type;
-
- while ((type = next()) != END_ELEMENT) {
- if (((1 << type) & MASK_GET_ELEMENT_TEXT) != 0) {
- if (mTokenState < mStTextThreshold) {
- finishToken(false);
- }
- mTextBuffer.contentsToStringBuffer(sb);
- continue;
- }
- if (type != COMMENT && type != PROCESSING_INSTRUCTION) {
- throw _constructUnexpectedInTyped(type);
- }
- }
- // Note: calls next() have validated text, no need for more validation
- return sb.toString();
- }
-
- /**
- * Returns type of the last event returned; or START_DOCUMENT before
- * any events has been explicitly returned.
- */
- public int getEventType()
- {
- /* Only complication -- multi-part coalesced text is to be reported
- * as CHARACTERS always, never as CDATA (StAX specs).
- */
- if (mCurrToken == CDATA) {
- if (mCfgCoalesceText || mCfgReportTextAsChars) {
- return CHARACTERS;
- }
- }
- return mCurrToken;
- }
-
- public String getLocalName()
- {
- // Note: for this we need not (yet) finish reading element
- if (mCurrToken == START_ELEMENT || mCurrToken == END_ELEMENT) {
- return mElementStack.getLocalName();
- }
- if (mCurrToken == ENTITY_REFERENCE) {
- /* 30-Sep-2005, TSa: Entity will be null in non-expanding mode
- * if no definition was found:
- */
- return (mCurrEntity == null) ? mCurrName: mCurrEntity.getName();
- }
- throw new IllegalStateException("Current state not START_ELEMENT, END_ELEMENT or ENTITY_REFERENCE");
- }
-
- // // // getLocation() defined in StreamScanner
-
- public QName getName()
- {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- return mElementStack.getCurrentElementName();
- }
-
- // // // Namespace access
-
- public NamespaceContext getNamespaceContext() {
- /* Unlike other getNamespaceXxx methods, this is available
- * for all events.
- * Note that the context is "live", ie. remains active (but not
- * static) even through calls to next(). StAX compliant apps
- * should not count on this behaviour, however.
- */
- return mElementStack;
- }
-
- public int getNamespaceCount() {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- return mElementStack.getCurrentNsCount();
- }
-
- public String getNamespacePrefix(int index) {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- // Internally it's marked as null, externally need to see "" or null, depending
- String p = mElementStack.getLocalNsPrefix(index);
- if (p == null) {
- return mReturnNullForDefaultNamespace ? null : XmlConsts.ATTR_NO_PREFIX;
- }
- return p;
- }
-
- public String getNamespaceURI() {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- // Internally it's marked as null, externally need to see ""
- String uri = mElementStack.getNsURI();
- return (uri == null) ? XmlConsts.ELEM_NO_NS_URI : uri;
- }
-
- public String getNamespaceURI(int index)
- {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- // Internally it's marked as null, externally need to see ""
- String uri = mElementStack.getLocalNsURI(index);
- return (uri == null) ? XmlConsts.ATTR_NO_NS_URI : uri;
- }
-
- public String getNamespaceURI(String prefix)
- {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- /* Note: this will need to return null if no URI found for
- * the prefix, so we can't mask it.
- */
- return mElementStack.getNamespaceURI(prefix);
- }
-
- public String getPIData() {
- if (mCurrToken != PROCESSING_INSTRUCTION) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_PI);
- }
- if (mTokenState <= TOKEN_STARTED) {
- safeFinishToken();
- }
- return mTextBuffer.contentsAsString();
- }
-
- public String getPITarget() {
- if (mCurrToken != PROCESSING_INSTRUCTION) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_PI);
- }
- // Target is always parsed automatically, not lazily...
- return mCurrName;
- }
-
- public String getPrefix() {
- if (mCurrToken != START_ELEMENT && mCurrToken != END_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_ELEM);
- }
- // Internally it's marked as null, externally need to see ""
- String p = mElementStack.getPrefix();
- return (p == null) ? XmlConsts.ELEM_NO_PREFIX : p;
- }
-
- public String getText()
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT) == 0) {
- throwNotTextual(mCurrToken);
- }
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- if (mCurrToken == ENTITY_REFERENCE) {
- return (mCurrEntity == null) ? null : mCurrEntity.getReplacementText();
- }
- if (mCurrToken == DTD) {
- /* 16-Aug-2004, TSa: Hmmh. Specs are bit ambiguous on whether this
- * should return just the internal subset, or the whole
- * thing...
- */
- return getDTDInternalSubset();
- }
- return mTextBuffer.contentsAsString();
- }
-
- public char[] getTextCharacters()
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT_XXX) == 0) {
- throwNotTextXxx(mCurrToken);
- }
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- if (mCurrToken == ENTITY_REFERENCE) {
- return mCurrEntity.getReplacementChars();
- }
- if (mCurrToken == DTD) {
- return getDTDInternalSubsetArray();
- }
- return mTextBuffer.getTextBuffer();
- }
-
- public int getTextCharacters(int sourceStart, char[] target, int targetStart, int len)
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT_XXX) == 0) {
- throwNotTextXxx(mCurrToken);
- }
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- return mTextBuffer.contentsToArray(sourceStart, target, targetStart, len);
- }
-
- public int getTextLength()
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT_XXX) == 0) {
- throwNotTextXxx(mCurrToken);
- }
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- return mTextBuffer.size();
- }
-
- public int getTextStart()
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT_XXX) == 0) {
- throwNotTextXxx(mCurrToken);
- }
- if (mTokenState < mStTextThreshold) {
- safeFinishToken();
- }
- return mTextBuffer.getTextStart();
- }
-
- public boolean hasName() {
- return (mCurrToken == START_ELEMENT) || (mCurrToken == END_ELEMENT);
- }
-
- public boolean hasNext() {
- /* 08-Oct-2005, TSa: In multi-doc mode, we have different
- * criteria...
- */
- return (mCurrToken != END_DOCUMENT)
- || (mParseState == STATE_MULTIDOC_HACK);
- }
-
- public boolean hasText() {
- return (((1 << mCurrToken) & MASK_GET_TEXT) != 0);
- }
-
- public boolean isAttributeSpecified(int index)
- {
- /* No need to check for ATTRIBUTE since we never return that...
- */
- if (mCurrToken != START_ELEMENT) {
- throw new IllegalStateException(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- return mAttrCollector.isSpecified(index);
- }
-
- public boolean isCharacters()
- {
- /* 21-Dec-2005, TSa: Changed for 3.0 to work the same way as stax
- * ref impl.
- */
- //return (mCurrToken == CHARACTERS || mCurrToken == CDATA || mCurrToken == SPACE);
- /* 21-Apr-2009, TSa: As per [WSTX-201], should be consistent with
- * what getEventType() returns (affects CDATA, SPACE, in
- * coalescing mode or when explicitly asked to return CDATA
- * as CHARACTERS)
- */
- return (getEventType() == CHARACTERS);
- }
-
- public boolean isEndElement() {
- return (mCurrToken == END_ELEMENT);
- }
-
- public boolean isStartElement() {
- return (mCurrToken == START_ELEMENT);
- }
-
- /**
- *
Precondition: the current event is START_ELEMENT.
- *
Postcondition: the current event is the corresponding END_ELEMENT.
- *this.
- */
- public DTDInfo getDTDInfo() throws XMLStreamException
- {
- /* Let's not allow it to be accessed during other events -- that
- * way callers won't count on it being available afterwards.
- */
- if (mCurrToken != DTD) {
- return null;
- }
- if (mTokenState < TOKEN_FULL_SINGLE) { // need to fully read it in now
- finishToken(false);
- }
- return this;
- }
-
- // // // StAX2, Additional location information
-
- /**
- * Location information is always accessible, for this reader.
- */
- public final LocationInfo getLocationInfo() {
- return this;
- }
-
- // // // StAX2, Pass-through text accessors
-
-
- /**
- * Method similar to {@link #getText()}, except
- * that it just uses provided Writer to write all textual content.
- * For further optimization, it may also be allowed to do true
- * pass-through, thus possibly avoiding one temporary copy of the
- * data.
- *getText will return
- * proper conntets. If false, reader is allowed to skip creation
- * of such copies: this can improve performance, but it also means
- * that further calls to getText is not guaranteed to
- * return meaningful data.
- *
- * @return Number of characters written to the reader
- */
- public int getText(Writer w, boolean preserveContents)
- throws IOException, XMLStreamException
- {
- if (((1 << mCurrToken) & MASK_GET_TEXT_WITH_WRITER) == 0) {
- throwNotTextual(mCurrToken);
- }
- /* May need to be able to do fully streaming... but only for
- * text events that have not yet been fully read; for other
- * types there's less benefit, and for fully read ones, we
- * already have everything ready.
- */
- if (!preserveContents) {
- if (mCurrToken == CHARACTERS) {
- int count = mTextBuffer.rawContentsTo(w);
- /* Let's also clear whatever was collected (as allowed by
- * method contract) previously, to both save memory, and
- * to ensure caller doesn't accidentally try to access it
- * (and get otherwise 'random' results).
- */
- mTextBuffer.resetWithEmpty();
- if (mTokenState < TOKEN_FULL_SINGLE) {
- count += readAndWriteText(w);
- }
- if (mCfgCoalesceText &&
- (mTokenState < TOKEN_FULL_COALESCED)) {
- if (mCfgCoalesceText) {
- count += readAndWriteCoalesced(w, false);
- }
- }
- return count;
- } else if (mCurrToken == CDATA) {
- int count = mTextBuffer.rawContentsTo(w);
- mTextBuffer.resetWithEmpty(); // same as with CHARACTERS
- if (mTokenState < TOKEN_FULL_SINGLE) {
- count += readAndWriteCData(w);
- }
- if (mCfgCoalesceText &&
- (mTokenState < TOKEN_FULL_COALESCED)) {
- if (mCfgCoalesceText) {
- count += readAndWriteCoalesced(w, true);
- }
- }
- return count;
- }
- }
- if (mTokenState < mStTextThreshold) {
- /* Otherwise, let's just finish the token; and due to guarantee
- * by streaming method, let's try ensure we get it all.
- */
- finishToken(false); // false -> shouldn't defer errors
- }
- if (mCurrToken == ENTITY_REFERENCE) {
- return mCurrEntity.getReplacementText(w);
- }
- if (mCurrToken == DTD) {
- char[] ch = getDTDInternalSubsetArray();
- if (ch != null) {
- w.write(ch);
- return ch.length;
- }
- return 0;
- }
- return mTextBuffer.rawContentsTo(w);
- }
-
- // // // StAX 2, Other accessors
-
- /**
- * @return Number of open elements in the stack; 0 when parser is in
- * prolog/epilog, 1 inside root element and so on.
- */
- public int getDepth() {
- /* Note: we can not necessarily use mCurrDepth, since it is
- * directly synchronized to the input (to catch unbalanced entity
- * expansion WRT element nesting), and not to actual token values
- * returned.
- */
- return mElementStack.getDepth();
- }
-
- /**
- * @return True, if cursor points to a start or end element that is
- * constructed from 'empty' element (ends with '/>');
- * false otherwise.
- */
- public boolean isEmptyElement() throws XMLStreamException
- {
- return (mCurrToken == START_ELEMENT) ? mStEmptyElem : false;
- }
-
- public NamespaceContext getNonTransientNamespaceContext()
- {
- // null -> no Location info, not needed with basic API
- return mElementStack.createNonTransientNsContext(null);
- }
-
- public String getPrefixedName()
- {
- switch (mCurrToken) {
- case START_ELEMENT:
- case END_ELEMENT:
- {
- String prefix = mElementStack.getPrefix();
- String ln = mElementStack.getLocalName();
-
- if (prefix == null) {
- return ln;
- }
- StringBuffer sb = new StringBuffer(ln.length() + 1 + prefix.length());
- sb.append(prefix);
- sb.append(':');
- sb.append(ln);
- return sb.toString();
- }
- case ENTITY_REFERENCE:
- return getLocalName();
- case PROCESSING_INSTRUCTION:
- return getPITarget();
- case DTD:
- return getDTDRootName();
-
- }
- throw new IllegalStateException("Current state not START_ELEMENT, END_ELEMENT, ENTITY_REFERENCE, PROCESSING_INSTRUCTION or DTD");
- }
-
- public void closeCompletely() throws XMLStreamException
- {
- closeAllInput(true);
- }
-
- /*
- ////////////////////////////////////////////////////
- // DTDInfo implementation (StAX 2)
- ////////////////////////////////////////////////////
- */
-
- /**
- *PROCESSING_INSTRUCTION; but may be
- * different in multi-doc mode, if we actually hit a secondary
- * xml declaration.
- */
- private final int readPIPrimary()
- throws XMLStreamException
- {
- // Ok, first we need the name:
- String target = parseFullName();
- mCurrName = target;
-
- if (target.length() == 0) {
- throwParseError(ErrorConsts.ERR_WF_PI_MISSING_TARGET);
- }
-
- // As per XML specs, #17, case-insensitive 'xml' is illegal:
- if (target.equalsIgnoreCase("xml")) {
- // 07-Oct-2005, TSa: Still legal in multi-doc mode...
- if (!mConfig.inputParsingModeDocuments()) {
- throwParseError(ErrorConsts.ERR_WF_PI_XML_TARGET, target, null);
- }
- // Ok, let's just verify we get space then
- char c = getNextCharFromCurrent(SUFFIX_IN_XML_DECL);
- if (!isSpaceChar(c)) {
- throwUnexpectedChar(c, "excepted a space in xml declaration after 'xml'");
- }
- return handleMultiDocStart(START_DOCUMENT);
- }
-
- // And then either white space before data, or end marker:
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextCharFromCurrent(SUFFIX_IN_PROC_INSTR);
- if (isSpaceChar(c)) { // Ok, space to skip
- mTokenState = TOKEN_STARTED;
- // Need to skip the WS...
- skipWS(c);
- } else { // Nope; apparently finishes right away...
- mTokenState = TOKEN_FULL_COALESCED;
- mTextBuffer.resetWithEmpty();
- // or does it?
- if (c != '?' || getNextCharFromCurrent(SUFFIX_IN_PROC_INSTR) != '>') {
- throwUnexpectedChar(c, ErrorConsts.ERR_WF_PI_XML_MISSING_SPACE);
- }
- }
-
- return PROCESSING_INSTRUCTION;
- }
-
- /**
- * Method that parses a processing instruction's data portion; at this
- * point target has been parsed.
- */
- private void readPI()
- throws XMLStreamException
- {
- int ptr = mInputPtr;
- int start = ptr;
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
-
- outer_loop:
- while (ptr < inputLen) {
- char c = inputBuf[ptr++];
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(ptr);
- } else if (c == '\r') {
- if (ptr < inputLen && !mNormalizeLFs) {
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- markLF(ptr);
- } else {
- --ptr; // pushback
- break;
- }
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '?') {
- // K; now just need '>' after zero or more '?'s
- while (true) {
- if (ptr >= inputLen) {
- /* end of buffer; need to push back at least one of
- * question marks (not all, since just one is needed
- * to close the PI)
- */
- --ptr;
- break outer_loop;
- }
- c = inputBuf[ptr++];
- if (c == '>') {
- mInputPtr = ptr;
- // Need to discard trailing '?>'
- mTextBuffer.resetWithShared(inputBuf, start, ptr-start-2);
- return;
- }
- if (c != '?') {
- // Not end, can continue, but need to push back last char, in case it's LF/CR
- --ptr;
- break;
- }
- }
- }
- }
-
- mInputPtr = ptr;
- // No point in trying to share... let's just append
- mTextBuffer.resetWithCopy(inputBuf, start, ptr-start);
- readPI2(mTextBuffer);
- }
-
- private void readPI2(TextBuffer tb)
- throws XMLStreamException
- {
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
- int inputPtr = mInputPtr;
-
- /* Output pointers; calls will also ensure that the buffer is
- * not shared, AND has room for one more char
- */
- char[] outBuf = tb.getCurrentSegment();
- int outPtr = tb.getCurrentSegmentSize();
-
- main_loop:
- while (true) {
- // Let's first ensure we have some data in there...
- if (inputPtr >= inputLen) {
- loadMoreFromCurrent(SUFFIX_IN_PROC_INSTR);
- inputBuf = mInputBuffer;
- inputPtr = mInputPtr;
- inputLen = mInputEnd;
- }
-
- // And then do chunks
- char c = inputBuf[inputPtr++];
- if (c < CHAR_SPACE) {
- if (c == '\n') {
- markLF(inputPtr);
- } else if (c == '\r') {
- mInputPtr = inputPtr;
- if (skipCRLF(c)) { // got 2 char LF
- if (!mNormalizeLFs) {
- // Special handling, to output 2 chars at a time:
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = mTextBuffer.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- }
- // And let's let default output the 2nd char, either way
- c = '\n';
- } else if (mNormalizeLFs) { // just \r, but need to convert
- c = '\n'; // For Mac text
- }
- /* Since skipCRLF() needs to peek(), buffer may have
- * changed, even if there was no CR+LF.
- */
- inputPtr = mInputPtr;
- inputBuf = mInputBuffer;
- inputLen = mInputEnd;
- } else if (c != '\t') {
- throwInvalidSpace(c);
- }
- } else if (c == '?') { // Ok, just need '>' after zero or more '?'s
- mInputPtr = inputPtr; // to allow us to call getNextChar
-
- qmLoop:
- while (true) {
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_PROC_INSTR);
- if (c == '>') { // got it!
- break main_loop;
- } else if (c == '?') {
- if (outPtr >= outBuf.length) { // need more room?
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- outBuf[outPtr++] = c;
- } else {
- /* Hmmh. Wasn't end mark after all. Thus, need to
- * fall back to normal processing, with one more
- * question mark (first one matched that wasn't
- * yet output),
- * reset variables, and go back to main loop.
- */
- inputPtr = --mInputPtr; // push back last char
- inputBuf = mInputBuffer;
- inputLen = mInputEnd;
- c = '?';
- break qmLoop;
- }
- }
- } // if (c == '?)
-
- // Need more room?
- if (outPtr >= outBuf.length) {
- outBuf = tb.finishCurrentSegment();
- outPtr = 0;
- }
- // Ok, let's add char to output:
- outBuf[outPtr++] = c;
-
- } // while (true)
-
- tb.setCurrentLength(outPtr);
- }
-
- /**
- * Method called to read the content of both current CDATA/CHARACTERS
- * events, and all following consequtive events into the text buffer.
- * At this point the current type is known, prefix (for CDATA) skipped,
- * and initial consequtive contents (if any) read in.
- *
- * @param deferErrors Flag to enable storing an exception to a
- * variable, instead of immediately throwing it. If true, will
- * just store the exception; if false, will not store, just throw.
- */
- protected void readCoalescedText(int currType, boolean deferErrors)
- throws XMLStreamException
- {
- boolean wasCData;
-
- // Ok; so we may need to combine adjacent text/CDATA chunks.
- if (currType == CHARACTERS || currType == SPACE) {
- readTextSecondary(Integer.MAX_VALUE, deferErrors);
- wasCData = false;
- } else if (currType == CDATA) {
- /* We may have actually really finished it, but just left
- * the 'unfinished' flag due to need to coalesce...
- */
- if (mTokenState <= TOKEN_PARTIAL_SINGLE) {
- readCDataSecondary(Integer.MAX_VALUE);
- }
- wasCData = true;
- } else {
- throw new IllegalStateException("Internal error: unexpected token "+tokenTypeDesc(mCurrToken)+"; expected CHARACTERS, CDATA or SPACE.");
- }
-
- // But how about additional text?
- while (!deferErrors || (mPendingException == null)) {
- if (mInputPtr >= mInputEnd) {
- mTextBuffer.ensureNotShared();
- if (!loadMore()) {
- // ??? Likely an error but let's just break
- break;
- }
- }
- // Let's peek, ie. not advance it yet
- char c = mInputBuffer[mInputPtr];
- if (c == '<') { // CDATA, maybe?
- // Need to distinguish ") or until
- * first 'hole' in text (buffer end, 2-char lf to convert, entity).
- *mRawAttrs) of the first attribute
- * instance that was created through default value expansion.
- */
- private final int mDefaultOffset;
-
- /*
- //////////////////////////////////////////////////////////////
- // Information that defines "Map-like" data structure used for
- // quick access to attribute values by fully-qualified name
- // (only used for "long" lists)
- //////////////////////////////////////////////////////////////
- */
-
- // // // For full explanation, see source for AttributeCollector
-
- private final int[] mAttrMap;
-
- private final int mAttrHashSize;
-
- private final int mAttrSpillEnd;
-
- /**
- * Method called to create "short" attribute list; list that has
- * only few entries, and can thus be searched for attributes using
- * linear search, without using any kind of Map structure.
- *DTDValidatorBase or returns true for
- * reallyValidating.
- *ValidationContext
- */
- public void reportValidationProblem(XMLValidationProblem prob)
- throws XMLStreamException
- {
- // !!! TBI: Fail-fast vs. deferred modes?
- /* For now let's implement basic functionality: warnings get
- * reported via XMLReporter, errors and fatal errors result in
- * immediate exceptions.
- */
- /* 27-May-2008, TSa: [WSTX-153] Above is incorrect: as per Stax
- * javadocs for XMLReporter, both warnings and non-fatal errors
- * (which includes all validation errors) should be reported via
- * XMLReporter interface, and only fatals should cause an
- * immediate stream exception (by-passing reporter)
- */
- if (prob.getSeverity() > XMLValidationProblem.SEVERITY_ERROR) {
- throw WstxValidationException.create(prob);
- }
- XMLReporter rep = mConfig.getXMLReporter();
- if (rep != null) {
- _reportProblem(rep, prob);
- } else {
- /* If no reporter, regular non-fatal errors are to be reported
- * as exceptions as well, for backwards compatibility
- */
- if (prob.getSeverity() >= XMLValidationProblem.SEVERITY_ERROR) {
- throw WstxValidationException.create(prob);
- }
- }
- }
-
- public void reportValidationProblem(String msg, int severity)
- throws XMLStreamException
- {
- reportValidationProblem(new XMLValidationProblem(getLastCharLocation(),
- msg, severity));
- }
-
- public void reportValidationProblem(String msg)
- throws XMLStreamException
- {
- reportValidationProblem(new XMLValidationProblem(getLastCharLocation(),
- msg,
- XMLValidationProblem.SEVERITY_ERROR));
- }
-
- public void reportValidationProblem(Location loc, String msg)
- throws XMLStreamException
- {
- reportValidationProblem(new XMLValidationProblem(loc, msg));
- }
-
- public void reportValidationProblem(String format, Object arg, Object arg2)
- throws XMLStreamException
- {
- reportValidationProblem(MessageFormat.format(format, new Object[] { arg, arg2 }));
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Other error reporting methods
- ///////////////////////////////////////////////////////////
- */
-
- protected WstxException constructWfcException(String msg)
- {
- return new WstxParsingException(msg, getLastCharLocation());
- }
-
- /**
- * Construct and return a {@link XMLStreamException} to throw
- * as a result of a failed Typed Access operation (but one not
- * caused by a Well-Formedness Constraint or Validation Constraint
- * problem)
- */
- /*
- protected WstxException _constructTypeException(String msg)
- {
- // Hmmh. Should there be a distinct sub-type?
- return new WstxParsingException(msg, getLastCharLocation());
- }
- */
-
- protected WstxException constructFromIOE(IOException ioe)
- {
- return new WstxIOException(ioe);
- }
-
- protected WstxException constructNullCharException()
- {
- return new WstxUnexpectedCharException("Illegal character (NULL, unicode 0) encountered: not valid in any content",
- getLastCharLocation(), CHAR_NULL);
- }
-
- protected void throwUnexpectedChar(int i, String msg)
- throws WstxException
- {
- char c = (char) i;
- String excMsg = "Unexpected character "+getCharDesc(c)+msg;
- throw new WstxUnexpectedCharException(excMsg, getLastCharLocation(), c);
- }
-
- protected void throwNullChar()
- throws WstxException
- {
- throw constructNullCharException();
- }
-
- protected void throwInvalidSpace(int i)
- throws WstxException
- {
- throwInvalidSpace(i, false);
- }
-
- protected WstxException throwInvalidSpace(int i, boolean deferErrors)
- throws WstxException
- {
- char c = (char) i;
- WstxException ex;
- if (c == CHAR_NULL) {
- ex = constructNullCharException();
- } else {
- String msg = "Illegal character ("+getCharDesc(c)+")";
- if (mXml11) {
- msg += " [note: in XML 1.1, it could be included via entity expansion]";
- }
- ex = new WstxUnexpectedCharException(msg, getLastCharLocation(), c);
- }
- if (!deferErrors) {
- throw ex;
- }
- return ex;
- }
-
- protected void throwUnexpectedEOF(String msg)
- throws WstxException
- {
- throw new WstxEOFException("Unexpected EOF"
- +(msg == null ? "" : msg),
- getLastCharLocation());
- }
-
- /**
- * Similar to {@link #throwUnexpectedEOF}, but only indicates ending
- * of an input block. Used when reading a token that can not span
- * input block boundaries (ie. can not continue past end of an
- * entity expansion).
- */
- protected void throwUnexpectedEOB(String msg)
- throws WstxException
- {
- throw new WstxEOFException("Unexpected end of input block"
- +(msg == null ? "" : msg),
- getLastCharLocation());
- }
-
- protected void throwFromIOE(IOException ioe)
- throws WstxException
- {
- throw new WstxIOException(ioe);
- }
-
- protected void throwFromStrE(XMLStreamException strex)
- throws WstxException
- {
- if (strex instanceof WstxException) {
- throw (WstxException) strex;
- }
- WstxException newEx = new WstxException(strex);
- ExceptionUtil.setInitCause(newEx, strex);
- throw newEx;
- }
-
- /**
- * Method called to report an error, when caller's signature only
- * allows runtime exceptions to be thrown.
- */
- protected void throwLazyError(Exception e)
- {
- if (e instanceof XMLStreamException) {
- WstxLazyException.throwLazily((XMLStreamException) e);
- }
- ExceptionUtil.throwRuntimeException(e);
- }
-
- protected String tokenTypeDesc(int type)
- {
- return ErrorConsts.tokenTypeDesc(type);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Input buffer handling
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Returns current input source this source uses.
- *
- *
- * If so, character value of entity is returned. Character 0 is returned
- * otherwise; if so, caller needs to do full resolution.
- *
- * Also note that this method is not heavily optimized, as it's not
- * likely to be a bottleneck for parsing.
- */
- protected final String parseSystemId(char quoteChar, boolean convertLFs,
- String errorMsg)
- throws XMLStreamException
- {
- char[] buf = getNameBuffer(-1);
- int ptr = 0;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(errorMsg);
- if (c == quoteChar) {
- break;
- }
- /* ??? 14-Jun-2004, TSa: Should we normalize linefeeds or not?
- * It seems like we should, for all input... so that's the way it
- * works.
- */
- if (c == '\n') {
- markLF();
- } else if (c == '\r') {
- if (peekNext() == '\n') {
- ++mInputPtr;
- if (!convertLFs) {
- /* The only tricky thing; need to preserve 2-char LF; need to
- * output one char from here, then can fall back to default:
- */
- if (ptr >= buf.length) {
- buf = expandBy50Pct(buf);
- }
- buf[ptr++] = '\r';
- }
- c = '\n';
- } else if (convertLFs) {
- c = '\n';
- }
- }
-
- // Other than that, let's just append it:
- if (ptr >= buf.length) {
- buf = expandBy50Pct(buf);
- }
- buf[ptr++] = c;
- }
-
- return (ptr == 0) ? "" : new String(buf, 0, ptr);
- }
-
- /**
- * Simple parsing method that parses system ids, which are generally
- * used in entities (from DOCTYPE declaration to internal/external
- * subsets).
- *
- * Also note that this method is not heavily optimized, as it's not
- * likely to be a bottleneck for parsing.
- */
- protected final String parsePublicId(char quoteChar, String errorMsg)
- throws XMLStreamException
- {
- char[] buf = getNameBuffer(-1);
- int ptr = 0;
- boolean spaceToAdd = false;
-
- while (true) {
- char c = (mInputPtr < mInputEnd) ?
- mInputBuffer[mInputPtr++] : getNextChar(errorMsg);
- if (c == quoteChar) {
- break;
- }
- if (c == '\n') {
- markLF();
- spaceToAdd = true;
- continue;
- } else if (c == '\r') {
- if (peekNext() == '\n') {
- ++mInputPtr;
- }
- spaceToAdd = true;
- continue;
- } else if (c == CHAR_SPACE) {
- spaceToAdd = true;
- continue;
- } else {
- // Verify it's a legal pubid char (see XML spec, #13, from 2.3)
- if ((c >= VALID_PUBID_CHAR_COUNT)
- || sPubidValidity[c] != PUBID_CHAR_VALID_B) {
- throwUnexpectedChar(c, " in public identifier");
- }
- }
-
- // Other than that, let's just append it:
- if (ptr >= buf.length) {
- buf = expandBy50Pct(buf);
- }
- /* Space-normalization means scrapping leading and trailing
- * white space, and coalescing remaining ws into single spaces.
- */
- if (spaceToAdd) { // pending white space to add?
- if (c == CHAR_SPACE) { // still a space; let's skip
- continue;
- }
- /* ok: if we have non-space, we'll either forget about
- * space(s) (if nothing has been output, ie. leading space),
- * or output a single space (in-between non-white space)
- */
- spaceToAdd = false;
- if (ptr > 0) {
- buf[ptr++] = CHAR_SPACE;
- if (ptr >= buf.length) {
- buf = expandBy50Pct(buf);
- }
- }
- }
- buf[ptr++] = c;
- }
-
- return (ptr == 0) ? "" : new String(buf, 0, ptr);
- }
-
- protected final void parseUntil(TextBuffer tb, char endChar, boolean convertLFs,
- String errorMsg)
- throws XMLStreamException
- {
- // Let's first ensure we have some data in there...
- if (mInputPtr >= mInputEnd) {
- loadMore(errorMsg);
- }
- while (true) {
- // Let's loop consequtive 'easy' spans:
- char[] inputBuf = mInputBuffer;
- int inputLen = mInputEnd;
- int ptr = mInputPtr;
- int startPtr = ptr;
- while (ptr < inputLen) {
- char c = inputBuf[ptr++];
- if (c == endChar) {
- int thisLen = ptr - startPtr - 1;
- if (thisLen > 0) {
- tb.append(inputBuf, startPtr, thisLen);
- }
- mInputPtr = ptr;
- return;
- }
- if (c == '\n') {
- mInputPtr = ptr; // markLF() requires this
- markLF();
- } else if (c == '\r') {
- if (!convertLFs && ptr < inputLen) {
- if (inputBuf[ptr] == '\n') {
- ++ptr;
- }
- mInputPtr = ptr;
- markLF();
- } else {
- int thisLen = ptr - startPtr - 1;
- if (thisLen > 0) {
- tb.append(inputBuf, startPtr, thisLen);
- }
- mInputPtr = ptr;
- c = getNextChar(errorMsg);
- if (c != '\n') {
- --mInputPtr; // pusback
- tb.append(convertLFs ? '\n' : '\r');
- } else {
- if (convertLFs) {
- tb.append('\n');
- } else {
- tb.append('\r');
- tb.append('\n');
- }
- }
- startPtr = ptr = mInputPtr;
- markLF();
- }
- }
- }
- int thisLen = ptr - startPtr;
- if (thisLen > 0) {
- tb.append(inputBuf, startPtr, thisLen);
- }
- loadMore(errorMsg);
- startPtr = ptr = mInputPtr;
- inputBuf = mInputBuffer;
- inputLen = mInputEnd;
- }
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////
- */
-
- private int resolveCharEnt(StringBuffer originalCharacters)
- throws XMLStreamException
- {
- int value = 0;
- char c = getNextChar(SUFFIX_IN_ENTITY_REF);
-
- if (originalCharacters != null) {
- originalCharacters.append(c);
- }
-
- if (c == 'x') { // hex
- while (true) {
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_ENTITY_REF);
- if (c == ';') {
- break;
- }
-
- if (originalCharacters != null) {
- originalCharacters.append(c);
- }
- value = value << 4;
- if (c <= '9' && c >= '0') {
- value += (c - '0');
- } else if (c >= 'a' && c <= 'f') {
- value += 10 + (c - 'a');
- } else if (c >= 'A' && c <= 'F') {
- value += 10 + (c - 'A');
- } else {
- throwUnexpectedChar(c, "; expected a hex digit (0-9a-fA-F).");
- }
- // Overflow?
- if (value > MAX_UNICODE_CHAR) {
- reportUnicodeOverflow();
- }
- }
- } else { // numeric (decimal)
- while (c != ';') {
- if (c <= '9' && c >= '0') {
- value = (value * 10) + (c - '0');
- // Overflow?
- if (value > MAX_UNICODE_CHAR) {
- reportUnicodeOverflow();
- }
- } else {
- throwUnexpectedChar(c, "; expected a decimal number.");
- }
- c = (mInputPtr < mInputEnd) ? mInputBuffer[mInputPtr++]
- : getNextCharFromCurrent(SUFFIX_IN_ENTITY_REF);
-
- if (originalCharacters != null && c != ';') {
- originalCharacters.append(c);
- }
- }
- }
- validateChar(value);
- return value;
- }
-
- /**
- * Method that will verify that expanded Unicode codepoint is a valid
- * XML content character.
- */
- private final void validateChar(int value)
- throws XMLStreamException
- {
- /* 24-Jan-2006, TSa: Ok, "high" Unicode chars are problematic,
- * need to be reported by a surrogate pair..
- */
- if (value >= 0xD800) {
- if (value < 0xE000) { // no surrogates via entity expansion
- reportIllegalChar(value);
- }
- if (value > 0xFFFF) {
- // Within valid range at all?
- if (value > MAX_UNICODE_CHAR) {
- reportUnicodeOverflow();
- }
- } else if (value >= 0xFFFE) { // 0xFFFE and 0xFFFF are illegal too
- reportIllegalChar(value);
- }
- // Ok, fine as is
- } else if (value < 32) {
- if (value == 0) {
- throwParseError("Invalid character reference: null character not allowed in XML content.");
- }
- // XML 1.1 allows most other chars; 1.0 does not:
- if (!mXml11 &&
- (value != 0x9 && value != 0xA && value != 0xD)) {
- reportIllegalChar(value);
- }
- }
- }
-
- protected final char[] getNameBuffer(int minSize)
- {
- char[] buf = mNameBuffer;
-
- if (buf == null) {
- mNameBuffer = buf = new char[(minSize > 48) ? (minSize+16) : 64];
- } else if (minSize >= buf.length) { // let's allow one char extra...
- int len = buf.length;
- len += (len >> 1); // grow by 50%
- mNameBuffer = buf = new char[(minSize >= len) ? (minSize+16) : len];
- }
- return buf;
- }
-
- protected final char[] expandBy50Pct(char[] buf)
- {
- int len = buf.length;
- char[] newBuf = new char[len + (len >> 1)];
- System.arraycopy(buf, 0, newBuf, 0, len);
- return newBuf;
- }
-
- /**
- * Method called to throw an exception indicating that a name that
- * should not be namespace-qualified (PI target, entity/notation name)
- * is one, and reader is namespace aware.
- */
- private void throwNsColonException(String name)
- throws XMLStreamException
- {
- throwParseError("Illegal name \"{0}\" (PI target, entity/notation name): can not contain a colon (XML Namespaces 1.0#6)", name, null);
- }
-
- private void throwRecursionError(String entityName)
- throws XMLStreamException
- {
- throwParseError("Illegal entity expansion: entity \"{0}\" expands itself recursively.", entityName, null);
- }
-
- private void reportUnicodeOverflow()
- throws XMLStreamException
- {
- throwParseError("Illegal character entity: value higher than max allowed (0x{0})", Integer.toHexString(MAX_UNICODE_CHAR), null);
- }
-
- private void reportIllegalChar(int value)
- throws XMLStreamException
- {
- throwParseError("Illegal character entity: expansion character (code 0x{0}", Integer.toHexString(value), null);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/TypedStreamReader.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/TypedStreamReader.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sr/TypedStreamReader.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sr/TypedStreamReader.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,754 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sr;
-
-import java.math.BigDecimal;
-import java.math.BigInteger;
-
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.typed.Base64Variant;
-import org.codehaus.stax2.typed.Base64Variants;
-import org.codehaus.stax2.typed.TypedArrayDecoder;
-import org.codehaus.stax2.typed.TypedValueDecoder;
-import org.codehaus.stax2.typed.TypedXMLStreamException;
-
-import org.codehaus.stax2.ri.Stax2Util;
-import org.codehaus.stax2.ri.typed.ValueDecoderFactory;
-import org.codehaus.stax2.ri.typed.CharArrayBase64Decoder;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.io.BranchingReaderSource;
-import com.ctc.wstx.io.InputBootstrapper;
-import com.ctc.wstx.io.WstxInputData;
-
-/**
- * Complete implementation of {@link org.codehaus.stax2.XMLStreamReader2},
- * including Typed Access API (Stax2 v3.0) implementation.
- * Only functionality missing is DTD validation, which is provided by a
- * specialized sub-class.
- */
-public class TypedStreamReader
- extends BasicStreamReader
-{
- /**
- * Mask of event types that are legal (starting) states
- * to call Typed Access API from.
- *
- */
- final protected static int MASK_TYPED_ACCESS_ARRAY =
- (1 << START_ELEMENT)
- | (1 << END_ELEMENT) // for convenience
- | (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE)
- // Not ok for PI or COMMENT? Let's assume so
- ;
-
- final protected static int MASK_TYPED_ACCESS_BINARY =
- (1 << START_ELEMENT) // note: END_ELEMENT handled separately
- | (1 << CHARACTERS) | (1 << CDATA) | (1 << SPACE)
- ;
-
- /**
- * Minimum length of text chunks to parse before base64 decoding.
- * Will try to limit it to fit within regular result buffers.
- */
- final static int MIN_BINARY_CHUNK = 2000;
-
- /**
- * Factory used for constructing decoders we need for typed access
- */
- protected ValueDecoderFactory _decoderFactory;
-
- /**
- * Lazily-constructed decoder object for decoding base64 encoded
- * element binary content.
- */
- protected CharArrayBase64Decoder _base64Decoder = null;
-
- /*
- ////////////////////////////////////////////////////
- // Instance construction
- ////////////////////////////////////////////////////
- */
-
- protected TypedStreamReader(InputBootstrapper bs,
- BranchingReaderSource input, ReaderCreator owner,
- ReaderConfig cfg, InputElementStack elemStack,
- boolean forER)
- throws XMLStreamException
- {
- super(bs, input, owner, cfg, elemStack, forER);
- }
-
- /**
- * Factory method for constructing readers.
- *
- * @param owner "Owner" of this reader, factory that created the reader;
- * needed for returning updated symbol table information after parsing.
- * @param input Input source used to read the XML document.
- * @param cfg Object that contains reader configuration info.
- */
- public static TypedStreamReader createStreamReader
- (BranchingReaderSource input, ReaderCreator owner, ReaderConfig cfg,
- InputBootstrapper bs, boolean forER)
- throws XMLStreamException
- {
-
- TypedStreamReader sr = new TypedStreamReader
- (bs, input, owner, cfg, createElementStack(cfg), forER);
- return sr;
- }
-
-
- /*
- ////////////////////////////////////////////////////////
- // TypedXMLStreamReader2 implementation, scalar elements
- ////////////////////////////////////////////////////////
- */
-
- public boolean getElementAsBoolean() throws XMLStreamException
- {
- ValueDecoderFactory.BooleanDecoder dec = _decoderFactory().getBooleanDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public int getElementAsInt() throws XMLStreamException
- {
- ValueDecoderFactory.IntDecoder dec = _decoderFactory().getIntDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public long getElementAsLong() throws XMLStreamException
- {
- ValueDecoderFactory.LongDecoder dec = _decoderFactory().getLongDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public float getElementAsFloat() throws XMLStreamException
- {
- ValueDecoderFactory.FloatDecoder dec = _decoderFactory().getFloatDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public double getElementAsDouble() throws XMLStreamException
- {
- ValueDecoderFactory.DoubleDecoder dec = _decoderFactory().getDoubleDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public BigInteger getElementAsInteger() throws XMLStreamException
- {
- ValueDecoderFactory.IntegerDecoder dec = _decoderFactory().getIntegerDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public BigDecimal getElementAsDecimal() throws XMLStreamException
- {
- ValueDecoderFactory.DecimalDecoder dec = _decoderFactory().getDecimalDecoder();
- getElementAs(dec);
- return dec.getValue();
- }
-
- public QName getElementAsQName() throws XMLStreamException
- {
- ValueDecoderFactory.QNameDecoder dec = _decoderFactory().getQNameDecoder(getNamespaceContext());
- getElementAs(dec);
- return _verifyQName(dec.getValue());
- }
-
- public final byte[] getElementAsBinary() throws XMLStreamException
- {
- return getElementAsBinary(Base64Variants.getDefaultVariant());
- }
-
- public byte[] getElementAsBinary(Base64Variant v) throws XMLStreamException
- {
- // note: code here is similar to Base64DecoderBase.aggregateAll(), see comments there
- Stax2Util.ByteAggregator aggr = _base64Decoder().getByteAggregator();
- byte[] buffer = aggr.startAggregation();
- while (true) {
- int offset = 0;
- int len = buffer.length;
-
- do {
- int readCount = readElementAsBinary(buffer, offset, len, v);
- if (readCount < 1) { // all done!
- return aggr.aggregateAll(buffer, offset);
- }
- offset += readCount;
- len -= readCount;
- } while (len > 0);
- buffer = aggr.addFullBlock(buffer);
- }
- }
-
- public void getElementAs(TypedValueDecoder tvd) throws XMLStreamException
- {
- if (mCurrToken != START_ELEMENT) {
- throwParseError(ErrorConsts.ERR_STATE_NOT_STELEM);
- }
- /* Ok, now: with START_ELEMENT we know that it's not partially
- * processed; that we are in-tree (not prolog or epilog).
