oscam (1.20-9519~r9292-ppa1~saucy) saucy; urgency=medium
* [r9292]
Add version and no. of cards information to gbx hello log.
* [r9291]
clockfix: Fix building with cmake.
* [r9290]
Further code restructuring for gbx.
* [r9289]
small optical changes
* [r9288]
clockfix: uclibc 0.9.31 is missing pthread_condattr_setclock().
* [r9287]
use checkbox not dropdown if possible
* [r9286]
visual change for stats
* [r9285]
clockfix: Fix build on Android.
Android's libc in NDK r9c do not ship with -lrt and pthread_condattr_setclock().
* [r9284]
clockfix: Do not use __GNUC_PREREQ macro because it may not exist.
* [r9283]
clockfix: glibc 2.3.6 in some ppc old toolchains do not have pthread_condattr_setclock().
Reported in the forum.
* [r9282]
clockfix: uclibc 0.9.30 do not support pthread_condattr_setclock().
Found by compiling OSCam for uclinux.
* [r9281]
Restore formatting of oscam-config.h
The massive code reordering made this header look like crap.
Restore the proper formatting.
This commit contains only whitespace changes.
* [r9280]
clockfix: Old uclibc versions do not have pthread_condattr_setclock().
uclibc <= 0.9.28 is known to not have pthread_condattr_setclock().
This was found when trying to compile CLOCKFIX enabled OSCam using
AZBOX toolchain.
Please report any other build failures related to missing
pthread_condattr_setclock() in the forum.
* [r9279]
clockfix: Convert clock type to enum.
Clean up of CLOCKFIX related code a bit.
* [r9278]
clockfix: Add basic clockfix infrastructure.
This just introduces the infrastructure and configuration
parts of CLOCKFIX patch. It doesn't change OSCam to actually
use CLOCK_MONOTONIC and doesn't switch time checks to milliseconds
instead of seconds.
This is just a preparation and shouldn't change OSCam's timing
behaviour.
WARNING: If you see build failure like this:
undefined reference to `pthread_condattr_setclock'
Report it in the forum and disable CLOCKFIX configuration.
`./config.sh -D CLOCKFIX`
Patch by: theparasol
* [r9277]
Replace free() with NULLFREE().
This changes were made as part of CLOCKFIX patch development.
Patch by: theparasol
* [r9276]
Added missing option for 'detect' parameter.
* [r9275]
revert last
sorry for last commit not all code is moved to lib
* [r9274]
remove redundant code
fix build warning
* [r9273]
add footer var for run as
* [r9272]
Remove receive buffer from struct gbox_data
* [r9271]
- Compile fix for stapi
* [r9270]
webif: Simplify system info code.
* [r9269]
Revert "init struct only if needet"
This reverts commit r9267.
p_stat_cur is used outside of the ifdef.
* [r9268]
stapi:
- Tryfix audiopid change:
Seems audio pid that is not selected as active by user fails to be slotted.
But it was registered as enable anyway, if user selected that audio pid later on it wont be enabled since dvbapi thought it was enabled already.
* [r9267]
init struct only if needet
* [r9266]
webif: Prevent possible format string exploit.
module-webif.c:4228:47: warning: format string is not a string literal (potentially insecure) [-Wformat-security]
* [r9265]
webif: Prevent warninigs about vararg functions that do not have format string.
module-webif.c:111:100: warning: data argument not used by format string [-Wformat-extra-args]
* [r9264]
webif: Add system info in status page.
Apply system_info_V9.patch from gismo2004.
* [r9263]
dvbapi: Remove questionable code.
The check is bogus because setting er->chid to 0x100000 makes it zero.
> er->chid = (chid != 0 ? chid : 0x10000); // if not zero apply, otherwise use no chid value 0x10000
module-dvbapi.c:3022:34: warning: implicit conversion from 'int' to 'uint16_t' (aka 'unsigned short') changes value from 65536 to 0 [-Wconstant-conversion]
* [r9262]
dvbapi: Prevent possibly bogus warning.
Older compilers can't see that the variable is most certainly being
initialized later.
module-dvbapi.c: In function `dvbapi_resort_ecmpids':
module-dvbapi.c:1799: warning: 'c' might be used uninitialized in this function
* [r9261]
dvbapi: Prevent bogus warning issued by old compilers.
abs() is guaranteed to return positive number so the warning is bogus.
module-dvbapi.c:896:12: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare]
* [r9260]
- Stapi use always indexer 1 exception for disable all on stop descrambling: use -1
- Reintroduced only enable / disable pid if nessecary (should at least stop log flooding!)
* [r9259]
no entitlements on second line
* [r9258]
fix long start wrong printf
* [r9257]
revert #9255 move to save html
* [r9256]
- This code is used to stop demuxers not to signal changes to pmtfile that means pid updates for a channel -> reverted!
* [r9255]
move entitlements to second line
* [r9254]
- Fresh ideas, sending logs by pm really helps -> TNX!
Now stapi devices use fixed indexer 1 since they dont use indexers at all.
Next is fixing the overkill of loglines, so plz send me in your personal -d128 logfiles by PM
* [r9253]
move out hopefully last part out of module-webif.c
* [r9252]
add unenc value for hover fix comment 3577
* [r9251]
fix buildwarning
* [r9250]
enable picon for proxy again
* [r9249]
webif: Fix reader picons url.
Thanks to khimtiki.
