SML/NJ language processing tools

 Tradition has it that when a new programming language is introduced,
 new scanner and parser generators are written in that language, and
 generate code for that language. Traditional also has it that the
 new tools are modeled after the old lex and yacc tools, both in terms
 of the algorithms used, and often the syntax as well. The language
 Standard ML is no exception: ml-lex and ml-yacc are the SML
 incarnations of the old Unix tools.
 .
 This package has two new tools, ml-ulex and ml-antlr, that follow
 tradition in separating scanning from parsing, but break from
 tradition in their implementation: ml-ulex is based on regular
 expression derivatives rather than subset-construction, and ml-antlr
 is based on LL(k) parsing rather than LALR(1) parsing.