- * The only possible complication would be:
- */
- if (mStEmptyElem) {
- /* And if so, we'll then get 'virtual' close tag; things
- * are simple as location info was set when dealing with
- * empty start element; and likewise, validation (if any)
- * has been taken care of
- */
- mStEmptyElem = false;
- mCurrToken = END_ELEMENT;
- _handleEmptyValue(tvd);
- return;
- }
- // First need to find a textual event
- while (true) {
- int type = next();
- if (type == END_ELEMENT) {
- _handleEmptyValue(tvd);
- return;
- }
- if (type == COMMENT || type == PROCESSING_INSTRUCTION) {
- continue;
- }
- if (((1 << type) & MASK_GET_ELEMENT_TEXT) == 0) {
- throwParseError("Expected a text token, got "+tokenTypeDesc(type)+".");
- }
- break;
- }
- if (mTokenState < TOKEN_FULL_SINGLE) {
- readCoalescedText(mCurrToken, false);
- }
- /* Ok: then a quick check; if it looks like we are directly
- * followed by the end tag, we need not construct String
- * quite yet.
- */
- if ((mInputPtr + 1) < mInputEnd &&
- mInputBuffer[mInputPtr] == '<' && mInputBuffer[mInputPtr+1] == '/') {
- // Note: next() has validated text, no need for more validation
- mInputPtr += 2;
- mCurrToken = END_ELEMENT;
- /* Can by-pass next(), nextFromTree(), in this case.
- * However, must do decoding first, and only then call
- * readEndElem(), since this latter call may invalidate
- * underlying input buffer (when end tag is at buffer
- * boundary)
- */
- try { // buffer now has all the data
- mTextBuffer.decode(tvd);
- } catch (IllegalArgumentException iae) {
- throw _constructTypeException(iae, mTextBuffer.contentsAsString());
- }
- readEndElem();
- return;
- }
-
- // Otherwise, we'll need to do slower processing
- int extra = 1 + (mTextBuffer.size() >> 1); // let's add 50% space
- StringBuffer sb = mTextBuffer.contentsAsStringBuffer(extra);
- int type;
-
- while ((type = next()) != END_ELEMENT) {
- if (((1 << type) & MASK_GET_ELEMENT_TEXT) != 0) {
- if (mTokenState < TOKEN_FULL_SINGLE) {
- readCoalescedText(type, false);
- }
- mTextBuffer.contentsToStringBuffer(sb);
- continue;
- }
- if (type != COMMENT && type != PROCESSING_INSTRUCTION) {
- throwParseError("Expected a text token, got "+tokenTypeDesc(type)+".");
- }
- }
- // Note: calls next() have validated text, no need for more validation
- String str = sb.toString();
- String tstr = Stax2Util.trimSpaces(str);
- if (tstr == null) {
- _handleEmptyValue(tvd);
- } else {
- try {
- tvd.decode(tstr);
- } catch (IllegalArgumentException iae) {
- throw _constructTypeException(iae, str);
- }
- }
- }
-
- /*
- ////////////////////////////////////////////////////////
- // TypedXMLStreamReader2 implementation, array elements
- ////////////////////////////////////////////////////////
- */
-
- public int readElementAsIntArray(int[] value, int from, int length) throws XMLStreamException
- {
- return readElementAsArray(_decoderFactory().getIntArrayDecoder(value, from, length));
- }
-
- public int readElementAsLongArray(long[] value, int from, int length) throws XMLStreamException
- {
- return readElementAsArray(_decoderFactory().getLongArrayDecoder(value, from, length));
- }
-
- public int readElementAsFloatArray(float[] value, int from, int length) throws XMLStreamException
- {
- return readElementAsArray(_decoderFactory().getFloatArrayDecoder(value, from, length));
- }
-
- public int readElementAsDoubleArray(double[] value, int from, int length) throws XMLStreamException
- {
- return readElementAsArray(_decoderFactory().getDoubleArrayDecoder(value, from, length));
- }
-
- /**
- * Method called to parse array of primitives.
- *finishDTD, to locate the specified
- * external DTD subset. Subset may be obtained from a cache, if cached
- * copy exists and is compatible; if not, it will be read from the
- * source identified by the public and/or system identifier passed.
- */
- private DTDSubset findDtdExtSubset(String pubId, String sysId,
- DTDSubset intSubset)
- throws XMLStreamException
- {
- boolean cache = hasConfigFlags(CFG_CACHE_DTDS);
- DTDId dtdId;
- try {
- dtdId = constructDtdId(pubId, sysId);
- } catch (IOException ioe) {
- throw constructFromIOE(ioe);
- }
-
- if (cache) {
- DTDSubset extSubset = findCachedSubset(dtdId, intSubset);
- if (extSubset != null) {
- return extSubset;
- }
- }
-
- // No useful cached copy? Need to read it then.
- /* For now, we do require system identifier; otherwise we don't
- * know how to resolve DTDs by public id. In future should
- * probably also have some simple catalog resolving facility?
- */
- if (sysId == null) {
- throwParseError("Can not resolve DTD with public id \"{0}\"; missing system identifier", mDtdPublicId, null);
- }
- WstxInputSource src = null;
-
- try {
- int xmlVersion = mDocXmlVersion;
- // 05-Feb-2006, TSa: If xmlVersion not explicitly known, defaults to 1.0
- if (xmlVersion == XmlConsts.XML_V_UNKNOWN) {
- xmlVersion = XmlConsts.XML_V_10;
- }
- /* null -> no explicit path context, use parent's
- * null -> not an entity expansion, no name.
- * Note, too, that we can NOT just pass mEntityResolver, since
- * that's the one used for general entities, whereas ext subset
- * should be resolved by the param entity resolver.
- */
- src = DefaultInputResolver.resolveEntity
- (mInput, null, null, pubId, sysId, mConfig.getDtdResolver(),
- mConfig, xmlVersion);
- } catch (FileNotFoundException fex) {
- /* Let's catch and rethrow this just so we get more meaningful
- * description (with input source position etc)
- */
- throwParseError("(was {0}) {1}", fex.getClass().getName(), fex.getMessage());
- } catch (IOException ioe) {
- throwFromIOE(ioe);
- }
-
- DTDSubset extSubset = FullDTDReader.readExternalSubset(src, mConfig, intSubset,
- hasConfigFlags(CFG_VALIDATE_AGAINST_DTD),
- mDocXmlVersion);
-
- if (cache) {
- /* Ok; can be cached, but only if it does NOT refer to
- * parameter entities defined in the internal subset (if
- * it does, there's no easy/efficient to check if it could
- * be used later on, plus it's unlikely it could be)
- */
- if (extSubset.isCachable()) {
- mOwner.addCachedDTD(dtdId, extSubset);
- }
- }
-
- return extSubset;
- }
-
- private DTDSubset findCachedSubset(DTDId id, DTDSubset intSubset)
- throws XMLStreamException
- {
- DTDSubset extSubset = mOwner.findCachedDTD(id);
- /* Ok, now; can use the cached copy iff it does not refer to
- * any parameter entities internal subset (if one exists)
- * defines:
- */
- if (extSubset != null) {
- if (intSubset == null || extSubset.isReusableWith(intSubset)) {
- return extSubset;
- }
- }
- return null;
- }
-
- /**
- * Method called to resolve path to external DTD subset, given
- * system identifier.
- */
- private URI resolveExtSubsetPath(String systemId) throws IOException
- {
- // Do we have a context to use for resolving?
- URL ctxt = (mInput == null) ? null : mInput.getSource();
-
- /* Ok, either got a context or not; let's create the URL based on
- * the id, and optional context:
- */
- if (ctxt == null) {
- /* Call will try to figure out if system id has the protocol
- * in it; if not, create a relative file, if it does, try to
- * resolve it.
- */
- return URLUtil.uriFromSystemId(systemId);
- }
- URL url = URLUtil.urlFromSystemId(systemId, ctxt);
- try {
- return new URI(url.toExternalForm());
- } catch (URISyntaxException e) { // should never occur...
- throw new IOException("Failed to construct URI for external subset, URL = "+url.toExternalForm()+": "+e.getMessage());
- }
- }
-
- protected DTDId constructDtdId(String pubId, String sysId)
- throws IOException
- {
- /* Following settings will change what gets stored as DTD, so
- * they need to separate cached instances too:
- */
- int significantFlags = mConfigFlags &
- (CFG_NAMESPACE_AWARE
- /* Let's optimize non-validating case; DTD info we need
- * is less if so (no need to store content specs for one)...
- * plus, eventual functionality may be different too.
- */
- | CFG_VALIDATE_AGAINST_DTD
- /* Also, whether we support dtd++ or not may change construction
- * of settings... (currently does not, but could)
- */
- | CFG_SUPPORT_DTDPP
- /* Also, basic xml:id support does matter -- xml:id attribute
- * type is verified only if it's enabled
- */
- | CFG_XMLID_TYPING
- );
- URI sysRef = (sysId == null || sysId.length() == 0) ? null :
- resolveExtSubsetPath(sysId);
-
- /* 29-Mar-2006, TSa: Apparently public ids are not always very
- * unique and/or can be mismatched with system ids, resulting
- * in false matches if using public ids. As a result, by default
- * Woodstox does NOT rely on public ids, when matching.
- */
- boolean usePublicId = (mConfigFlags & CFG_CACHE_DTDS_BY_PUBLIC_ID) != 0;
- if (usePublicId && pubId != null && pubId.length() > 0) {
- return DTDId.construct(pubId, sysRef, significantFlags, mXml11);
- }
- if (sysRef == null) {
- return null;
- }
- return DTDId.constructFromSystemId(sysRef, significantFlags, mXml11);
- }
-
- protected DTDId constructDtdId(URI sysId)
- throws IOException
- {
- int significantFlags = mConfigFlags &
- (CFG_NAMESPACE_AWARE
- /* Let's optimize non-validating case; DTD info we need
- * is less if so (no need to store content specs for one)
- */
- | CFG_VALIDATE_AGAINST_DTD
- /* Also, whether we support dtd++ or not may change construction
- * of settings... (currently does not, but could)
- */
- | CFG_SUPPORT_DTDPP
- );
- return DTDId.constructFromSystemId(sysId, significantFlags, mXml11);
- }
-
- /*
- ///////////////////////////////////////////////////////////////////////
- // Private methods, DTD validation support
- ///////////////////////////////////////////////////////////////////////
- */
-
- /**
- * Method called by lower-level parsing code when invalid content
- * (anything inside element with 'empty' content spec; text inside
- * non-mixed element etc) is found during basic scanning. Note
- * that actual DTD element structure problems are not reported
- * through this method.
- */
- protected void reportInvalidContent(int evtType)
- throws XMLStreamException
- {
- switch (mVldContent) {
- case XMLValidator.CONTENT_ALLOW_NONE:
- reportValidationProblem(ErrorConsts.ERR_VLD_EMPTY,
- mElementStack.getTopElementDesc(),
- ErrorConsts.tokenTypeDesc(evtType));
- break;
- case XMLValidator.CONTENT_ALLOW_WS:
- case XMLValidator.CONTENT_ALLOW_WS_NONSTRICT: // should this ever occur?
- reportValidationProblem(ErrorConsts.ERR_VLD_NON_MIXED,
- mElementStack.getTopElementDesc(), null);
- break;
- case XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT:
- case XMLValidator.CONTENT_ALLOW_ANY_TEXT:
- /* Not 100% sure if this should ever happen... depends on
- * interpretation of 'any' content model?
- */
- reportValidationProblem(ErrorConsts.ERR_VLD_ANY,
- mElementStack.getTopElementDesc(),
- ErrorConsts.tokenTypeDesc(evtType));
- break;
- default: // should never occur:
- throwParseError("Internal error: trying to report invalid content for "+evtType);
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/stax/package.html libwoodstox-java-5.1.0/src/java/com/ctc/wstx/stax/package.html
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/stax/package.html 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/stax/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,3 +0,0 @@
-makeChild method
- */
- SymbolTable mSymbols = mRootSymbols;
-
- /*
- ///////////////////////////////////////////////////////////
- // Life-cycle:
- ///////////////////////////////////////////////////////////
- */
-
- public WstxInputFactory() {
- mConfig = ReaderConfig.createFullDefaults();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // ReaderCreator implementation
- ///////////////////////////////////////////////////////////
- */
-
- // // // Configuration access methods:
-
- /**
- * Method readers created by this factory call, if DTD caching is
- * enabled, to see if an external DTD (subset) has been parsed
- * and cached earlier.
- */
- public synchronized DTDSubset findCachedDTD(DTDId id)
- {
- return (mDTDCache == null) ?
- null : (DTDSubset) mDTDCache.find(id);
- }
-
- // // // Callbacks for updating shared information
-
- /**
- * Method individual parsers call to pass back symbol table that
- * they updated, which may be useful for other parser to reuse, instead
- * of previous base symbol table.
- *close()
- * is called. Will be true for input sources that are automatically
- * managed by stream reader (input streams created for
- * {@link java.net.URL} and {@link java.io.File} arguments, or when
- * configuration settings indicate auto-closing is to be enabled
- * (the default value is false as per Stax 1.0 specs).
- */
- private XMLStreamReader2 doCreateSR(ReaderConfig cfg, String systemId, InputBootstrapper bs,
- URL src, boolean forER,
- boolean autoCloseInput)
- throws XMLStreamException
- {
- /* Automatic closing of input: will happen always for some input
- * types (ones application has no direct access to; but can also
- * be explicitly enabled.
- */
- if (!autoCloseInput) {
- autoCloseInput = cfg.willAutoCloseInput();
- }
-
- Reader r;
- try {
- r = bs.bootstrapInput(cfg, true, XmlConsts.XML_V_UNKNOWN);
- if (bs.declaredXml11()) {
- cfg.enableXml11(true);
- }
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
-
- /* null -> no public id available
- * false -> don't close the reader when scope is closed.
- */
- BranchingReaderSource input = InputSourceFactory.constructDocumentSource
- (cfg, bs, null, systemId, src, r, autoCloseInput);
-
- return ValidatingStreamReader.createValidatingStreamReader(input, this, cfg, bs, forER);
- }
-
- /**
- * Method that is eventually called to create a (full) stream read
- * instance.
- *close()
- * is called. Will be true for input sources that are automatically
- * managed by stream reader (input streams created for
- * {@link java.net.URL} and {@link java.io.File} arguments, or when
- * configuration settings indicate auto-closing is to be enabled
- * (the default value is false as per Stax 1.0 specs).
- */
- public XMLStreamReader2 createSR(ReaderConfig cfg, String systemId, InputBootstrapper bs,
- boolean forER,
- boolean autoCloseInput)
- throws XMLStreamException
- {
- // 16-Aug-2004, TSa: Maybe we have a context?
- URL src = cfg.getBaseURL();
-
- // If not, maybe we can derive it from system id?
- if ((src == null) && (systemId != null && systemId.length() > 0)) {
- try {
- src = URLUtil.urlFromSystemId(systemId);
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
- }
- return doCreateSR(cfg, systemId, bs, src, forER, autoCloseInput);
- }
-
- protected XMLStreamReader2 createSR(String systemId, InputStream in, String enc,
- boolean forER,
- boolean autoCloseInput)
- throws XMLStreamException
- {
- // sanity check:
- if (in == null) {
- throw new IllegalArgumentException("Null InputStream is not a valid argument");
- }
-
- ReaderConfig cfg = createPrivateConfig();
- if (enc == null || enc.length() == 0) {
- return createSR(cfg, systemId, StreamBootstrapper.getInstance
- (null, systemId, in), forER, autoCloseInput);
- }
-
- /* !!! 17-Feb-2006, TSa: We don't yet know if it's xml 1.0 or 1.1;
- * so have to specify 1.0 (which is less restrictive WRT input
- * streams). Would be better to let bootstrapper deal with it
- * though:
- */
- Reader r = DefaultInputResolver.constructOptimizedReader(cfg, in, false, enc);
- return createSR(cfg, systemId, ReaderBootstrapper.getInstance
- (null, systemId, r, enc), forER, autoCloseInput);
- }
-
- protected XMLStreamReader2 createSR(ReaderConfig cfg, URL src, boolean forER, boolean autoCloseInput)
- throws XMLStreamException
- {
- try {
- return createSR(cfg, src, URLUtil.inputStreamFromURL(src),
- forER, autoCloseInput);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- private XMLStreamReader2 createSR(ReaderConfig cfg, URL src, InputStream in,
- boolean forER,
- boolean autoCloseInput)
- throws XMLStreamException
- {
- String systemId = src.toExternalForm();
- return doCreateSR(cfg, systemId,
- StreamBootstrapper.getInstance(null, systemId, in),
- src, forER, autoCloseInput);
- }
-
- protected XMLStreamReader2 createSR(String systemId, Reader r,
- boolean forER,
- boolean autoCloseInput)
- throws XMLStreamException
- {
- return createSR(createPrivateConfig(), systemId,
- ReaderBootstrapper.getInstance
- (null, systemId, r, null), forER, autoCloseInput);
- }
-
- protected XMLStreamReader2 createSR(File f, boolean forER, boolean autoCloseInput)
- throws XMLStreamException
- {
- ReaderConfig cfg = createPrivateConfig();
- try {
- /* 18-Nov-2008, TSa: If P_BASE_URL is set, and File reference is
- * relative, let's resolve against base...
- */
- if (!f.isAbsolute()) {
- URL base = cfg.getBaseURL();
- if (base != null) {
- URL src = new URL(base, f.getPath());
- return createSR(cfg, src, URLUtil.inputStreamFromURL(src), forER, autoCloseInput);
- }
- }
- return createSR(cfg, f.toURL(), new FileInputStream(f), forER, autoCloseInput);
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
- }
-
- /**
- * Another internal factory method, used when dealing with a generic
- * Source base type. One thing worth noting is that 'auto-closing'
- * will be enabled if the input source or Reader is constructed (and
- * thus owned) by Woodstox.
- *
- * @param forER True, if the reader is being constructed to be used
- * by an event reader; false if it is not (or the purpose is not known)
- */
- protected XMLStreamReader2 createSR(javax.xml.transform.Source src,
- boolean forER)
- throws XMLStreamException
- {
- ReaderConfig cfg = createPrivateConfig();
- Reader r = null;
- InputStream in = null;
- String pubId = null;
- String sysId = null;
- String encoding = null;
- boolean autoCloseInput;
-
- InputBootstrapper bs = null;
-
- if (src instanceof Stax2Source) {
- Stax2Source ss = (Stax2Source) src;
- sysId = ss.getSystemId();
- pubId = ss.getPublicId();
- encoding = ss.getEncoding();
-
- try {
- /* 11-Nov-2008, TSa: Let's add optimized handling for byte-block
- * source
- */
- if (src instanceof Stax2ByteArraySource) {
- Stax2ByteArraySource bas = (Stax2ByteArraySource) src;
- bs = StreamBootstrapper.getInstance(pubId, sysId, bas.getBuffer(), bas.getBufferStart(), bas.getBufferEnd());
- } else {
- in = ss.constructInputStream();
- if (in == null) {
- r = ss.constructReader();
- }
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- /* Caller has no direct access to stream/reader, Woodstox
- * owns it and thus has to close too
- */
- autoCloseInput = true;
- } else if (src instanceof StreamSource) {
- StreamSource ss = (StreamSource) src;
- sysId = ss.getSystemId();
- pubId = ss.getPublicId();
- in = ss.getInputStream();
- if (in == null) {
- r = ss.getReader();
- }
- /* Caller still has access to stream/reader; no need to
- * force auto-close-input
- */
- autoCloseInput = cfg.willAutoCloseInput();
- } else if (src instanceof SAXSource) {
- SAXSource ss = (SAXSource) src;
- /* 28-Jan-2006, TSa: Not a complete implementation, but maybe
- * even this might help...
- */
- sysId = ss.getSystemId();
- InputSource isrc = ss.getInputSource();
- if (isrc != null) {
- encoding = isrc.getEncoding();
- in = isrc.getByteStream();
- if (in == null) {
- r = isrc.getCharacterStream();
- }
- }
- /* Caller still has access to stream/reader; no need to
- * force auto-close-input
- */
- autoCloseInput = cfg.willAutoCloseInput();
- } else if (src instanceof DOMSource) {
- DOMSource domSrc = (DOMSource) src;
- // SymbolTable not used by the DOM-based 'reader':
- return WstxDOMWrappingReader.createFrom(domSrc, cfg);
- } else {
- throw new IllegalArgumentException("Can not instantiate Stax reader for XML source type "+src.getClass()+" (unrecognized type)");
- }
- if (bs == null) { // may have already created boostrapper...
- if (r != null) {
- bs = ReaderBootstrapper.getInstance(pubId, sysId, r, encoding);
- } else if (in != null) {
- bs = StreamBootstrapper.getInstance(pubId, sysId, in);
- } else if (sysId != null && sysId.length() > 0) {
- /* 26-Dec-2008, TSa: If we must construct URL from system id,
- * it means caller will not have access to resulting
- * stream, thus we will force auto-closing.
- */
- autoCloseInput = true;
- try {
- return createSR(cfg, URLUtil.urlFromSystemId(sysId),
- forER, autoCloseInput);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- } else {
- throw new XMLStreamException("Can not create Stax reader for the Source passed -- neither reader, input stream nor system id was accessible; can not use other types of sources (like embedded SAX streams)");
- }
- }
- return createSR(cfg, sysId, bs, forER, autoCloseInput);
- }
-
- protected XMLEventAllocator createEventAllocator()
- {
- // Explicitly set allocate?
- if (mAllocator != null) {
- return mAllocator.newInstance();
- }
-
- /* Complete or fast one? Note: standard allocator is designed
- * in such a way that newInstance() need not be called (calling
- * it wouldn't do anything, anyway)
- */
- return mConfig.willPreserveLocation() ?
- DefaultEventAllocator.getDefaultInstance()
- : DefaultEventAllocator.getFastInstance();
- }
-
- /**
- * Method called to construct a copy of the factory's configuration
- * object, such that two will be unlinked (changes to one are not
- * reflect in the other).
- *
- *
- */
-public class WstxOutputFactory
- extends XMLOutputFactory2
- implements OutputConfigFlags
-{
- /*
- ///////////////////////////////////////////////////////////
- // Actual storage of configuration settings
- ///////////////////////////////////////////////////////////
- */
-
- protected final WriterConfig mConfig;
-
- /*
- ///////////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////////
- */
-
- public WstxOutputFactory() {
- mConfig = WriterConfig.createFullDefaults();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLOutputFactory API
- ///////////////////////////////////////////////////////////
- */
-
- public XMLEventWriter createXMLEventWriter(OutputStream out)
- throws XMLStreamException
- {
- return createXMLEventWriter(out, null);
- }
-
- public XMLEventWriter createXMLEventWriter(OutputStream out, String enc)
- throws XMLStreamException
- {
- if (out == null) {
- throw new IllegalArgumentException("Null OutputStream is not a valid argument");
- }
- return new Stax2EventWriterImpl(createSW(out, null, enc, false));
- }
-
- public XMLEventWriter createXMLEventWriter(javax.xml.transform.Result result)
- throws XMLStreamException
- {
- return new Stax2EventWriterImpl(createSW(result));
- }
-
- public XMLEventWriter createXMLEventWriter(Writer w)
- throws XMLStreamException
- {
- if (w == null) {
- throw new IllegalArgumentException("Null Writer is not a valid argument");
- }
- return new Stax2EventWriterImpl(createSW(null, w, null, false));
- }
-
- public XMLStreamWriter createXMLStreamWriter(OutputStream out)
- throws XMLStreamException
- {
- return createXMLStreamWriter(out, null);
- }
-
- public XMLStreamWriter createXMLStreamWriter(OutputStream out, String enc)
- throws XMLStreamException
- {
- if (out == null) {
- throw new IllegalArgumentException("Null OutputStream is not a valid argument");
- }
- return createSW(out, null, enc, false);
- }
-
- public XMLStreamWriter createXMLStreamWriter(javax.xml.transform.Result result)
- throws XMLStreamException
- {
- return createSW(result);
- }
-
- public XMLStreamWriter createXMLStreamWriter(Writer w)
- throws XMLStreamException
- {
- if (w == null) {
- throw new IllegalArgumentException("Null Writer is not a valid argument");
- }
- return createSW(null, w, null, false);
- }
-
- public Object getProperty(String name)
- {
- return mConfig.getProperty(name);
- }
-
- public boolean isPropertySupported(String name) {
- return mConfig.isPropertySupported(name);
- }
-
- public void setProperty(String name, Object value)
- {
- mConfig.setProperty(name, value);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Stax2 extensions
- ///////////////////////////////////////////////////////////
- */
-
- // // // Stax2 additional (encoding-aware) factory methods
-
- public XMLEventWriter createXMLEventWriter(Writer w, String enc)
- throws XMLStreamException
- {
- return new Stax2EventWriterImpl(createSW(null, w, enc, false));
- }
-
- public XMLEventWriter createXMLEventWriter(XMLStreamWriter sw)
- throws XMLStreamException
- {
- XMLStreamWriter2 sw2 = Stax2WriterAdapter.wrapIfNecessary(sw);
- return new Stax2EventWriterImpl(sw2);
- }
-
- public XMLStreamWriter2 createXMLStreamWriter(Writer w, String enc)
- throws XMLStreamException
- {
- return createSW(null, w, enc, false);
- }
-
- // // // Stax2 "Profile" mutators
-
- public void configureForXmlConformance()
- {
- mConfig.configureForXmlConformance();
- }
-
- public void configureForRobustness()
- {
- mConfig.configureForRobustness();
- }
-
- public void configureForSpeed()
- {
- mConfig.configureForSpeed();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Woodstox-specific configuration access
- ///////////////////////////////////////////////////////////
- */
-
- public WriterConfig getConfig() {
- return mConfig;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods:
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Bottleneck factory method used internally; needs to take care of passing
- * proper settings to stream writer.
- *
- * @param requireAutoClose Whether this result will always require
- * auto-close be enabled (true); or only if application has
- * requested it (false)
- */
- private XMLStreamWriter2 createSW(OutputStream out, Writer w, String enc,
- boolean requireAutoClose)
- throws XMLStreamException
- {
- /* Need to ensure that the configuration object is not shared
- * any more; otherwise later changes via factory could be
- * visible half-way through output...
- */
- WriterConfig cfg = mConfig.createNonShared();
- XmlWriter xw;
-
- boolean autoCloseOutput = requireAutoClose || mConfig.willAutoCloseOutput();
-
- if (w == null) {
- if (enc == null) {
- enc = WstxOutputProperties.DEFAULT_OUTPUT_ENCODING;
- } else {
- /* Canonical ones are interned, so we may have
- * normalized encoding already...
- */
- if (enc != CharsetNames.CS_UTF8
- && enc != CharsetNames.CS_ISO_LATIN1
- && enc != CharsetNames.CS_US_ASCII) {
- enc = CharsetNames.normalize(enc);
- }
- }
-
- try {
- if (enc == CharsetNames.CS_UTF8) {
- w = new UTF8Writer(cfg, out, autoCloseOutput);
- xw = new BufferingXmlWriter(w, cfg, enc, autoCloseOutput, out, 16);
- } else if (enc == CharsetNames.CS_ISO_LATIN1) {
- xw = new ISOLatin1XmlWriter(out, cfg, autoCloseOutput);
- } else if (enc == CharsetNames.CS_US_ASCII) {
- xw = new AsciiXmlWriter(out, cfg, autoCloseOutput);
- } else {
- w = new OutputStreamWriter(out, enc);
- xw = new BufferingXmlWriter(w, cfg, enc, autoCloseOutput, out, -1);
- }
- } catch (IOException ex) {
- throw new XMLStreamException(ex);
- }
- } else {
- // we may still be able to figure out the encoding:
- if (enc == null) {
- enc = CharsetNames.findEncodingFor(w);
- }
- try {
- xw = new BufferingXmlWriter(w, cfg, enc, autoCloseOutput, null, -1);
- } catch (IOException ex) {
- throw new XMLStreamException(ex);
- }
- }
-
- return createSW(enc, cfg, xw);
- }
-
- /**
- * Called by {@link #createSW(OutputStream, Writer, String, boolean)} after all of the nessesary configuration
- * logic is complete.
- */
- protected XMLStreamWriter2 createSW(String enc, WriterConfig cfg, XmlWriter xw) {
- if (cfg.willSupportNamespaces()) {
- if (cfg.automaticNamespacesEnabled()) {
- return new RepairingNsStreamWriter(xw, enc, cfg);
- }
- return new SimpleNsStreamWriter(xw, enc, cfg);
- }
- return new NonNsStreamWriter(xw, enc, cfg);
- }
-
- private XMLStreamWriter2 createSW(Result res)
- throws XMLStreamException
- {
- OutputStream out = null;
- Writer w = null;
- String encoding = null;
- boolean requireAutoClose;
- String sysId = null;
-
- if (res instanceof Stax2Result) {
- Stax2Result sr = (Stax2Result) res;
- try {
- out = sr.constructOutputStream();
- if (out == null) {
- w = sr.constructWriter();
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- // yes, it's required since caller has no access to stream/writer:
- requireAutoClose = true;
- } else if (res instanceof StreamResult) {
- StreamResult sr = (StreamResult) res;
- out = sr.getOutputStream();
- sysId = sr.getSystemId();
- if (out == null) {
- w = sr.getWriter();
- }
- /* Caller owns it, only auto-close if requested to do so:
- * (except that for system-id-only, it'll still be required,
- * see code below)
- */
- requireAutoClose = false;
- } else if (res instanceof SAXResult) {
- SAXResult sr = (SAXResult) res;
- sysId = sr.getSystemId();
- if (sysId == null || sysId.length() == 0) {
- throw new XMLStreamException("Can not create a stream writer for a SAXResult that does not have System Id (support for using SAX input source not implemented)");
- }
- requireAutoClose = true;
- } else if (res instanceof DOMResult) {
- return WstxDOMWrappingWriter.createFrom(mConfig.createNonShared(), (DOMResult) res);
- } else {
- throw new IllegalArgumentException("Can not instantiate a writer for XML result type "+res.getClass()+" (unrecognized type)");
- }
-
- if (out != null) {
- return createSW(out, null, encoding, requireAutoClose);
- }
- if (w != null) {
- return createSW(null, w, encoding, requireAutoClose);
- }
- if (sysId != null && sysId.length() > 0) {
- /* 26-Dec-2008, TSa: If we must construct URL from system id,
- * it means caller will not have access to resulting
- * stream, thus we will force auto-closing.
- */
- requireAutoClose = true;
- try {
- out = URLUtil.outputStreamFromURL(URLUtil.urlFromSystemId(sysId));
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- return createSW(out, null, encoding, requireAutoClose);
- }
- throw new XMLStreamException("Can not create Stax writer for passed-in Result -- neither writer, output stream or system id was accessible");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/AsciiXmlWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/AsciiXmlWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/AsciiXmlWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/AsciiXmlWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,758 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.*;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.io.CharsetNames;
-
-/**
- * Concrete implementation of {@link EncodingXmlWriter} used when output
- * is to be encoded using 7-bit ascii (US-ASCII) encoding.
- */
-public final class AsciiXmlWriter
- extends EncodingXmlWriter
-{
- public AsciiXmlWriter(OutputStream out, WriterConfig cfg, boolean autoclose)
- throws IOException
- {
- super(out, cfg, CharsetNames.CS_US_ASCII, autoclose);
- }
-
- public void writeRaw(char[] cbuf, int offset, int len)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
-
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- if (mCheckContent) {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = cbuf[offset];
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- } else {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- mOutputBuffer[ptr++] = (byte) cbuf[offset];
- }
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- public void writeRaw(String str, int offset, int len)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- if (mCheckContent) {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = str.charAt(offset);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- } else {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- mOutputBuffer[ptr++] = (byte) str.charAt(offset);
- }
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected void writeAttrValue(String data)
- throws IOException
- {
- int offset = 0;
- int len = data.length();
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data.charAt(offset++);
- sec = calcSurrogate(sec);
- mOutputPtr = ptr;
- ptr = writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- /* Need to quote all white space except for regular
- * space chars, to preserve them (round-tripping)
- */
- if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- } else if (c != '\n' && c != '\t') {
- if (mCheckContent) {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c);
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- }
- }
- // fall-through to char entity output
- } else if (c < 0x7F) {
- if (c != '<' && c != '&' && c != '"') {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- // otherwise fall back on quoting
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data.charAt(offset++));
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- mOutputPtr = ptr;
- ptr = writeAsEntity(c);
- len = data.length() - offset;
- continue main_loop;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected void writeAttrValue(char[] data, int offset, int len)
- throws IOException
- {
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data[offset++];
- sec = calcSurrogate(sec);
- mOutputPtr = ptr;
- ptr = writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data[offset++];
- if (c < 32) {
- /* Need to quote all white space except for regular
- * space chars, to preserve them (round-tripping)
- */
- if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- } else if (c != '\n' && c != '\t') {
- if (mCheckContent) {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c);
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- }
- }
- // fall-through to char entity output
- } else if (c < 0x7F) {
- if (c != '<' && c != '&' && c != '"') {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- // otherwise fall back on quoting
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data[offset++]);
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. contiguous data we can output
- */
- mOutputPtr = ptr;
- ptr = writeAsEntity(c);
- max -= (inEnd - offset); // since we didn't loop completely
- break inner_loop;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected int writeCDataContent(String data)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- } else if (c == '>') { // embedded "]]>"?
- if (offset > 2 && data.charAt(offset-2) == ']'
- && data.charAt(offset-3) == ']') {
- if (!mFixContent) {
- return offset-3;
- }
- /* Relatively easy fix; just need to close this
- * section, and open a new one...
- */
- mOutputPtr = ptr;
- writeCDataEnd();
- writeCDataStart();
- writeAscii(BYTE_GT);
- ptr = mOutputPtr;
- /* No guarantees there's as much free room in the
- * output buffer, thus, need to restart loop:
- */
- len = data.length() - offset;
- continue main_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writeCDataContent(char[] cbuf, int start, int len)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- if (!mCheckContent) {
- writeRaw(cbuf, start, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
- int offset = start;
-
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = cbuf[offset++];
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- } else if (c == '>') { // embedded "]]>"?
- if (offset >= (start+3) && cbuf[offset-2] == ']'
- && cbuf[offset-3] == ']') {
- if (!mFixContent) {
- return offset-3;
- }
- /* Relatively easy fix; just need to close this
- * section, and open a new one...
- */
- mOutputPtr = ptr;
- writeCDataEnd();
- writeCDataStart();
- writeAscii(BYTE_GT);
- ptr = mOutputPtr;
- /* No guarantees there's as much free room in the
- * output buffer, thus, need to restart loop:
- */
- max -= (inEnd - offset);
- break inner_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writeCommentContent(String data)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
-
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- } else if (c == '-') { // embedded "--"?
- if (offset > 1 && data.charAt(offset-2) == '-') {
- if (!mFixContent) {
- return offset-2;
- }
- /* Quite easy to fix: just add an extra space
- * in front. There will be room for that char;
- * but may need to take that the following '-'
- * also fits.
- */
- mOutputBuffer[ptr++] = ' ';
- if (ptr >= mOutputBuffer.length) { // whops. need to flush
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- }
- mOutputBuffer[ptr++] = BYTE_HYPHEN;
- /* Also, since we did output an extra char, better
- * restart the loop (since max calculation is now
- * off)
- */
- max -= (inEnd - offset);
- break inner_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writePIData(String data)
- throws IOException, XMLStreamException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = data.charAt(offset);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- mOutputPtr = ptr;
- if (c > 0x7F) { // 0x7F invalid in xml11
- handleInvalidAsciiChar(c);
- } else if (mXml11) {
- c = handleInvalidChar(c);
- }
- } else if (c == '>') { // enclosed end marker ("?>")?
- if (offset > 0 && data.charAt(offset-1) == '?') {
- return offset-2;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected void writeTextContent(String data)
- throws IOException
- {
- int offset = 0;
- int len = data.length();
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - mOutputPtr;
- if (max < 1) { // output buffer full?
- flushBuffer();
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data.charAt(offset++);
- sec = calcSurrogate(sec);
- writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n' || c == '\t') { // TODO: line count
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- } else if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- } else if (!mXml11 || c == 0) { // ok in xml1.1, as entity
- if (mCheckContent) {
- c = handleInvalidChar(c);
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- // otherwise... well, I guess we can just escape it
- }
- // \r, or xml1.1 + other whitespace, need to escape
- } else if (c < 0x7F) {
- if (c != '<' && c != '&') {
- if (c != '>' || (offset > 1 && data.charAt(offset-2) != ']')) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- }
- // otherwise fall back on quoting
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data.charAt(offset++));
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- writeAsEntity(c);
- len = data.length() - offset;
- continue main_loop;
- }
- len -= max;
- }
- }
-
- protected void writeTextContent(char[] cbuf, int offset, int len)
- throws IOException
- {
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - mOutputPtr;
- if (max < 1) { // output buffer full?
- flushBuffer();
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = cbuf[offset++];
- sec = calcSurrogate(sec);
- writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = cbuf[offset++];
- if (c < 32) {
- if (c == '\n' || c == '\t') { // TODO: line count
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- } else if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- } else if (!mXml11 || c == 0) { // ok in xml1.1, as entity
- if (mCheckContent) {
- c = handleInvalidChar(c);
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- // otherwise... well, I guess we can just try to escape it
- }
- // \r, or xml1.1 + other whitespace, need to escape
- } else if (c < 0x7F) {
- if (c !='<' && c != '&') {
- /* Since we can be conservative, it doesn't matter
- * if second check is not exact
- */
- if (c != '>' || (offset > 1 && cbuf[offset-2] != ']')) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- }
- // otherwise fall back on quoting
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(cbuf[offset++]);
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- writeAsEntity(c);
- max -= (inEnd - offset);
- break inner_loop;
- }
- len -= max;
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////
- */
-
- protected void handleInvalidAsciiChar(int c)
- throws IOException
- {
- // First, let's flush any output we may have, to make debugging easier
- flush();
-
- /* 17-May-2006, TSa: Would really be useful if we could throw
- * XMLStreamExceptions; esp. to indicate actual output location.
- * However, this causes problem with methods that call us and
- * can only throw IOExceptions (when invoked via Writer proxy).
- * Need to figure out how to resolve this.
- */
- throw new IOException("Invalid XML character (0x"+Integer.toHexString(c)+"); can only be output using character entity when using US-ASCII encoding");
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BaseNsStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BaseNsStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BaseNsStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BaseNsStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,753 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE,
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.events.StartElement;
-
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-
-import com.ctc.wstx.api.EmptyElementHandler;
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.exc.WstxIOException;
-import com.ctc.wstx.util.DefaultXmlSymbolTable;
-
-/**
- * Mid-level base class of namespace-aware stream writers. Contains
- * shared functionality between repairing and non-repairing implementations.
- */
-public abstract class BaseNsStreamWriter
- extends TypedStreamWriter
-{
- /*
- ////////////////////////////////////////////////////
- // Constants
- ////////////////////////////////////////////////////
- */
-
- final protected static String sPrefixXml = DefaultXmlSymbolTable.getXmlSymbol();
-
- final protected static String sPrefixXmlns = DefaultXmlSymbolTable.getXmlnsSymbol();
-
- final protected static String ERR_NSDECL_WRONG_STATE =
- "Trying to write a namespace declaration when there is no open start element.";
-
-
- /*
- ////////////////////////////////////////////////////
- // Configuration (options, features)
- ////////////////////////////////////////////////////
- */
-
- // // // Additional specific config flags base class doesn't have
-
- /**
- * True, if writer needs to automatically output namespace declarations
- * (we are in repairing mode)
- */
- final protected boolean mAutomaticNS;
-
- final protected EmptyElementHandler mEmptyElementHandler;
-
- /*
- ////////////////////////////////////////////////////
- // State information
- ////////////////////////////////////////////////////
- */
-
- protected SimpleOutputElement mCurrElem = SimpleOutputElement.createRoot();
-
- /**
- * Optional "root" namespace context that application can set. If so,
- * it can be used to lookup namespace/prefix mappings
- */
- protected NamespaceContext mRootNsContext = null;
-
- /*
- ////////////////////////////////////////////////////
- // Pool for recycling SimpleOutputElement instances
- ////////////////////////////////////////////////////
- */
-
- /* Note: although pooling of cheap objects like SimpleOutputElement
- * is usually not a good idea, here profiling showed it to be a
- * significant improvement. As long as instances are ONLY reused
- * within context of a single writer, they stay in cheap ("Eden")
- * GC area, and thus it should be a win.
- */
- protected SimpleOutputElement mOutputElemPool = null;
-
- /**
- * Although pooled objects are small, let's limit the pool size
- * nonetheless, to optimize memory usage for deeply nested
- * documents. In general, even just low number like 4 levels gets
- * decent return, but 8 should get 99% hit rate.
- */
- final static int MAX_POOL_SIZE = 8;
-
- protected int mPoolSize = 0;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle (ctors)
- ////////////////////////////////////////////////////
- */
-
- public BaseNsStreamWriter(XmlWriter xw, String enc, WriterConfig cfg,
- boolean repairing)
- {
- super(xw, enc, cfg);
- mAutomaticNS = repairing;
- mEmptyElementHandler = cfg.getEmptyElementHandler();
- }
-
- /*
- ////////////////////////////////////////////////////
- // XMLStreamWriter API
- ////////////////////////////////////////////////////
- */
-
- public NamespaceContext getNamespaceContext() {
- return mCurrElem;
- }
-
- public String getPrefix(String uri) {
- return mCurrElem.getPrefix(uri);
- }
-
- public abstract void setDefaultNamespace(String uri)
- throws XMLStreamException;
-
- /**
- *String.charAt() (which
- * perhaps surprisingly is often case, and especially significant
- * for longer buffers).
- */
- protected char[] mCopyBuffer = null;
-
- /*
- ///////////////////////////////////////////////////////////
- // Per-factory configuration (options, features)
- ///////////////////////////////////////////////////////////
- */
-
- protected final WriterConfig mConfig;
-
- // // // Specialized configuration flags, extracted from config flags:
-
- protected final boolean mCfgCDataAsText;
- protected final boolean mCfgCopyDefaultAttrs;
- protected final boolean mCfgAutomaticEmptyElems;
-
- // NOTE: can not be final, may be enabled when schema (etc) validation enabled
-
- protected boolean mCheckStructure;
- protected boolean mCheckAttrs;
-
- /*
- ///////////////////////////////////////////////////////////
- // Per-writer configuration
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Encoding to use; may be passed from the factory (when
- * a method that defines encoding is used), updated by
- * a call to {@link #writeStartDocument}, or null if
- * neither. Is passed to the escaping writer factory to
- * allow escaping writers to do additional escaping if
- * necessary (like encapsulating non-ascii chars in a doc
- * encoded usig ascii).
- */
- protected String mEncoding;
-
- /**
- * Optional validator to use for validating output against
- * one or more schemas, and/or for safe pretty-printing (indentation).
- */
- protected XMLValidator mValidator = null;
-
- /**
- * Since XML 1.1 has some differences to 1.0, we need to keep a flag
- * to indicate if we were to output XML 1.1 document.
- */
- protected boolean mXml11 = false;
-
- /**
- * Custom validation problem handler, if any.
- */
- protected ValidationProblemHandler mVldProbHandler = null;
-
- /*
- ////////////////////////////////////////////////////
- // State information
- ////////////////////////////////////////////////////
- */
-
- protected int mState = STATE_PROLOG;
-
- /**
- * Flag that is set to true first time something has been output.
- * Generally needed to keep track of whether XML declaration
- * (START_DOCUMENT) can be output or not.
- */
- protected boolean mAnyOutput = false;
-
- /**
- * Flag that is set during time that a start element is "open", ie.
- * START_ELEMENT has been output (and possibly zero or more name
- * space declarations and attributes), before other main-level
- * constructs have been output.
- */
- protected boolean mStartElementOpen = false;
-
- /**
- * Flag that indicates that current element is an empty element (one
- * that is explicitly defined as one, by calling a method -- NOT one
- * that just happens to be empty).
- * This is needed to know what to do when next non-ns/attr node
- * is output; normally a new context is opened, but for empty
- * elements not.
- */
- protected boolean mEmptyElement = false;
-
- /**
- * State value used with validation, to track types of content
- * that is allowed at this point in output stream. Only used if
- * validation is enabled: if so, value is determined via validation
- * callbacks.
- */
- protected int mVldContent = XMLValidator.CONTENT_ALLOW_ANY_TEXT;
-
- /**
- * Value passed as the expected root element, when using the multiple
- * argument {@link #writeDTD} method. Will be used in structurally
- * validating mode (and in dtd-validating mode, since that automatically
- * enables structural validation as well, to pre-filter well-formedness
- * errors that validators might have trouble dealing with).
- */
- protected String mDtdRootElem = null;
-
- protected boolean mReturnNullForDefaultNamespace;
-
- /*
- ////////////////////////////////////////////////////
- // Life-cycle
- ////////////////////////////////////////////////////
- */
-
- protected BaseStreamWriter(XmlWriter xw, String enc, WriterConfig cfg)
- {
- mWriter = xw;
- mEncoding = enc;
- mConfig = cfg;
-
- int flags = cfg.getConfigFlags();
-
- mCheckStructure = (flags & OutputConfigFlags.CFG_VALIDATE_STRUCTURE) != 0;
- mCheckAttrs = (flags & OutputConfigFlags.CFG_VALIDATE_ATTR) != 0;
-
- mCfgAutomaticEmptyElems = (flags & OutputConfigFlags.CFG_AUTOMATIC_EMPTY_ELEMENTS) != 0;
- mCfgCDataAsText = (flags & OutputConfigFlags.CFG_OUTPUT_CDATA_AS_TEXT) != 0;
- mCfgCopyDefaultAttrs = (flags & OutputConfigFlags.CFG_COPY_DEFAULT_ATTRS) != 0;
-
- Object value = getProperty(WstxInputProperties.P_RETURN_NULL_FOR_DEFAULT_NAMESPACE);
- mReturnNullForDefaultNamespace = (value instanceof Boolean) && ((Boolean) value).booleanValue();
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter API
- ///////////////////////////////////////////////////////////
- */
-
- public void close()
- throws XMLStreamException
- {
- /* 19-Jul-2004, TSa: Hmmh. Let's actually close all still open
- * elements, starting with currently open start (-> empty)
- * element, if one exists, and then closing scopes by adding
- * matching end elements.
- */
- _finishDocument(false);
- }
-
- public void flush()
- throws XMLStreamException
- {
- /* Note: there have been changes to exact scope of flushing
- * (with Woodstox versions 2.x and 3.x); but the current
- * one of just flushing the underlying OutputStream or Writer
- * should be the interpretation compatible with the Stax specs.
- */
- try {
- mWriter.flush();
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
- }
-
- public abstract NamespaceContext getNamespaceContext();
-
- public abstract String getPrefix(String uri);
-
- public Object getProperty(String name)
- {
- /* These properties just exist for interoperability with
- * toolkits that were designed to work with Sun's parser (which
- * introduced properties)
- */
- if (name.equals(WstxOutputProperties.P_OUTPUT_UNDERLYING_STREAM)) {
- return mWriter.getOutputStream();
- }
- if (name.equals(WstxOutputProperties.P_OUTPUT_UNDERLYING_WRITER)) {
- return mWriter.getWriter();
- }
- return mConfig.getProperty(name);
- }
-
- public abstract void setDefaultNamespace(String uri)
- throws XMLStreamException;
-
- public abstract void setNamespaceContext(NamespaceContext context)
- throws XMLStreamException;
-
- public abstract void setPrefix(String prefix, String uri)
- throws XMLStreamException;
-
- public abstract void writeAttribute(String localName, String value)
- throws XMLStreamException;
-
- public abstract void writeAttribute(String nsURI, String localName,
- String value)
- throws XMLStreamException;
-
- public abstract void writeAttribute(String prefix, String nsURI,
- String localName, String value)
- throws XMLStreamException;
-
- public void writeCData(String data)
- throws XMLStreamException
- {
- /* 02-Dec-2004, TSa: Maybe the writer is to "re-direct" these
- * writes as normal text? (sometimes useful to deal with broken
- * XML parsers, for example)
- */
- if (mCfgCDataAsText) {
- writeCharacters(data);
- return;
- }
-
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- verifyWriteCData();
- if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT
- && mValidator != null) {
- /* Last arg is false, since we do not know if more text
- * may be added with additional calls
- */
- mValidator.validateText(data, false);
- }
- int ix;
- try {
- ix = mWriter.writeCData(data);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- if (ix >= 0) { // unfixable problems?
- reportNwfContent(ErrorConsts.WERR_CDATA_CONTENT, DataUtil.Integer(ix));
- }
- }
-
- public void writeCharacters(char[] text, int start, int len)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- /* Not legal outside main element tree, except if it's all
- * white space
- */
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- if (!StringUtil.isAllWhitespace(text, start, len)) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_NONWS_TEXT);
- }
- }
- }
- // 08-Dec-2005, TSa: validator-based validation?
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) { // never ok
- reportInvalidContent(CHARACTERS);
- } else { // all-ws is ok...
- if (!StringUtil.isAllWhitespace(text, start, len)) {
- reportInvalidContent(CHARACTERS);
- }
- }
- } else if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT) {
- if (mValidator != null) {
- /* Last arg is false, since we do not know if more text
- * may be added with additional calls
- */
- mValidator.validateText(text, start, len, false);
- }
- }
-
- if (len > 0) { // minor optimization
- try {
- /* 21-Jun-2006, TSa: Fixing [WSTX-59]: no quoting can be done
- * outside of element tree.
- */
- if (inPrologOrEpilog()) {
- mWriter.writeRaw(text, start, len);
- } else {
- mWriter.writeCharacters(text, start, len);
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
- }
-
- public void writeCharacters(String text)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- // Need to validate structure?
- if (mCheckStructure) {
- // Not valid in prolog/epilog, except if it's all white space:
- if (inPrologOrEpilog()) {
- if (!StringUtil.isAllWhitespace(text)) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_NONWS_TEXT);
- }
- }
- }
-
- /* 08-Dec-2005, TSa: validator-based validation?
- * Note: although it'd be good to check validity first, we
- * do not know allowed textual content before actually writing
- * pending start element (if any)... so can't call this earlier
- */
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) { // never ok
- reportInvalidContent(CHARACTERS);
- } else { // all-ws is ok...
- if (!StringUtil.isAllWhitespace(text)) {
- reportInvalidContent(CHARACTERS);
- }
- }
- } else if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT) {
- if (mValidator != null) {
- /* Last arg is false, since we do not know if more text
- * may be added with additional calls
- */
- mValidator.validateText(text, false);
- }
- }
-
- // Ok, let's just write it out
- /* 21-Jun-2006, TSa: Fixing [WSTX-59]: no quoting can be done
- * outside of element tree.
- */
- if (inPrologOrEpilog()) {
- try {
- mWriter.writeRaw(text);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- return;
- }
-
- /* Now, would it pay off to make an intermediate copy?
- * String.getChars (which uses System.arraycopy()) is
- * very fast compared to access via String.charAt.
- */
- int len = text.length();
- if (len >= MIN_ARRAYCOPY) {
- char[] buf = getCopyBuffer();
-
- int offset = 0;
- while (len > 0) {
- int thisLen = (len > buf.length) ? buf.length : len;
- text.getChars(offset, offset+thisLen, buf, 0);
- try {
- mWriter.writeCharacters(buf, 0, thisLen);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- offset += thisLen;
- len -= thisLen;
- }
- } else { // nope, let's just access String using charAt().
- try {
- mWriter.writeCharacters(text);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
- }
-
- public void writeComment(String data)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- // 08-Dec-2005, TSa: validator-based validation?
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- reportInvalidContent(COMMENT);
- }
-
- /* No structural validation needed per se, for comments; they are
- * allowed anywhere in XML content. However, content may need to
- * be checked (by XmlWriter)
- */
- int ix;
- try {
- ix = mWriter.writeComment(data);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- if (ix >= 0) {
- reportNwfContent(ErrorConsts.WERR_COMMENT_CONTENT, DataUtil.Integer(ix));
- }
- }
-
- public abstract void writeDefaultNamespace(String nsURI)
- throws XMLStreamException;
-
- public void writeDTD(String dtd)
- throws XMLStreamException
- {
- verifyWriteDTD();
- mDtdRootElem = ""; // marker to verify only one is output
- try {
- mWriter.writeDTD(dtd);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- /* 20-Dec-2005, TSa: Should we try to decipher what was actually
- * written, for validation?
- */
- }
-
- public abstract void writeEmptyElement(String localName)
- throws XMLStreamException;
-
- public abstract void writeEmptyElement(String nsURI, String localName)
- throws XMLStreamException;
-
- public abstract void writeEmptyElement(String prefix, String localName, String nsURI)
- throws XMLStreamException;
-
- public void writeEndDocument() throws XMLStreamException
- {
- _finishDocument(false);
- }
-
- public abstract void writeEndElement() throws XMLStreamException;
-
- public void writeEntityRef(String name)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- // Structurally, need to check we are not in prolog/epilog.
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- reportNwfStructure("Trying to output an entity reference outside main element tree (in prolog or epilog)");
- }
- }
- // 08-Dec-2005, TSa: validator-based validation?
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- /* May be char entity, general entity; whatever it is it's
- * invalid!
- */
- reportInvalidContent(ENTITY_REFERENCE);
- }
-
- //if (mValidator != null) {
- /* !!! 11-Dec-2005, TSa: Should be able to use DTD based validators
- * to check if entity has been declared...
- */
- //}
-
- try {
- mWriter.writeEntityReference(name);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public abstract void writeNamespace(String prefix, String nsURI)
- throws XMLStreamException;
-
- public void writeProcessingInstruction(String target)
- throws XMLStreamException
- {
- writeProcessingInstruction(target, null);
- }
-
- public void writeProcessingInstruction(String target, String data)
- throws XMLStreamException
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- // Structurally, PIs are always ok (content might not be)
- // 08-Dec-2005, TSa: validator-based validation?
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) {
- reportInvalidContent(PROCESSING_INSTRUCTION);
- }
- int ix;
- try {
- ix = mWriter.writePI(target, data);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- if (ix >= 0) {
- throw new XMLStreamException("Illegal input: processing instruction content has embedded '?>' in it (index "+ix+")");
- }
- }
-
- public void writeStartDocument()
- throws XMLStreamException
- {
- /* 03-Feb-2005, TSa: As per StAX 1.0 specs, version should
- * be "1.0", and encoding "utf-8" (yes, lower case... it's
- * wrong, but specs mandate it)
- */
- /* 11-Jan-2006, TSa: Let's actually rather use whatever was passed
- * in, if anything; only if none then default to something else.
- * Plus, what the heck; let's use properly capitalized value
- * too (and ignore faulty def in stax specs).
- */
- if (mEncoding == null) {
- mEncoding = WstxOutputProperties.DEFAULT_OUTPUT_ENCODING;
- }
- writeStartDocument(mEncoding, WstxOutputProperties.DEFAULT_XML_VERSION);
- }
-
- public void writeStartDocument(String version)
- throws XMLStreamException
- {
- writeStartDocument(mEncoding, version);
- }
-
- public void writeStartDocument(String encoding, String version)
- throws XMLStreamException
- {
- doWriteStartDocument(version, encoding, null);
- }
-
- protected void doWriteStartDocument(String version, String encoding,
- String standAlone)
- throws XMLStreamException
- {
- /* Not legal to output XML declaration if there has been ANY
- * output prior... that is, if we validate the structure.
- */
- if (mCheckStructure) {
- if (mAnyOutput) {
- reportNwfStructure("Can not output XML declaration, after other output has already been done.");
- }
- }
-
- mAnyOutput = true;
-
- if (mConfig.willValidateContent()) {
- // !!! 06-May-2004, TSa: Should validate encoding?
- /*if (encoding != null) {
- }*/
- if (version != null && version.length() > 0) {
- if (!(version.equals(XmlConsts.XML_V_10_STR)
- || version.equals(XmlConsts.XML_V_11_STR))) {
- reportNwfContent("Illegal version argument ('"+version
- +"'); should only use '"+XmlConsts.XML_V_10_STR
- +"' or '"+XmlConsts.XML_V_11_STR+"'");
- }
- }
- }
-
- if (version == null || version.length() == 0) {
- version = WstxOutputProperties.DEFAULT_XML_VERSION;
- }
-
- /* 04-Feb-2006, TSa: Need to know if we are writing XML 1.1
- * document...
- */
- mXml11 = XmlConsts.XML_V_11_STR.equals(version);
- if (mXml11) {
- mWriter.enableXml11();
- }
-
- if (encoding != null && encoding.length() > 0) {
- /* 03-May-2005, TSa: But what about conflicting encoding? Let's
- * only update encoding, if it wasn't set.
- */
- if (mEncoding == null || mEncoding.length() == 0) {
- mEncoding = encoding;
- }
- }
- try {
- mWriter.writeXmlDeclaration(version, encoding, standAlone);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public abstract void writeStartElement(String localName)
- throws XMLStreamException;
-
- public abstract void writeStartElement(String nsURI, String localName)
- throws XMLStreamException;
-
- public abstract void writeStartElement(String prefix, String localName,
- String nsURI)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////////////////////////
- // XMLStreamWriter2 methods (StAX2)
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method that essentially copies event that the specified reader has
- * just read.
- *
- * @param sr Stream reader to use for accessing event to copy
- * @param preserveEventData If true, writer is not allowed to change
- * the state of the reader (so that all the data associated with the
- * current event has to be preserved); if false, writer is allowed
- * to use methods that may cause some data to be discarded. Setting
- * this to false may improve the performance, since it may allow
- * full no-copy streaming of data, especially textual contents.
- */
- public void copyEventFromReader(XMLStreamReader2 sr, boolean preserveEventData)
- throws XMLStreamException
- {
- try {
- switch (sr.getEventType()) {
- case START_DOCUMENT:
- {
- String version = sr.getVersion();
- /* No real declaration? If so, we don't want to output
- * anything, to replicate as closely as possible the
- * source document
- */
- if (version == null || version.length() == 0) {
- ; // no output if no real input
- } else {
- if (sr.standaloneSet()) {
- writeStartDocument(sr.getVersion(),
- sr.getCharacterEncodingScheme(),
- sr.isStandalone());
- } else {
- writeStartDocument(sr.getCharacterEncodingScheme(),
- sr.getVersion());
- }
- }
- }
- return;
-
- case END_DOCUMENT:
- writeEndDocument();
- return;
-
- // Element start/end events:
- case START_ELEMENT:
- if (sr instanceof StreamReaderImpl) {
- StreamReaderImpl impl = (StreamReaderImpl) sr;
- copyStartElement(impl.getInputElementStack(), impl.getAttributeCollector());
- } else { // otherwise impl from Stax ref. impl (Stax2WriterImpl) has to do:
- super.copyStartElement(sr);
- }
- return;
-
- case END_ELEMENT:
- writeEndElement();
- return;
-
- case SPACE:
- {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- /* No need to write as chars, should be pure space
- * (caller should have verified); also, no escaping
- * necessary.
- */
- sr.getText(wrapAsRawWriter(), preserveEventData);
- }
- return;
-
- case CDATA:
-
- // First; is this to be changed to 'normal' text output?
- if (!mCfgCDataAsText) {
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- // Not legal outside main element tree:
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_CDATA);
- }
- }
- /* Note: no need to check content, since reader is assumed
- * to have verified it to be valid XML.
- */
- mWriter.writeCDataStart();
- sr.getText(wrapAsRawWriter(), preserveEventData);
- mWriter.writeCDataEnd();
- return;
- }
- // fall down if it is to be converted...
-
- case CHARACTERS:
- {
- // Let's just assume content is fine... not 100% reliably
- // true, but usually is (not true if input had a root
- // element surrounding text, but omitted for output)
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- sr.getText(wrapAsTextWriter(), preserveEventData);
- }
- return;
-
- case COMMENT:
- {
- mAnyOutput = true;
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- // No need to check for content (embedded '--'); reader
- // is assumed to have verified it's ok (otherwise should
- // have thrown an exception for non-well-formed XML)
- mWriter.writeCommentStart();
- sr.getText(wrapAsRawWriter(), preserveEventData);
- mWriter.writeCommentEnd();
- }
- return;
-
- case PROCESSING_INSTRUCTION:
- {
- mWriter.writePIStart(sr.getPITarget(), true);
- sr.getText(wrapAsRawWriter(), preserveEventData);
- mWriter.writePIEnd();
- }
- return;
-
- case DTD:
- {
- DTDInfo info = sr.getDTDInfo();
- if (info == null) {
- // Hmmmh. It is legal for this to happen, for
- // non-DTD-aware readers. But what is the right
- // thing to do here?
- throwOutputError("Current state DOCTYPE, but not DTDInfo Object returned -- reader doesn't support DTDs?");
- }
- // Could optimize this a bit (stream the int. subset
- // possible), but it's never going to occur more than
- // once per document, so it's probably not much of a
- // bottleneck, ever
- writeDTD(info);
- }
- return;
-
- case ENTITY_REFERENCE:
- writeEntityRef(sr.getLocalName());
- return;
-
- case ATTRIBUTE:
- case NAMESPACE:
- case ENTITY_DECLARATION:
- case NOTATION_DECLARATION:
- // Let's just fall back to throw the exception
- }
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- throw new XMLStreamException("Unrecognized event type ("
- +sr.getEventType()+"); not sure how to copy");
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // StAX2, output handling
- ///////////////////////////////////////////////////////////
- */
-
- public void closeCompletely()
- throws XMLStreamException
- {
- _finishDocument(true);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // StAX2, config
- ///////////////////////////////////////////////////////////
- */
-
- // NOTE: getProperty() defined in Stax 1.0 interface
-
- public boolean isPropertySupported(String name) {
- // !!! TBI: not all these properties are really supported
- return mConfig.isPropertySupported(name);
- }
-
- /**
- * @param name Name of the property to set
- * @param value Value to set property to.
- *
- * @return True, if the specified property was succesfully
- * set to specified value; false if its value was not changed
- */
- public boolean setProperty(String name, Object value)
- {
- /* Note: can not call local method, since it'll return false for
- * recognized but non-mutable properties
- */
- return mConfig.setProperty(name, value);
- }
-
- public XMLValidator validateAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- XMLValidator vld = schema.createValidator(this);
-
- if (mValidator == null) {
- /* Need to enable other validation modes? Structural validation
- * should always be done when we have other validators as well,
- * as well as attribute uniqueness checks.
- */
- mCheckStructure = true;
- mCheckAttrs = true;
- mValidator = vld;
- } else {
- mValidator = new ValidatorPair(mValidator, vld);
- }
- return vld;
- }
-
- public XMLValidator stopValidatingAgainst(XMLValidationSchema schema)
- throws XMLStreamException
- {
- XMLValidator[] results = new XMLValidator[2];
- XMLValidator found = null;
- if (ValidatorPair.removeValidator(mValidator, schema, results)) { // found
- found = results[0];
- mValidator = results[1];
- found.validationCompleted(false);
- if (mValidator == null) {
- resetValidationFlags();
- }
- }
- return found;
- }
-
- public XMLValidator stopValidatingAgainst(XMLValidator validator)
- throws XMLStreamException
- {
- XMLValidator[] results = new XMLValidator[2];
- XMLValidator found = null;
- if (ValidatorPair.removeValidator(mValidator, validator, results)) { // found
- found = results[0];
- mValidator = results[1];
- found.validationCompleted(false);
- if (mValidator == null) {
- resetValidationFlags();
- }
- }
- return found;
- }
-
- public ValidationProblemHandler setValidationProblemHandler(ValidationProblemHandler h)
- {
- ValidationProblemHandler oldH = mVldProbHandler;
- mVldProbHandler = h;
- return oldH;
- }
-
- private void resetValidationFlags()
- {
- int flags = mConfig.getConfigFlags();
- mCheckStructure = (flags & CFG_VALIDATE_STRUCTURE) != 0;
- mCheckAttrs = (flags & CFG_VALIDATE_ATTR) != 0;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // StAX2, other accessors, mutators
- ///////////////////////////////////////////////////////////
- */
-
- public XMLStreamLocation2 getLocation()
- {
- return new WstxInputLocation(null, // no parent
- null, null, // pub/sys ids not yet known
- mWriter.getAbsOffset(),
- mWriter.getRow(), mWriter.getColumn());
- }
-
- public String getEncoding() {
- return mEncoding;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // StAX2, output methods
- ///////////////////////////////////////////////////////////
- */
-
- public void writeCData(char[] cbuf, int start, int len)
- throws XMLStreamException
- {
- /* 02-Dec-2004, TSa: Maybe the writer is to "re-direct" these
- * writes as normal text? (sometimes useful to deal with broken
- * XML parsers, for example)
- */
- if (mCfgCDataAsText) {
- writeCharacters(cbuf, start, len);
- return;
- }
-
- mAnyOutput = true;
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- verifyWriteCData();
- if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT
- && mValidator != null) {
- /* Last arg is false, since we do not know if more text
- * may be added with additional calls
- */
- mValidator.validateText(cbuf, start, len, false);
- }
- int ix;
- try {
- ix = mWriter.writeCData(cbuf, start, len);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- if (ix >= 0) { // problems that could not to be fixed?
- throwOutputError(ErrorConsts.WERR_CDATA_CONTENT, DataUtil.Integer(ix));
- }
- }
-
- public void writeDTD(DTDInfo info)
- throws XMLStreamException
- {
- writeDTD(info.getDTDRootName(), info.getDTDSystemId(),
- info.getDTDPublicId(), info.getDTDInternalSubset());
- }
-
- public void writeDTD(String rootName, String systemId, String publicId,
- String internalSubset)
- throws XMLStreamException
- {
- verifyWriteDTD();
- mDtdRootElem = rootName;
- try {
- mWriter.writeDTD(rootName, systemId, publicId, internalSubset);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public abstract void writeFullEndElement() throws XMLStreamException;
-
- public void writeStartDocument(String version, String encoding,
- boolean standAlone)
- throws XMLStreamException
- {
- doWriteStartDocument(version, encoding, standAlone ? "yes" : "no");
- }
-
- public void writeRaw(String text)
- throws XMLStreamException
- {
- mAnyOutput = true;
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- try {
- mWriter.writeRaw(text, 0, text.length());
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public void writeRaw(String text, int start, int offset)
- throws XMLStreamException
- {
- mAnyOutput = true;
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- try {
- mWriter.writeRaw(text, start, offset);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public void writeRaw(char[] text, int start, int offset)
- throws XMLStreamException
- {
- mAnyOutput = true;
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- try {
- mWriter.writeRaw(text, start, offset);
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- public void writeSpace(String text)
- throws XMLStreamException
- {
- /* For now, let's just use writeRaw(): otherwise would need
- * to add it to all backend writers:
- */
- writeRaw(text);
- }
-
- public void writeSpace(char[] text, int offset, int length)
- throws XMLStreamException
- {
- // For now, let's just use writeRaw() (see above)
- writeRaw(text, offset, length);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // ValidationContext interface (StAX2, validation)
- ///////////////////////////////////////////////////////////
- */
-
- public String getXmlVersion() {
- return mXml11 ? XmlConsts.XML_V_11_STR : XmlConsts.XML_V_10_STR;
- }
-
- public abstract QName getCurrentElementName();
-
- public abstract String getNamespaceURI(String prefix);
-
- /**
- * As of now, there is no way to specify the base URI. Could be improved
- * in future, if xml:base is supported.
- */
- public String getBaseUri() {
- return null;
- }
-
- public Location getValidationLocation() {
- return getLocation();
- }
-
- public void reportProblem(XMLValidationProblem prob)
- throws XMLStreamException
- {
- // Custom handler set? If so, it'll take care of it:
- if (mVldProbHandler != null) {
- mVldProbHandler.reportProblem(prob);
- return;
- }
-
- /* For now let's implement basic functionality: warnings get
- * reported via XMLReporter, errors and fatal errors result in
- * immediate exceptions.
- */
- /* 27-May-2008, TSa: [WSTX-153] Above is incorrect: as per Stax
- * javadocs for XMLReporter, both warnings and non-fatal errors
- * (which includes all validation errors) should be reported via
- * XMLReporter interface, and only fatals should cause an
- * immediate stream exception (by-passing reporter)
- */
- if (prob.getSeverity() > XMLValidationProblem.SEVERITY_ERROR) {
- throw WstxValidationException.create(prob);
- }
- XMLReporter rep = mConfig.getProblemReporter();
- if (rep != null) {
- doReportProblem(rep, prob);
- } else {
- /* If no reporter, regular non-fatal errors are to be reported
- * as exceptions as well, for backwards compatibility
- */
- if (prob.getSeverity() >= XMLValidationProblem.SEVERITY_ERROR) {
- throw WstxValidationException.create(prob);
- }
- }
- }
-
- /**
- * Adding default attribute values does not usually make sense on
- * output side, so the implementation is a NOP for now.
- */
- public int addDefaultAttribute(String localName, String uri, String prefix,
- String value)
- {
- // nothing to do, but to indicate we didn't add it...
- return -1;
- }
-
- // // // Notation/entity access: not (yet?) implemented
-
- public boolean isNotationDeclared(String name) { return false; }
-
- public boolean isUnparsedEntityDeclared(String name) { return false; }
-
-
- // // // Attribute access: not yet implemented:
-
- /* !!! TODO: Implement attribute access (iff validate-attributes
- * enabled?
- */
-
- public int getAttributeCount() { return 0; }
-
- public String getAttributeLocalName(int index) { return null; }
-
- public String getAttributeNamespace(int index) { return null; }
-
- public String getAttributePrefix(int index) { return null; }
-
- public String getAttributeValue(int index) { return null; }
-
- public String getAttributeValue(String nsURI, String localName) {
- return null;
- }
-
- public String getAttributeType(int index) {
- return "";
- }
-
- public int findAttributeIndex(String nsURI, String localName) {
- return -1;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods (ie not part of public API)
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method that can be called to get a wrapper instance that
- * can be used to essentially call the writeRaw
- * method via regular Writer interface.
- */
- public final Writer wrapAsRawWriter()
- {
- return mWriter.wrapAsRawWriter();
- }
-
- /**
- * Method that can be called to get a wrapper instance that
- * can be used to essentially call the writeCharacters
- * method via regular Writer interface.
- */
- public final Writer wrapAsTextWriter()
- {
- return mWriter.wrapAsTextWriter();
- }
-
- /**
- * Method that is used by output classes to determine whether we
- * are in validating mode.
- *writeCharacters; not
- * good... but done to optimize white-space cases.
- */
- public void writeCharacters(Characters ch)
- throws XMLStreamException
- {
- // Need to finish an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
-
- /* Not legal outside main element tree, except if it's all
- * white space
- */
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- if (!ch.isIgnorableWhiteSpace() && !ch.isWhiteSpace()) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_NONWS_TEXT);
- }
- }
- }
-
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- if (mVldContent == XMLValidator.CONTENT_ALLOW_NONE) { // never ok
- reportInvalidContent(CHARACTERS);
- } else { // all-ws is ok...
- if (!ch.isIgnorableWhiteSpace() && !ch.isWhiteSpace()) {
- reportInvalidContent(CHARACTERS);
- }
- }
- } else if (mVldContent == XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT) {
- if (mValidator != null) {
- /* Last arg is false, since we do not know if more text
- * may be added with additional calls
- */
- mValidator.validateText(ch.getData(), false);
- }
- }
-
- // Ok, let's just write it out:
- try {
- mWriter.writeCharacters(ch.getData());
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
- }
-
- /**
- * Method called to close an open start element, when another
- * main-level element (not namespace declaration or attribute)
- * is being output; except for end element which is handled differently.
- */
- protected abstract void closeStartElement(boolean emptyElem)
- throws XMLStreamException;
-
- protected final boolean inPrologOrEpilog() {
- return (mState != STATE_TREE);
- }
-
- /**
- * @param forceRealClose If true, will force calling of close() on the
- * underlying physical result (stream, writer). If false, will let
- * XmlWriter decide whether to call close(); will be done if auto-closing
- * enabled, but not otherwise.
- */
- private final void _finishDocument(boolean forceRealClose)
- throws XMLStreamException
- {
- // Is tree still open?
- if (mState != STATE_EPILOG) {
- if (mCheckStructure && mState == STATE_PROLOG) {
- reportNwfStructure("Trying to write END_DOCUMENT when document has no root (ie. trying to output empty document).");
- }
- // 20-Jul-2004, TSa: Need to close the open sub-tree, if it exists...
- // First, do we have an open start element?
- if (mStartElementOpen) {
- closeStartElement(mEmptyElement);
- }
- // Then, one by one, need to close open scopes:
- /* 17-Nov-2008, TSa: that is, if we are allowed to do it
- * (see [WSTX-165])
- */
- if (mState != STATE_EPILOG && mConfig.automaticEndElementsEnabled()) {
- do {
- writeEndElement();
- } while (mState != STATE_EPILOG);
- }
- }
-
- /* And finally, inform the underlying writer that it should flush
- * and release its buffers, and close components it uses if any.
- */
- char[] buf = mCopyBuffer;
- if (buf != null) {
- mCopyBuffer = null;
- mConfig.freeMediumCBuffer(buf);
- }
- try {
- mWriter.close(forceRealClose);
- } catch (IOException ie) {
- throw new WstxIOException(ie);
- }
- }
-
- /**
- * Implementation-dependant method called to fully copy START_ELEMENT
- * event that the passed-in stream reader points to
- */
- public abstract void copyStartElement(InputElementStack elemStack,
- AttributeCollector attrCollector)
- throws IOException, XMLStreamException;
-
- /**
- * Method called before writing a QName via Typed Access API.
- * In namespace-repairing mode it should take appropriate actions
- * to ensure that the given namespace URI is bound to a namespace
- * and return whatever it maps to. In non-repairing work no additional
- * work is to be done and methods
- *
- * @return Prefix to use when writing out given QName as an element
- * or attribute value
- */
- public abstract String validateQNamePrefix(QName name)
- throws XMLStreamException;
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods, validation
- ///////////////////////////////////////////////////////////
- */
-
- protected final void verifyWriteCData()
- throws XMLStreamException
- {
- // Not legal outside main element tree:
- if (mCheckStructure) {
- if (inPrologOrEpilog()) {
- reportNwfStructure(ErrorConsts.WERR_PROLOG_CDATA);
- }
- }
- // 08-Dec-2005, TSa: validator-based validation?
- if (mVldContent <= XMLValidator.CONTENT_ALLOW_WS) {
- // there's no ignorable white space CDATA...
- reportInvalidContent(CDATA);
- }
- }
-
- protected final void verifyWriteDTD()
- throws XMLStreamException
- {
- // 20-Nov-2004, TSa: can check that we are in prolog
- if (mCheckStructure) {
- if (mState != STATE_PROLOG) {
- throw new XMLStreamException("Can not write DOCTYPE declaration (DTD) when not in prolog any more (state "+mState+"; start element(s) written)");
- }
- // 20-Dec-2005, TSa: and that we only output one...
- if (mDtdRootElem != null) {
- throw new XMLStreamException("Trying to write multiple DOCTYPE declarations");
- }
- }
- }
-
- protected void verifyRootElement(String localName, String prefix)
- throws XMLStreamException
- {
- /* Note: this check is bit lame, due to DOCTYPE declaration (and DTD
- * in general) being namespace-ignorant...
- */
- if (isValidating()) {
- /* 17-Mar-2006, TSa: Ideally, this should be a validity
- * problem?
- */
- if (mDtdRootElem != null && mDtdRootElem.length() > 0) {
- String wrongElem = null;
-
- /* Ugh. It is possible that we just don't know the prefix --
- * in repairing mode it's assigned after this check. So for
- * now, let's only verify the local name
- */
- if (localName.equals(mDtdRootElem)) {
- // good
- } else {
- int lnLen = localName.length();
- int oldLen = mDtdRootElem.length();
-
- if (oldLen > lnLen
- && mDtdRootElem.endsWith(localName)
- && mDtdRootElem.charAt(oldLen - lnLen - 1) == ':') {
- // good also
- } else {
- if (prefix == null) { // doesn't and won't have one
- wrongElem = localName;
- } else if (prefix.length() == 0) { // don't know what it'd be
- wrongElem = "[unknown]:"+localName;
- } else {
- wrongElem = prefix + ":" + localName;
- }
- }
- }
- if (wrongElem != null) {
- reportValidationProblem(ErrorConsts.ERR_VLD_WRONG_ROOT, wrongElem, mDtdRootElem);
- }
- }
- }
- mState = STATE_TREE;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods, basic output problem reporting
- ///////////////////////////////////////////////////////////
- */
-
- protected static void throwOutputError(String msg)
- throws XMLStreamException
- {
- throw new XMLStreamException(msg);
- }
-
- protected static void throwOutputError(String format, Object arg)
- throws XMLStreamException
- {
- String msg = MessageFormat.format(format, new Object[] { arg });
- throwOutputError(msg);
- }
-
- /**
- * Method called when an illegal method (namespace-specific method
- * on non-ns writer) is called by the application.
- */
- protected static void reportIllegalMethod(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- /**
- * This is the method called when an output method call violates
- * structural well-formedness checks
- * and {@link WstxOutputProperties#P_OUTPUT_VALIDATE_STRUCTURE} is
- * is enabled.
- */
- protected static void reportNwfStructure(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- protected static void reportNwfStructure(String msg, Object arg)
- throws XMLStreamException
- {
- throwOutputError(msg, arg);
- }
-
- /**
- * This is the method called when an output method call violates
- * content well-formedness checks
- * and {@link WstxOutputProperties#P_OUTPUT_VALIDATE_CONTENT} is
- * is enabled.
- */
- protected static void reportNwfContent(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- protected static void reportNwfContent(String msg, Object arg)
- throws XMLStreamException
- {
- throwOutputError(msg, arg);
- }
-
- /**
- * This is the method called when an output method call violates
- * attribute well-formedness checks (trying to output dup attrs)
- * and {@link WstxOutputProperties#P_OUTPUT_VALIDATE_NAMES} is
- * is enabled.
- */
- protected static void reportNwfAttr(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- protected static void reportNwfAttr(String msg, Object arg)
- throws XMLStreamException
- {
- throwOutputError(msg, arg);
- }
-
- protected static void throwFromIOE(IOException ioe)
- throws XMLStreamException
- {
- throw new WstxIOException(ioe);
- }
-
- protected static void reportIllegalArg(String msg)
- throws IllegalArgumentException
- {
- throw new IllegalArgumentException(msg);
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods, output validation problem reporting
- ///////////////////////////////////////////////////////////
- */
-
- protected void reportInvalidContent(int evtType)
- throws XMLStreamException
- {
- switch (mVldContent) {
- case XMLValidator.CONTENT_ALLOW_NONE:
- reportValidationProblem(ErrorConsts.ERR_VLD_EMPTY,
- getTopElementDesc(),
- ErrorConsts.tokenTypeDesc(evtType));
- break;
- case XMLValidator.CONTENT_ALLOW_WS:
- reportValidationProblem(ErrorConsts.ERR_VLD_NON_MIXED,
- getTopElementDesc());
- break;
- case XMLValidator.CONTENT_ALLOW_VALIDATABLE_TEXT:
- case XMLValidator.CONTENT_ALLOW_ANY_TEXT:
- /* Not 100% sure if this should ever happen... depends on
- * interpretation of 'any' content model?
- */
- reportValidationProblem(ErrorConsts.ERR_VLD_ANY,
- getTopElementDesc(),
- ErrorConsts.tokenTypeDesc(evtType));
- break;
- default: // should never occur:
- reportValidationProblem("Internal error: trying to report invalid content for "+evtType);
- }
- }
-
- public void reportValidationProblem(String msg, Location loc, int severity)
- throws XMLStreamException
- {
- reportProblem(new XMLValidationProblem(loc, msg, severity));
- }
-
- public void reportValidationProblem(String msg, int severity)
- throws XMLStreamException
- {
- reportProblem(new XMLValidationProblem(getValidationLocation(),
- msg, severity));
- }
-
- public void reportValidationProblem(String msg)
- throws XMLStreamException
- {
- reportProblem(new XMLValidationProblem(getValidationLocation(),
- msg,
- XMLValidationProblem.SEVERITY_ERROR));
- }
-
- public void reportValidationProblem(Location loc, String msg)
- throws XMLStreamException
- {
- reportProblem(new XMLValidationProblem(getValidationLocation(), msg));
- }
-
- public void reportValidationProblem(String format, Object arg)
- throws XMLStreamException
- {
- String msg = MessageFormat.format(format, new Object[] { arg });
- reportProblem(new XMLValidationProblem(getValidationLocation(),
- msg));
- }
-
- public void reportValidationProblem(String format, Object arg, Object arg2)
- throws XMLStreamException
- {
- String msg = MessageFormat.format(format, new Object[] { arg, arg2 });
- reportProblem(new XMLValidationProblem(getValidationLocation(), msg));
- }
-
- protected void doReportProblem(XMLReporter rep, String probType, String msg, Location loc)
- throws XMLStreamException
- {
- if (loc == null) {
- loc = getLocation();
- }
- doReportProblem(rep, new XMLValidationProblem(loc, msg, XMLValidationProblem.SEVERITY_ERROR, probType));
- }
-
- protected void doReportProblem(XMLReporter rep, XMLValidationProblem prob)
- throws XMLStreamException
- {
- if (rep != null) {
- Location loc = prob.getLocation();
- if (loc == null) {
- loc = getLocation();
- prob.setLocation(loc);
- }
- // Backwards-compatibility fix: add non-null type, if missing:
- if (prob.getType() == null) {
- prob.setType(ErrorConsts.WT_VALIDATION);
- }
- // [WSTX-154]: was catching and dropping thrown exception: shouldn't.
- rep.report(prob.getMessage(), prob.getType(), prob, loc);
- }
- }
-
- /**
- * Method needed for error message generation
- */
- protected abstract String getTopElementDesc();
-
- /*
- ///////////////////////////////////////////////////////////
- // Package methods, other
- ///////////////////////////////////////////////////////////
- */
-
- protected final char[] getCopyBuffer()
- {
- char[] buf = mCopyBuffer;
- if (buf == null) {
- mCopyBuffer = buf = mConfig.allocMediumCBuffer(DEFAULT_COPYBUFFER_LEN);
- }
- return buf;
- }
-
- protected final char[] getCopyBuffer(int minLen)
- {
- char[] buf = mCopyBuffer;
- if (buf == null || minLen > buf.length) {
- mCopyBuffer = buf = mConfig.allocMediumCBuffer(Math.max(DEFAULT_COPYBUFFER_LEN, minLen));
- }
- return buf;
- }
-
- public String toString()
- {
- return "[StreamWriter: "+getClass()+", underlying outputter: "
- +((mWriter == null) ? "NULL" : mWriter.toString()+"]");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BufferingXmlWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BufferingXmlWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/BufferingXmlWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/BufferingXmlWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1666 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.*;
-
-import javax.xml.stream.XMLStreamConstants;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.io.CharsetNames;
-import com.ctc.wstx.io.CompletelyCloseable;
-
-/**
- * Concrete implementation of {@link XmlWriter} that will dispatch writes
- * to another writer (of type {@link java.io.Writer}, and will NOT handle
- * encoding. It will, however, do basic buffering such that the underlying
- * Writer need (and thus, should) not do buffering.
- *out) is using, if known. Needed to support
- * (optional) access to the underlying stream
- */
- public BufferingXmlWriter(Writer out, WriterConfig cfg, String enc,
- boolean autoclose,
- OutputStream outs, int bitsize)
- throws IOException
- {
- super(cfg, enc, autoclose);
- mOut = out;
- mOutputBuffer = cfg.allocFullCBuffer(DEFAULT_BUFFER_SIZE);
- mOutputBufLen = mOutputBuffer.length;
- mSmallWriteSize = DEFAULT_SMALL_SIZE;
- mOutputPtr = 0;
-
- mUnderlyingStream = outs;
-
- // Let's use double-quotes, as usual; alternative is apostrophe
- mEncQuoteChar = '"';
- mEncQuoteEntity = """;
- /* Note: let's actually exclude couple of illegal chars for
- * unicode-based encoders. But we do not have to worry about
- * surrogates quite here, fortunately.
- */
- if (bitsize < 1) {
- bitsize = guessEncodingBitSize(enc);
- }
- mEncHighChar = ((bitsize < 16) ? (1 << bitsize) : 0xFFFE);
- }
-
- protected int getOutputPtr() {
- return mOutputPtr;
- }
-
- /*
- ////////////////////////////////////////////////
- // Raw access to underlying output objects
- ////////////////////////////////////////////////
- */
-
- final protected OutputStream getOutputStream()
- {
- return mUnderlyingStream;
- }
-
- final protected Writer getWriter()
- {
- return mOut;
- }
-
- /*
- ////////////////////////////////////////////////
- // Low-level (pass-through) methods
- ////////////////////////////////////////////////
- */
-
- public void close(boolean forceRealClose)
- throws IOException
- {
- flush();
- mTextWriter = null;
- mAttrValueWriter = null;
-
- // Buffers to free?
- char[] buf = mOutputBuffer;
- if (buf != null) {
- mOutputBuffer = null;
- mConfig.freeFullCBuffer(buf);
- }
- // Plus may need to close the actual writer
- if (forceRealClose || mAutoCloseOutput) {
- /* 14-Nov-2008, TSa: To resolve [WSTX-163], need to have a way
- * to force UTF8Writer to close the underlying stream...
- */
- if (mOut instanceof CompletelyCloseable) {
- ((CompletelyCloseable)mOut).closeCompletely();
- } else {
- mOut.close();
- }
- }
- }
-
- public final void flush()
- throws IOException
- {
- flushBuffer();
- mOut.flush();
- }
-
- public void writeRaw(char[] cbuf, int offset, int len)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
-
- // First; is the new request small or not? If yes, needs to be buffered
- if (len < mSmallWriteSize) { // yup
- // Does it fit in with current buffer? If not, need to flush first
- if ((mOutputPtr + len) > mOutputBufLen) {
- flushBuffer();
- }
- System.arraycopy(cbuf, offset, mOutputBuffer, mOutputPtr, len);
- mOutputPtr += len;
- return;
- }
-
- // Ok, not a small request. But buffer may have existing content?
- int ptr = mOutputPtr;
- if (ptr > 0) {
- // If it's a small chunk, need to fill enough before flushing
- if (ptr < mSmallWriteSize) {
- /* Also, if we are to copy any stuff, let's make sure
- * that we either copy it all in one chunk, or copy
- * enough for non-small chunk, flush, and output remaining
- * non-small chink (former possible if chunk we were requested
- * to output is only slightly over 'small' size)
- */
- int needed = (mSmallWriteSize - ptr);
-
- // Just need minimal copy:
- System.arraycopy(cbuf, offset, mOutputBuffer, ptr, needed);
- mOutputPtr = ptr + needed;
- len -= needed;
- offset += needed;
- }
- flushBuffer();
- }
-
- // And then we'll just write whatever we have left:
- mOut.write(cbuf, offset, len);
- }
-
- /**
- * Method called to output typed values (int, long, double, float etc)
- * that are known not to contain any escapable characters, or anything
- * else beyond 7-bit ascii range.
- */
- public final void writeRawAscii(char[] cbuf, int offset, int len)
- throws IOException
- {
- // Can't optimize any further with buffering writer, so:
- writeRaw(cbuf, offset, len);
- }
-
- public void writeRaw(String str)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
- final int len = str.length();
-
- // First; is the new request small or not? If yes, needs to be buffered
- if (len < mSmallWriteSize) { // yup
- // Does it fit in with current buffer? If not, need to flush first
- if ((mOutputPtr + len) >= mOutputBufLen) {
- flushBuffer();
- }
- str.getChars(0, len, mOutputBuffer, mOutputPtr);
- mOutputPtr += len;
- return;
- }
- // Otherwise, let's just call the main method
- writeRaw(str, 0, len);
- }
-
- public void writeRaw(String str, int offset, int len)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
-
- // First; is the new request small or not? If yes, needs to be buffered
- if (len < mSmallWriteSize) { // yup
- // Does it fit in with current buffer? If not, need to flush first
- if ((mOutputPtr + len) >= mOutputBufLen) {
- flushBuffer();
- }
- str.getChars(offset, offset+len, mOutputBuffer, mOutputPtr);
- mOutputPtr += len;
- return;
- }
-
- // Ok, not a small request. But buffer may have existing content?
- int ptr = mOutputPtr;
- if (ptr > 0) {
- // If it's a small chunk, need to fill enough before flushing
- if (ptr < mSmallWriteSize) {
- /* Also, if we are to copy any stuff, let's make sure
- * that we either copy it all in one chunk, or copy
- * enough for non-small chunk, flush, and output remaining
- * non-small chunk (former possible if chunk we were requested
- * to output is only slightly over 'small' size)
- */
- int needed = (mSmallWriteSize - ptr);
-
- // Just need minimal copy:
- str.getChars(offset, offset+needed, mOutputBuffer, ptr);
- mOutputPtr = ptr + needed;
- len -= needed;
- offset += needed;
- }
- flushBuffer();
- }
-
- // And then we'll just write whatever we have left:
- mOut.write(str, offset, len);
- }
-
- /*
- ////////////////////////////////////////////////
- // "Trusted" low-level output methods
- ////////////////////////////////////////////////
- */
-
- public final void writeCDataStart()
- throws IOException
- {
- fastWriteRaw("");
- }
-
- public final void writeCommentStart()
- throws IOException
- {
- fastWriteRaw("");
- }
-
- public final void writePIStart(String target, boolean addSpace)
- throws IOException
- {
- fastWriteRaw('<', '?');
- fastWriteRaw(target);
- if (addSpace) {
- fastWriteRaw(' ');
- }
- }
-
- public final void writePIEnd()
- throws IOException
- {
- fastWriteRaw('?', '>');
- }
-
- /*
- ////////////////////////////////////////////////
- // Higher-level output methods, text output
- ////////////////////////////////////////////////
- */
-
- public int writeCData(String data)
- throws IOException
- {
- if (mCheckContent) {
- int ix = verifyCDataContent(data);
- if (ix >= 0) {
- if (!mFixContent) { // Can we fix it?
- return ix;
- }
- // Yes we can! (...Bob the Builder...)
- writeSegmentedCData(data, ix);
- return -1;
- }
- }
- fastWriteRaw("");
- return -1;
- }
-
- public int writeCData(char[] cbuf, int offset, int len)
- throws IOException
- {
- if (mCheckContent) {
- int ix = verifyCDataContent(cbuf, offset, len);
- if (ix >= 0) {
- if (!mFixContent) { // Can we fix it?
- return ix;
- }
- // Yes we can! (...Bob the Builder...)
- writeSegmentedCData(cbuf, offset, len, ix);
- return -1;
- }
- }
- fastWriteRaw("");
- return -1;
- }
-
- public void writeCharacters(String text)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
- if (mTextWriter != null) { // custom escaping?
- mTextWriter.write(text);
- return;
- }
- int inPtr = 0;
- final int len = text.length();
- int highChar = mEncHighChar;
-
- main_loop:
- while (true) {
- String ent = null;
-
- inner_loop:
- while (true) {
- if (inPtr >= len) {
- break main_loop;
- }
- char c = text.charAt(inPtr++);
- if (c <= HIGHEST_ENCODABLE_TEXT_CHAR) {
- if (c <= 0x0020) {
- if (c != ' ' && c != '\n' && c != '\t') { // fine as is
- if (c == '\r') {
- if (mEscapeCR) {
- break inner_loop;
- }
- } else {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c); // throws an error usually
- } else {
- break inner_loop; // need quoting
- }
- }
- }
- } else if (c == '<') {
- ent = "<";
- break inner_loop;
- } else if (c == '&') {
- ent = "&";
- break inner_loop;
- } else if (c == '>') {
- // Let's be conservative; and if there's any
- // change it might be part of "]]>" quote it
- if (inPtr < 2 || text.charAt(inPtr-2) == ']') {
- ent = ">";
- break inner_loop;
- }
- }
- } else if (c >= highChar) {
- break inner_loop;
- }
- if (mOutputPtr >= mOutputBufLen) {
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = c;
- }
- if (ent != null) {
- writeRaw(ent);
- } else {
- writeAsEntity(text.charAt(inPtr-1));
- }
- }
- }
-
- public void writeCharacters(char[] cbuf, int offset, int len)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
-
- if (mTextWriter != null) { // custom escaping?
- mTextWriter.write(cbuf, offset, len);
- } else { // nope, default:
- len += offset;
- do {
- int c = 0;
- int highChar = mEncHighChar;
- int start = offset;
- String ent = null;
-
- for (; offset < len; ++offset) {
- c = cbuf[offset];
- if (c <= HIGHEST_ENCODABLE_TEXT_CHAR) {
- if (c == '<') {
- ent = "<";
- break;
- } else if (c == '&') {
- ent = "&";
- break;
- } else if (c == '>') {
- /* Let's be conservative; and if there's any
- * change it might be part of "]]>" quote it
- */
- if ((offset == start) || cbuf[offset-1] == ']') {
- ent = ">";
- break;
- }
- } else if (c < 0x0020) {
- if (c == '\n' || c == '\t') { // fine as is
- ;
- } else if (c == '\r') {
- if (mEscapeCR) {
- break;
- }
- } else {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c);
- // Hmmh. This is very inefficient, but...
- ent = String.valueOf((char) c);
- }
- break; // need quoting
- }
- }
- } else if (c >= highChar) {
- break;
- }
- // otherwise ok
- }
- int outLen = offset - start;
- if (outLen > 0) {
- writeRaw(cbuf, start, outLen);
- }
- if (ent != null) {
- writeRaw(ent);
- ent = null;
- } else if (offset < len) {
- writeAsEntity(c);
- }
- } while (++offset < len);
- }
- }
-
- /**
- * Method that will try to output the content as specified. If
- * the content passed in has embedded "--" in it, it will either
- * add an intervening space between consequtive hyphens (if content
- * fixing is enabled), or return the offset of the first hyphen in
- * multi-hyphen sequence.
- */
- public int writeComment(String data)
- throws IOException
- {
- if (mCheckContent) {
- int ix = verifyCommentContent(data);
- if (ix >= 0) {
- if (!mFixContent) { // Can we fix it?
- return ix;
- }
- // Yes we can! (...Bob the Builder...)
- writeSegmentedComment(data, ix);
- return -1;
- }
- }
- fastWriteRaw("");
- return -1;
- }
-
- public void writeDTD(String data)
- throws IOException
- {
- writeRaw(data);
- }
-
- public void writeDTD(String rootName, String systemId, String publicId,
- String internalSubset)
- throws IOException, XMLStreamException
- {
- fastWriteRaw(" 0) {
- fastWriteRaw(' ', '[');
- fastWriteRaw(internalSubset);
- fastWriteRaw(']');
- }
- fastWriteRaw('>');
- }
-
- public void writeEntityReference(String name)
- throws IOException, XMLStreamException
- {
- if (mCheckNames) {
- verifyNameValidity(name, mNsAware);
- }
- fastWriteRaw('&');
- fastWriteRaw(name);
- fastWriteRaw(';');
- }
-
- public void writeXmlDeclaration(String version, String encoding, String standalone)
- throws IOException
- {
- fastWriteRaw(" 0) {
- fastWriteRaw(" encoding='");
- fastWriteRaw(encoding);
- fastWriteRaw('\'');
- }
- if (standalone != null) {
- fastWriteRaw(" standalone='");
- fastWriteRaw(standalone);
- fastWriteRaw('\'');
- }
- fastWriteRaw('?', '>');
- }
-
- public int writePI(String target, String data)
- throws IOException, XMLStreamException
- {
- if (mCheckNames) {
- // As per namespace specs, can not have colon(s)
- verifyNameValidity(target, mNsAware);
- }
- fastWriteRaw('<', '?');
- fastWriteRaw(target);
- if (data != null && data.length() > 0) {
- if (mCheckContent) {
- int ix = data.indexOf('?');
- if (ix >= 0) {
- ix = data.indexOf("?>", ix);
- if (ix >= 0) {
- return ix;
- }
- }
- }
- fastWriteRaw(' ');
- // Data may be longer, let's call regular writeRaw method
- writeRaw(data);
- }
- fastWriteRaw('?', '>');
- return -1;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, elements
- ////////////////////////////////////////////////////
- */
-
- public void writeStartTagStart(String localName)
- throws IOException, XMLStreamException
- {
- if (mCheckNames) {
- verifyNameValidity(localName, mNsAware);
- }
-
- int ptr = mOutputPtr;
- int extra = (mOutputBufLen - ptr) - (1 + localName.length());
- if (extra < 0) { // split on boundary, slower
- fastWriteRaw('<');
- fastWriteRaw(localName);
- } else {
- char[] buf = mOutputBuffer;
- buf[ptr++] = '<';
- int len = localName.length();
- localName.getChars(0, len, buf, ptr);
- mOutputPtr = ptr+len;
- }
- }
-
- public void writeStartTagStart(String prefix, String localName)
- throws IOException, XMLStreamException
- {
- if (prefix == null || prefix.length() == 0) { // shouldn't happen
- writeStartTagStart(localName);
- return;
- }
-
- if (mCheckNames) {
- verifyNameValidity(prefix, mNsAware);
- verifyNameValidity(localName, mNsAware);
- }
-
- int ptr = mOutputPtr;
- int len = prefix.length();
- int extra = (mOutputBufLen - ptr) - (2 + localName.length() + len);
- if (extra < 0) { // across buffer boundary, slow case
- fastWriteRaw('<');
- fastWriteRaw(prefix);
- fastWriteRaw(':');
- fastWriteRaw(localName);
- } else { // fast case, all inlined
- char[] buf = mOutputBuffer;
- buf[ptr++] = '<';
- prefix.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = ':';
- len = localName.length();
- localName.getChars(0, len, buf, ptr);
- mOutputPtr = ptr+len;
- }
- }
-
- public void writeStartTagEnd()
- throws IOException
- {
- fastWriteRaw('>');
- }
-
- public void writeStartTagEmptyEnd()
- throws IOException
- {
- int ptr = mOutputPtr;
- if ((ptr + 3) >= mOutputBufLen) {
- if (mOut == null) {
- return;
- }
- flushBuffer();
- ptr = mOutputPtr;
- }
- char[] buf = mOutputBuffer;
- if (mAddSpaceAfterEmptyElem) {
- buf[ptr++] = ' ';
- }
- buf[ptr++] = '/';
- buf[ptr++] = '>';
- mOutputPtr = ptr;
- }
-
- public void writeEndTag(String localName)
- throws IOException
- {
- int ptr = mOutputPtr;
- int extra = (mOutputBufLen - ptr) - (3 + localName.length());
- if (extra < 0) {
- fastWriteRaw('<', '/');
- fastWriteRaw(localName);
- fastWriteRaw('>');
- } else {
- char[] buf = mOutputBuffer;
- buf[ptr++] = '<';
- buf[ptr++] = '/';
- int len = localName.length();
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '>';
- mOutputPtr = ptr;
- }
- }
-
- public void writeEndTag(String prefix, String localName)
- throws IOException
- {
- if (prefix == null || prefix.length() == 0) {
- writeEndTag(localName);
- return;
- }
- int ptr = mOutputPtr;
- int len = prefix.length();
- int extra = (mOutputBufLen - ptr) - (4 + localName.length() + len);
- if (extra < 0) {
- fastWriteRaw('<', '/');
- /* At this point, it is assumed caller knows that end tag
- * matches with start tag, and that it (by extension) has been
- * validated if and as necessary
- */
- fastWriteRaw(prefix);
- fastWriteRaw(':');
- fastWriteRaw(localName);
- fastWriteRaw('>');
- } else {
- char[] buf = mOutputBuffer;
- buf[ptr++] = '<';
- buf[ptr++] = '/';
- prefix.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = ':';
- len = localName.length();
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '>';
- mOutputPtr = ptr;
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, attributes/ns
- ////////////////////////////////////////////////////
- */
-
- public void writeAttribute(String localName, String value)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(localName, mNsAware);
- }
- int len = localName.length();
- if (((mOutputBufLen - mOutputPtr) - (3 + len)) < 0) {
- fastWriteRaw(' ');
- fastWriteRaw(localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- len = (value == null) ? 0 : value.length();
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, 0, len);
- } else { // nope, default
- writeAttrValue(value, len);
- }
- }
- fastWriteRaw('"');
- }
-
- public void writeAttribute(String localName, char[] value, int offset, int vlen)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(localName, mNsAware);
- }
- int len = localName.length();
- if (((mOutputBufLen - mOutputPtr) - (3 + len)) < 0) {
- fastWriteRaw(' ');
- fastWriteRaw(localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- if (vlen > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, offset, vlen);
- } else { // nope, default
- writeAttrValue(value, offset, vlen);
- }
- }
- fastWriteRaw('"');
- }
-
- public void writeAttribute(String prefix, String localName, String value)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(prefix, mNsAware);
- verifyNameValidity(localName, mNsAware);
- }
- int len = prefix.length();
- if (((mOutputBufLen - mOutputPtr) - (4 + localName.length() + len)) < 0) {
- fastWriteRaw(' ');
- if (len > 0) {
- fastWriteRaw(prefix);
- fastWriteRaw(':');
- }
- fastWriteRaw(localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- prefix.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = ':';
- len = localName.length();
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- len = (value == null) ? 0 : value.length();
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, 0, len);
- } else { // nope, default
- writeAttrValue(value, len);
- }
- }
- fastWriteRaw('"');
- }
-
- public void writeAttribute(String prefix, String localName, char[] value, int offset, int vlen)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(prefix, mNsAware);
- verifyNameValidity(localName, mNsAware);
- }
- int len = prefix.length();
- if (((mOutputBufLen - mOutputPtr) - (4 + localName.length() + len)) < 0) {
- fastWriteRaw(' ');
- if (len > 0) {
- fastWriteRaw(prefix);
- fastWriteRaw(':');
- }
- fastWriteRaw(localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- prefix.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = ':';
- len = localName.length();
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
- if (vlen > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, offset, vlen);
- } else { // nope, default
- writeAttrValue(value, offset, vlen);
- }
- }
- fastWriteRaw('"');
- }
-
- private final void writeAttrValue(String value, int len)
- throws IOException
- {
- int inPtr = 0;
- final char qchar = mEncQuoteChar;
- int highChar = mEncHighChar;
-
- main_loop:
- while (true) { // main_loop
- String ent = null;
-
- inner_loop:
- while (true) {
- if (inPtr >= len) {
- break main_loop;
- }
- char c = value.charAt(inPtr++);
- if (c <= HIGHEST_ENCODABLE_ATTR_CHAR) { // special char?
- if (c < 0x0020) { // tab, cr/lf need encoding too
- if (c == '\r') {
- if (mEscapeCR) {
- break inner_loop; // quoting
- }
- } else if (c != '\n' && c != '\t'
- && (!mXml11 || c == 0)) {
- c = handleInvalidChar(c);
- } else {
- break inner_loop; // need quoting
- }
- } else if (c == qchar) {
- ent = mEncQuoteEntity;
- break inner_loop;
- } else if (c == '<') {
- ent = "<";
- break inner_loop;
- } else if (c == '&') {
- ent = "&";
- break inner_loop;
- }
- } else if (c >= highChar) { // out of range, have to escape
- break inner_loop;
- }
- if (mOutputPtr >= mOutputBufLen) {
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = c;
- }
- if (ent != null) {
- writeRaw(ent);
- } else {
- writeAsEntity(value.charAt(inPtr-1));
- }
- }
- }
-
- private final void writeAttrValue(char[] value, int offset, int len)
- throws IOException
- {
- len += offset;
- final char qchar = mEncQuoteChar;
- int highChar = mEncHighChar;
-
- main_loop:
- while (true) { // main_loop
- String ent = null;
-
- inner_loop:
- while (true) {
- if (offset >= len) {
- break main_loop;
- }
- char c = value[offset++];
- if (c <= HIGHEST_ENCODABLE_ATTR_CHAR) { // special char?
- if (c < 0x0020) { // tab, cr/lf need encoding too
- if (c == '\r') {
- if (mEscapeCR) {
- break inner_loop; // quoting
- }
- } else if (c != '\n' && c != '\t'
- && (!mXml11 || c == 0)) {
- c = handleInvalidChar(c);
- } else {
- break inner_loop; // need quoting
- }
- } else if (c == qchar) {
- ent = mEncQuoteEntity;
- break inner_loop;
- } else if (c == '<') {
- ent = "<";
- break inner_loop;
- } else if (c == '&') {
- ent = "&";
- break inner_loop;
- }
- } else if (c >= highChar) { // out of range, have to escape
- break inner_loop;
- }
- if (mOutputPtr >= mOutputBufLen) {
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = c;
- }
- if (ent != null) {
- writeRaw(ent);
- } else {
- writeAsEntity(value[offset-1]);
- }
- }
- }
-
- /*
- ////////////////////////////////////////////////
- // Methods used by Typed Access API
- ////////////////////////////////////////////////
- */
-
- public final void writeTypedElement(AsciiValueEncoder enc)
- throws IOException
- {
- if (mOut == null) {
- return;
- }
-
- int free = mOutputBufLen - mOutputPtr;
- if (enc.bufferNeedsFlush(free)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- // If no flushing needed, indicates that all data was encoded
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
- }
-
- public final void writeTypedElement(AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- int free = mOutputBufLen - mOutputPtr;
- if (enc.bufferNeedsFlush(free)) {
- flush();
- }
- int start = mOutputPtr;
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- // False -> can't be sure it's the whole remaining text
- validator.validateText(mOutputBuffer, start, mOutputPtr, false);
- if (enc.isCompleted()) {
- break;
- }
- flush();
- start = mOutputPtr;
- }
- }
-
- public void writeTypedAttribute(String localName, AsciiValueEncoder enc)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(localName, mNsAware);
- }
- int len = localName.length();
- if ((mOutputPtr + 3 + len) > mOutputBufLen) {
- fastWriteRaw(' ');
- fastWriteRaw(localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- localName.getChars(0, len, buf, ptr);
- ptr += len;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- int free = mOutputBufLen - mOutputPtr;
- if (enc.bufferNeedsFlush(free)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
- fastWriteRaw('"');
- }
-
- public void writeTypedAttribute(String prefix, String localName,
- AsciiValueEncoder enc)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (mCheckNames) {
- verifyNameValidity(prefix, mNsAware);
- verifyNameValidity(localName, mNsAware);
- }
- int plen = prefix.length();
- int llen = localName.length();
-
- if ((mOutputPtr + 4 + plen + llen) > mOutputBufLen) {
- writePrefixedName(prefix, localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- if (plen > 0) {
- prefix.getChars(0, plen, buf, ptr);
- ptr += plen;
- buf[ptr++] = ':';
-
- }
- localName.getChars(0, llen, buf, ptr);
- ptr += llen;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- int free = mOutputBufLen - mOutputPtr;
- if (enc.bufferNeedsFlush(free)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
-
- fastWriteRaw('"');
- }
-
- public void writeTypedAttribute(String prefix, String localName, String nsURI,
- AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException
- {
- if (mOut == null) {
- return;
- }
- if (prefix == null) {
- prefix = "";
- }
- if (nsURI == null) {
- nsURI = "";
- }
- int plen = prefix.length();
- if (mCheckNames) {
- if (plen > 0) {
- verifyNameValidity(prefix, mNsAware);
- }
- verifyNameValidity(localName, mNsAware);
- }
- if (((mOutputBufLen - mOutputPtr) - (4 + localName.length() + plen)) < 0) {
- writePrefixedName(prefix, localName);
- fastWriteRaw('=', '"');
- } else {
- int ptr = mOutputPtr;
- char[] buf = mOutputBuffer;
- buf[ptr++] = ' ';
- if (plen > 0) {
- prefix.getChars(0, plen, buf, ptr);
- ptr += plen;
- buf[ptr++] = ':';
-
- }
- int llen = localName.length();
- localName.getChars(0, llen, buf, ptr);
- ptr += llen;
- buf[ptr++] = '=';
- buf[ptr++] = '"';
- mOutputPtr = ptr;
- }
-
- /* Tricky here is this: attributes to validate can not be
- * split (validators expect complete values). So, if value
- * won't fit as is, may need to aggregate using StringBuilder
- */
- int free = mOutputBufLen - mOutputPtr;
- if (enc.bufferNeedsFlush(free)) {
- flush();
- }
- int start = mOutputPtr;
-
- // First, let's see if one call is enough
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- if (enc.isCompleted()) { // yup
- validator.validateAttribute(localName, nsURI, prefix, mOutputBuffer, start, mOutputPtr);
- return;
- }
-
- // If not, must combine first
- StringBuffer sb = new StringBuffer(mOutputBuffer.length << 1);
- sb.append(mOutputBuffer, start, mOutputPtr-start);
- while (true) {
- flush();
- start = mOutputPtr;
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBufLen);
- sb.append(mOutputBuffer, start, mOutputPtr-start);
- // All done?
- if (enc.isCompleted()) {
- break;
- }
- }
- fastWriteRaw('"');
-
- // Then validate
- String valueStr = sb.toString();
- validator.validateAttribute(localName, nsURI, prefix, valueStr);
- }
-
- protected final void writePrefixedName(String prefix, String localName)
- throws IOException
- {
- fastWriteRaw(' ');
- if (prefix.length() > 0) {
- fastWriteRaw(prefix);
- fastWriteRaw(':');
- }
- fastWriteRaw(localName);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, buffering
- ////////////////////////////////////////////////////
- */
-
- private final void flushBuffer()
- throws IOException
- {
- if (mOutputPtr > 0 && mOutputBuffer != null) {
- int ptr = mOutputPtr;
- // Need to update location info, to keep it in sync
- mLocPastChars += ptr;
- mLocRowStartOffset -= ptr;
- mOutputPtr = 0;
- mOut.write(mOutputBuffer, 0, ptr);
- }
- }
-
- private final void fastWriteRaw(char c)
- throws IOException
- {
- if (mOutputPtr >= mOutputBufLen) {
- if (mOut == null) {
- return;
- }
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = c;
- }
-
- private final void fastWriteRaw(char c1, char c2)
- throws IOException
- {
- if ((mOutputPtr + 1) >= mOutputBufLen) {
- if (mOut == null) {
- return;
- }
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = c1;
- mOutputBuffer[mOutputPtr++] = c2;
- }
-
- private final void fastWriteRaw(String str)
- throws IOException
- {
- int len = str.length();
- int ptr = mOutputPtr;
- if ((ptr + len) >= mOutputBufLen) {
- if (mOut == null) {
- return;
- }
- /* It's even possible that String is longer than the buffer (not
- * likely, possible). If so, let's just call the full
- * method:
- */
- if (len > mOutputBufLen) {
- writeRaw(str);
- return;
- }
- flushBuffer();
- ptr = mOutputPtr;
- }
- str.getChars(0, len, mOutputBuffer, ptr);
- mOutputPtr = ptr+len;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods, content verification/fixing
- ////////////////////////////////////////////////////
- */
-
- /**
- * @return Index at which a problem was found, if any; -1 if there's
- * no problem.
- */
- protected int verifyCDataContent(String content)
- {
- if (content != null && content.length() >= 3) {
- int ix = content.indexOf(']');
- if (ix >= 0) {
- return content.indexOf("]]>", ix);
- }
- }
- return -1;
- }
-
- protected int verifyCDataContent(char[] c, int start, int end)
- {
- if (c != null) {
- start += 2;
- /* Let's do simple optimization for search...
- * (simple bayer-moore - like algorithm)
- */
- while (start < end) {
- char ch = c[start];
- if (ch == ']') {
- ++start; // let's just move by one in this case
- continue;
- }
- if (ch == '>') { // match?
- if (c[start-1] == ']'
- && c[start-2] == ']') {
- return start-2;
- }
- }
- start += 2;
- }
- }
- return -1;
- }
-
- protected int verifyCommentContent(String content)
- {
- int ix = content.indexOf('-');
- if (ix >= 0) {
- /* actually, it's illegal to just end with '-' too, since
- * that would cause invalid end marker '--->'
- */
- if (ix < (content.length() - 1)) {
- ix = content.indexOf("--", ix);
- }
- }
- return ix;
- }
-
- protected void writeSegmentedCData(String content, int index)
- throws IOException
- {
- /* It's actually fairly easy, just split "]]>" into 2 pieces;
- * for each ']]>'; first one containing "]]", second one ">"
- * (as long as necessary)
- */
- int start = 0;
- while (index >= 0) {
- fastWriteRaw("");
- start = index+2;
- index = content.indexOf("]]>", start);
- }
- // Ok, then the last segment
- fastWriteRaw("");
- }
-
- protected void writeSegmentedCData(char[] c, int start, int len, int index)
- throws IOException
- {
- int end = start + len;
- while (index >= 0) {
- fastWriteRaw("");
- start = index+2;
- index = verifyCDataContent(c, start, end);
- }
- // Ok, then the last segment
- fastWriteRaw("");
- }
-
- protected void writeSegmentedComment(String content, int index)
- throws IOException
- {
- int len = content.length();
- // First the special case (last char is hyphen):
- if (index == (len-1)) {
- fastWriteRaw("");
- return;
- }
-
- /* Fixing comments is more difficult than that of CDATA segments';
- * this because CDATA can still contain embedded ']]'s, but
- * comment neither allows '--' nor ending with '-->'; which means
- * that it's impossible to just split segments. Instead we'll do
- * something more intrusive, and embed single spaces between all
- * '--' character pairs... it's intrusive, but comments are not
- * supposed to contain any data, so that should be fine (plus
- * at least result is valid, unlike contents as is)
- */
- fastWriteRaw("");
- }
-
- /**
- * Method used to figure out which part of the Unicode char set the
- * encoding can natively support. Values returned are 7, 8 and 16,
- * to indicate (respectively) "ascii", "ISO-Latin" and "native Unicode".
- * These just best guesses, but should work ok for the most common
- * encodings.
- */
- public static int guessEncodingBitSize(String enc)
- {
- if (enc == null || enc.length() == 0) { // let's assume default is UTF-8...
- return 16;
- }
-
- // Let's see if we can find a normalized name, first:
- enc = CharsetNames.normalize(enc);
-
- // Ok, first, do we have known ones; starting with most common:
- if (enc == CharsetNames.CS_UTF8) {
- return 16; // meaning up to 2^16 can be represented natively
- } else if (enc == CharsetNames.CS_ISO_LATIN1) {
- return 8;
- } else if (enc == CharsetNames.CS_US_ASCII) {
- return 7;
- } else if (enc == CharsetNames.CS_UTF16
- || enc == CharsetNames.CS_UTF16BE
- || enc == CharsetNames.CS_UTF16LE
- || enc == CharsetNames.CS_UTF32BE
- || enc == CharsetNames.CS_UTF32LE) {
- return 16;
- }
-
- /* Above and beyond well-recognized names, it might still be
- * good to have more heuristics for as-of-yet unhandled cases...
- * But, it's probably easier to only assume 8-bit clean (could
- * even make it just 7, let's see how this works out)
- */
- return 8;
- }
-
- protected final void writeAsEntity(int c)
- throws IOException
- {
- char[] buf = mOutputBuffer;
- int ptr = mOutputPtr;
- if ((ptr + 10) >= buf.length) { // &#x [up to 6 hex digits] ;
- flushBuffer();
- ptr = mOutputPtr;
- }
- buf[ptr++] = '&';
-
- // Can use more optimal notation for 8-bit ascii stuff:
- if (c < 256) {
- /* Also; although not really mandatory, let's also
- * use pre-defined entities where possible.
- */
- if (c == '&') {
- buf[ptr++] = 'a';
- buf[ptr++] = 'm';
- buf[ptr++] = 'p';
- } else if (c == '<') {
- buf[ptr++] = 'l';
- buf[ptr++] = 't';
- } else if (c == '>') {
- buf[ptr++] = 'g';
- buf[ptr++] = 't';
- } else if (c == '\'') {
- buf[ptr++] = 'a';
- buf[ptr++] = 'p';
- buf[ptr++] = 'o';
- buf[ptr++] = 's';
- } else if (c == '"') {
- buf[ptr++] = 'q';
- buf[ptr++] = 'u';
- buf[ptr++] = 'o';
- buf[ptr++] = 't';
- } else {
- buf[ptr++] = '#';;
- buf[ptr++] = 'x';;
- // Can use shortest quoting for tab, cr, lf:
- if (c >= 16) {
- int digit = (c >> 4);
- buf[ptr++] = (char) ((digit < 10) ? ('0' + digit) : (('a' - 10) + digit));
- c &= 0xF;
- }
- buf[ptr++] = (char) ((c < 10) ? ('0' + c) : (('a' - 10) + c));
- }
- } else {
- buf[ptr++] = '#';
- buf[ptr++] = 'x';
-
- // Ok, let's write the shortest possible sequence then:
- int shift = 20;
- int origPtr = ptr;
-
- do {
- int digit = (c >> shift) & 0xF;
- if (digit > 0 || (ptr != origPtr)) {
- buf[ptr++] = (char) ((digit < 10) ? ('0' + digit) : (('a' - 10) + digit));
- }
- shift -= 4;
- } while (shift > 0);
- c &= 0xF;
- buf[ptr++] = (char) ((c < 10) ? ('0' + c) : (('a' - 10) + c));
- }
- buf[ptr++] = ';';
- mOutputPtr = ptr;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/EncodingXmlWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/EncodingXmlWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/EncodingXmlWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/EncodingXmlWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,934 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.*;
-import javax.xml.stream.XMLStreamException;
-
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.XmlConsts;
-
-/**
- * Intermediate base class used when outputting to streams that use
- * an encoding that is compatible with 7-bit single-byte Ascii encoding.
- * That means it can be used for UTF-8, ISO-Latin1 and pure Ascii.
- *writeAscii method, since it is the most convenient
- * place to catch cases where a text segment ends with an unmatched
- * surrogate pair half.
- */
-public abstract class EncodingXmlWriter
- extends XmlWriter
-{
- /**
- * Let's use a typical default to have a compromise between large
- * enough chunks to output, and minimizing memory overhead.
- * 4k should be close enough to a physical page to work out
- * acceptably, without causing excessive (if temporary) memory usage.
- */
- final static int DEFAULT_BUFFER_SIZE = 4000;
-
- final static byte BYTE_SPACE = (byte) ' ';
- final static byte BYTE_COLON = (byte) ':';
- final static byte BYTE_SEMICOLON = (byte) ';';
- final static byte BYTE_LBRACKET = (byte) '[';
- final static byte BYTE_RBRACKET = (byte) ']';
- final static byte BYTE_QMARK = (byte) '?';
- final static byte BYTE_EQ = (byte) '=';
- final static byte BYTE_SLASH = (byte) '/';
- final static byte BYTE_HASH = (byte) '#';
- final static byte BYTE_HYPHEN = (byte) '-';
-
- final static byte BYTE_LT = (byte) '<';
- final static byte BYTE_GT = (byte) '>';
- final static byte BYTE_AMP = (byte) '&';
- final static byte BYTE_QUOT = (byte) '"';
- final static byte BYTE_APOS = (byte) '\'';
-
- final static byte BYTE_A = (byte) 'a';
- final static byte BYTE_G = (byte) 'g';
- final static byte BYTE_L = (byte) 'l';
- final static byte BYTE_M = (byte) 'm';
- final static byte BYTE_O = (byte) 'o';
- final static byte BYTE_P = (byte) 'p';
- final static byte BYTE_Q = (byte) 'q';
- final static byte BYTE_S = (byte) 's';
- final static byte BYTE_T = (byte) 't';
- final static byte BYTE_U = (byte) 'u';
- final static byte BYTE_X = (byte) 'x';
-
- /*
- ////////////////////////////////////////////////
- // Output state, buffering
- ////////////////////////////////////////////////
- */
-
- /**
- * Actual output stream to use for outputting encoded content as
- * bytes.
- */
- private final OutputStream mOut;
-
- protected byte[] mOutputBuffer;
-
- protected int mOutputPtr;
-
- /**
- * In case a split surrogate pair is output (which can only successfully
- * occur with either writeRaw or
- * writeCharacters), the first part is temporarily stored
- * within this member variable.
- */
- protected int mSurrogate = 0;
-
- /*
- ////////////////////////////////////////////////
- //
- ////////////////////////////////////////////////
- */
-
- public EncodingXmlWriter(OutputStream out, WriterConfig cfg, String encoding,
- boolean autoclose)
- throws IOException
- {
- super(cfg, encoding, autoclose);
- mOut = out;
- mOutputBuffer = cfg.allocFullBBuffer(DEFAULT_BUFFER_SIZE);
- mOutputPtr = 0;
- }
-
- /**
- * This method is needed by the super class, to calculate hard
- * byte/char offsets.
- */
- protected int getOutputPtr() {
- return mOutputPtr;
- }
-
- /*
- ////////////////////////////////////////////////
- // Partial API implementation
- ////////////////////////////////////////////////
- */
-
- final protected OutputStream getOutputStream()
- {
- return mOut;
- }
-
- final protected Writer getWriter()
- {
- // No writers are involved with these implementations...
- return null;
- }
-
- public void close(boolean forceRealClose)
- throws IOException
- {
- flush();
-
- // Buffers to free?
- byte[] buf = mOutputBuffer;
- if (buf != null) {
- mOutputBuffer = null;
- mConfig.freeFullBBuffer(buf);
- }
- // Plus may need to close the actual stream
- if (forceRealClose || mAutoCloseOutput) {
- /* 14-Nov-2008, TSa: Wrt [WSTX-163]; no need to
- * check whether mOut implements CompletelyCloseable
- * (unlike with BufferingXmlWriter)
- */
- mOut.close();
- }
- }
-
- public final void flush()
- throws IOException
- {
- flushBuffer();
- mOut.flush();
- }
-
- public abstract void writeRaw(char[] cbuf, int offset, int len)
- throws IOException;
-
- public abstract void writeRaw(String str, int offset, int len)
- throws IOException;
-
- /*
- //////////////////////////////////////////////////
- // "Trusted" low-level output methods (that do not
- // need to verify validity of input)
- //////////////////////////////////////////////////
- */
-
- public final void writeCDataStart()
- throws IOException
- {
- writeAscii("");
- }
-
- public final void writeCommentStart()
- throws IOException
- {
- writeAscii("");
- }
-
- public final void writePIStart(String target, boolean addSpace)
- throws IOException
- {
- writeAscii(BYTE_LT, BYTE_QMARK);
- writeRaw(target);
- if (addSpace) {
- writeAscii(BYTE_SPACE);
- }
- }
-
- public final void writePIEnd()
- throws IOException
- {
- writeAscii(BYTE_QMARK, BYTE_GT);
- }
-
- /*
- ////////////////////////////////////////////////
- // Higher-level output methods, text output
- ////////////////////////////////////////////////
- */
-
- public int writeCData(String data)
- throws IOException
- {
- writeAscii("= 0) {
- return ix;
- }
- writeAscii("]]>");
- return -1;
- }
-
- public int writeCData(char[] cbuf, int offset, int len)
- throws IOException
- {
- writeAscii("= 0) {
- return ix;
- }
- writeAscii("]]>");
- return -1;
- }
-
- public final void writeCharacters(String data)
- throws IOException
- {
- // Note: may get second part of a surrogate
- if (mTextWriter != null) { // custom escaping?
- mTextWriter.write(data);
- } else { // nope, default:
- writeTextContent(data);
- }
- }
-
- public final void writeCharacters(char[] cbuf, int offset, int len)
- throws IOException
- {
- // Note: may get second part of a surrogate
- if (mTextWriter != null) { // custom escaping?
- mTextWriter.write(cbuf, offset, len);
- } else { // nope, default:
- writeTextContent(cbuf, offset, len);
- }
- }
-
- /**
- * Method that will try to output the content as specified. If
- * the content passed in has embedded "--" in it, it will either
- * add an intervening space between consequtive hyphens (if content
- * fixing is enabled), or return the offset of the first hyphen in
- * multi-hyphen sequence.
- */
- public int writeComment(String data)
- throws IOException
- {
- writeAscii("");
- return -1;
- }
-
- public void writeDTD(String data)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- writeRaw(data, 0, data.length());
- }
-
- public void writeDTD(String rootName, String systemId, String publicId,
- String internalSubset)
- throws IOException, XMLStreamException
- {
- writeAscii(" 0) {
- writeAscii(BYTE_SPACE, BYTE_LBRACKET);
- writeRaw(internalSubset, 0, internalSubset.length());
- writeAscii(BYTE_RBRACKET);
- }
- writeAscii(BYTE_GT);
- }
-
- public void writeEntityReference(String name)
- throws IOException, XMLStreamException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- writeAscii(BYTE_AMP);
- writeName(name);
- writeAscii(BYTE_SEMICOLON);
- }
-
- public void writeXmlDeclaration(String version, String encoding, String standalone)
- throws IOException
- {
- writeAscii(" 0) {
- writeAscii(" encoding='");
- // Should be ascii, but let's play it safe:
- writeRaw(encoding, 0, encoding.length());
- writeAscii(BYTE_APOS);
- }
- if (standalone != null) {
- writeAscii(" standalone='");
- writeAscii(standalone);
- writeAscii(BYTE_APOS);
- }
- writeAscii(BYTE_QMARK, BYTE_GT);
- }
-
- public int writePI(String target, String data)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_LT, BYTE_QMARK);
- writeName(target);
- if (data != null && data.length() > 0) {
- writeAscii(BYTE_SPACE);
- int ix = writePIData(data);
- if (ix >= 0) { // embedded "?>"?
- return ix;
- }
- }
- writeAscii(BYTE_QMARK, BYTE_GT);
- return -1;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, elements
- ////////////////////////////////////////////////////
- */
-
- public void writeStartTagStart(String localName)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_LT);
- writeName(localName);
- }
-
- public void writeStartTagStart(String prefix, String localName)
- throws IOException, XMLStreamException
- {
- if (prefix == null || prefix.length() == 0) {
- writeStartTagStart(localName);
- return;
- }
- writeAscii(BYTE_LT);
- writeName(prefix);
- writeAscii(BYTE_COLON);
- writeName(localName);
- }
-
- public void writeStartTagEnd()
- throws IOException
- {
- writeAscii(BYTE_GT);
- }
-
- public void writeStartTagEmptyEnd()
- throws IOException
- {
- if (mAddSpaceAfterEmptyElem) {
- writeAscii(" />");
- } else {
- writeAscii(BYTE_SLASH, BYTE_GT);
- }
- }
-
- public void writeEndTag(String localName)
- throws IOException
- {
- writeAscii(BYTE_LT, BYTE_SLASH);
- /* At this point, it is assumed caller knows that end tag
- * matches with start tag, and that it (by extension) has been
- * validated if and as necessary
- */
- writeNameUnchecked(localName);
- writeAscii(BYTE_GT);
- }
-
- public void writeEndTag(String prefix, String localName)
- throws IOException
- {
- writeAscii(BYTE_LT, BYTE_SLASH);
- /* At this point, it is assumed caller knows that end tag
- * matches with start tag, and that it (by extension) has been
- * validated if and as necessary
- */
- if (prefix != null && prefix.length() > 0) {
- writeNameUnchecked(prefix);
- writeAscii(BYTE_COLON);
- }
- writeNameUnchecked(localName);
- writeAscii(BYTE_GT);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, attributes/ns
- ////////////////////////////////////////////////////
- */
-
- public void writeAttribute(String localName, String value)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- int len = value.length();
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, 0, len);
- } else { // nope, default
- writeAttrValue(value);
- }
- }
- writeAscii(BYTE_QUOT);
- }
-
- public void writeAttribute(String localName, char[] value, int offset, int len)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, offset, len);
- } else { // nope, default
- writeAttrValue(value, offset, len);
- }
- }
- writeAscii(BYTE_QUOT);
- }
-
- public void writeAttribute(String prefix, String localName, String value)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(prefix);
- writeAscii(BYTE_COLON);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- int len = value.length();
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, 0, len);
- } else { // nope, default
- writeAttrValue(value);
- }
- }
- writeAscii(BYTE_QUOT);
- }
-
- public void writeAttribute(String prefix, String localName, char[] value, int offset, int len)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(prefix);
- writeAscii(BYTE_COLON);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- if (len > 0) {
- if (mAttrValueWriter != null) { // custom escaping?
- mAttrValueWriter.write(value, offset, len);
- } else { // nope, default
- writeAttrValue(value, offset, len);
- }
- }
- writeAscii(BYTE_QUOT);
- }
-
- /*
- ////////////////////////////////////////////////
- // Methods used by Typed Access API
- ////////////////////////////////////////////////
- */
-
- /**
- * Non-validating version of typed write method
- */
- public final void writeTypedElement(AsciiValueEncoder enc)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- if (enc.bufferNeedsFlush(mOutputBuffer.length - mOutputPtr)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBuffer.length);
- // If no flushing needed, indicates that all data was encoded
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
- }
-
- /**
- * Validating version of typed write method
- */
- public final void writeTypedElement(AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
-
- /* Ok, this gets trickier: can't use efficient direct-to-bytes
- * encoding since validator won't be able to use it. Instead
- * have to use temporary copy buffer.
- */
- final int copyBufferLen = copyBuffer.length;
-
- // Copy buffer should never be too small, no need to check up front
- do {
- int ptr = enc.encodeMore(copyBuffer, 0, copyBufferLen);
-
- // False -> can't be sure it's the whole remaining text
- validator.validateText(copyBuffer, 0, ptr, false);
- writeRawAscii(copyBuffer, 0, ptr);
- } while (!enc.isCompleted());
- }
-
- public void writeTypedAttribute(String localName, AsciiValueEncoder enc)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- if (enc.bufferNeedsFlush(mOutputBuffer.length - mOutputPtr)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBuffer.length);
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
- writeAscii(BYTE_QUOT);
- }
-
- public void writeTypedAttribute(String prefix, String localName,
- AsciiValueEncoder enc)
- throws IOException, XMLStreamException
- {
- writeAscii(BYTE_SPACE);
- writeName(prefix);
- writeAscii(BYTE_COLON);
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- if (enc.bufferNeedsFlush(mOutputBuffer.length - mOutputPtr)) {
- flush();
- }
- while (true) {
- mOutputPtr = enc.encodeMore(mOutputBuffer, mOutputPtr, mOutputBuffer.length);
- if (enc.isCompleted()) {
- break;
- }
- flush();
- }
- writeAscii(BYTE_QUOT);
- }
-
- public void writeTypedAttribute(String prefix, String localName, String nsURI,
- AsciiValueEncoder enc,
- XMLValidator validator, char[] copyBuffer)
- throws IOException, XMLStreamException
- {
- boolean hasPrefix = (prefix != null && prefix.length() > 0);
- if (nsURI == null) {
- nsURI = "";
- }
- //validator.validateAttribute(localName, nsURI, (hasPrefix ? prefix: ""), buf, offset, len);
-
- writeAscii(BYTE_SPACE);
- if (hasPrefix) {
- writeName(prefix);
- writeAscii(BYTE_COLON);
- }
- writeName(localName);
- writeAscii(BYTE_EQ, BYTE_QUOT);
-
- /* Ok, this gets trickier: can't use efficient direct-to-bytes
- * encoding since validator won't be able to use it. Instead
- * have to use temporary copy buffer.
- * In addition, attributes to validate can not be
- * split (validators expect complete values). So, if value
- * won't fit as is, may need to aggregate using StringBuilder
- */
- final int copyBufferLen = copyBuffer.length;
-
- // First, let's see if one call is enough
- int last = enc.encodeMore(copyBuffer, 0, copyBufferLen);
- writeRawAscii(copyBuffer, 0, last);
- if (enc.isCompleted()) {
- validator.validateAttribute(localName, nsURI, prefix, copyBuffer, 0, last);
- return;
- }
-
- // If not, must combine first
- StringBuffer sb = new StringBuffer(copyBufferLen << 1);
- sb.append(copyBuffer, 0, last);
- do {
- last = enc.encodeMore(copyBuffer, 0, copyBufferLen);
- writeRawAscii(copyBuffer, 0, last);
- sb.append(copyBuffer, 0, last);
- } while (!enc.isCompleted());
-
- writeAscii(BYTE_QUOT);
-
- // Then validate
- String valueStr = sb.toString();
- validator.validateAttribute(localName, nsURI, prefix, valueStr);
-
- return;
- }
-
- /*
- ////////////////////////////////////////////////
- // Methods for sub-classes to use
- ////////////////////////////////////////////////
- */
-
- protected final void flushBuffer()
- throws IOException
- {
- if (mOutputPtr > 0 && mOutputBuffer != null) {
- int ptr = mOutputPtr;
- mOutputPtr = 0;
- mOut.write(mOutputBuffer, 0, ptr);
- }
- }
-
- protected final void writeAscii(byte b)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- if (mOutputPtr >= mOutputBuffer.length) {
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = b;
- }
-
- protected final void writeAscii(byte b1, byte b2)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- if ((mOutputPtr + 1) >= mOutputBuffer.length) {
- flushBuffer();
- }
- mOutputBuffer[mOutputPtr++] = b1;
- mOutputBuffer[mOutputPtr++] = b2;
- }
-
- protected final void writeAscii(String str)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
-
- int len = str.length();
- int ptr = mOutputPtr;
- byte[] buf = mOutputBuffer;
- if ((ptr + len) >= buf.length) {
- /* It's even possible that String is longer than the buffer (not
- * likely, possible). If so, let's just call the full
- * method:
- */
- if (len > buf.length) {
- writeRaw(str, 0, len);
- return;
- }
- flushBuffer();
- ptr = mOutputPtr;
- }
- mOutputPtr += len;
- for (int i = 0; i < len; ++i) {
- buf[ptr++] = (byte)str.charAt(i);
- }
- }
-
- public final void writeRawAscii(char[] buf, int offset, int len)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- int ptr = mOutputPtr;
- byte[] dst = mOutputBuffer;
- if ((ptr + len) >= dst.length) {
- if (len > dst.length) {
- writeRaw(buf, offset, len);
- return;
- }
- flushBuffer();
- ptr = mOutputPtr;
- }
- mOutputPtr += len;
- for (int i = 0; i < len; ++i) {
- dst[ptr+i] = (byte)buf[offset+i];
- }
- }
-
- /**
- * Entity writing can be optimized quite nicely, since it only
- * needs to output ascii characters.
- *
- * @return New value of mOutputPtr
- */
- protected final int writeAsEntity(int c)
- throws IOException
- {
- byte[] buf = mOutputBuffer;
- int ptr = mOutputPtr;
- if ((ptr + 10) >= buf.length) { // &#x [up to 6 hex digits] ;
- flushBuffer();
- ptr = mOutputPtr;
- }
- buf[ptr++] = BYTE_AMP;
-
- // Can use more optimal notation for 8-bit ascii stuff:
- if (c < 256) {
- /* Also; although not really mandatory, let's also
- * use pre-defined entities where possible.
- */
- if (c == '&') {
- buf[ptr++] = BYTE_A;
- buf[ptr++] = BYTE_M;
- buf[ptr++] = BYTE_P;
- } else if (c == '<') {
- buf[ptr++] = BYTE_L;
- buf[ptr++] = BYTE_T;
- } else if (c == '>') {
- buf[ptr++] = BYTE_G;
- buf[ptr++] = BYTE_T;
- } else if (c == '\'') {
- buf[ptr++] = BYTE_A;
- buf[ptr++] = BYTE_P;
- buf[ptr++] = BYTE_O;
- buf[ptr++] = BYTE_S;
- } else if (c == '"') {
- buf[ptr++] = BYTE_Q;
- buf[ptr++] = BYTE_U;
- buf[ptr++] = BYTE_O;
- buf[ptr++] = BYTE_T;
- } else {
- buf[ptr++] = BYTE_HASH;
- buf[ptr++] = BYTE_X;
- // Can use shortest quoting for tab, cr, lf:
- if (c >= 16) {
- int digit = (c >> 4);
- buf[ptr++] = (byte) ((digit < 10) ? ('0' + digit) : (('a' - 10) + digit));
- c &= 0xF;
- }
- buf[ptr++] = (byte) ((c < 10) ? ('0' + c) : (('a' - 10) + c));
- }
- } else {
- buf[ptr++] = BYTE_HASH;
- buf[ptr++] = BYTE_X;
-
- // Ok, let's write the shortest possible sequence then:
- int shift = 20;
- int origPtr = ptr;
-
- do {
- int digit = (c >> shift) & 0xF;
- if (digit > 0 || (ptr != origPtr)) {
- buf[ptr++] = (byte) ((digit < 10) ? ('0' + digit) : (('a' - 10) + digit));
- }
- shift -= 4;
- } while (shift > 0);
- c &= 0xF;
- buf[ptr++] = (byte) ((c < 10) ? ('0' + c) : (('a' - 10) + c));
- }
- buf[ptr++] = BYTE_SEMICOLON;
- mOutputPtr = ptr;
- return ptr;
- }
-
- protected final void writeName(String name)
- throws IOException, XMLStreamException
- {
- if (mCheckNames) {
- verifyNameValidity(name, mNsAware);
- }
- // TODO: maybe we could reuse some previously encoded names?
- writeRaw(name, 0, name.length());
- }
-
- protected final void writeNameUnchecked(String name)
- throws IOException
- {
- writeRaw(name, 0, name.length());
- }
-
- protected final int calcSurrogate(int secondSurr)
- throws IOException
- {
- // First, let's verify first surrogate is valid:
- int firstSurr = mSurrogate;
- mSurrogate = 0;
- if (firstSurr < SURR1_FIRST || firstSurr > SURR1_LAST) {
- throwUnpairedSurrogate(firstSurr);
- }
-
- // Then that the second one is:
- if ((secondSurr < SURR2_FIRST) || (secondSurr > SURR2_LAST)) {
- throwUnpairedSurrogate(secondSurr);
- }
- int ch = 0x10000 + ((firstSurr - SURR1_FIRST) << 10) + (secondSurr - SURR2_FIRST);
- if (ch > XmlConsts.MAX_UNICODE_CHAR) {
- throw new IOException("Illegal surrogate character pair, resulting code 0x"+Integer.toHexString(ch)+" above legal XML character range");
- }
- return ch;
- }
-
- protected final void throwUnpairedSurrogate()
- throws IOException
- {
- int surr = mSurrogate;
- mSurrogate = 0;
- throwUnpairedSurrogate(surr);
- }
-
- protected final void throwUnpairedSurrogate(int code)
- throws IOException
-{
- // Let's flush to make debugging easier
- flush();
- throw new IOException("Unpaired surrogate character (0x"+Integer.toHexString(code)+")");
-}
-
- /*
- ////////////////////////////////////////////////
- // Abstract methods for sub-classes to define
- ////////////////////////////////////////////////
- */
-
- protected abstract void writeAttrValue(String data)
- throws IOException;
-
- protected abstract void writeAttrValue(char[] value, int offset, int len)
- throws IOException;
-
- protected abstract int writeCDataContent(String data)
- throws IOException;
-
- protected abstract int writeCDataContent(char[] cbuf, int start, int len)
- throws IOException;
-
- protected abstract int writeCommentContent(String data)
- throws IOException;
-
- protected abstract int writePIData(String data)
- throws IOException, XMLStreamException;
-
- protected abstract void writeTextContent(String data)
- throws IOException;
-
- protected abstract void writeTextContent(char[] cbuf, int start, int len)
- throws IOException;
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/ISOLatin1XmlWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/ISOLatin1XmlWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/ISOLatin1XmlWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/ISOLatin1XmlWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,791 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.*;
-
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.io.CharsetNames;
-
-/**
- * Concrete implementation of {@link EncodingXmlWriter} used when output
- * is to be encoded using ISO-8859-1, aka ISO-Latin1 encoding.
- *writeRaw
- * methods.
- */
-public final class ISOLatin1XmlWriter
- extends EncodingXmlWriter
-{
- public ISOLatin1XmlWriter(OutputStream out, WriterConfig cfg, boolean autoclose)
- throws IOException
- {
- super(out, cfg, CharsetNames.CS_ISO_LATIN1, autoclose);
- }
-
- public void writeRaw(char[] cbuf, int offset, int len)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
-
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- if (mCheckContent) {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = cbuf[offset];
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- } else {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- mOutputBuffer[ptr++] = (byte) cbuf[offset];
- }
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- public void writeRaw(String str, int offset, int len)
- throws IOException
- {
- if (mSurrogate != 0) {
- throwUnpairedSurrogate();
- }
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- if (mCheckContent) {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = str.charAt(offset);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- } else {
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- mOutputBuffer[ptr++] = (byte) str.charAt(offset);
- }
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected void writeAttrValue(String data)
- throws IOException
- {
- int offset = 0;
- int len = data.length();
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data.charAt(offset++);
- sec = calcSurrogate(sec);
- mOutputPtr = ptr;
- ptr = writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- /* Need to quote all white space except for regular
- * space chars, to preserve them (round-tripping)
- */
- if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- } else if (c != '\n' && c != '\t') {
- if (mCheckContent) {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c);
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- }
- }
- // fall-through to char entity output
- } else if (c < 0x7F) {
- if (c != '<' && c != '&' && c != '"') {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- // otherwise fall back on quoting
- } else if (c > 0x9F && c <= 0xFF) {
- mOutputBuffer[ptr++] = (byte) c;
- continue; // [WSTX-88]
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data.charAt(offset++));
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- mOutputPtr = ptr;
- ptr = writeAsEntity(c);
- len = data.length() - offset;
- continue main_loop;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected void writeAttrValue(char[] data, int offset, int len)
- throws IOException
- {
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data[offset++];
- sec = calcSurrogate(sec);
- mOutputPtr = ptr;
- ptr = writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data[offset++];
- if (c < 32) {
- /* Need to quote all white space except for regular
- * space chars, to preserve them (round-tripping)
- */
- if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- } else if (c != '\n' && c != '\t') {
- if (mCheckContent) {
- if (!mXml11 || c == 0) {
- c = handleInvalidChar(c);
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- }
- }
- // fall-through to char entity output
- } else if (c < 0x7F) {
- if (c != '<' && c != '&' && c != '"') {
- mOutputBuffer[ptr++] = (byte) c;
- continue;
- }
- // otherwise fall back on quoting
- } else if (c > 0x9F && c <= 0xFF) {
- mOutputBuffer[ptr++] = (byte) c;
- continue; // [WSTX-88]
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data[offset++]);
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. contiguous data we can output
- */
- mOutputPtr = ptr;
- ptr = writeAsEntity(c);
- max -= (inEnd - offset); // since we didn't loop completely
- break inner_loop;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- }
-
- protected int writeCDataContent(String data)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
- int ptr = mOutputPtr;
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- } else if (c == '>') { // embedded "]]>"?
- if (offset > 2 && data.charAt(offset-2) == ']'
- && data.charAt(offset-3) == ']') {
- if (!mFixContent) {
- return offset-3;
- }
- /* Relatively easy fix; just need to close this
- * section, and open a new one...
- */
- mOutputPtr = ptr;
- writeCDataEnd();
- writeCDataStart();
- writeAscii(BYTE_GT);
- ptr = mOutputPtr;
- /* No guarantees there's as much free room in the
- * output buffer, thus, need to restart loop:
- */
- len = data.length() - offset;
- continue main_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writeCDataContent(char[] cbuf, int start, int len)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- if (!mCheckContent) {
- writeRaw(cbuf, start, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
- int offset = start;
-
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = cbuf[offset++];
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- } else if (c == '>') { // embedded "]]>"?
- if (offset >= (start+3) && cbuf[offset-2] == ']'
- && cbuf[offset-3] == ']') {
- if (!mFixContent) {
- return offset-3;
- }
- /* Relatively easy fix; just need to close this
- * section, and open a new one...
- */
- mOutputPtr = ptr;
- writeCDataEnd();
- writeCDataStart();
- writeAscii(BYTE_GT);
- ptr = mOutputPtr;
- /* No guarantees there's as much free room in the
- * output buffer, thus, need to restart loop:
- */
- max -= (inEnd - offset);
- break inner_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writeCommentContent(String data)
- throws IOException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
-
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- } else if (c == '-') { // embedded "--"?
- if (offset > 1 && data.charAt(offset-2) == '-') {
- if (!mFixContent) {
- return offset-2;
- }
- /* Quite easy to fix: just add an extra space
- * in front. There will be room for that char;
- * but may need to take that the following '-'
- * also fits.
- */
- mOutputBuffer[ptr++] = ' ';
- if (ptr >= mOutputBuffer.length) { // whops. need to flush
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- }
- mOutputBuffer[ptr++] = BYTE_HYPHEN;
- /* Also, since we did output an extra char, better
- * restart the loop (since max calculation is now
- * off)
- */
- max -= (inEnd - offset);
- break inner_loop;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected int writePIData(String data)
- throws IOException, XMLStreamException
- {
- // Note: mSurrogate can not be non-zero at this point, no need to check
-
- int offset = 0;
- int len = data.length();
- if (!mCheckContent) {
- writeRaw(data, offset, len);
- return -1;
- }
-
- int ptr = mOutputPtr;
- while (len > 0) {
- int max = mOutputBuffer.length - ptr;
- if (max < 1) { // output buffer full?
- mOutputPtr = ptr;
- flushBuffer();
- ptr = 0;
- max = mOutputBuffer.length;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- for (int inEnd = offset + max; offset < inEnd; ++offset) {
- int c = data.charAt(offset);
- if (c < 32) {
- if (c == '\n') {
- // !!! TBI: line nr
- } else if (c == '\r') {
- // !!! TBI: line nr (and skipping \n that may follow)
- } else if (c != '\t') {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- } else if (c > 0x7E) {
- if (c > 0xFF) {
- mOutputPtr = ptr;
- handleInvalidLatinChar(c);
- } else if (mXml11) {
- if (c < 0x9F && c != 0x85) {
- mOutputPtr = ptr;
- c = handleInvalidChar(c);
- }
- }
- } else if (c == '>') { // enclosed end marker ("?>")?
- if (offset > 0 && data.charAt(offset-1) == '?') {
- return offset-2;
- }
- }
- mOutputBuffer[ptr++] = (byte) c;
- }
- len -= max;
- }
- mOutputPtr = ptr;
- return -1;
- }
-
- protected void writeTextContent(String data)
- throws IOException
- {
- int offset = 0;
- int len = data.length();
-
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - mOutputPtr;
- if (max < 1) { // output buffer full?
- flushBuffer();
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = data.charAt(offset++);
- sec = calcSurrogate(sec);
- writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = data.charAt(offset++);
- if (c < 32) {
- if (c == '\n' || c == '\t') { // TODO: line count
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- } else if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- } else if (!mXml11 || c == 0) { // ok in xml1.1, as entity
- if (mCheckContent) {
- c = handleInvalidChar(c);
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- // otherwise... well, I guess we can just escape it
- }
- // \r, or xml1.1 + other whitespace, need to escape
- } else if (c < 0x7F) {
- if (c != '<' && c != '&') {
- if (c != '>' || (offset > 1 && data.charAt(offset-2) != ']')) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- }
- // otherwise fall back on quoting
- } else if (c > 0x9F && c <= 0xFF) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue; // [WSTX-88]
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(data.charAt(offset++));
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- writeAsEntity(c);
- len = data.length() - offset;
- continue main_loop;
- }
- len -= max;
- }
- }
-
- protected void writeTextContent(char[] cbuf, int offset, int len)
- throws IOException
- {
- main_loop:
- while (len > 0) {
- int max = mOutputBuffer.length - mOutputPtr;
- if (max < 1) { // output buffer full?
- flushBuffer();
- max = mOutputBuffer.length;
- }
- // Do we start with a surrogate?
- if (mSurrogate != 0) {
- int sec = cbuf[offset++];
- sec = calcSurrogate(sec);
- writeAsEntity(sec);
- --len;
- continue main_loop;
- }
- // How much can we output?
- if (max > len) {
- max = len;
- }
- inner_loop:
- for (int inEnd = offset + max; offset < inEnd; ) {
- int c = cbuf[offset++];
- if (c < 32) {
- if (c == '\n' || c == '\t') { // TODO: line count
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- } else if (c == '\r') {
- if (!mEscapeCR) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- } else if (!mXml11 || c == 0) { // ok in xml1.1, as entity
- if (mCheckContent) {
- c = handleInvalidChar(c);
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- // otherwise... well, I guess we can just escape it
- }
- // \r, or xml1.1 + other whitespace, need to escape
- } else if (c < 0x7F) {
- if (c !='<' && c != '&') {
- /* Since we can be conservative, it doesn't matter
- * if second check is not exact
- */
- if (c != '>' || (offset > 1 && cbuf[offset-2] != ']')) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue;
- }
- }
- // otherwise fall back on quoting
- } else if (c > 0x9F && c <= 0xFF) {
- mOutputBuffer[mOutputPtr++] = (byte) c;
- continue; // [WSTX-88]
- } else {
- // Surrogate?
- if (c >= SURR1_FIRST && c <= SURR2_LAST) {
- mSurrogate = c;
- // Last char needs special handling:
- if (offset == inEnd) {
- break inner_loop;
- }
- c = calcSurrogate(cbuf[offset++]);
- // Let's fall down to entity output
- }
- }
- /* Has to be escaped as char entity; as such, also need
- * to re-calc max. continguous data we can output
- */
- writeAsEntity(c);
- max -= (inEnd - offset);
- break inner_loop;
- }
- len -= max;
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Internal methods
- ////////////////////////////////////////////////////
- */
-
- protected void handleInvalidLatinChar(int c)
- throws IOException
- {
- // First, let's flush any output we may have, to make debugging easier
- flush();
-
- /* 17-May-2006, TSa: Would really be useful if we could throw
- * XMLStreamExceptions; esp. to indicate actual output location.
- * However, this causes problem with methods that call us and
- * can only throw IOExceptions (when invoked via Writer proxy).
- * Need to figure out how to resolve this.
- */
- throw new IOException("Invalid XML character (0x"+Integer.toHexString(c)+"); can only be output using character entity when using ISO-8859-1 encoding");
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/NonNsStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/NonNsStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/NonNsStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/NonNsStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,553 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE which is
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.util.*;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.events.Attribute;
-import javax.xml.stream.events.StartElement;
-
-import org.codehaus.stax2.ri.typed.AsciiValueEncoder;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.cfg.XmlConsts;
-import com.ctc.wstx.sr.AttributeCollector;
-import com.ctc.wstx.sr.InputElementStack;
-import com.ctc.wstx.util.EmptyNamespaceContext;
-import com.ctc.wstx.util.StringVector;
-
-/**
- * Implementation of {@link XMLStreamWriter} used when namespace support
- * is not enabled. This means that only local names are used for elements
- * and attributes; and if rudimentary namespace declarations need to be
- * output, they are output using attribute writing methods.
- */
-public class NonNsStreamWriter
- extends TypedStreamWriter
-{
- /*
- ////////////////////////////////////////////////////
- // State information
- ////////////////////////////////////////////////////
- */
-
- /**
- * Stack of currently open start elements; only local names
- * are included.
- */
- final StringVector mElements;
-
- /**
- * Container for attribute names for current element; used only
- * if uniqueness of attribute names is to be enforced.
- *setPrefix, writeNamespace
- * and writeStartElement; the order writers expect is
- * bit different from the order in which element information is
- * passed in.
- */
- public final void copyStartElement(InputElementStack elemStack, AttributeCollector ac)
- throws IOException, XMLStreamException
- {
- /* In case of repairing stream writer, we can actually just
- * go ahead and first output the element: stream writer should
- * be able to resolve namespace mapping for the element
- * automatically, as necessary.
- */
- String prefix = elemStack.getPrefix();
- String uri = elemStack.getNsURI();
- writeStartElement(prefix, elemStack.getLocalName(), uri);
-
- /* 04-Sep-2006, TSa: Although we could really just ignore all
- * namespace declarations, some apps prefer (or even expect...)
- * that ns bindings are preserved as much as possible. So, let's
- * just try to output them as they are (could optimize and skip
- * ones related to the start element [same prefix or URI], but
- * for now let's not bother)
- */
- int nsCount = elemStack.getCurrentNsCount();
- if (nsCount > 0) { // yup, got some...
- for (int i = 0; i < nsCount; ++i) {
- writeNamespace(elemStack.getLocalNsPrefix(i), elemStack.getLocalNsURI(i));
- }
- }
-
- /* And then let's just output attributes, if any (whether to copy
- * implicit, aka "default" attributes, is configurable)
- */
- int attrCount = mCfgCopyDefaultAttrs ? ac.getCount() : ac.getSpecifiedCount();
-
- /* Unlike in non-ns and simple-ns modes, we can not simply literally
- * copy the attributes here. It is possible that some namespace
- * prefixes have been remapped... so need to be bit more careful.
- */
- if (attrCount > 0) {
- for (int i = 0; i < attrCount; ++i) {
- // First; need to make sure that the prefix-to-ns mapping
- // attribute has is valid... and can not output anything
- // before that's done (since remapping will output a namespace
- // declaration!)
- uri = ac.getURI(i);
- prefix = ac.getPrefix(i);
-
- // With attributes, missing/empty prefix always means 'no
- // namespace', can take a shortcut:
- if (prefix == null || prefix.length() == 0) {
- ;
- } else {
- // and otherwise we'll always have a prefix as attributes
- // can not make use of the def. namespace...
- prefix = findOrCreateAttrPrefix(prefix, uri, mCurrElem);
- }
- /* Hmmh. Since the prefix we use may be different from what
- * collector has, we can not use pass-through method of
- * the collector, but need to call XmlWriter directly:
- */
- if (prefix == null || prefix.length() == 0) {
- mWriter.writeAttribute(ac.getLocalName(i), ac.getValue(i));
- } else {
- mWriter.writeAttribute(prefix, ac.getLocalName(i), ac.getValue(i));
- }
- }
- }
- }
-
- public String validateQNamePrefix(QName name)
- throws XMLStreamException
- {
- /* Gets bit more complicated: we need to ensure that given URI
- * is properly bound...
- */
- String uri = name.getNamespaceURI();
- String suggPrefix = name.getPrefix();
- String actPrefix = validateElemPrefix(suggPrefix, uri, mCurrElem);
- if (actPrefix == null) { // no suitable prefix, must bind
- /* Need to ensure that we'll pass "" as prefix, not null, so
- * that it is understood as "I want to use the default NS", not
- * as "whatever prefix, I don't care"
- */
- if (suggPrefix == null) {
- suggPrefix = "";
- }
- actPrefix = generateElemPrefix(suggPrefix, uri, mCurrElem);
- if (actPrefix == null || actPrefix.length() == 0) { // def NS
- writeDefaultNamespace(uri);
- } else {
- writeNamespace(actPrefix, uri);
- }
- }
- return actPrefix;
- }
-
- /*
- ///////////////////////////////////////////////////////////
- // Internal methods
- ///////////////////////////////////////////////////////////
- */
-
- /**
- * Method called to find an existing prefix for the given namespace,
- * if any exists in the scope. If one is found, it's returned (including
- * "" for the current default namespace); if not, null is returned.
- *
- * @param nsURI URI of namespace for which we need a prefix
- */
- protected final String findElemPrefix(String nsURI, SimpleOutputElement elem)
- throws XMLStreamException
- {
- /* Special case: empty NS URI can only be bound to the empty
- * prefix...
- */
- if (nsURI == null || nsURI.length() == 0) {
- String currDefNsURI = elem.getDefaultNsUri();
- if (currDefNsURI != null && currDefNsURI.length() > 0) {
- // Nope; won't do... has to be re-bound, but not here:
- return null;
- }
- return "";
- }
- return mCurrElem.getPrefix(nsURI);
- }
-
- /**
- * Method called after {@link #findElemPrefix} has returned null,
- * to create and bind a namespace mapping for specified namespace.
- */
- protected final String generateElemPrefix(String suggPrefix, String nsURI,
- SimpleOutputElement elem)
- throws XMLStreamException
- {
- /* Ok... now, since we do not have an existing mapping, let's
- * see if we have a preferred prefix to use.
- */
- /* Except if we need the empty namespace... that can only be
- * bound to the empty prefix:
- */
- if (nsURI == null || nsURI.length() == 0) {
- return "";
- }
-
- /* Ok; with elements this is easy: the preferred prefix can
- * ALWAYS be used, since it can mask preceding bindings:
- */
- if (suggPrefix == null) {
- // caller wants this URI to map as the default namespace?
- if (mSuggestedDefNs != null && mSuggestedDefNs.equals(nsURI)) {
- suggPrefix = "";
- } else {
- suggPrefix = (mSuggestedPrefixes == null) ? null:
- (String) mSuggestedPrefixes.get(nsURI);
- if (suggPrefix == null) {
- /* 16-Oct-2005, TSa: We have 2 choices here, essentially;
- * could make elements always try to override the def
- * ns... or can just generate new one. Let's do latter
- * for now.
- */
- if (mAutoNsSeq == null) {
- mAutoNsSeq = new int[1];
- mAutoNsSeq[0] = 1;
- }
- suggPrefix = elem.generateMapping(mAutomaticNsPrefix, nsURI,
- mAutoNsSeq);
- }
- }
- }
-
- // Ok; let's let the caller deal with bindings
- return suggPrefix;
- }
-
- /**
- * Method called to somehow find a prefix for given namespace, to be
- * used for a new start element; either use an existing one, or
- * generate a new one. If a new mapping needs to be generated,
- * it will also be automatically bound, and necessary namespace
- * declaration output.
- *
- * @param suggPrefix Suggested prefix to bind, if any; may be null
- * to indicate "no preference"
- * @param nsURI URI of namespace for which we need a prefix
- * @param elem Currently open start element, on which the attribute
- * will be added.
- */
- protected final String findOrCreateAttrPrefix(String suggPrefix, String nsURI,
- SimpleOutputElement elem)
- throws XMLStreamException
- {
- if (nsURI == null || nsURI.length() == 0) {
- /* Attributes never use the default namespace; missing
- * prefix always leads to the empty ns... so nothing
- * special is needed here.
- */
- return null;
- }
- // Maybe the suggested prefix is properly bound?
- if (suggPrefix != null) {
- int status = elem.isPrefixValid(suggPrefix, nsURI, false);
- if (status == SimpleOutputElement.PREFIX_OK) {
- return suggPrefix;
- }
- /* Otherwise, if the prefix is unbound, let's just bind
- * it -- if caller specified a prefix, it probably prefers
- * binding that prefix even if another prefix already existed?
- * The remaining case (already bound to another URI) we don't
- * want to touch, at least not yet: it may or not be safe
- * to change binding, so let's just not try it.
- */
- if (status == SimpleOutputElement.PREFIX_UNBOUND) {
- elem.addPrefix(suggPrefix, nsURI);
- doWriteNamespace(suggPrefix, nsURI);
- return suggPrefix;
- }
- }
-
- // If not, perhaps there's another existing binding available?
- String prefix = elem.getExplicitPrefix(nsURI);
- if (prefix != null) { // already had a mapping for the URI... cool.
- return prefix;
- }
-
- /* Nope, need to create one. First, let's see if there's a
- * preference...
- */
- if (suggPrefix != null) {
- prefix = suggPrefix;
- } else if (mSuggestedPrefixes != null) {
- prefix = (String) mSuggestedPrefixes.get(nsURI);
- // note: def ns is never added to suggested prefix map
- }
-
- if (prefix != null) {
- /* Can not use default namespace for attributes.
- * Also, re-binding is tricky for attributes; can't
- * re-bind anything that's bound on this scope... or
- * used in this scope. So, to simplify life, let's not
- * re-bind anything for attributes.
- */
- if (prefix.length() == 0
- || (elem.getNamespaceURI(prefix) != null)) {
- prefix = null;
- }
- }
-
- if (prefix == null) {
- if (mAutoNsSeq == null) {
- mAutoNsSeq = new int[1];
- mAutoNsSeq[0] = 1;
- }
- prefix = mCurrElem.generateMapping(mAutomaticNsPrefix, nsURI,
- mAutoNsSeq);
- }
-
- // Ok; so far so good: let's now bind and output the namespace:
- elem.addPrefix(prefix, nsURI);
- doWriteNamespace(prefix, nsURI);
- return prefix;
- }
-
- private final String validateElemPrefix(String prefix, String nsURI,
- SimpleOutputElement elem)
- throws XMLStreamException
- {
- /* 06-Feb-2005, TSa: Special care needs to be taken for the
- * "empty" (or missing) namespace:
- * (see comments from findOrCreatePrefix())
- */
- if (nsURI == null || nsURI.length() == 0) {
- String currURL = elem.getDefaultNsUri();
- if (currURL == null || currURL.length() == 0) {
- // Ok, good:
- return "";
- }
- // Nope, needs to be re-bound:
- return null;
- }
-
- int status = elem.isPrefixValid(prefix, nsURI, true);
- if (status == SimpleOutputElement.PREFIX_OK) {
- return prefix;
- }
-
- /* Hmmh... now here's bit of dilemma: that particular prefix is
- * either not bound, or is masked... but it is possible some other
- * prefix would be bound. Should we search for another one, or
- * try to re-define suggested one? Let's do latter, for now;
- * caller can then (try to) bind the preferred prefix:
- */
- return null;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/SimpleNsStreamWriter.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/SimpleNsStreamWriter.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/SimpleNsStreamWriter.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/SimpleNsStreamWriter.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,336 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in the file LICENSE,
- * included with the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.IOException;
-import java.util.Iterator;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamWriter;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.events.Attribute;
-import javax.xml.stream.events.Namespace;
-import javax.xml.stream.events.StartElement;
-
-import com.ctc.wstx.api.WriterConfig;
-import com.ctc.wstx.cfg.ErrorConsts;
-import com.ctc.wstx.sr.AttributeCollector;
-import com.ctc.wstx.sr.InputElementStack;
-
-/**
- * Namespace-aware implementation of {@link XMLStreamWriter}, that does
- * not do namespace repairing, ie doesn't try to resolve possible
- * conflicts between prefixes and namespace URIs, or automatically
- * create namespace bindings.
- */
-public class SimpleNsStreamWriter
- extends BaseNsStreamWriter
-{
- /*
- ////////////////////////////////////////////////////
- // Life-cycle (ctors)
- ////////////////////////////////////////////////////
- */
-
- public SimpleNsStreamWriter(XmlWriter xw, String enc, WriterConfig cfg)
- {
- super(xw, enc, cfg, false);
- }
-
- /*
- ////////////////////////////////////////////////////
- // XMLStreamWriter API
- ////////////////////////////////////////////////////
- */
-
- //public NamespaceContext getNamespaceContext()
- //public void setNamespaceContext(NamespaceContext context)
- //public String getPrefix(String uri)
- //public void setPrefix(String prefix, String uri)
-
- //public void writeAttribute(String localName, String value)
-
- public void writeAttribute(String nsURI, String localName, String value)
- throws XMLStreamException
- {
- // No need to set mAnyOutput, nor close the element
- if (!mStartElementOpen) {
- throwOutputError(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- String prefix = mCurrElem.getExplicitPrefix(nsURI);
- if (!mReturnNullForDefaultNamespace && prefix == null) {
- throwOutputError("Unbound namespace URI '" + nsURI + "'");
- }
- doWriteAttr(localName, nsURI, prefix, value);
- }
-
- public void writeAttribute(String prefix, String nsURI,
- String localName, String value)
- throws XMLStreamException
- {
- if (!mStartElementOpen) {
- throwOutputError(ErrorConsts.WERR_ATTR_NO_ELEM);
- }
- doWriteAttr(localName, nsURI, prefix, value);
- }
-
- //public void writeEmptyElement(String localName) throws XMLStreamException
- //public void writeEmptyElement(String nsURI, String localName) throws XMLStreamException
- //public void writeEmptyElement(String prefix, String localName, String nsURI) throws XMLStreamException
-
- //public void writeEndElement() throws XMLStreamException
-
- public void writeDefaultNamespace(String nsURI)
- throws XMLStreamException
- {
- if (!mStartElementOpen) {
- throwOutputError(ERR_NSDECL_WRONG_STATE);
- }
- // 27-Mar-2007, TSa: Apparently TCK expects a binding to be added
- setDefaultNamespace(nsURI);
- doWriteDefaultNs(nsURI);
- }
-
- public void writeNamespace(String prefix, String nsURI)
- throws XMLStreamException
- {
- if (prefix == null || prefix.length() == 0
- || prefix.equals(XMLConstants.XMLNS_ATTRIBUTE)) {
- writeDefaultNamespace(nsURI);
- return;
- }
-
- // No need to set mAnyOutput, and shouldn't close the element.
- // But element needs to be open, obviously.
- if (!mStartElementOpen) {
- throwOutputError(ERR_NSDECL_WRONG_STATE);
- }
- /* 05-Feb-2005, TSa: Also, as per namespace specs; the 'empty'
- * namespace URI can not be bound as a non-default namespace
- * (ie. for any actual prefix)
- */
- /* 04-Feb-2005, TSa: Namespaces 1.1 does allow this, though,
- * so for xml 1.1 documents we need to allow it
- */
- if (!mXml11) {
- if (nsURI.length() == 0) {
- throwOutputError(ErrorConsts.ERR_NS_EMPTY);
- }
- // 01-Apr-2005, TSa: Can we (and do we want to) verify NS consistency?
- }
- // 27-Mar-2007, TSa: Apparently TCK expects a binding to be added
- setPrefix(prefix, nsURI);
- doWriteNamespace(prefix, nsURI);
- }
-
- /*
- ////////////////////////////////////////////////////
- // Package methods:
- ////////////////////////////////////////////////////
- */
-
- public void setDefaultNamespace(String uri)
- throws XMLStreamException
- {
- mCurrElem.setDefaultNsUri(uri);
- }
-
- public void doSetPrefix(String prefix, String uri)
- throws XMLStreamException
- {
- mCurrElem.addPrefix(prefix, uri);
- }
-
- public void writeStartElement(StartElement elem)
- throws XMLStreamException
- {
- QName name = elem.getName();
- Iterator it = elem.getNamespaces();
-
- while (it.hasNext()) {
- Namespace ns = (Namespace) it.next();
- // First need to 'declare' namespace:
- String prefix = ns.getPrefix();
- if (prefix == null || prefix.length() == 0) {
- setDefaultNamespace(ns.getNamespaceURI());
- } else {
- setPrefix(prefix, ns.getNamespaceURI());
- }
- }
-
- /* Outputting element itself is fairly easy. The main question
- * is whether namespaces match. Let's use simple heuristics:
- * if writer is to do automatic prefix matching, let's only
- * pass explicit prefix (not default one); otherwise we'll
- * pass all parameters as is.
- */
- /* Quick check first though: if URI part of QName is null, it's
- * assumed element will just use whatever is current default
- * namespace....
- */
- String nsURI = name.getNamespaceURI();
- if (nsURI == null) {
- writeStartElement(name.getLocalPart());
- } else {
- String prefix = name.getPrefix();
- writeStartElement(prefix, name.getLocalPart(), nsURI);
- }
-
- // And now we need to output namespaces (including default), if any:
- it = elem.getNamespaces();
- while (it.hasNext()) {
- Namespace ns = (Namespace) it.next();
- String prefix = ns.getPrefix();
- if (prefix == null || prefix.length() == 0) {
- writeDefaultNamespace(ns.getNamespaceURI());
- } else {
- writeNamespace(prefix, ns.getNamespaceURI());
- }
- }
-
-
- // And finally, need to output attributes as well:
-
- it = elem.getAttributes();
- while (it.hasNext()) {
- Attribute attr = (Attribute) it.next();
- name = attr.getName();
- nsURI = name.getNamespaceURI();
- // In non-default/empty namespace?
- if (nsURI != null && nsURI.length() > 0) {
- writeAttribute(name.getPrefix(), nsURI,
- name.getLocalPart(), attr.getValue());
- } else {
- writeAttribute(name.getLocalPart(), attr.getValue());
- }
- }
- }
-
- //public void writeEndElement(QName name) throws XMLStreamException
-
- protected void writeStartOrEmpty(String localName, String nsURI)
- throws XMLStreamException
- {
- // Need a prefix...
- String prefix = mCurrElem.getPrefix(nsURI);
- if (prefix == null) {
- throw new XMLStreamException("Unbound namespace URI '"+nsURI+"'");
- }
- checkStartElement(localName, prefix);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, prefix);
- }
-
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, prefix, localName, nsURI);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(prefix, localName, nsURI);
- }
- doWriteStartTag(prefix, localName);
- }
-
- protected void writeStartOrEmpty(String prefix, String localName, String nsURI)
- throws XMLStreamException
- {
- checkStartElement(localName, prefix);
- if (mValidator != null) {
- mValidator.validateElementStart(localName, nsURI, prefix);
- }
-
- if (mOutputElemPool != null) {
- SimpleOutputElement newCurr = mOutputElemPool;
- mOutputElemPool = newCurr.reuseAsChild(mCurrElem, prefix, localName, nsURI);
- --mPoolSize;
- mCurrElem = newCurr;
- } else {
- mCurrElem = mCurrElem.createChild(prefix, localName, nsURI);
- }
- doWriteStartTag(prefix, localName);
- }
-
- /**
- * Element copier method implementation suitable to be used with
- * namespace-aware writers in non-repairing (explicit namespaces) mode.
- * The trickiest thing is having to properly
- * order calls to setPrefix, writeNamespace
- * and writeStartElement; the order writers expect is
- * bit different from the order in which element information is
- * passed in.
- */
- public final void copyStartElement(InputElementStack elemStack,
- AttributeCollector attrCollector)
- throws IOException, XMLStreamException
- {
- // Any namespace declarations/bindings?
- int nsCount = elemStack.getCurrentNsCount();
- if (nsCount > 0) { // yup, got some...
- /* First, need to (or at least, should?) add prefix bindings:
- * (may not be 100% required, but probably a good thing to do,
- * just so that app code has access to prefixes then)
- */
- for (int i = 0; i < nsCount; ++i) {
- String prefix = elemStack.getLocalNsPrefix(i);
- String uri = elemStack.getLocalNsURI(i);
- if (prefix == null || prefix.length() == 0) { // default NS
- setDefaultNamespace(uri);
- } else {
- setPrefix(prefix, uri);
- }
- }
- }
-
- writeStartElement(elemStack.getPrefix(),
- elemStack.getLocalName(),
- elemStack.getNsURI());
-
- if (nsCount > 0) {
- // And then output actual namespace declarations:
- for (int i = 0; i < nsCount; ++i) {
- String prefix = elemStack.getLocalNsPrefix(i);
- String uri = elemStack.getLocalNsURI(i);
-
- if (prefix == null || prefix.length() == 0) { // default NS
- writeDefaultNamespace(uri);
- } else {
- writeNamespace(prefix, uri);
- }
- }
- }
-
- /* And then let's just output attributes, if any (whether to copy
- * implicit, aka "default" attributes, is configurable)
- */
- int attrCount = mCfgCopyDefaultAttrs ?
- attrCollector.getCount() :
- attrCollector.getSpecifiedCount();
-
- if (attrCount > 0) {
- for (int i = 0; i < attrCount; ++i) {
- attrCollector.writeAttribute(i, mWriter);
- }
- }
- }
-
- public String validateQNamePrefix(QName name)
- {
- // Good as is, let's not complicate things
- return name.getPrefix();
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/SimpleOutputElement.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/SimpleOutputElement.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/SimpleOutputElement.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/SimpleOutputElement.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,367 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2005 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.util.*;
-
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.namespace.QName;
-import javax.xml.stream.XMLStreamException;
-
-import com.ctc.wstx.compat.QNameCreator;
-import com.ctc.wstx.util.BijectiveNsMap;
-
-/**
- * Class that encapsulates information about a specific element in virtual
- * output stack for namespace-aware writers.
- * It provides support for URI-to-prefix mappings as well as namespace
- * mapping generation.
- *
- *
- */
-public abstract class XmlWriter
-{
- protected final static int SURR1_FIRST = 0xD800;
- protected final static int SURR1_LAST = 0xDBFF;
- protected final static int SURR2_FIRST = 0xDC00;
- protected final static int SURR2_LAST = 0xDFFF;
-
- protected final static char DEFAULT_QUOTE_CHAR = '"';
-
- protected final WriterConfig mConfig;
- protected final String mEncoding;
-
- // // // Operating mode: base class needs to know whether
- // // // namespaces are support (for entity/PI target validation)
-
- protected final boolean mNsAware;
-
- protected final boolean mCheckStructure;
- protected final boolean mCheckContent;
- protected final boolean mCheckNames;
- protected final boolean mFixContent;
-
- /**
- * Whether to escape CR (\r) character.
- */
- final boolean mEscapeCR;
-
- /**
- * Whether to add a space after empty element (before closing "/>")
- * or not.
- */
- final boolean mAddSpaceAfterEmptyElem;
-
- /**
- * Flag that defines whether close() on this writer should call
- * close on the underlying output object (stream, writer)
- */
- protected final boolean mAutoCloseOutput;
-
- /**
- * Optional escaping writer used for escaping characters like '<'
- * '&' and '>' in textual content.
- * Constructed if calling code has
- * installed a special escaping writer factory for text content.
- * Null if the default escaper is to be used.
- */
- protected Writer mTextWriter;
-
- /**
- * Optional escaping writer used for escaping characters like '"'
- * '&' and '<' in attribute values.
- * Constructed if calling code has
- * installed a special escaping writer factory for text content.
- * Null if the default escaper is to be used.
- */
- protected Writer mAttrValueWriter;
-
- /**
- * Indicates whether output is to be compliant; if false, is to be
- * xml 1.0 compliant, if true, xml 1.1 compliant.
- */
- protected boolean mXml11 = false;
-
- /**
- * Lazy-constructed wrapper object, which will route all calls to
- * Writer API, to matching writeRaw methods of this
- * XmlWriter instance.
- */
- protected XmlWriterWrapper mRawWrapper = null;
-
- /**
- * Lazy-constructed wrapper object, which will route all calls to
- * Writer API, to matching writeCharacters methods of this
- * XmlWriter instance.
- */
- protected XmlWriterWrapper mTextWrapper = null;
-
- /*
- ///////////////////////////////////////////////////////
- // Output location info
- ///////////////////////////////////////////////////////
- */
-
- /**
- * Number of characters output prior to currently buffered output
- */
- protected int mLocPastChars = 0;
-
- protected int mLocRowNr = 1;
-
- /**
- * Offset of the first character on this line. May be negative, if
- * the offset was in a buffer that has been flushed out.
- */
- protected int mLocRowStartOffset = 0;
-
- /*
- ///////////////////////////////////////////////////////
- // Life-cycle
- ///////////////////////////////////////////////////////
- */
-
- protected XmlWriter(WriterConfig cfg, String encoding, boolean autoclose)
- throws IOException
- {
- mConfig = cfg;
- mEncoding = encoding;
- mAutoCloseOutput = autoclose;
- int flags = cfg.getConfigFlags();
- mNsAware = (flags & OutputConfigFlags.CFG_ENABLE_NS) != 0;
- mCheckStructure = (flags & OutputConfigFlags.CFG_VALIDATE_STRUCTURE) != 0;
- mCheckContent = (flags & OutputConfigFlags.CFG_VALIDATE_CONTENT) != 0;
- mCheckNames = (flags & OutputConfigFlags.CFG_VALIDATE_NAMES) != 0;
- mFixContent = (flags & OutputConfigFlags.CFG_FIX_CONTENT) != 0;
- mEscapeCR = (flags & OutputConfigFlags.CFG_ESCAPE_CR) != 0;
- mAddSpaceAfterEmptyElem = (flags & OutputConfigFlags.CFG_ADD_SPACE_AFTER_EMPTY_ELEM) != 0;
-
- // Has caller requested any custom text or attr value escaping?
-
- EscapingWriterFactory f = mConfig.getTextEscaperFactory();
- if (f == null) {
- mTextWriter = null;
- } else {
- String enc = (mEncoding == null || mEncoding.length() == 0) ?
- WstxOutputProperties.DEFAULT_OUTPUT_ENCODING : mEncoding;
- mTextWriter = f.createEscapingWriterFor(wrapAsRawWriter(), enc);
- }
-
- f = mConfig.getAttrValueEscaperFactory();
- if (f == null) {
- mAttrValueWriter = null;
- } else {
- String enc = (mEncoding == null || mEncoding.length() == 0) ?
- WstxOutputProperties.DEFAULT_OUTPUT_ENCODING : mEncoding;
- mAttrValueWriter = f.createEscapingWriterFor(wrapAsRawWriter(), enc);
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Extra configuration
- ////////////////////////////////////////////////////
- */
-
- public void enableXml11() {
- mXml11 = true;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Access to underlying physical output destinations
- ////////////////////////////////////////////////////
- */
-
- /**
- * @return Underlying OutputStream used for physical output,
- * if the writer was constructed using one
- */
- protected abstract OutputStream getOutputStream();
-
- /**
- * @return Underlying Writer used for physical output,
- * if the writer was constructed with one, or one was
- * created to be used with an OutputStream.
- */
- protected abstract Writer getWriter();
-
- /*
- ////////////////////////////////////////////////////
- // Basic methods for communicating with underlying
- // stream or writer
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method called to flush the buffer(s), and close the output
- * sink (stream or writer) if enabled (auto-closing) or
- * forced.
- */
- public abstract void close(boolean forceRealClose) throws IOException;
-
- public abstract void flush()
- throws IOException;
-
- public abstract void writeRaw(String str, int offset, int len)
- throws IOException;
-
- public void writeRaw(String str)
- throws IOException
- {
- writeRaw(str, 0, str.length());
- }
-
- public abstract void writeRaw(char[] cbuf, int offset, int len)
- throws IOException;
-
- /**
- * Like {@link #writeRaw}, but caller guarantees that the contents
- * additionally are known to be in 7-bit ascii range.
- */
- public abstract void writeRawAscii(char[] cbuf, int offset, int len)
- throws IOException;
-
- /*
- ////////////////////////////////////////////////////
- // Raw, non-verifying write methods; used when
- // directly copying trusted content
- ////////////////////////////////////////////////////
- */
-
- public abstract void writeCDataStart()
- throws IOException;
-
- public abstract void writeCDataEnd()
- throws IOException;
-
- public abstract void writeCommentStart()
- throws IOException;
-
- public abstract void writeCommentEnd()
- throws IOException;
-
- public abstract void writePIStart(String target, boolean addSpace)
- throws IOException;
-
- public abstract void writePIEnd()
- throws IOException;
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, textual:
- ////////////////////////////////////////////////////
- */
-
- /**
- * @param data Contents of the CDATA section to write out
-
- * @return offset of the (first) illegal content segment ("]]>") in
- * passed content and not in repairing mode; or -1 if none or is
- * repairing
- */
- public abstract int writeCData(String data)
- throws IOException, XMLStreamException;
-
- public abstract int writeCData(char[] cbuf, int offset, int len)
- throws IOException, XMLStreamException;
-
- public abstract void writeCharacters(String data)
- throws IOException;
-
- public abstract void writeCharacters(char[] cbuf, int offset, int len)
- throws IOException;
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, non-textual, non-elem/attr:
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method that will try to output the content as specified. If
- * the content passed in has embedded "--" in it, it will either
- * add an intervening space between consequtive hyphens (if content
- * fixing is enabled), or return the offset of the first hyphen in
- * multi-hyphen sequence.
- */
- public abstract int writeComment(String data)
- throws IOException, XMLStreamException;
-
- /**
- * Older "legacy" output method for outputting DOCTYPE declaration.
- * Assumes that the passed-in String contains a complete DOCTYPE
- * declaration properly quoted.
- */
- public abstract void writeDTD(String data)
- throws IOException, XMLStreamException;
-
- public abstract void writeDTD(String rootName,
- String systemId, String publicId,
- String internalSubset)
- throws IOException, XMLStreamException;
-
- public abstract void writeEntityReference(String name)
- throws IOException, XMLStreamException;
-
- public abstract int writePI(String target, String data)
- throws IOException, XMLStreamException;
-
- public abstract void writeXmlDeclaration(String version, String enc, String standalone)
- throws IOException;
-
- /*
- ////////////////////////////////////////////////////
- // Write methods, elements
- ////////////////////////////////////////////////////
- */
-
- /**
- *writeRaw
- * method.
- */
- public final Writer wrapAsRawWriter()
- {
- if (mRawWrapper == null) {
- mRawWrapper = XmlWriterWrapper.wrapWriteRaw(this);
- }
- return mRawWrapper;
- }
-
- public final Writer wrapAsTextWriter()
- {
- if (mTextWrapper == null) {
- mTextWrapper = XmlWriterWrapper.wrapWriteCharacters(this);
- }
- return mTextWrapper;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Helper methods for sub-classes
- ////////////////////////////////////////////////////
- */
-
- /**
- * Method called to verify that the name is a legal XML name.
- */
- public final void verifyNameValidity(String name, boolean checkNs)
- throws XMLStreamException
- {
- /* No empty names... caller must have dealt with optional arguments
- * prior to calling this method
- */
- if (name == null || name.length() == 0) {
- reportNwfName(ErrorConsts.WERR_NAME_EMPTY);
- }
- int illegalIx = WstxInputData.findIllegalNameChar(name, checkNs, mXml11);
- if (illegalIx >= 0) {
- if (illegalIx == 0) {
- reportNwfName(ErrorConsts.WERR_NAME_ILLEGAL_FIRST_CHAR,
- WstxInputData.getCharDesc(name.charAt(0)));
- }
- reportNwfName(ErrorConsts.WERR_NAME_ILLEGAL_CHAR,
- WstxInputData.getCharDesc(name.charAt(illegalIx)));
- }
- }
-
- /**
- * This is the method called when an output method call violates
- * name well-formedness checks
- * and {@link WstxOutputProperties#P_OUTPUT_VALIDATE_NAMES} is
- * is enabled.
- */
- protected void reportNwfName(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- protected void reportNwfName(String msg, Object arg)
- throws XMLStreamException
- {
- throwOutputError(msg, arg);
- }
-
- protected void reportNwfContent(String msg)
- throws XMLStreamException
- {
- throwOutputError(msg);
- }
-
- protected void throwOutputError(String msg)
- throws XMLStreamException
- {
- // First, let's flush any output we may have, to make debugging easier
- try {
- flush();
- } catch (IOException ioe) {
- throw new WstxIOException(ioe);
- }
-
- throw new XMLStreamException(msg);
- }
-
- protected void throwOutputError(String format, Object arg)
- throws XMLStreamException
- {
- String msg = MessageFormat.format(format, new Object[] { arg });
- throwOutputError(msg);
- }
-
- /**
- * Method called to handle invalid character in textual content requested
- * to be output. Content may be part of textual events (CHARACTER, CDATA),
- * attribute value, COMMENT content or PROCESSING_INSTRUCTION data.
- * The default behavior is to just throw an exception, but this can
- * be configured via property {@link WstxOutputProperties#P_OUTPUT_INVALID_CHAR_HANDLER}.
- */
- protected char handleInvalidChar(int c)
- throws IOException
- {
- // First, let's flush any output we may have, to make debugging easier
- flush();
- InvalidCharHandler h = mConfig.getInvalidCharHandler();
- if (h == null) {
- h = InvalidCharHandler.FailingHandler.getInstance();
- }
- return h.convertInvalidChar(c);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/XmlWriterWrapper.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/XmlWriterWrapper.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/sw/XmlWriterWrapper.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/sw/XmlWriterWrapper.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,172 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004- Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.sw;
-
-import java.io.*;
-
-/**
- * This is a simple wrapper class, which decorates an {@link XmlWriter}
- * to look like a Writer. This is necessary to implement a (legacy)
- * character quoting system introduced for Woodstox 2.0, which relies
- * on having a Writer to use for outputting.
- */
-public abstract class XmlWriterWrapper
- extends Writer
-{
- protected final XmlWriter mWriter;
-
- private char[] mBuffer = null;
-
- public static XmlWriterWrapper wrapWriteRaw(XmlWriter xw)
- {
- return new RawWrapper(xw);
- }
-
- public static XmlWriterWrapper wrapWriteCharacters(XmlWriter xw)
- {
- return new TextWrapper(xw);
- }
-
- protected XmlWriterWrapper(XmlWriter writer)
- {
- mWriter = writer;
- }
-
- public final void close()
- throws IOException
- {
- mWriter.close(false);
- }
-
- public final void flush()
- throws IOException
- {
- mWriter.flush();
- }
-
- /* !!! 30-Nov-2006, TSa: Due to co-variance between Appendable and
- * Writer, this would not compile with javac 1.5, in 1.4 mode
- * (source and target set to "1.4". Not a huge deal, but since
- * the base impl is just fine, no point in overriding it.
- */
- /*
- public final Writer append(char c)
- throws IOException
- {
- if (mBuffer == null) {
- mBuffer = new char[1];
- }
- mBuffer[0] = (char) c;
- write(mBuffer, 0, 1);
- return this;
- }
- */
-
- public final void write(char[] cbuf)
- throws IOException
- {
- write(cbuf, 0, cbuf.length);
- }
-
- public abstract void write(char[] cbuf, int off, int len)
- throws IOException;
-
- public final void write(int c)
- throws IOException
- {
- if (mBuffer == null) {
- mBuffer = new char[1];
- }
- mBuffer[0] = (char) c;
- write(mBuffer, 0, 1);
- }
-
- public abstract void write(String str)
- throws IOException;
-
- public abstract void write(String str, int off, int len)
- throws IOException;
-
- /*
- //////////////////////////////////////////////////
- // Implementation classes
- //////////////////////////////////////////////////
- */
-
- /**
- * This wrapper directs calls to writeRaw methods. Thus,
- * it is a "vanilla" writer, and no escaping is done.
- */
- private final static class RawWrapper
- extends XmlWriterWrapper
- {
- protected RawWrapper(XmlWriter writer)
- {
- super(writer);
- }
-
- public void write(char[] cbuf, int off, int len)
- throws IOException
- {
- mWriter.writeRaw(cbuf, off, len);
- }
-
- public void write(String str, int off, int len)
- throws IOException
- {
- mWriter.writeRaw(str, off, len);
- }
-
- public final void write(String str)
- throws IOException
- {
- mWriter.writeRaw(str, 0, str.length());
- }
- }
-
- /**
- * This wrapper directs calls to writeCharacters methods.
- * This means that text content escaping (and, possibly, validation)
- * is done, using default or custom escaping code.
- */
- private static class TextWrapper
- extends XmlWriterWrapper
- {
- protected TextWrapper(XmlWriter writer)
- {
- super(writer);
- }
-
- public void write(char[] cbuf, int off, int len)
- throws IOException
- {
- mWriter.writeCharacters(cbuf, off, len);
- }
-
- public void write(String str)
- throws IOException
- {
- mWriter.writeCharacters(str);
- }
-
- public void write(String str, int off, int len)
- throws IOException
- {
- mWriter.writeCharacters(str.substring(off, off+len));
- }
- }
-}
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ArgUtil.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ArgUtil.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/ArgUtil.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/ArgUtil.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,58 +0,0 @@
-package com.ctc.wstx.util;
-
-/**
- * Simple static utility class that contains (static) utility methods useful
- * when parsing non-typesafe arguments (String-only configuration, command
- * line args).
- */
-public final class ArgUtil
-{
- private ArgUtil() { }
-
- public static boolean convertToBoolean(String prop, Object value)
- {
- if (value == null) {
- return false;
- }
- if (value instanceof Boolean) {
- return ((Boolean) value).booleanValue();
- }
- if (value instanceof String) {
- String str = (String) value;
- if (str.equalsIgnoreCase("false")) {
- return false;
- }
- if (str.equalsIgnoreCase("true")) {
- return true;
- }
- throw new IllegalArgumentException("Invalid String value for property '"+prop+"': expected Boolean value.");
- }
- throw new IllegalArgumentException("Invalid value type ("+value.getClass()+") for property '"+prop+"': expected Boolean value.");
- }
-
- public static int convertToInt(String prop, Object value, int minValue)
- {
- int i;
-
- if (value == null) {
- i = 0;
- } else if (value instanceof Number) {
- i = ((Number) value).intValue();
- } else if (value instanceof String) {
- try {
- i = Integer.parseInt((String) value);
- } catch (NumberFormatException nex) {
- throw new IllegalArgumentException("Invalid String value for property '"+prop+"': expected a number (Integer).");
- }
- } else {
- throw new IllegalArgumentException("Invalid value type ("+value.getClass()+") for property '"+prop+"': expected Integer value.");
- }
-
- if (i < minValue) {
- throw new IllegalArgumentException("Invalid numeric value ("+i
- +") for property '"+prop
- +"': minimum is "+minValue+".");
- }
- return i;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/BaseNsContext.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/BaseNsContext.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/BaseNsContext.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/BaseNsContext.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,132 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.util;
-
-import java.io.IOException;
-import java.io.Writer;
-import java.util.Iterator;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamWriter;
-
-import org.codehaus.stax2.ri.SingletonIterator;
-
-import com.ctc.wstx.cfg.ErrorConsts;
-
-/**
- * Abstract base class that defines extra features defined by most
- * NamespaceContext implementations Wodstox uses.
- */
-public abstract class BaseNsContext
- implements NamespaceContext
-{
- /**
- * This is the URI returned for default namespace, when it hasn't
- * been explicitly declared; could be either "" or null.
- */
- protected final static String UNDECLARED_NS_URI = "";
-
- /*
- /////////////////////////////////////////////
- // NamespaceContext API
- /////////////////////////////////////////////
- */
-
- public final String getNamespaceURI(String prefix)
- {
- /* First the known offenders; invalid args, 2 predefined xml namespace
- * prefixes
- */
- if (prefix == null) {
- throw new IllegalArgumentException(ErrorConsts.ERR_NULL_ARG);
- }
- if (prefix.length() > 0) {
- if (prefix.equals(XMLConstants.XML_NS_PREFIX)) {
- return XMLConstants.XML_NS_URI;
- }
- if (prefix.equals(XMLConstants.XMLNS_ATTRIBUTE)) {
- return XMLConstants.XMLNS_ATTRIBUTE_NS_URI;
- }
- }
- return doGetNamespaceURI(prefix);
- }
-
- public final String getPrefix(String nsURI)
- {
- /* First the known offenders; invalid args, 2 predefined xml namespace
- * prefixes
- */
- if (nsURI == null || nsURI.length() == 0) {
- throw new IllegalArgumentException("Illegal to pass null/empty prefix as argument.");
- }
- if (nsURI.equals(XMLConstants.XML_NS_URI)) {
- return XMLConstants.XML_NS_PREFIX;
- }
- if (nsURI.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- return XMLConstants.XMLNS_ATTRIBUTE;
- }
- return doGetPrefix(nsURI);
- }
-
- public final Iterator getPrefixes(String nsURI)
- {
- /* First the known offenders; invalid args, 2 predefined xml namespace
- * prefixes
- */
- if (nsURI == null || nsURI.length() == 0) {
- throw new IllegalArgumentException("Illegal to pass null/empty prefix as argument.");
- }
- if (nsURI.equals(XMLConstants.XML_NS_URI)) {
- return new SingletonIterator(XMLConstants.XML_NS_PREFIX);
- }
- if (nsURI.equals(XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
- return new SingletonIterator(XMLConstants.XMLNS_ATTRIBUTE);
- }
-
- return doGetPrefixes(nsURI);
- }
-
- /*
- /////////////////////////////////////////////
- // Extended API
- /////////////////////////////////////////////
- */
-
- public abstract Iterator getNamespaces();
-
- /**
- * Method called by the matching start element class to
- * output all namespace declarations active in current namespace
- * scope, if any.
- */
- public abstract void outputNamespaceDeclarations(Writer w) throws IOException;
-
- public abstract void outputNamespaceDeclarations(XMLStreamWriter w) throws XMLStreamException;
-
- /*
- /////////////////////////////////////////////////
- // Template methods sub-classes need to implement
- /////////////////////////////////////////////////
- */
-
- public abstract String doGetNamespaceURI(String prefix);
-
- public abstract String doGetPrefix(String nsURI);
-
- public abstract Iterator doGetPrefixes(String nsURI);
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/BijectiveNsMap.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/BijectiveNsMap.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/BijectiveNsMap.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/BijectiveNsMap.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,326 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2005 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.util;
-
-import java.util.*;
-
-import javax.xml.XMLConstants;
-import javax.xml.namespace.NamespaceContext;
-
-import com.ctc.wstx.util.DataUtil;
-
-/**
- * Helper class that implements "bijective map" (Map that allows use of values
- * as keys and vice versa, bidirectional access), and is specifically
- * used for storing namespace binding information.
- * One thing worth noting is that Strings stored are NOT assumed to have
- * been unified (interned) -- if they were, different implementation would
- * be more optimal.
- *mBuckets.length - 1, when mBuckets.length is
- * a power of two.
- */
- protected int mIndexMask;
-
- /*
- ////////////////////////////////////////
- // Linked list info
- ////////////////////////////////////////
- */
-
- protected ElementId mHead;
-
- protected ElementId mTail;
-
- /*
- ////////////////////////////////////////
- // Life-cycle:
- ////////////////////////////////////////
- */
-
- public ElementIdMap()
- {
- this(DEFAULT_SIZE);
- }
-
- /**
- * This constructor is mainly used for testing, as it can be sized
- * appropriately to test rehashing etc.
- */
- public ElementIdMap(int initialSize)
- {
- int actual = MIN_SIZE;
- while (actual < initialSize) {
- actual += actual;
- }
- mTable = new ElementId[actual];
- // Mask is easy to calc for powers of two.
- mIndexMask = actual - 1;
- mSize = 0;
- mSizeThreshold = (actual * FILL_PCT) / 100;
- mHead = mTail = null;
- }
-
- /*
- ////////////////////////////////////////////////////
- // Public API
- ////////////////////////////////////////////////////
- */
-
- public ElementId getFirstUndefined()
- {
- /* Since the linked list is pruned to always start with
- * the first (in doc order) undefined id, we can just
- * return head:
- */
- return mHead;
- }
-
- /**
- * Method called when a reference to id is encountered. If so, need
- * to check if specified id entry (ref or definiton) exists; and if not,
- * to add a reference marker.
- */
- public ElementId addReferenced(char[] buffer, int start, int len, int hash,
- Location loc, PrefixedName elemName, PrefixedName attrName)
- {
- int index = (hash & mIndexMask);
- ElementId id = mTable[index];
-
- while (id != null) {
- if (id.idMatches(buffer, start, len)) { // found existing one
- return id;
- }
- id = id.nextColliding();
- }
-
- // Not found, need to create a placeholder...
-
- // But first, do we need more room?
- if (mSize >= mSizeThreshold) {
- rehash();
- // Index changes, for the new entr:
- index = (hash & mIndexMask);
- }
- ++mSize;
-
- // Ok, then, let's create the entry
- String idStr = new String(buffer, start, len);
- id = new ElementId(idStr, loc, false, elemName, attrName);
-
- // First, let's link it to Map; all ids have to be connected
- id.setNextColliding(mTable[index]);
- mTable[index] = id;
-
- // And then add the undefined entry at the end of list
- if (mHead == null) {
- mHead = mTail = id;
- } else {
- mTail.linkUndefined(id);
- mTail = id;
- }
- return id;
- }
-
- public ElementId addReferenced(String idStr,
- Location loc, PrefixedName elemName, PrefixedName attrName)
- {
- int hash = calcHash(idStr);
- int index = (hash & mIndexMask);
- ElementId id = mTable[index];
-
- while (id != null) {
- if (id.idMatches(idStr)) { // found existing one
- return id;
- }
- id = id.nextColliding();
- }
-
- // Not found, need to create a placeholder...
-
- // But first, do we need more room?
- if (mSize >= mSizeThreshold) {
- rehash();
- // Index changes, for the new entr:
- index = (hash & mIndexMask);
- }
- ++mSize;
-
- // Ok, then, let's create the entry
- id = new ElementId(idStr, loc, false, elemName, attrName);
-
- // First, let's link it to Map; all ids have to be connected
- id.setNextColliding(mTable[index]);
- mTable[index] = id;
-
- // And then add the undefined entry at the end of list
- if (mHead == null) {
- mHead = mTail = id;
- } else {
- mTail.linkUndefined(id);
- mTail = id;
- }
- return id;
- }
-
- /**
- * Method called when an id definition is encountered. If so, need
- * to check if specified id entry (ref or definiton) exists. If not,
- * need to add the definition marker. If it does exist, need to
- * 'upgrade it', if it was a reference marker; otherwise need to
- * just return the old entry, and expect caller to check for dups
- * and report the error.
- */
- public ElementId addDefined(char[] buffer, int start, int len, int hash,
- Location loc, PrefixedName elemName, PrefixedName attrName)
- {
- int index = (hash & mIndexMask);
- ElementId id = mTable[index];
-
- while (id != null) {
- if (id.idMatches(buffer, start, len)) {
- break;
- }
- id = id.nextColliding();
- }
-
- /* Not found, can just add it to the Map; no need to add to the
- * linked list as it's not undefined
- */
- if (id == null) {
- // First, do we need more room?
- if (mSize >= mSizeThreshold) {
- rehash();
- index = (hash & mIndexMask);
- }
- ++mSize;
- String idStr = new String(buffer, start, len);
- id = new ElementId(idStr, loc, true, elemName, attrName);
- id.setNextColliding(mTable[index]);
- mTable[index] = id;
- } else {
- /* If already defined, nothing additional to do (we could
- * signal an error here, though... for now, we'll let caller
- * do that
- */
- if (id.isDefined()) {
- ;
- } else {
- /* Not defined, just need to upgrade, and possibly remove from
- * the linked list.
- */
- id.markDefined(loc);
-
- /* Ok; if it was the first undefined, need to unlink it, as
- * well as potentially next items.
- */
- if (id == mHead) {
- do {
- mHead = mHead.nextUndefined();
- } while (mHead != null && mHead.isDefined());
-
- // Did we clear up all undefined ids?
- if (mHead == null) {
- mTail = null;
- }
- }
- }
- }
-
- return id;
- }
-
- public ElementId addDefined(String idStr,
- Location loc, PrefixedName elemName, PrefixedName attrName)
- {
- int hash = calcHash(idStr);
- int index = (hash & mIndexMask);
- ElementId id = mTable[index];
-
- while (id != null) {
- if (id.idMatches(idStr)) {
- break;
- }
- id = id.nextColliding();
- }
-
- /* Not found, can just add it to the Map; no need to add to the
- * linked list as it's not undefined
- */
- if (id == null) {
- if (mSize >= mSizeThreshold) { // need more room
- rehash();
- index = (hash & mIndexMask);
- }
- ++mSize;
- id = new ElementId(idStr, loc, true, elemName, attrName);
- id.setNextColliding(mTable[index]);
- mTable[index] = id;
- } else {
- /* If already defined, nothing additional to do (we could
- * signal an error here, though... for now, we'll let caller
- * do that
- */
- if (id.isDefined()) {
- ;
- } else {
- /* Not defined, just need to upgrade, and possibly remove from
- * the linked list.
- */
- id.markDefined(loc);
-
- /* Ok; if it was the first undefined, need to unlink it, as
- * well as potentially next items.
- */
- if (id == mHead) {
- do {
- mHead = mHead.nextUndefined();
- } while (mHead != null && mHead.isDefined());
- if (mHead == null) { // cleared up all undefined ids?
- mTail = null;
- }
- }
- }
- }
-
- return id;
- }
-
- /**
- * Implementation of a hashing method for variable length
- * Strings. Most of the time intention is that this calculation
- * is done by caller during parsing, not here; however, sometimes
- * it needs to be done for parsed "String" too.
- *com.ctc package.
-
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/PrefixedName.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/PrefixedName.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/PrefixedName.java 2012-04-24 04:38:20.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/PrefixedName.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,185 +0,0 @@
-/* Woodstox XML processor
- *
- * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
- *
- * Licensed under the License specified in file LICENSE, included with
- * the source code.
- * You may not use this file except in compliance with the License.
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package com.ctc.wstx.util;
-
-import javax.xml.namespace.QName;
-
-/**
- * Simple key Object to be used for storing/accessing of potentially namespace
- * scoped element and attribute names.
- *
- * Finally, rehashing is also more expensive, as hash codes are not
- * stored; rehashing requires all entries' hash codes to be recalculated.
- * Reason for not storing hash codes is reduced memory usage, hoping
- * for better memory locality.
- *mBuckets.length - 1, when mBuckets.length is
- * a power of two.
- */
- protected int mIndexMask;
-
- /*
- ////////////////////////////////////////
- // Information about concurrency
- ////////////////////////////////////////
- */
-
- /**
- * Version of this table instance; used when deriving new concurrently
- * used versions from existing 'master' instance.
- */
- protected int mThisVersion;
-
- /**
- * Flag that indicates if any changes have been made to the data;
- * used to both determine if bucket array needs to be copied when
- * (first) change is made, and potentially if updated bucket list
- * is to be resync'ed back to master instance.
- */
- protected boolean mDirty;
-
- /*
- ////////////////////////////////////////
- // Life-cycle:
- ////////////////////////////////////////
- */
-
- /**
- * Method for constructing a master symbol table instance; this one
- * will create master instance with default size, and with interning
- * enabled.
- */
- public SymbolTable() {
- this(true);
- }
-
- /**
- * Method for constructing a master symbol table instance.
- */
- public SymbolTable(boolean internStrings) {
- this(internStrings, DEFAULT_TABLE_SIZE);
- }
-
- /**
- * Method for constructing a master symbol table instance.
- */
- public SymbolTable(boolean internStrings, int initialSize) {
- this(internStrings, initialSize, DEFAULT_FILL_FACTOR);
- }
-
- /**
- * Main method for constructing a master symbol table instance; will
- * be called by other public constructors.
- *
- * @param internStrings Whether Strings to add are intern()ed or not
- * @param initialSize Minimum initial size for bucket array; internally
- * will always use a power of two equal to or bigger than this value.
- * @param fillFactor Maximum fill factor allowed for bucket table;
- * when more entries are added, table will be expanded.
- */
- public SymbolTable(boolean internStrings, int initialSize,
- float fillFactor)
- {
- mInternStrings = internStrings;
- // Let's start versions from 1
- mThisVersion = 1;
- // And we'll also set flags so no copying of buckets is needed:
- mDirty = true;
-
- // No point in requesting funny initial sizes...
- if (initialSize < 1) {
- throw new IllegalArgumentException("Can not use negative/zero initial size: "+initialSize);
- }
- /* Initial size has to be a power of two. Also, let's not honour
- * sizes that are ridiculously small...
- */
- {
- int currSize = 4;
- while (currSize < initialSize) {
- currSize += currSize;
- }
- initialSize = currSize;
- }
-
- mSymbols = new String[initialSize];
- mBuckets = new Bucket[initialSize >> 1];
- // Mask is easy to calc for powers of two.
- mIndexMask = initialSize - 1;
- mSize = 0;
-
- // Sanity check for fill factor:
- if (fillFactor < 0.01f) {
- throw new IllegalArgumentException("Fill factor can not be lower than 0.01.");
- }
- if (fillFactor > 10.0f) { // just to catch stupid values, ie. useless from performance perspective
- throw new IllegalArgumentException("Fill factor can not be higher than 10.0.");
- }
- mSizeThreshold = (int) (initialSize * fillFactor + 0.5);
- }
-
- /**
- * Internal constructor used when creating child instances.
- */
- private SymbolTable(boolean internStrings, String[] symbols,
- Bucket[] buckets, int size, int sizeThreshold,
- int indexMask, int version)
- {
- mInternStrings = internStrings;
- mSymbols = symbols;
- mBuckets = buckets;
- mSize = size;
- mSizeThreshold = sizeThreshold;
- mIndexMask = indexMask;
- mThisVersion = version;
-
- // Need to make copies of arrays, if/when adding new entries
- mDirty = false;
- }
-
- /**
- * "Factory" method; will create a new child instance of this symbol
- * table. It will be a copy-on-write instance, ie. it will only use
- * read-only copy of parent's data, but when changes are needed, a
- * copy will be created.
- *StringBuilder instead.
- */
- private StringBuffer mBuilder = null;
-
- public TextAccumulator() { }
-
- public boolean hasText() {
- return (mBuilder != null) || (mText != null);
- }
-
- public void addText(String text)
- {
- int len = text.length();
- if (len > 0) {
- // Any prior text?
- if (mText != null) {
- mBuilder = new StringBuffer(mText.length() + len);
- mBuilder.append(mText);
- mText = null;
- }
- if (mBuilder != null) {
- mBuilder.append(text);
- } else {
- mText = text;
- }
- }
- }
-
- public void addText(char[] buf, int start, int end)
- {
- int len = end-start;
- if (len > 0) {
- // Any prior text?
- if (mText != null) {
- mBuilder = new StringBuffer(mText.length() + len);
- mBuilder.append(mText);
- mText = null;
- } else if (mBuilder == null) {
- /* more efficient to use a builder than a string; and although
- * could use a char array, StringBuilder has the benefit of
- * being able to share the array, eventually.
- */
- mBuilder = new StringBuffer(len);
- }
- mBuilder.append(buf, start, end-start);
- }
- }
-
- public String getAndClear()
- {
- String result;
-
- if (mText != null) {
- result = mText;
- mText = null;
- } else if (mBuilder != null) {
- result = mBuilder.toString();
- mBuilder = null;
- } else {
- result = "";
- }
- return result;
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextBuffer.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextBuffer.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/TextBuffer.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/TextBuffer.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,1406 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.io.*;
-import java.util.ArrayList;
-import javax.xml.stream.Location;
-import javax.xml.stream.XMLStreamException;
-
-import org.xml.sax.ContentHandler;
-import org.xml.sax.SAXException;
-import org.xml.sax.ext.LexicalHandler;
-
-import org.codehaus.stax2.typed.Base64Variant;
-import org.codehaus.stax2.typed.TypedArrayDecoder;
-import org.codehaus.stax2.typed.TypedValueDecoder;
-import org.codehaus.stax2.typed.TypedXMLStreamException;
-import org.codehaus.stax2.validation.XMLValidator;
-
-import org.codehaus.stax2.ri.typed.CharArrayBase64Decoder;
-
-import com.ctc.wstx.api.ReaderConfig;
-import com.ctc.wstx.dtd.DTDEventListener;
-import com.ctc.wstx.sr.InputProblemReporter;
-import com.ctc.wstx.util.StringUtil;
-
-/**
- * TextBuffer is a class similar to {@link StringBuffer}, with
- * following differences:
- *
- *
- *end - start equals word length (similar
- * to the way String.substring() has).
- *
- * @return (Shared) string instance of the word, if it exists in
- * the word set; null if not.
- */
- public String find(char[] str, final int start, final int end)
- {
- char[] data = mData;
-
- // 03-Jan-2006, TSa: Special case; one entry
- if (data == null) {
- return findFromOne(str, start, end);
- }
-
- int ptr = 0; // pointer to compressed set data
- int offset = start;
-
- while (true) {
- // End of input String? Need to match the runt entry!
- if (offset == end) {
- if (data[ptr+1] == CHAR_NULL) {
- return mWords[data[ptr+2] - NEGATIVE_OFFSET];
- }
- return null;
- }
-
- int count = data[ptr++];
- // Need to find the branch to follow, if any
- char c = str[offset++];
-
- inner_block:
- do { // dummy loop, need to have break
- // Linear or binary search?
- if (count < MIN_BINARY_SEARCH) {
- // always at least two branches; never less
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- break inner_block;
- }
- if (data[ptr+2] == c) {
- ptr = (int) data[ptr+3];
- break inner_block;
- }
- int branchEnd = ptr + (count << 1);
- // Starts from entry #3, if such exists
- for (ptr += 4; ptr < branchEnd; ptr += 2) {
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- break inner_block;
- }
- }
- return null; // No match!
- } else { // Ok, binary search:
- int low = 0;
- int high = count-1;
- int mid;
-
- while (low <= high) {
- mid = (low + high) >> 1;
- int ix = ptr + (mid << 1);
- int diff = data[ix] - c;
- if (diff > 0) { // char was 'higher', need to go down
- high = mid-1;
- } else if (diff < 0) { // lower, need to go up
- low = mid+1;
- } else { // match (so far)
- ptr = (int) data[ix+1];
- break inner_block;
- }
- }
- return null; // No match!
- }
- } while (false);
-
- // Ok; now, is it the end?
- if (ptr >= NEGATIVE_OFFSET) {
- String word = mWords[ptr - NEGATIVE_OFFSET];
- int expLen = (end - start);
- if (word.length() != expLen) {
- return null;
- }
- for (int i = offset - start; offset < end; ++i, ++offset) {
- if (word.charAt(i) != str[offset]) {
- return null;
- }
- }
- return word;
- }
- }
- // never gets here
- }
-
- private String findFromOne(char[] str, final int start, final int end)
- {
- String word = mWords[0];
- int len = end-start;
- if (word.length() != len) {
- return null;
- }
- for (int i = 0; i < len; ++i) {
- if (word.charAt(i) != str[start+i]) {
- return null;
- }
- }
- return word;
- }
-
- /**
- * @return (Shared) string instance of the word, if it exists in
- * the word set; null if not.
- */
- public String find(String str)
- {
- char[] data = mData;
-
- // 03-Jan-2006, TSa: Special case; one entry
- if (data == null) {
- String word = mWords[0];
- return word.equals(str) ? word : null;
- }
-
- int ptr = 0; // pointer to compressed set data
- int offset = 0;
- int end = str.length();
-
- while (true) {
- // End of input String? Need to match the runt entry!
- if (offset == end) {
- if (data[ptr+1] == CHAR_NULL) {
- return mWords[data[ptr+2] - NEGATIVE_OFFSET];
- }
- return null;
- }
-
- int count = data[ptr++];
- // Need to find the branch to follow, if any
- char c = str.charAt(offset++);
-
- inner_block:
- do { // dummy loop, need to have break
- // Linear or binary search?
- if (count < MIN_BINARY_SEARCH) {
- // always at least two branches; never less
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- break inner_block;
- }
- if (data[ptr+2] == c) {
- ptr = (int) data[ptr+3];
- break inner_block;
- }
- int branchEnd = ptr + (count << 1);
- // Starts from entry #3, if such exists
- for (ptr += 4; ptr < branchEnd; ptr += 2) {
- if (data[ptr] == c) {
- ptr = (int) data[ptr+1];
- break inner_block;
- }
- }
- return null; // No match!
- } else { // Ok, binary search:
- int low = 0;
- int high = count-1;
- int mid;
-
- while (low <= high) {
- mid = (low + high) >> 1;
- int ix = ptr + (mid << 1);
- int diff = data[ix] - c;
- if (diff > 0) { // char was 'higher', need to go down
- high = mid-1;
- } else if (diff < 0) { // lower, need to go up
- low = mid+1;
- } else { // match (so far)
- ptr = (int) data[ix+1];
- break inner_block;
- }
- }
- return null; // No match!
- }
- } while (false);
-
- // Ok; now, is it the end?
- if (ptr >= NEGATIVE_OFFSET) {
- String word = mWords[ptr - NEGATIVE_OFFSET];
- if (word.length() != str.length()) {
- return null;
- }
- for (; offset < end; ++offset) {
- if (word.charAt(offset) != str.charAt(offset)) {
- return null;
- }
- }
- return word;
- }
- }
- // never gets here
- }
-
- /*
- ////////////////////////////////////////////////
- // Re-defined public methods
- ////////////////////////////////////////////////
- */
-
- public String toString()
- {
- StringBuffer sb = new StringBuffer(16 + (mWords.length << 3));
- for (int i = 0, len = mWords.length; i < len; ++i) {
- if (i > 0) {
- sb.append(", ");
- }
- sb.append(mWords[i]);
- }
- return sb.toString();
- }
-
- /*
- ////////////////////////////////////////////////
- // Private methods
- ////////////////////////////////////////////////
- */
-
- /*
- ////////////////////////////////////////////////
- // Helper classes
- ////////////////////////////////////////////////
- */
-
- private final static class Builder
- {
- final String[] mWords;
-
- char[] mData;
-
- /**
- * Number of characters currently used from mData
- */
- int mSize;
-
- public Builder(TreeSet wordSet)
- {
- int wordCount = wordSet.size();
-
- mWords = new String[wordCount];
- wordSet.toArray(mWords);
-
- /* 03-Jan-2006, TSa: Special case: just one entry; if so,
- * let's leave char array null, and just have the String
- * array with one entry.
- */
- if (wordCount < 2) {
- if (wordCount == 0) {
- throw new IllegalArgumentException(); // not legal
- }
- mData = null;
- } else {
- /* Let's guess approximate size we should need, assuming
- * average word length of 6 characters, overhead matching
- * compression (ie. about 1-to-1 ratio overall)
- */
- int size = wordCount * 6;
- if (size < 256) {
- size = 256;
- }
- mData = new char[size];
- }
- }
-
- /**
- * @return Raw character data that contains compressed structure
- * of the word set
- */
- public WordResolver construct()
- {
- char[] result;
-
- /* 03-Jan-2006, TSa: Special case: just one entry; if so,
- * let's leave char array null, and just have the String
- * array with one entry.
- */
- if (mData == null) {
- result = null;
- } else {
- constructBranch(0, 0, mWords.length);
-
- // Too big?
- if (mSize > NEGATIVE_OFFSET) {
- return null;
- }
-
- result = new char[mSize];
- System.arraycopy(mData, 0, result, 0, mSize);
- }
-
- return new WordResolver(mWords, result);
- }
-
- /**
- * Method that is called recursively to build the data
- * representation for a branch, ie. part of word set tree
- * that still has more than one ending
- *
- * @param charIndex Index of the character in words to consider
- * for this round
- * @param start Index of the first word to be processed
- * @param end Index of the word after last word to be processed
- * (so that number of words is end - start - 1
- */
- private void constructBranch(int charIndex, int start, int end)
- {
- // If more than one entry, need to divide into groups
-
- // First, need to add placeholder for branch count:
- if (mSize >= mData.length) {
- expand(1);
- }
- mData[mSize++] = 0; // placeholder!
- /* structStart will point to second char of first entry
- * (which will temporarily have entry count, eventually 'link'
- * to continuation)
- */
- int structStart = mSize + 1;
- int groupCount = 0;
- int groupStart = start;
- String[] words = mWords;
- boolean gotRunt;
-
- /* First thing we need to do is a special check for the
- * first entry -- it may be "runt" word, one that has no
- * more chars but also has a longer version ("id" vs.
- * "identifier"). If so, it needs to be marked; this is done
- * by adding a special entry before other entries (since such
- * entry would always be ordered first alphabetically)
- */
- if (words[groupStart].length() == charIndex) { // yup, got one:
- if ((mSize + 2) > mData.length) {
- expand(2);
- }
- /* First null marks the "missing" char (or, end-of-word);
- * and then we need the index
- */
- mData[mSize++] = CHAR_NULL;
- mData[mSize++] = (char) (NEGATIVE_OFFSET + groupStart);
-
- // Ok, let's then ignore that entry
- ++groupStart;
- ++groupCount;
- gotRunt = true;
- } else {
- gotRunt = false;
- }
-
- // Ok, then, let's find the ('real') groupings:
- while (groupStart < end) {
- // Inner loop, let's find the group:
- char c = words[groupStart].charAt(charIndex);
- int j = groupStart+1;
- while (j < end && words[j].charAt(charIndex) == c) {
- ++j;
- }
- /* Ok, let's store the char in there, along with count;
- * count will be needed in second, and will then get
- * overwritten with actual data later on
- */
- if ((mSize + 2) > mData.length) {
- expand(2);
- }
- mData[mSize++] = c;
- mData[mSize++] = (char) (j - groupStart); // entries in group
- groupStart = j;
- ++groupCount;
- }
-
- /* Ok, groups found; need to loop through them, recursively
- * calling branch and/or leaf methods
- */
- // first let's output the header, ie. group count:
- mData[structStart-2] = (char) groupCount;
- groupStart = start;
-
- // Do we have the "runt" to skip?
- if (gotRunt) {
- structStart += 2;
- ++groupStart;
- }
-
- int structEnd = mSize;
- ++charIndex;
- for (; structStart < structEnd; structStart += 2) {
- groupCount = (int) mData[structStart]; // no sign expansion, is ok
- /* Ok, count gotten, can either create a branch (if more than
- * one entry) or leaf (just one entry)
- */
- if (groupCount == 1) {
- mData[structStart] = (char) (NEGATIVE_OFFSET + groupStart);
- } else {
- mData[structStart] = (char) mSize;
- constructBranch(charIndex, groupStart,
- groupStart + groupCount);
- }
- groupStart += groupCount;
- }
-
- // done!
- }
-
- private char[] expand(int needSpace)
- {
- char[] old = mData;
- int len = old.length;
- int newSize = len + ((len < 4096) ? len : (len >> 1));
-
- /* Let's verify we get enough; should always be true but
- * better safe than sorry
- */
- if (newSize < (mSize + needSpace)) {
- newSize = mSize + needSpace + 64;
- }
- mData = new char[newSize];
- System.arraycopy(old, 0, mData, 0, len);
- return mData;
- }
- }
-
- /*
- ////////////////////////////////////////////////////
- // Simple test driver, useful for debugging
- // (uncomment if needed -- commented out so it won't
- // affect coverage testing)
- ////////////////////////////////////////////////////
- */
-
- /*
- public static void main(String[] args)
- {
- if (args.length < 2) {
- System.err.println("Usage: "+WordResolver.class+" word1 [word2] ... [wordN] keyword");
- System.exit(1);
- }
- String key = args[args.length-1];
- TreeSet words = new TreeSet();
- for (int i = 0; i < args.length-1; ++i) {
- words.add(args[i]);
- }
-
- WordResolver set = WordResolver.constructInstance(words);
-
-//outputData(set.mData);
-
- // Ok, and then the test!
- char[] keyA = new char[key.length() + 4];
- key.getChars(0, key.length(), keyA, 2);
- //System.out.println("Word '"+key+"' found via array search: "+WordResolver.find(data, keyA, 2, key.length() + 2));
- System.out.println("Word '"+key+"' found via array search: "+set.find(keyA, 2, key.length() + 2));
- }
-
- static void outputData(char[] data)
- {
- for (int i = 0; i < data.length; ++i) {
- char c = data[i];
- System.out.print(Integer.toHexString(i)+" ["+Integer.toHexString(c)+"]");
- if (c > 32 && c <= 127) { // printable char (letter)
- System.out.println(" -> '"+c+"'");
- } else {
- System.out.println();
- }
- }
- }
- */
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/WordSet.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/WordSet.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/WordSet.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/WordSet.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,476 +0,0 @@
-package com.ctc.wstx.util;
-
-import java.util.*;
-
-/**
- * An efficient (both memory and time) implementation of a Set used to
- * verify that a given
- * word is contained within the set. The general usage pattern is expected
- * to be such that most checks are positive, ie. that the word indeed
- * is contained in the set.
- *
- *
- *end - start - 1
- */
- private void constructBranch(int charIndex, int start, int end)
- {
- // If more than one entry, need to divide into groups
-
- // First, need to add placeholder for branch count:
- if (mSize >= mData.length) {
- expand(1);
- }
- mData[mSize++] = 0; // placeholder!
- /* structStart will point to second char of first entry
- * (which will temporarily have entry count, eventually 'link'
- * to continuation)
- */
- int structStart = mSize + 1;
- int groupCount = 0;
- int groupStart = start;
- String[] words = mWords;
-
- /* First thing we need to do is a special check for the
- * first entry -- it may be "runt" word, one that has no
- * more chars but also has a longer version ("id" vs.
- * "identifier"). If there is such a word, it'll always
- * be first in alphabetic ordering:
- */
- if (words[groupStart].length() == charIndex) { // yup, got one:
- if ((mSize + 2) > mData.length) {
- expand(2);
- }
- /* Nulls mark both imaginary branching null char and
- * "missing link" to the rest
- */
- mData[mSize++] = CHAR_NULL;
- mData[mSize++] = CHAR_NULL;
-
- // Ok, let's then ignore that entry
- ++groupStart;
- ++groupCount;
- }
-
- // Ok, then, let's find the ('real') groupings:
- while (groupStart < end) {
- // Inner loop, let's find the group:
- char c = words[groupStart].charAt(charIndex);
- int j = groupStart+1;
- while (j < end && words[j].charAt(charIndex) == c) {
- ++j;
- }
- /* Ok, let's store the char in there, along with count;
- * count will be needed in second, and will then get
- * overwritten with actual data later on
- */
- if ((mSize + 2) > mData.length) {
- expand(2);
- }
- mData[mSize++] = c;
- mData[mSize++] = (char) (j - groupStart); // entries in group
- groupStart = j;
- ++groupCount;
- }
-
- /* Ok, groups found; need to loop through them, recursively
- * calling branch and/or leaf methods
- */
- // first let's output the header, ie. group count:
- mData[structStart-2] = (char) groupCount;
- groupStart = start;
-
- // Do we have the "runt" to skip?
- if (mData[structStart] == CHAR_NULL) {
- structStart += 2;
- ++groupStart;
- }
-
- int structEnd = mSize;
- ++charIndex;
- for (; structStart < structEnd; structStart += 2) {
- groupCount = (int) mData[structStart]; // no sign expansion, is ok
- // Ok, count gotten, can now put the 'link' (pointer) in there
- mData[structStart] = (char) mSize;
- if (groupCount == 1) {
- /* One optimization; if it'd lead to a single runt
- * entry, we can just add 'null' link:
- */
- String word = words[groupStart];
- if (word.length() == charIndex) {
- mData[structStart] = CHAR_NULL;
- } else { // otherwise, let's just create end state:
- constructLeaf(charIndex, groupStart);
- }
- } else {
- constructBranch(charIndex, groupStart,
- groupStart + groupCount);
- }
- groupStart += groupCount;
- }
-
- // done!
- }
-
- /**
- * Method called to add leaf entry to word set; basically
- * "here is the rest of the only matching word"
- */
- private void constructLeaf(int charIndex, int wordIndex)
- {
- String word = mWords[wordIndex];
- int len = word.length();
- char[] data = mData;
-
- // need room for 1 header char, rest of the word
- if ((mSize + len + 1) >= data.length) {
- data = expand(len+1);
- }
-
- data[mSize++] = (char) (NEGATIVE_OFFSET + (len - charIndex));
- for (; charIndex < len; ++charIndex) {
- data[mSize++] = word.charAt(charIndex);
- }
- }
-
- private char[] expand(int needSpace)
- {
- char[] old = mData;
- int len = old.length;
- int newSize = len + ((len < 4096) ? len : (len >> 1));
-
- /* Let's verify we get enough; should always be true but
- * better safe than sorry
- */
- if (newSize < (mSize + needSpace)) {
- newSize = mSize + needSpace + 64;
- }
- mData = new char[newSize];
- System.arraycopy(old, 0, mData, 0, len);
- return mData;
- }
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/XmlChars.java libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/XmlChars.java
--- libwoodstox-java-4.1.3/src/java/com/ctc/wstx/util/XmlChars.java 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/com/ctc/wstx/util/XmlChars.java 1970-01-01 00:00:00.000000000 +0000
@@ -1,532 +0,0 @@
-package com.ctc.wstx.util;
-
-/**
- * Simple utility class that encapsulates logic of determining validity
- * of characters outside basic 7-bit range of Unicode, for XML 1.0
- */
-public final class XmlChars
-{
- /* We don't need full 64k bits... (0x80 - 0x312C) / 32. But to
- * simplify things, let's just include first 0x80 entries in there etc
- */
- final static int SIZE = (0x3140 >> 5); // 32 bits per int
-
- final static int[] sXml10StartChars = new int[SIZE];
- static {
- SETBITS(sXml10StartChars, 0xC0, 0xD6);
- SETBITS(sXml10StartChars, 0xD8, 0xF6);
- SETBITS(sXml10StartChars, 0xF8, 0xFF);
- SETBITS(sXml10StartChars, 0x100, 0x131);
- SETBITS(sXml10StartChars, 0x134, 0x13e);
- SETBITS(sXml10StartChars, 0x141, 0x148);
- SETBITS(sXml10StartChars, 0x14a, 0x17e);
- SETBITS(sXml10StartChars, 0x180, 0x1c3);
- SETBITS(sXml10StartChars, 0x1cd, 0x1f0);
- SETBITS(sXml10StartChars, 0x1f4, 0x1f5);
- SETBITS(sXml10StartChars, 0x1fa, 0x217);
- SETBITS(sXml10StartChars, 0x250, 0x2a8);
- SETBITS(sXml10StartChars, 0x2bb, 0x2c1);
- SETBITS(sXml10StartChars, 0x386);
- SETBITS(sXml10StartChars, 0x388, 0x38a);
- SETBITS(sXml10StartChars, 0x38c);
- SETBITS(sXml10StartChars, 0x38e, 0x3a1);
- SETBITS(sXml10StartChars, 0x3a3, 0x3ce);
- SETBITS(sXml10StartChars, 0x3d0, 0x3d6);
- SETBITS(sXml10StartChars, 0x3da);
- SETBITS(sXml10StartChars, 0x3dc);
- SETBITS(sXml10StartChars, 0x3de);
- SETBITS(sXml10StartChars, 0x3e0);
- SETBITS(sXml10StartChars, 0x3e2, 0x3f3);
- SETBITS(sXml10StartChars, 0x401, 0x40c);
- SETBITS(sXml10StartChars, 0x40e, 0x44f);
- SETBITS(sXml10StartChars, 0x451, 0x45c);
- SETBITS(sXml10StartChars, 0x45e, 0x481);
- SETBITS(sXml10StartChars, 0x490, 0x4c4);
- SETBITS(sXml10StartChars, 0x4c7, 0x4c8);
- SETBITS(sXml10StartChars, 0x4cb, 0x4cc);
- SETBITS(sXml10StartChars, 0x4d0, 0x4eb);
- SETBITS(sXml10StartChars, 0x4ee, 0x4f5);
- SETBITS(sXml10StartChars, 0x4f8, 0x4f9);
-
- SETBITS(sXml10StartChars, 0x531, 0x556);
- SETBITS(sXml10StartChars, 0x559);
- SETBITS(sXml10StartChars, 0x561, 0x586);
- SETBITS(sXml10StartChars, 0x5d0, 0x5ea);
- SETBITS(sXml10StartChars, 0x5f0, 0x5f2);
- SETBITS(sXml10StartChars, 0x621, 0x63a);
- SETBITS(sXml10StartChars, 0x641, 0x64a);
- SETBITS(sXml10StartChars, 0x671, 0x6b7);
- SETBITS(sXml10StartChars, 0x6ba, 0x6be);
- SETBITS(sXml10StartChars, 0x6c0, 0x6ce);
- SETBITS(sXml10StartChars, 0x6d0, 0x6d3);
- SETBITS(sXml10StartChars, 0x6d5);
-
- SETBITS(sXml10StartChars, 0x6e5, 0x6e6);
- SETBITS(sXml10StartChars, 0x905, 0x939);
- SETBITS(sXml10StartChars, 0x93d);
- SETBITS(sXml10StartChars, 0x958, 0x961);
- SETBITS(sXml10StartChars, 0x985, 0x98c);
- SETBITS(sXml10StartChars, 0x98f, 0x990);
- SETBITS(sXml10StartChars, 0x993, 0x9a8);
- SETBITS(sXml10StartChars, 0x9aa, 0x9b0);
- SETBITS(sXml10StartChars, 0x9b2);
- SETBITS(sXml10StartChars, 0x9b6, 0x9b9);
- SETBITS(sXml10StartChars, 0x9dc);
- SETBITS(sXml10StartChars, 0x9dd);
- SETBITS(sXml10StartChars, 0x9df, 0x9e1);
- SETBITS(sXml10StartChars, 0x9f0); SETBITS(sXml10StartChars, 0x9f1);
- SETBITS(sXml10StartChars, 0xA05, 0xA0A);
- SETBITS(sXml10StartChars, 0xA0F); SETBITS(sXml10StartChars, 0xA10);
- SETBITS(sXml10StartChars, 0xA13, 0xA28);
- SETBITS(sXml10StartChars, 0xA2A, 0xA30);
- SETBITS(sXml10StartChars, 0xA32); SETBITS(sXml10StartChars, 0xA33);
- SETBITS(sXml10StartChars, 0xA35); SETBITS(sXml10StartChars, 0xA36);
- SETBITS(sXml10StartChars, 0xA38); SETBITS(sXml10StartChars, 0xA39);
- SETBITS(sXml10StartChars, 0xA59, 0xA5C);
- SETBITS(sXml10StartChars, 0xA5E);
- SETBITS(sXml10StartChars, 0xA72, 0xA74);
- SETBITS(sXml10StartChars, 0xA85, 0xA8B);
- SETBITS(sXml10StartChars, 0xA8D);
- SETBITS(sXml10StartChars, 0xA8F, 0xA91);
- SETBITS(sXml10StartChars, 0xA93, 0xAA8);
- SETBITS(sXml10StartChars, 0xAAA, 0xAB0);
- SETBITS(sXml10StartChars, 0xAB2, 0xAB3);
- SETBITS(sXml10StartChars, 0xAB5, 0xAB9);
- SETBITS(sXml10StartChars, 0xABD);
- SETBITS(sXml10StartChars, 0xAE0);
- SETBITS(sXml10StartChars, 0xB05, 0xB0C);
- SETBITS(sXml10StartChars, 0xB0F); SETBITS(sXml10StartChars, 0xB10);
- SETBITS(sXml10StartChars, 0xB13, 0xB28);
-
- SETBITS(sXml10StartChars, 0xB2A, 0xB30);
- SETBITS(sXml10StartChars, 0xB32); SETBITS(sXml10StartChars, 0xB33);
- SETBITS(sXml10StartChars, 0xB36, 0xB39);
- SETBITS(sXml10StartChars, 0xB3D);
- SETBITS(sXml10StartChars, 0xB5C); SETBITS(sXml10StartChars, 0xB5D);
- SETBITS(sXml10StartChars, 0xB5F, 0xB61);
- SETBITS(sXml10StartChars, 0xB85, 0xB8A);
- SETBITS(sXml10StartChars, 0xB8E, 0xB90);
-
- SETBITS(sXml10StartChars, 0xB92, 0xB95);
- SETBITS(sXml10StartChars, 0xB99, 0xB9A);
- SETBITS(sXml10StartChars, 0xB9C);
- SETBITS(sXml10StartChars, 0xB9E); SETBITS(sXml10StartChars, 0xB9F);
- SETBITS(sXml10StartChars, 0xBA3); SETBITS(sXml10StartChars, 0xBA4);
- SETBITS(sXml10StartChars, 0xBA8, 0xBAA);
- SETBITS(sXml10StartChars, 0xBAE, 0xBB5);
- SETBITS(sXml10StartChars, 0xBB7, 0xBB9);
- SETBITS(sXml10StartChars, 0xC05, 0xC0C);
- SETBITS(sXml10StartChars, 0xC0E, 0xC10);
-
- SETBITS(sXml10StartChars, 0xC12, 0xC28);
- SETBITS(sXml10StartChars, 0xC2A, 0xC33);
- SETBITS(sXml10StartChars, 0xC35, 0xC39);
- SETBITS(sXml10StartChars, 0xC60); SETBITS(sXml10StartChars, 0xC61);
- SETBITS(sXml10StartChars, 0xC85, 0xC8C);
- SETBITS(sXml10StartChars, 0xC8E, 0xC90);
- SETBITS(sXml10StartChars, 0xC92, 0xCA8);
- SETBITS(sXml10StartChars, 0xCAA, 0xCB3);
- SETBITS(sXml10StartChars, 0xCB5, 0xCB9);
- SETBITS(sXml10StartChars, 0xCDE);
- SETBITS(sXml10StartChars, 0xCE0); SETBITS(sXml10StartChars, 0xCE1);
- SETBITS(sXml10StartChars, 0xD05, 0xD0C);
- SETBITS(sXml10StartChars, 0xD0E, 0xD10);
- SETBITS(sXml10StartChars, 0xD12, 0xD28);
- SETBITS(sXml10StartChars, 0xD2A, 0xD39);
- SETBITS(sXml10StartChars, 0xD60); SETBITS(sXml10StartChars, 0xD61);
- SETBITS(sXml10StartChars, 0xE01, 0xE2E);
- SETBITS(sXml10StartChars, 0xE30);
- SETBITS(sXml10StartChars, 0xE32); SETBITS(sXml10StartChars, 0xE33);
- SETBITS(sXml10StartChars, 0xE40, 0xE45);
- SETBITS(sXml10StartChars, 0xE81); SETBITS(sXml10StartChars, 0xE82);
- SETBITS(sXml10StartChars, 0xE84);
- SETBITS(sXml10StartChars, 0xE87); SETBITS(sXml10StartChars, 0xE88);
- SETBITS(sXml10StartChars, 0xE8A); SETBITS(sXml10StartChars, 0xE8D);
- SETBITS(sXml10StartChars, 0xE94, 0xE97);
- SETBITS(sXml10StartChars, 0xE99, 0xE9F);
- SETBITS(sXml10StartChars, 0xEA1, 0xEA3);
- SETBITS(sXml10StartChars, 0xEA5); SETBITS(sXml10StartChars, 0xEA7);
- SETBITS(sXml10StartChars, 0xEAA); SETBITS(sXml10StartChars, 0xEAB);
- SETBITS(sXml10StartChars, 0xEAD); SETBITS(sXml10StartChars, 0xEAE);
- SETBITS(sXml10StartChars, 0xEB0);
- SETBITS(sXml10StartChars, 0xEB2); SETBITS(sXml10StartChars, 0xEB3);
- SETBITS(sXml10StartChars, 0xEBD);
-
- SETBITS(sXml10StartChars, 0xEC0, 0xEC4);
- SETBITS(sXml10StartChars, 0xF40, 0xF47);
- SETBITS(sXml10StartChars, 0xF49, 0xF69);
- SETBITS(sXml10StartChars, 0x10a0, 0x10c5);
- SETBITS(sXml10StartChars, 0x10d0, 0x10f6);
- SETBITS(sXml10StartChars, 0x1100);
- SETBITS(sXml10StartChars, 0x1102, 0x1103);
- SETBITS(sXml10StartChars, 0x1105, 0x1107);
- SETBITS(sXml10StartChars, 0x1109);
- SETBITS(sXml10StartChars, 0x110b, 0x110c);
- SETBITS(sXml10StartChars, 0x110e, 0x1112);
- SETBITS(sXml10StartChars, 0x113c);
- SETBITS(sXml10StartChars, 0x113e);
- SETBITS(sXml10StartChars, 0x1140);
- SETBITS(sXml10StartChars, 0x114c);
- SETBITS(sXml10StartChars, 0x114e);
- SETBITS(sXml10StartChars, 0x1150);
- SETBITS(sXml10StartChars, 0x1154, 0x1155);
- SETBITS(sXml10StartChars, 0x1159);
- SETBITS(sXml10StartChars, 0x115f, 0x1161);
- SETBITS(sXml10StartChars, 0x1163);
- SETBITS(sXml10StartChars, 0x1165);
- SETBITS(sXml10StartChars, 0x1167);
- SETBITS(sXml10StartChars, 0x1169);
- SETBITS(sXml10StartChars, 0x116d, 0x116e);
- SETBITS(sXml10StartChars, 0x1172, 0x1173);
- SETBITS(sXml10StartChars, 0x1175);
- SETBITS(sXml10StartChars, 0x119e);
- SETBITS(sXml10StartChars, 0x11a8);
- SETBITS(sXml10StartChars, 0x11ab);
- SETBITS(sXml10StartChars, 0x11ae, 0x11af);
- SETBITS(sXml10StartChars, 0x11b7, 0x11b8);
- SETBITS(sXml10StartChars, 0x11ba);
- SETBITS(sXml10StartChars, 0x11bc, 0x11c2);
- SETBITS(sXml10StartChars, 0x11eb);
- SETBITS(sXml10StartChars, 0x11f0);
- SETBITS(sXml10StartChars, 0x11f9);
- SETBITS(sXml10StartChars, 0x1e00, 0x1e9b);
- SETBITS(sXml10StartChars, 0x1ea0, 0x1ef9);
- SETBITS(sXml10StartChars, 0x1f00, 0x1f15);
- SETBITS(sXml10StartChars, 0x1f18, 0x1f1d);
- SETBITS(sXml10StartChars, 0x1f20, 0x1f45);
- SETBITS(sXml10StartChars, 0x1f48, 0x1f4d);
- SETBITS(sXml10StartChars, 0x1f50, 0x1f57);
- SETBITS(sXml10StartChars, 0x1f59);
- SETBITS(sXml10StartChars, 0x1f5b);
- SETBITS(sXml10StartChars, 0x1f5d);
- SETBITS(sXml10StartChars, 0x1f5f, 0x1f7d);
- SETBITS(sXml10StartChars, 0x1f80, 0x1fb4);
- SETBITS(sXml10StartChars, 0x1fb6, 0x1fbc);
- SETBITS(sXml10StartChars, 0x1fbe);
- SETBITS(sXml10StartChars, 0x1fc2, 0x1fc4);
- SETBITS(sXml10StartChars, 0x1fc6, 0x1fcc);
- SETBITS(sXml10StartChars, 0x1fd0, 0x1fd3);
- SETBITS(sXml10StartChars, 0x1fd6, 0x1fdb);
- SETBITS(sXml10StartChars, 0x1fe0, 0x1fec);
- SETBITS(sXml10StartChars, 0x1ff2, 0x1ff4);
- SETBITS(sXml10StartChars, 0x1ff6, 0x1ffc);
- SETBITS(sXml10StartChars, 0x2126);
- SETBITS(sXml10StartChars, 0x212a, 0x212b);
- SETBITS(sXml10StartChars, 0x212e);
- SETBITS(sXml10StartChars, 0x2180, 0x2182);
- SETBITS(sXml10StartChars, 0x3041, 0x3094);
- SETBITS(sXml10StartChars, 0x30a1, 0x30fa);
- SETBITS(sXml10StartChars, 0x3105, 0x312c);
- // note: AC00 - D7A3 handled separately
-
- // [86] Ideographic (but note: > 0x312c handled separately)
- SETBITS(sXml10StartChars, 0x3007);
- SETBITS(sXml10StartChars, 0x3021, 0x3029);
- }
-
- final static int[] sXml10Chars = new int[SIZE];
- static {
- // Let's start with all valid start chars:
- System.arraycopy(sXml10StartChars, 0, sXml10Chars, 0, SIZE);
-
- // [87] CombiningChar ::=
- SETBITS(sXml10Chars, 0x300, 0x345);
- SETBITS(sXml10Chars, 0x360, 0x361);
- SETBITS(sXml10Chars, 0x483, 0x486);
- SETBITS(sXml10Chars, 0x591, 0x5a1);
- SETBITS(sXml10Chars, 0x5a3, 0x5b9);
- SETBITS(sXml10Chars, 0x5bb, 0x5bd);
- SETBITS(sXml10Chars, 0x5bf);
-
- SETBITS(sXml10Chars, 0x5c1, 0x5c2);
- SETBITS(sXml10Chars, 0x5c4);
- SETBITS(sXml10Chars, 0x64b, 0x652);
- SETBITS(sXml10Chars, 0x670);
- SETBITS(sXml10Chars, 0x6d6, 0x6dc);
- SETBITS(sXml10Chars, 0x6dd, 0x6df);
- SETBITS(sXml10Chars, 0x6e0, 0x6e4);
- SETBITS(sXml10Chars, 0x6e7, 0x6e8);
- SETBITS(sXml10Chars, 0x6ea, 0x6ed);
-
- SETBITS(sXml10Chars, 0x901, 0x903);
- SETBITS(sXml10Chars, 0x93c);
- SETBITS(sXml10Chars, 0x93e, 0x94c);
- SETBITS(sXml10Chars, 0x94d);
- SETBITS(sXml10Chars, 0x951, 0x954);
- SETBITS(sXml10Chars, 0x962); SETBITS(sXml10Chars, 0x963);
- SETBITS(sXml10Chars, 0x981, 0x983);
- SETBITS(sXml10Chars, 0x9bc);
- SETBITS(sXml10Chars, 0x9be); SETBITS(sXml10Chars, 0x9bf);
- SETBITS(sXml10Chars, 0x9c0, 0x9c4);
- SETBITS(sXml10Chars, 0x9c7); SETBITS(sXml10Chars, 0x9c8);
- SETBITS(sXml10Chars, 0x9cb, 0x9cd);
- SETBITS(sXml10Chars, 0x9d7);
- SETBITS(sXml10Chars, 0x9e2); SETBITS(sXml10Chars, 0x9e3);
- SETBITS(sXml10Chars, 0xA02);
- SETBITS(sXml10Chars, 0xA3C);
- SETBITS(sXml10Chars, 0xA3E); SETBITS(sXml10Chars, 0xA3F);
- SETBITS(sXml10Chars, 0xA40, 0xA42);
- SETBITS(sXml10Chars, 0xA47); SETBITS(sXml10Chars, 0xA48);
- SETBITS(sXml10Chars, 0xA4B, 0xA4D);
- SETBITS(sXml10Chars, 0xA70); SETBITS(sXml10Chars, 0xA71);
- SETBITS(sXml10Chars, 0xA81, 0xA83);
- SETBITS(sXml10Chars, 0xABC);
- SETBITS(sXml10Chars, 0xABE, 0xAC5);
- SETBITS(sXml10Chars, 0xAC7, 0xAC9);
- SETBITS(sXml10Chars, 0xACB, 0xACD);
- SETBITS(sXml10Chars, 0xB01, 0xB03);
- SETBITS(sXml10Chars, 0xB3C);
- SETBITS(sXml10Chars, 0xB3E, 0xB43);
- SETBITS(sXml10Chars, 0xB47); SETBITS(sXml10Chars, 0xB48);
- SETBITS(sXml10Chars, 0xB4B, 0xB4D);
- SETBITS(sXml10Chars, 0xB56); SETBITS(sXml10Chars, 0xB57);
- SETBITS(sXml10Chars, 0xB82); SETBITS(sXml10Chars, 0xB83);
- SETBITS(sXml10Chars, 0xBBE, 0xBC2);
- SETBITS(sXml10Chars, 0xBC6, 0xBC8);
- SETBITS(sXml10Chars, 0xBCA, 0xBCD);
- SETBITS(sXml10Chars, 0xBD7);
- SETBITS(sXml10Chars, 0xC01, 0xC03);
- SETBITS(sXml10Chars, 0xC3E, 0xC44);
- SETBITS(sXml10Chars, 0xC46, 0xC48);
- SETBITS(sXml10Chars, 0xC4A, 0xC4D);
- SETBITS(sXml10Chars, 0xC55, 0xC56);
- SETBITS(sXml10Chars, 0xC82, 0xC83);
- SETBITS(sXml10Chars, 0xCBE, 0xCC4);
- SETBITS(sXml10Chars, 0xCC6, 0xCC8);
- SETBITS(sXml10Chars, 0xCCA, 0xCCD);
- SETBITS(sXml10Chars, 0xCD5, 0xCD6);
- SETBITS(sXml10Chars, 0xD02, 0xD03);
- SETBITS(sXml10Chars, 0xD3E, 0xD43);
- SETBITS(sXml10Chars, 0xD46, 0xD48);
- SETBITS(sXml10Chars, 0xD4A, 0xD4D);
- SETBITS(sXml10Chars, 0xD57);
- SETBITS(sXml10Chars, 0xE31);
- SETBITS(sXml10Chars, 0xE34, 0xE3A);
- SETBITS(sXml10Chars, 0xE47, 0xE4E);
- SETBITS(sXml10Chars, 0xEB1);
- SETBITS(sXml10Chars, 0xEB4, 0xEB9);
- SETBITS(sXml10Chars, 0xEBB, 0xEBC);
- SETBITS(sXml10Chars, 0xEC8, 0xECD);
- SETBITS(sXml10Chars, 0xF18, 0xF19);
- SETBITS(sXml10Chars, 0xF35); SETBITS(sXml10Chars, 0xF37);
- SETBITS(sXml10Chars, 0xF39);
- SETBITS(sXml10Chars, 0xF3E); SETBITS(sXml10Chars, 0xF3F);
- SETBITS(sXml10Chars, 0xF71, 0xF84);
- SETBITS(sXml10Chars, 0xF86, 0xF8B);
- SETBITS(sXml10Chars, 0xF90, 0xF95);
- SETBITS(sXml10Chars, 0xF97);
- SETBITS(sXml10Chars, 0xF99, 0xFAD);
- SETBITS(sXml10Chars, 0xFB1, 0xFB7);
- SETBITS(sXml10Chars, 0xFB9);
- SETBITS(sXml10Chars, 0x20D0, 0x20DC);
- SETBITS(sXml10Chars, 0x20E1);
- SETBITS(sXml10Chars, 0x302A, 0x302F);
- SETBITS(sXml10Chars, 0x3099); SETBITS(sXml10Chars, 0x309A);
- // [88] Digit:
- SETBITS(sXml10Chars, 0x660, 0x669);
- SETBITS(sXml10Chars, 0x6f0, 0x6f9);
- SETBITS(sXml10Chars, 0x966, 0x96f);
- SETBITS(sXml10Chars, 0x9e6, 0x9ef);
- SETBITS(sXml10Chars, 0xa66, 0xa6f);
- SETBITS(sXml10Chars, 0xae6, 0xaef);
- SETBITS(sXml10Chars, 0xb66, 0xb6f);
- SETBITS(sXml10Chars, 0xbe7, 0xbef);
- SETBITS(sXml10Chars, 0xc66, 0xc6f);
- SETBITS(sXml10Chars, 0xce6, 0xcef);
- SETBITS(sXml10Chars, 0xd66, 0xd6f);
- SETBITS(sXml10Chars, 0xe50, 0xe59);
- SETBITS(sXml10Chars, 0xed0, 0xed9);
- SETBITS(sXml10Chars, 0xf20, 0xf29);
-
- // [89] Extender:
- SETBITS(sXml10Chars, 0xb7);
- SETBITS(sXml10Chars, 0x2d0);
- SETBITS(sXml10Chars, 0x2d1);
- SETBITS(sXml10Chars, 0x387);
- SETBITS(sXml10Chars, 0x640);
- SETBITS(sXml10Chars, 0xE46);
- SETBITS(sXml10Chars, 0xEC6);
- SETBITS(sXml10Chars, 0x3005);
- SETBITS(sXml10Chars, 0x3031, 0x3035);
- SETBITS(sXml10Chars, 0x309d, 0x309e);
- SETBITS(sXml10Chars, 0x30fc, 0x30fe);
- }
-
- private XmlChars() { }
-
- public final static boolean is10NameStartChar(char c)
- {
- // First, let's deal with outliers
- if (c > 0x312C) { // Most valid chars are below this..
- if (c < 0xAC00) {
- return (c >= 0x4E00 && c <= 0x9FA5); // valid ideograms
- }
- if (c <= 0xD7A3) { // 0xAC00 - 0xD7A3, valid base chars
- return true;
- }
- /* As to surrogate pairs... let's do the bare minimum;
- * 0xD800 - 0xDBFF (high surrogate) are ok; low surrogates
- * can only follow high one
- */
- return (c <= 0xDBFF && c >= 0xD800);
- }
- // but then we'll just need to use the table...
- int ix = (int) c;
- return (sXml10StartChars[ix >> 5] & (1 << (ix & 31))) != 0;
- }
-
- public final static boolean is10NameChar(char c)
- {
- // First, let's deal with outliers
- if (c > 0x312C) { // Most valid chars are below this..
- if (c < 0xAC00) {
- return (c >= 0x4E00 && c <= 0x9FA5); // valid ideograms
- }
- if (c <= 0xD7A3) { // 0xAC00 - 0xD7A3, valid base chars
- return true;
- }
- /* As to surrogate pairs... let's do the bare minimum;
- * 0xD800 - 0xDFFF (high, low surrogate) are ok (need to
- * check pairing in future)
- */
- return (c >= 0xD800 && c <= 0xDFFF);
- }
- // but then we'll just need to use the table...
- int ix = (int) c;
- return (sXml10Chars[ix >> 5] & (1 << (ix & 31))) != 0;
- }
-
- public final static boolean is11NameStartChar(char c)
- {
- // Others are checked block-by-block:
- if (c <= 0x2FEF) {
- if (c < 0x300) {
- if (c < 0x00C0) { // 8-bit ctrl chars
- return false;
- }
- // most of the rest are fine...
- return (c != 0xD7 && c != 0xF7);
- }
- if (c >= 0x2C00) {
- // 0x2C00 - 0x2FEF are ok
- return true;
- }
- if (c < 0x370 || c > 0x218F) {
- // 0x300 - 0x36F, 0x2190 - 0x2BFF invalid
- return false;
- }
- if (c < 0x2000) {
- // 0x370 - 0x37D, 0x37F - 0x1FFF are ok
- return (c != 0x37E);
- }
- if (c >= 0x2070) {
- // 0x2070 - 0x218F are ok
- return (c <= 0x218F);
- }
- // And finally, 0x200C - 0x200D
- return (c == 0x200C || c == 0x200D);
- }
-
- // 0x3000 and above:
- if (c >= 0x3001) {
- /* Hmmh, let's allow high surrogates here, without checking
- * that they are properly followed... crude basic support,
- * I know, but allows valid combinations, just doesn't catch
- * invalid ones
- */
- if (c <= 0xDBFF) { // 0x3001 - 0xD7FF (chars),
- // 0xD800 - 0xDBFF (high surrogate) are ok (unlike DC00-DFFF)
- return true;
- }
- if (c >= 0xF900 && c <= 0xFFFD) {
- /* Check above removes low surrogate (since one can not
- * START an identifier), and byte-order markers..
- */
- return (c <= 0xFDCF || c >= 0xFDF0);
- }
- }
-
- return false;
- }
-
- public final static boolean is11NameChar(char c)
- {
- // Others are checked block-by-block:
- if (c <= 0x2FEF) {
- if (c < 0x2000) { // only 8-bit ctrl chars and 0x37E to filter out
- return (c >= 0x00C0 && c != 0x37E) || (c == 0xB7);
- }
- if (c >= 0x2C00) {
- // 0x100 - 0x1FFF, 0x2C00 - 0x2FEF are ok
- return true;
- }
- if (c < 0x200C || c > 0x218F) {
- // 0x2000 - 0x200B, 0x2190 - 0x2BFF invalid
- return false;
- }
- if (c >= 0x2070) {
- // 0x2070 - 0x218F are ok
- return true;
- }
- // And finally, 0x200C - 0x200D, 0x203F - 0x2040 are ok
- return (c == 0x200C || c == 0x200D
- || c == 0x203F || c == 0x2040);
- }
-
- // 0x3000 and above:
- if (c >= 0x3001) {
- /* Hmmh, let's allow surrogate heres, without checking that
- * they have proper ordering. For non-first name chars, both are
- * ok, for valid names. Crude basic support,
- * I know, but allows valid combinations, just doesn't catch
- * invalid ones
- */
- if (c <= 0xDFFF) { // 0x3001 - 0xD7FF (chars),
- // 0xD800 - 0xDFFF (high, low surrogate) are ok:
- return true;
- }
- if (c >= 0xF900 && c <= 0xFFFD) {
- /* Check above removes other invalid chars (below valid
- * range), and byte-order markers (0xFFFE, 0xFFFF).
- */
- return (c <= 0xFDCF || c >= 0xFDF0);
- }
- }
-
- return false;
- }
-
- private static void SETBITS(int[] array, int start, int end)
- {
- int bit1 = (start & 31);
- int bit2 = (end & 31);
- start >>= 5;
- end >>= 5;
-
- /* Ok; this is not perfectly optimal, but should be good enough...
- * we'll only do one-by-one at the ends.
- */
- if (start == end) {
- for (; bit1 <= bit2; ++bit1) {
- array[start] |= (1 << bit1);
- }
- } else {
- for (int bit = bit1; bit <= 31; ++bit) {
- array[start] |= (1 << bit);
- }
- while (++start < end) {
- array[start] = -1;
- }
- for (int bit = 0; bit <= bit2; ++bit) {
- array[end] |= (1 << bit);
- }
- }
- }
-
- private static void SETBITS(int[] array, int point) {
- int ix = (point >> 5);
- int bit = (point & 31);
-
- array[ix] |= (1 << bit);
- }
-}
diff -Nru libwoodstox-java-4.1.3/src/java/org/codehaus/stax2/ri/package.html libwoodstox-java-5.1.0/src/java/org/codehaus/stax2/ri/package.html
--- libwoodstox-java-4.1.3/src/java/org/codehaus/stax2/ri/package.html 2012-04-24 04:38:21.000000000 +0000
+++ libwoodstox-java-5.1.0/src/java/org/codehaus/stax2/ri/package.html 1970-01-01 00:00:00.000000000 +0000
@@ -1,5 +0,0 @@
-XmlWriter can only throw IOExceptions.
+ */
+public interface InvalidCharHandler
+{
+ public char convertInvalidChar(int invalidChar) throws IOException;
+
+ /**
+ * This handler implementation just throws an exception for
+ * all invalid characters encountered. It is the default handler
+ * used if nothing else has been specified.
+ */
+ public static class FailingHandler
+ implements InvalidCharHandler
+ {
+ public final static int SURR1_FIRST = 0xD800;
+ public final static int SURR1_LAST = 0xDBFF;
+ public final static int SURR2_FIRST = 0xDC00;
+ public final static int SURR2_LAST = 0xDFFF;
+
+ private final static FailingHandler sInstance = new FailingHandler();
+
+ protected FailingHandler() { }
+
+ public static FailingHandler getInstance() { return sInstance; }
+
+ @Override
+ public char convertInvalidChar(int c) throws IOException
+ {
+ /* 17-May-2006, TSa: Would really be useful if we could throw
+ * XMLStreamExceptions; esp. to indicate actual output location.
+ * However, this causes problem with methods that call us and
+ * can only throw IOExceptions (when invoked via Writer proxy).
+ * Need to figure out how to resolve this.
+ */
+ if (c == 0) {
+ throw new IOException("Invalid null character in text to output");
+ }
+ if (c < ' ' || (c >= 0x7F && c <= 0x9F)) {
+ String msg = "Invalid white space character (0x"+Integer.toHexString(c)+") in text to output (in xml 1.1, could output as a character entity)";
+ throw new IOException(msg);
+ }
+ if (c > 0x10FFFF) {
+ throw new IOException("Illegal unicode character point (0x"+Integer.toHexString(c)+") to output; max is 0x10FFFF as per RFC 3629");
+ }
+ /* Surrogate pair in non-quotable (not text or attribute value)
+ * content, and non-unicode encoding (ISO-8859-x, Ascii)?
+ */
+ if (c >= SURR1_FIRST && c <= SURR2_LAST) {
+ throw new IOException("Illegal surrogate pair -- can only be output via character entities, which are not allowed in this content");
+ }
+ throw new IOException("Invalid XML character (0x"+Integer.toHexString(c)+") in text to output");
+ }
+ }
+
+ /**
+ * Alternative to the default handler, this handler converts all invalid
+ * characters to the specified output character. That character will
+ * not be further verified or modified by the stream writer.
+ */
+ public static class ReplacingHandler
+ implements InvalidCharHandler
+ {
+ final char mReplacementChar;
+
+ public ReplacingHandler(char c) {
+ mReplacementChar = c;
+ }
+
+ @Override
+ public char convertInvalidChar(int c) throws IOException {
+ return mReplacementChar;
+ }
+ }
+}
+
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/api/package.html libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/api/package.html
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/api/package.html 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/api/package.html 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,7 @@
+ThreadLocal contains a {@link SoftRerefence}
+ * to a {@link BufferRecycler} used to provide a low-cost
+ * buffer recycling between Reader instances.
+ */
+ final static ThreadLocal
+ * None.
+ *
+ *
+ *
+ */
+ public void configureForConvenience()
+ {
+ // StAX (1.0) settings:
+ doCoalesceText(true);
+ doReplaceEntityRefs(true);
+
+ // StAX2:
+ doReportCData(false);
+ doReportPrologWhitespace(false);
+ /* Also, knowing exact locations is nice esp. for error
+ * reporting purposes
+ */
+ doPreserveLocation(true);
+
+ // Woodstox-specific:
+
+ /* Also, we can force errors to be reported in timely manner:
+ * (once again, at potential expense of performance)
+ */
+ doParseLazily(false);
+ }
+
+ /**
+ * Method to call to make the Reader created be as fast as possible reading
+ * documents, especially for long-running processes where caching is
+ * likely to help.
+ *
+ * See {@link XMLInputFactory2#configureForSpeed} for
+ * required settings for standard StAX/StAX properties.
+ *XMLStreamFactory2.P_LAZY_PARSING (to allow for synchronous
+ * error notification by forcing full XML events to be completely
+ * parsed when reader's next() is called)
+ *
+ *
+ */
+ public void configureForSpeed()
+ {
+ // StAX (1.0):
+ doCoalesceText(false);
+
+ // StAX2:
+ doPreserveLocation(false);
+ doReportPrologWhitespace(false);
+ //doInternNames(true); // this is a NOP
+ doInternNsURIs(true);
+ doXmlIdUniqChecks(false);
+
+ // Woodstox-specific:
+ doCacheDTDs(true);
+ doParseLazily(true);
+
+ /* If we let Reader decide sizes of text segments, it should be
+ * able to optimize it better, thus low min value. This value
+ * is only used in cases where text is at buffer boundary, or
+ * where entity prevents using consequtive chars from input buffer:
+ */
+ setShortestReportedTextSegment(16);
+ setInputBufferLength(8000); // 16k input buffer
+ }
+
+ /**
+ * Method to call to minimize the memory usage of the stream/event reader;
+ * both regarding Objects created, and the temporary memory usage during
+ * parsing.
+ * This generally incurs some performance penalties, due to using
+ * smaller input buffers.
+ *P_CACHE_DTDS.
+ * XMLStremaFactory2.P_LAZY_PARSING (can improve performance
+ * especially when skipping text segments)
+ * P_MIN_TEXT_SEGMENT, to allow
+ * reader to optimize segment length it uses (and possibly avoids
+ * one copy operation in the process)
+ * P_INPUT_BUFFER_LENGTH a bit from default,
+ * to allow for longer consequtive read operations; also reduces cases
+ * where partial text segments are on input buffer boundaries.
+ *
+ *
+ */
+ public void configureForLowMemUsage()
+ {
+ // StAX (1.0)
+ doCoalesceText(false);
+
+ // StAX2:
+
+ doPreserveLocation(false); // can reduce temporary mem usage
+
+ // Woodstox-specific:
+ doCacheDTDs(false);
+ doParseLazily(true); // can reduce temporary mem usage
+ doXmlIdUniqChecks(false); // enabling would increase mem usage
+ setShortestReportedTextSegment(ReaderConfig.DEFAULT_SHORTEST_TEXT_SEGMENT);
+ setInputBufferLength(512); // 1k input buffer
+ // Text buffer need not be huge, as we do not coalesce
+ }
+
+ /**
+ * Method to call to make Reader try to preserve as much of input
+ * formatting as possible, so that round-tripping would be as lossless
+ * as possible.
+ *P_CACHE_DTDS
+ * P_PARSE_LAZILY
+ * P_MIN_TEXT_SEGMENT to the (somewhat low)
+ * default value.
+ * P_INPUT_BUFFER_LENGTH a bit from the default
+ *
+ *
+ */
+ public void configureForRoundTripping()
+ {
+ // StAX (1.0)
+ doCoalesceText(false);
+ doReplaceEntityRefs(false);
+
+ // StAX2:
+ doReportCData(true);
+ doReportPrologWhitespace(true);
+
+ // Woodstox specific settings
+ doTreatCharRefsAsEnts(true);
+ doNormalizeLFs(false);
+
+ // effectively prevents from reporting partial segments:
+ setShortestReportedTextSegment(Integer.MAX_VALUE);
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////////////////
+ // Buffer recycling:
+ ///////////////////////////////////////////////////////////////////////
+ */
+
+ public char[] allocSmallCBuffer(int minSize)
+ {
+ if (mCurrRecycler != null) {
+ char[] result = mCurrRecycler.getSmallCBuffer(minSize);
+ if (result != null) {
+ return result;
+ }
+ }
+ // Nope; no recycler, or it has no suitable buffers, let's create:
+ return new char[minSize];
+ }
+
+ public void freeSmallCBuffer(char[] buffer)
+ {
+ // Need to create (and assign) the buffer?
+ if (mCurrRecycler == null) {
+ mCurrRecycler = createRecycler();
+ }
+ mCurrRecycler.returnSmallCBuffer(buffer);
+ }
+
+ public char[] allocMediumCBuffer(int minSize)
+ {
+ if (mCurrRecycler != null) {
+ char[] result = mCurrRecycler.getMediumCBuffer(minSize);
+ if (result != null) {
+ return result;
+ }
+ }
+ return new char[minSize];
+ }
+
+ public void freeMediumCBuffer(char[] buffer)
+ {
+ if (mCurrRecycler == null) {
+ mCurrRecycler = createRecycler();
+ }
+ mCurrRecycler.returnMediumCBuffer(buffer);
+ }
+
+ public char[] allocFullCBuffer(int minSize)
+ {
+ if (mCurrRecycler != null) {
+ char[] result = mCurrRecycler.getFullCBuffer(minSize);
+ if (result != null) {
+ return result;
+ }
+ }
+ return new char[minSize];
+ }
+
+ public void freeFullCBuffer(char[] buffer)
+ {
+ // Need to create (and assign) the buffer?
+ if (mCurrRecycler == null) {
+ mCurrRecycler = createRecycler();
+ }
+ mCurrRecycler.returnFullCBuffer(buffer);
+ }
+
+ public byte[] allocFullBBuffer(int minSize)
+ {
+ if (mCurrRecycler != null) {
+ byte[] result = mCurrRecycler.getFullBBuffer(minSize);
+ if (result != null) {
+ return result;
+ }
+ }
+ return new byte[minSize];
+ }
+
+ public void freeFullBBuffer(byte[] buffer)
+ {
+ // Need to create (and assign) the buffer?
+ if (mCurrRecycler == null) {
+ mCurrRecycler = createRecycler();
+ }
+ mCurrRecycler.returnFullBBuffer(buffer);
+ }
+
+ private BufferRecycler createRecycler()
+ {
+ BufferRecycler recycler = new BufferRecycler();
+ // No way to reuse/reset SoftReference, have to create new always:
+ mRecyclerRef.set(new SoftReferenceP_MIN_TEXT_SEGMENT to the maximum value so
+ * that all original text segment chunks are reported without
+ * segmentation (but without coalescing with adjacent CDATA segments)
+ * P_TREAT_CHAR_REFS_AS_ENTS to true, so the all the
+ * original character references are reported with their position,
+ * original text, and the replacement text.
+ * ThreadLocal contains a {@link SoftRerefence}
+ * to a {@link BufferRecycler} used to provide a low-cost
+ * buffer recycling between Reader instances.
+ */
+ final static ThreadLocal
+ *
+ *PARSING_MODE_DOCUMENT) allows parsing of only
+ * well-formed XML documents, but the other two modes allow more lenient
+ * parsing. Fragment mode allows parsing of XML content that does not
+ * have a single root element (can have zero or more), nor can have
+ * XML or DOCTYPE declarations: this may be useful if parsing a subset
+ * of a full XML document. Multi-document
+ * (PARSING_MODE_DOCUMENTS) mode on the other hand allows
+ * parsing of a stream that contains multiple consequtive well-formed
+ * documents, with possibly multiple XML and DOCTYPE declarations.
+ *close or
+ * closeCompletely, or implicitly by a call
+ * to writeEndDocument.
+ *javax.xml.transform.dom.DOMResult.
+ *
+ *
+ *
+ * @author Tatu Saloranta
+ * @author Dan Diephouse
+ */
+public class WstxDOMWrappingWriter
+ extends DOMWrappingWriter
+{
+ /*
+ ///////////////////////////////////////////////////////////
+ // Constants
+ ///////////////////////////////////////////////////////////
+ */
+
+ final protected static String ERR_NSDECL_WRONG_STATE =
+ "Trying to write a namespace declaration when there is no open start element.";
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Configuration
+ ///////////////////////////////////////////////////////////
+ */
+
+ protected final WriterConfig mConfig;
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // State
+ ///////////////////////////////////////////////////////////
+ */
+
+ /**
+ * This element is the current context element, under which
+ * all other nodes are added, until matching end element
+ * is output. Null outside of the main element tree.
+ *writeEmptyElement) will never become
+ * current element.
+ */
+ protected DOMOutputElement mCurrElem;
+
+ /**
+ * This element is non-null right after a call to
+ * either writeStartElement and
+ * writeEmptyElement, and can be used to
+ * add attributes and namespace declarations.
+ *writeEmptyElement) will
+ * become open element but NOT current element. Conversely,
+ * regular elements will remain current element when
+ * non elements are written (text, comments, PI), but
+ * not the open element.
+ */
+ protected DOMOutputElement mOpenElement;
+
+ /**
+ * for NsRepairing mode
+ */
+ protected int[] mAutoNsSeq;
+ protected String mSuggestedDefNs = null;
+ protected String mAutomaticNsPrefix;
+
+ /**
+ * Map that contains URI-to-prefix entries that point out suggested
+ * prefixes for URIs. These are populated by calls to
+ * {@link #setPrefix}, and they are only used as hints for binding;
+ * if there are conflicts, repairing writer can just use some other
+ * prefix.
+ */
+ HashMapmUndeclaredEntity
+ */
+ private String mValue = null;
+
+ /**
+ * For now, let's only keep track of the first undeclared entity:
+ * can be extended if necessary.
+ */
+ private UndeclaredEntity mUndeclaredEntity = null;
+
+ /*
+ ////////////////////////////////////////////////////
+ // Life-cycle (creation, configuration)
+ ////////////////////////////////////////////////////
+ */
+
+ private DefaultAttrValue(int defValueType)
+ {
+ mDefValueType = defValueType;
+ }
+
+ public static DefaultAttrValue constructImplied() { return sImplied; }
+ public static DefaultAttrValue constructRequired() { return sRequired; }
+
+ public static DefaultAttrValue constructFixed() {
+ return new DefaultAttrValue(DEF_FIXED);
+ }
+
+ public static DefaultAttrValue constructOptional() {
+ return new DefaultAttrValue(DEF_DEFAULT);
+ }
+
+ public void setValue(String v) {
+ mValue = v;
+ }
+
+ public void addUndeclaredPE(String name, Location loc)
+ {
+ addUndeclaredEntity(name, loc, true);
+ }
+
+ public void addUndeclaredGE(String name, Location loc)
+ {
+ addUndeclaredEntity(name, loc, false);
+ }
+
+ public void reportUndeclared(ValidationContext ctxt, XMLValidator dtd)
+ throws XMLStreamException
+ {
+ mUndeclaredEntity.reportUndeclared(ctxt, dtd);
+ }
+
+ /*
+ ////////////////////////////////////////////////////
+ // Accessors:
+ ////////////////////////////////////////////////////
+ */
+
+ public boolean hasUndeclaredEntities() {
+ return (mUndeclaredEntity != null);
+ }
+
+ public String getValue() {
+ return mValue;
+ }
+
+ /**
+ * @return Expanded default value String, if there were no problems
+ * (no undeclared entities), or null to indicate there were problems.
+ * In latter case, caller is to figure out exact type of the problem
+ * and report this appropriately to the application.
+ */
+ public String getValueIfOk()
+ {
+ return (mUndeclaredEntity == null) ? mValue : null;
+ }
+
+ public boolean isRequired() {
+ return (this == sRequired);
+ }
+
+ public boolean isFixed() {
+ return (mDefValueType == DEF_FIXED);
+ }
+
+ public boolean hasDefaultValue() {
+ return (mDefValueType == DEF_DEFAULT)
+ || (mDefValueType == DEF_FIXED);
+ }
+
+ /**
+ * Method used by the element to figure out if attribute needs "special"
+ * checking; basically if it's required, and/or has a default value.
+ * In both cases missing the attribute has specific consequences, either
+ * exception or addition of a default value.
+ */
+ public boolean isSpecial() {
+ // Only non-special if #IMPLIED
+ return (this != sImplied);
+ }
+
+ /*
+ ////////////////////////////////////////////////////
+ // Internal methods
+ ////////////////////////////////////////////////////
+ */
+
+ private void addUndeclaredEntity(String name, Location loc, boolean isPe)
+ {
+ if (mUndeclaredEntity == null) {
+ mUndeclaredEntity = new UndeclaredEntity(name, loc, isPe);
+ }
+ }
+
+ /*
+ ////////////////////////////////////////////////////
+ // Helper class(es):
+ ////////////////////////////////////////////////////
+ */
+
+ final static class UndeclaredEntity
+ {
+ final String mName;
+ final boolean mIsPe;
+ final Location mLocation;
+
+ UndeclaredEntity(String name, Location loc, boolean isPe)
+ {
+ mName = name;
+ mIsPe = isPe;
+ mLocation = loc;
+ }
+
+ public void reportUndeclared(ValidationContext ctxt, XMLValidator dtd)
+ throws XMLStreamException
+ {
+ String msg = MessageFormat.format(ErrorConsts.ERR_DTD_UNDECLARED_ENTITY, new Object[] { (mIsPe ? "parsed" : "general"), mName });
+ XMLValidationProblem prob = new XMLValidationProblem
+ (mLocation, msg, XMLValidationProblem.SEVERITY_FATAL);
+ prob.setReporter(dtd);
+ ctxt.reportProblem(prob);
+ }
+ }
+}
diff -Nru libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DFAState.java libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DFAState.java
--- libwoodstox-java-4.1.3/src/main/java/com/ctc/wstx/dtd/DFAState.java 1970-01-01 00:00:00.000000000 +0000
+++ libwoodstox-java-5.1.0/src/main/java/com/ctc/wstx/dtd/DFAState.java 2018-03-31 01:33:28.000000000 +0000
@@ -0,0 +1,204 @@
+/* Woodstox XML processor
+ *
+ * Copyright (c) 2004 Tatu Saloranta, tatu.saloranta@iki.fi
+ *
+ * Licensed under the License specified in the file LICENSE which is
+ * included with the source code.
+ * You may not use this file except in compliance with the License.
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.ctc.wstx.dtd;
+
+import java.util.*;
+
+import com.ctc.wstx.util.PrefixedName;
+
+/**
+ * Class that represents a state in DFA used for validating complex
+ * DTD content models.
+ */
+public final class DFAState
+{
+ final int mIndex;
+ final boolean mAccepting;
+
+ BitSet mTokenSet;
+
+ HashMapvalidateElementStart,
+ * validateAttribute,
+ * validateElementAndAttributes calls.
+ */
+ protected DTDElement mCurrElem = null;
+
+ /**
+ * Stack of element definitions matching the current active element stack.
+ * Instances are elements definitions read from DTD.
+ */
+ protected DTDElement[] mElems = null;
+
+ /**
+ * Number of elements in {@link #mElems}.
+ */
+ protected int mElemCount = 0;
+
+ /**
+ * Attribute definitions for attributes the current element may have
+ */
+ protected HashMapequals()
+ * method (but only within a single subset!). This also reduces memory
+ * usage to some extent.
+ */
+ private PrefixedName readDTDQName(char firstChar)
+ throws XMLStreamException
+ {
+ String prefix, localName;
+
+ if (!mCfgNsEnabled) {
+ prefix = null;
+ localName = parseFullName(firstChar);
+ } else {
+ localName = parseLocalName(firstChar);
+ /* Hmmh. This is tricky; should only read from the current
+ * scope, but it is ok to hit end-of-block if it was a PE
+ * expansion...
+ */
+ char c = dtdNextIfAvailable();
+ if (c == CHAR_NULL) { // end-of-block
+ // ok, that's it...
+ prefix = null;
+ } else {
+ if (c == ':') { // Ok, got namespace and local name
+ prefix = localName;
+ c = dtdNextFromCurr();
+ localName = parseLocalName(c);
+ } else {
+ prefix = null;
+ --mInputPtr;
+ }
+ }
+ }
+
+ return findSharedName(prefix, localName);
+ }
+
+ private char readArity()
+ throws XMLStreamException
+ {
+ char c = (mInputPtr < mInputEnd) ?
+ mInputBuffer[mInputPtr++] : getNextChar(getErrorMsg());
+ if (c == '?' || c == '*' || c == '+') {
+ return c;
+ }
+ // Hmmh, not recognized, let's put it back:
+ --mInputPtr;
+
+ // Default is 'just one'
+ return ' ';
+ }
+
+ /**
+ * Method that reads and pre-processes replacement text for an internal
+ * entity (parameter or generic).
+ */
+ private char[] parseEntityValue(String id, Location loc, char quoteChar)
+ throws XMLStreamException
+ {
+ /* 25-Jun-2004, TSa: Let's first mark current input source as the
+ * scope, so we can both make sure it ends in this input
+ * context (file), and that embedded single/double quotes
+ * in potentially expanded entities do not end the value
+ * definition (as per XML 1.0/3, 4.4.5)
+ */
+ WstxInputSource currScope = mInput;
+
+ /* 18-Jul-2004, TSa: Also, let's see if parameter entities are
+ * allowed; they are only legal outside of main internal subset
+ * (ie. main XML input) file (or to be precise; they are legal
+ * in the int. subset only as complete declarations)
+ */
+ //boolean allowPEs = mIsExternal || (mInput != mRootInput);
+
+ TextBuffer tb = mValueBuffer;
+ if (tb == null) {
+ tb = TextBuffer.createTemporaryBuffer();
+ }
+ tb.resetInitialized();
+
+ char[] outBuf = tb.getCurrentSegment();
+ int outPtr = tb.getCurrentSegmentSize();
+
+ while (true) {
+ if (mInputPtr >= mInputEnd) {
+ loadMoreScoped(currScope, id, loc);
+ }
+ char c = mInputBuffer[mInputPtr++];
+
+ // Let's get most normal chars 'skipped' first
+ if (c >= CHAR_FIRST_PURE_TEXT) {
+ ;
+ } else if (c == quoteChar) {
+ // Only end if we are in correct scope:
+ if (mInput == currScope) {
+ break;
+ }
+ } else if (c == '&') { // char entity that needs to be replaced?
+ /* 06-Sep-2004, TSa: We can NOT expand pre-defined entities, as
+ * XML specs consider them 'real' (non-char) entities.
+ * And expanding them would cause problems with entities
+ * that have such entities.
+ */
+ int d = resolveCharOnlyEntity(false);
+ // Did we get a real char entity?
+ if (d != 0) {
+ if (d <= 0xFFFF) {
+ c = (char) d;
+ } else {
+ // Need more room?
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ d -= 0x10000;
+ outBuf[outPtr++] = (char) ((d >> 10) + 0xD800);;
+ c = (char) ((d & 0x3FF) + 0xDC00);
+ }
+ } else {
+ /* 11-Feb-2006, TSa: Even so, must verify that the
+ * entity reference is well-formed.
+ */
+ boolean first = true;
+ while (true) {
+ if (outPtr >= outBuf.length) { // need more room?
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ outBuf[outPtr++] = c; // starting with '&'
+ if (mInputPtr >= mInputEnd) {
+ loadMoreScoped(currScope, id, loc);
+ }
+ c = mInputBuffer[mInputPtr++];
+ if (c == ';') {
+ break;
+ }
+ if (first) {
+ first = false;
+ if (isNameStartChar(c)) {
+ continue;
+ }
+ } else {
+ if (isNameChar(c)) {
+ continue;
+ }
+ }
+ if (c == ':' && !mCfgNsEnabled) {
+ continue; // fine in non-ns mode
+ }
+ if (first) { // missing name
+ throwDTDUnexpectedChar(c, "; expected entity name after '&'");
+ }
+ throwDTDUnexpectedChar(c, "; expected semi-colon after entity name");
+ }
+ // we can just fall through to let semicolon be added
+ }
+ // Either '&' itself, or expanded char entity
+ } else if (c == '%') { // param entity?
+ expandPE();
+ // Need to loop over, no char available yet
+ continue;
+ } else if (c < CHAR_SPACE) {
+ if (c == '\n') {
+ markLF();
+ } else if (c == '\r') {
+ if (skipCRLF(c)) {
+ if (!mNormalizeLFs) {
+ // Special handling, to output 2 chars at a time:
+ if (outPtr >= outBuf.length) { // need more room?
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ outBuf[outPtr++] = c;
+ }
+ c = '\n';
+ } else {
+ if (mNormalizeLFs) { // Mac LF
+ c = '\n';
+ }
+ }
+ } else if (c != '\t') {
+ throwInvalidSpace(c);
+ }
+ }
+
+ // Need more room?
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ // Ok, let's add char to output:
+ outBuf[outPtr++] = c;
+ }
+ tb.setCurrentLength(outPtr);
+
+ // Ok, now need the closing '>':
+ char c = skipDtdWs(true);
+ if (c != '>') {
+ throwDTDUnexpectedChar(c, "; expected closing '>' after ENTITY declaration");
+ }
+ char[] result = tb.contentsAsArray();
+ mValueBuffer = tb; // recycle, if needed later on
+
+ return result;
+ }
+
+ /**
+ * This method is similar to {@link #parseEntityValue} in some ways,
+ * but has some notable differences, due to the way XML specs define
+ * differences. Main differences are that parameter entities are not
+ * allowed (or rather, recognized as entities), and that general
+ * entities need to be verified, but NOT expanded right away.
+ * Whether forward references are allowed or not is an open question
+ * right now.
+ */
+ private void parseAttrDefaultValue(DefaultAttrValue defVal, char quoteChar, PrefixedName attrName,
+ Location loc, boolean gotFixed)
+ throws XMLStreamException
+ {
+ if (quoteChar != '"' && quoteChar != '\'') { // caller doesn't test it
+ String msg = "; expected a single or double quote to enclose the default value";
+ if (!gotFixed) {
+ msg += ", or one of keywords (#REQUIRED, #IMPLIED, #FIXED)";
+ }
+ msg += " (for attribute '"+attrName+"')";
+ throwDTDUnexpectedChar(quoteChar, msg);
+ }
+
+ /* Let's mark the current input source as the scope, so we can both
+ * make sure it ends in this input context (DTD subset), and that
+ * embedded single/double quotes in potentially expanded entities do
+ * not end the value definition (as per XML 1.0/3, 4.4.5)
+ */
+ WstxInputSource currScope = mInput;
+
+ TextBuffer tb = mValueBuffer;
+ if (tb == null) {
+ tb = TextBuffer.createTemporaryBuffer();
+ }
+ tb.resetInitialized();
+
+ int outPtr = 0;
+ char[] outBuf = tb.getCurrentSegment();
+ int outLen = outBuf.length;
+
+ /* One more note: this is mostly cut'n pasted from stream reader's
+ * parseNormalizedAttrValue...
+ */
+ main_loop:
+
+ while (true) {
+ if (mInputPtr >= mInputEnd) {
+ boolean check = (mInput == currScope);
+ loadMore(getErrorMsg());
+ // Did we get out of the scope?
+ if (check && (mInput != currScope)) {
+ _reportWFCViolation("Unterminated attribute default value for attribute '"
+ +attrName+"' (definition started at "
+ +loc+")");
+ }
+ }
+ char c = mInputBuffer[mInputPtr++];
+
+ // Let's do a quick for most attribute content chars:
+ if (c < CHAR_FIRST_PURE_TEXT) {
+ if (c <= CHAR_SPACE) {
+ if (c == '\n') {
+ markLF();
+ } else if (c == '\r') {
+ c = getNextChar(SUFFIX_IN_DEF_ATTR_VALUE);
+ if (c != '\n') { // nope, not 2-char lf (Mac?)
+ --mInputPtr;
+ c = mNormalizeLFs ? '\n' : '\r';
+ } else {
+ // Fine if we are to normalize lfs
+ /* !!! 20-Jan-2007, TSa: Hmmh. Not sure if and
+ * how to preserve: for now, let's assume there's
+ * no need.
+ */
+ /*
+ if (!mNormalizeLFs) {
+ if (outPtr >= outLen) { // need more room?
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ outLen = outBuf.length;
+ }
+ outBuf[outPtr++] = '\r';
+ // c is fine to continue
+ }
+ */
+ }
+ markLF();
+ } else if (c != CHAR_SPACE && c != '\t') {
+ throwInvalidSpace(c);
+ }
+ c = CHAR_SPACE;
+ } else if (c == quoteChar) {
+ /* It is possible to get these via expanded entities;
+ * need to make sure this is the same input level as
+ * the one that had starting quote
+ */
+ if (mInput == currScope) {
+ break;
+ }
+ } else if (c == '&') { // an entity of some sort...
+ /* Will need to expand char entities and pre-defd
+ * int. entities (amp, lt, apos, gt): first method
+ * is just more optimized than the second
+ */
+ int d;
+ if (inputInBuffer() >= 3) {
+ d = resolveSimpleEntity(true);
+ } else {
+ d = resolveCharOnlyEntity(true);
+ }
+ // Only get null if it's a 'real' general entity...
+ if (d == 0) {
+ c = getNextChar(SUFFIX_IN_ENTITY_REF);
+ String id = parseEntityName(c);
+ try {
+ mCurrAttrDefault = defVal;
+ mExpandingPE = false;
+ expandEntity(id, false, ENTITY_EXP_GE);
+ } finally {
+ mCurrAttrDefault = null;
+ }
+ // Ok, should have updated the input source by now
+ continue main_loop;
+ }
+
+ if (c <= 0xFFFF) {
+
+ } else{
+ if (d <= 0xFFFF) {
+ c = (char) d;
+ } else {
+ // Need more room?
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ d -= 0x10000;
+ outBuf[outPtr++] = (char) ((d >> 10) + 0xD800);;
+ c = (char) ((d & 0x3FF) + 0xDC00);
+ }
+ }
+ } else if (c == '<') {
+ throwDTDUnexpectedChar(c, SUFFIX_IN_DEF_ATTR_VALUE);
+ }
+ } // if (c < CHAR_FIRST_PURE_TEXT)
+
+ // Ok, let's just add char in, whatever it was
+ if (outPtr >= outLen) { // need more room?
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ outLen = outBuf.length;
+ }
+ outBuf[outPtr++] = c;
+ }
+
+ tb.setCurrentLength(outPtr);
+ defVal.setValue(tb.contentsAsString());
+ mValueBuffer = tb;
+ }
+
+ /**
+ * Method similar to {@link #skipPI}, but one that does basic
+ * well-formedness checks.
+ */
+ protected void readPI()
+ throws XMLStreamException
+ {
+ String target = parseFullName();
+ if (target.length() == 0) {
+ _reportWFCViolation(ErrorConsts.ERR_WF_PI_MISSING_TARGET);
+ }
+ if (target.equalsIgnoreCase("xml")) {
+ _reportWFCViolation(ErrorConsts.ERR_WF_PI_XML_TARGET, target);
+ }
+
+ char c = dtdNextFromCurr();
+ // Ok, need a space between target and data nonetheless
+ if (!isSpaceChar(c)) { // except if it ends right away
+ if (c != '?' || dtdNextFromCurr() != '>') {
+ throwUnexpectedChar(c, ErrorConsts.ERR_WF_PI_XML_MISSING_SPACE);
+ }
+ if (mEventListener != null) {
+ mEventListener.dtdProcessingInstruction(target, "");
+ }
+ } else if (mEventListener == null) {
+ /* Otherwise, not that much to check since we don't care about
+ * the contents.
+ */
+ while (true) {
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ if (c == '?') {
+ do {
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ } while (c == '?');
+ if (c == '>') {
+ break;
+ }
+ }
+ if (c < CHAR_SPACE) {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ } else if (c != '\t') {
+ throwInvalidSpace(c);
+ }
+ }
+ }
+ } else {
+ // 24-Nov-2006, TSa: Actually, someone does care...
+ // First, need to skip extra space (if any)
+ while (c <= CHAR_SPACE) {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ } else if (c != '\t' && c != ' ') {
+ throwInvalidSpace(c);
+ }
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ }
+
+ TextBuffer tb = getTextBuffer();
+ char[] outBuf = tb.getCurrentSegment();
+ int outPtr = 0;
+
+ while (true) {
+ if (c == '?') {
+ while (true) {
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ if (c != '?') {
+ break;
+ }
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ outBuf[outPtr++] = c;
+ }
+ if (c == '>') {
+ break;
+ }
+ // Need to push back char that follows '?', output '?'
+ --mInputPtr;
+ c = '?';
+ } else if (c < CHAR_SPACE) {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ c = '\n';
+ } else if (c != '\t') {
+ throwInvalidSpace(c);
+ }
+ }
+ // Need more room?
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ // Ok, let's add char to output:
+ outBuf[outPtr++] = c;
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ }
+ tb.setCurrentLength(outPtr);
+ String data = tb.contentsAsString();
+ mEventListener.dtdProcessingInstruction(target, data);
+ }
+ }
+
+ /**
+ * Method similar to {@link #skipComment}, but that has to collect
+ * contents, to be reported for a SAX handler.
+ */
+ protected void readComment(DTDEventListener l)
+ throws XMLStreamException
+ {
+ TextBuffer tb = getTextBuffer();
+ char[] outBuf = tb.getCurrentSegment();
+ int outPtr = 0;
+
+ while (true) {
+ char c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : dtdNextFromCurr();
+ if (c < CHAR_SPACE) {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ c = '\n';
+ } else if (c != '\t') {
+ throwInvalidSpace(c);
+ }
+ } else if (c == '-') {
+ c = dtdNextFromCurr();
+ if (c == '-') { // Ok, has to be end marker then:
+ // Either get '>' or error:
+ c = dtdNextFromCurr();
+ if (c != '>') {
+ throwParseError(ErrorConsts.ERR_HYPHENS_IN_COMMENT);
+ }
+ break;
+ }
+ c = '-';
+ --mInputPtr; // need to push back the second char read
+ }
+ // Need more room?
+ if (outPtr >= outBuf.length) {
+ outBuf = tb.finishCurrentSegment();
+ outPtr = 0;
+ }
+ // Ok, let's add char to output:
+ outBuf[outPtr++] = c;
+ }
+ tb.setCurrentLength(outPtr);
+ tb.fireDtdCommentEvent(l);
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Internal methods, conditional blocks:
+ ///////////////////////////////////////////////////////////
+ */
+
+ private void checkInclusion()
+ throws XMLStreamException
+ {
+ String keyword;
+
+ // INCLUDE/IGNORE not allowed in internal subset...
+ /* 18-Jul-2004, TSa: Except if it's in an expanded parsed external
+ * entity...
+ */
+ if (!mIsExternal && mInput == mRootInput) {
+ _reportWFCViolation("Internal DTD subset can not use (INCLUDE/IGNORE) directives (except via external entities)");
+ }
+
+ char c = skipDtdWs(true);
+ if (c != 'I') {
+ // let's obtain the keyword for error reporting purposes:
+ keyword = readDTDKeyword(String.valueOf(c));
+ } else {
+ c = dtdNextFromCurr();
+ if (c == 'G') {
+ keyword = checkDTDKeyword("NORE");
+ if (keyword == null) {
+ handleIgnored();
+ return;
+ }
+ keyword = "IG"+keyword;
+ } else if (c == 'N') {
+ keyword = checkDTDKeyword("CLUDE");
+ if (keyword == null) {
+ handleIncluded();
+ return;
+ }
+ keyword = "IN"+keyword;
+ } else {
+ --mInputPtr;
+ keyword = readDTDKeyword("I");
+ }
+ }
+
+ // If we get here, it was an error...
+ _reportWFCViolation("Unrecognized directive '"+keyword+"'; expected either 'IGNORE' or 'INCLUDE'");
+ }
+
+ private void handleIncluded()
+ throws XMLStreamException
+ {
+ char c = skipDtdWs(false);
+ if (c != '[') {
+ throwDTDUnexpectedChar(c, "; expected '[' to follow 'INCLUDE' directive");
+ }
+ ++mIncludeCount;
+ }
+
+ private void handleIgnored()
+ throws XMLStreamException
+ {
+ char c = skipDtdWs(false);
+ int count = 1; // Nesting of IGNORE/INCLUDE sections we have to match
+
+ if (c != '[') {
+ throwDTDUnexpectedChar(c, "; expected '[' to follow 'IGNORE' directive");
+ }
+
+ /* Ok; now, let's just skip until we get the closing ']]>'
+ */
+ String errorMsg = getErrorMsg();
+ while (true) {
+ c = (mInputPtr < mInputEnd)
+ ? mInputBuffer[mInputPtr++] : getNextChar(errorMsg);
+ if (c < CHAR_SPACE) {
+ if (c == '\n' || c == '\r') {
+ skipCRLF(c);
+ } else if (c != '\t') {
+ throwInvalidSpace(c);
+ }
+ } else if (c == ']') { // closing?
+ if (getNextChar(errorMsg) == ']'
+ && getNextChar(errorMsg) == '>') {
+ if (--count < 1) { // done!
+ return;
+ }
+ // nested ignores, let's just continue
+ } else {
+ --mInputPtr; // need to push one char back, may be '<'
+ }
+ } else if (c == '<') {
+ if (getNextChar(errorMsg) == '!'
+ && getNextChar(errorMsg) == '[') {
+ // Further nesting, sweet
+ ++count;
+ } else {
+ --mInputPtr; // need to push one char back, may be '<'
+ }
+ }
+ }
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Internal methods, validation, exceptions
+ ///////////////////////////////////////////////////////////
+ */
+
+ private void _reportUndefinedNotationRefs()
+ throws XMLStreamException
+ {
+ int count = mNotationForwardRefs.size();
+
+ String id = mNotationForwardRefs.keySet().iterator().next();
+ String msg = ""+count+" referenced notation"+((count == 1) ? "":"s")+" undefined: first one '"+id+"'";
+ _reportVCViolation(msg);
+ }
+
+ private void _reportBadDirective(String dir)
+ throws XMLStreamException
+ {
+ String msg = "Unrecognized DTD directive ''; expected ATTLIST, ELEMENT, ENTITY or NOTATION";
+ if (mCfgSupportDTDPP) {
+ msg += " (or, for DTD++, TARGETNS)";
+ }
+ _reportWFCViolation(msg);
+ }
+
+ private void _reportVCViolation(String msg)
+ throws XMLStreamException
+ {
+ /* 01-Sep-2006, TSa: Not 100% sure what's the right way to do it --
+ * they are errors (non-fatal, but not warnings), but the way
+ * base class handles things, we probably better 'downgrade'
+ * them to warnings in non-validating mode.
+ */
+ if (mCfgFullyValidating) {
+ reportValidationProblem(msg, XMLValidationProblem.SEVERITY_ERROR);
+ } else {
+ reportValidationProblem(msg, XMLValidationProblem.SEVERITY_WARNING);
+ }
+ }
+
+ private void _reportWFCViolation(String msg)
+ throws XMLStreamException
+ {
+ throwParseError(msg);
+ }
+
+ private void _reportWFCViolation(String format, Object arg)
+ throws XMLStreamException
+ {
+ throwParseError(format, arg, null);
+ }
+
+ private void throwDTDElemError(String msg, Object elem)
+ throws XMLStreamException
+ {
+ _reportWFCViolation(elemDesc(elem) + ": " + msg);
+ }
+
+ private void throwDTDAttrError(String msg, DTDElement elem, PrefixedName attrName)
+ throws XMLStreamException
+ {
+ _reportWFCViolation(attrDesc(elem, attrName) + ": " + msg);
+ }
+
+ private void throwDTDUnexpectedChar(int i, String extraMsg)
+ throws XMLStreamException
+ {
+ if (extraMsg == null) {
+ throwUnexpectedChar(i, getErrorMsg());
+ }
+ throwUnexpectedChar(i, getErrorMsg()+extraMsg);
+ }
+
+ private void throwForbiddenPE()
+ throws XMLStreamException
+ {
+ _reportWFCViolation("Can not have parameter entities in the internal subset, except for defining complete declarations (XML 1.0, #2.8, WFC 'PEs In Internal Subset')");
+ }
+
+ private String elemDesc(Object elem) {
+ return "Element <"+elem+">)";
+ }
+
+ private String attrDesc(Object elem, PrefixedName attrName) {
+ return "Attribute '"+attrName+"' (of element <"+elem+">)";
+ }
+
+ private String entityDesc(WstxInputSource input) {
+ return "Entity &"+input.getEntityId()+";";
+ }
+
+ /*
+ ///////////////////////////////////////////////////////////
+ // Internal methods, main-level declaration parsing
+ ///////////////////////////////////////////////////////////
+ */
+
+ /**
+ *
+ *