* [r9248]
webif: Refactor setting of reader and user names in status page /part 2/.
* [r9247]
webif: Refactor setting of reader and user names in status page /part 1/.
* [r9246]
webif: Use proper user name variable in status.
* [r9245]
webif: Use proper reader name variable in status.
* [r9244]
webif: Cleanup reader name initialization.
* [r9243]
webif: Encode reader name properly.
* [r9242]
webif: Fix user name encoding in URLs.
This should fix ticket #3579.
* [r9241]
webif: Cleanup user name encoding.
* [r9240]
webif: Close tag.
* [r9239]
entitlements also used with CCCAM
* [r9238]
urlencode links with + - usw
tryfix #3579
* [r9237]
fixed userconfig userpicon
* [r9236]
webif: Reword shutdown page.
Reword and also replace urlencoded symbols with normal ones.
' (single quote) and / (slash) do not need to be urlencoded when
used in normal text paragraph.
* [r9235]
webif: More fixes for reader names containing strange characters.
* [r9234]
webif: Encode & (amersand) symbol correctly.
& must be properly encoded as & in html.
* [r9233]
- Stapi: More errorlogging and new attempt to fix audio switching
* [r9232]
webif: Correctly initialize reader and user labels in status.
Commit r9231 fixed label initialization but missed the fact that
the needed variables were not initialized.
* [r9231]
webif: Use URL encoded parameters when adding strings in URLs.
This fixes ticket #3577.
* [r9230]
- jet another tryfix stapi dynamic streampids
Seams not all stapi receivers work the same, rev 9214 reported working and not working at all :(
Rebased again on rev 9173 but now stop_descrambling will disable all streampids like it was in trunk before dynamic streampids for stapi
* [r9229]
- module-gbox.c. some code cosmetics.
* [r9228]
fix hover info
* [r9227]
continue automation cardmhz only applicable with autospeed = 0
for irdeto the value F convertion factor wich should be 558 according atr table
must indeed be 618 that's indeed magic
* [r9226]
forget to set vars
fix #3578
* [r9225]
fix build warnings
* [r9224]
replace png with full resizable svg from BIT
no depcrated insecure ie8 supported!!! dont use ie8 !!!
* [r9223]
enable reader directedit on label
enable edit for reader & user on nopicon mode
* [r9222]
trying increased worketu for t14 irdeto cards v2 and tripple
zigo nl does not work on v2 and tripple (sometimes well)
try increased worketu by those readers
* [r9221]
edit on label on userconfig
* [r9220]
small change default N by init
* [r9219]
temporary revert a part off 9209
* [r9218]
move log out of module-webif.c
+ fix #3576
* [r9217]
move headers out of module-webif.c
+ small bugfix
* [r9216]
set proxy caid:srvid to 0000:0000 until there is a propper solution
* [r9215]
enabling bitbang mode
* [r9214]
- Fix for loop on stop decoding created by copy / pasting too much code to be recycled ;)
* [r9213]
- More fix for wrong use of unset last_caid + last_srvid
* [r9212]
- Fix for setting wrong last caid & srvid on readers
* [r9211]
- Another tryfix stapi dynamic streampids (tnx ngc1927 for log!)
rev 9208 stapi -> standby -> wakeup: have to change channels otherwise no descrambling
The stapi receiver simply shutsdown removing all pid en descrambler associations
and doesnt let dvbapi do it. However the streampids in dvbapi are still registered as enabled -> no descrambling any more on this channel!
Rebased the patch to workaround this fact, hope I got it right this time :)
* [r9210]
fix 3574
* [r9209]
smartreader protocol revert to static mutex lock
Adapted irdeto init convertion factor to iso table value
Added extra log output to trace init error on T14 cards
* [r9208]
- Next tryfix stapi dynamic streampids
* [r9207]
clean out current channel
* [r9206]
fix cccam select
thanx gismo2004
* [r9205]
add keepalive
add CacheEx CW Check
add wait time
thanks to gismo2004 for this patch
* [r9204]
add statistics-link for nonpicon status-webif in lbvaluereader
* [r9203]
addstatistics function to reader-lbvalue
add show picture for by XXX if schow picon is enabled
* [r9202]
revert to 9200 as this is working...
@skyndas: PLEASE stop this commit war! when you apply a patch update it to latest trunk before commit to avoid problems like this. if you have a problem with a commit or the person who did it discuss this personally without breaking things.
* [r9201]
* [r9200]
fix trunk
* [r9199]
* [r9198]
change more printf to addVar
* [r9197]
Update for constant CW file format in documentation.
* [r9196]
fix build warning
* [r9195]
fix 3571
* [r9194]
fix more printf to addVar
* [r9193]
Fix wrong handling of gbx card level.
* [r9192]
remove useless </span>
thanx capncook
* [r9191]
forget one file to update
* [r9190]
enable directedit for reader/proxy/user
move user out of module-webif.c
* [r9189]
change printf to addVar
* [r9188]
move status buttons out of module-webif.c
* [r9187]
move notify out of module-webif.c
* [r9186]
move usericon out of webif add missing info
* [r9185]
Include poll.h instead of sys/poll.h
poll.h is the correct header to get poll(2) definition.
Found by trying to compile oscam with musl libc.
* [r9184]
Partly revert r8909:
In constcw module, we are now supporting the standard format again which is CAID:PROVIDER:SID:PMTPID:ECMPID::CW.
For Enigma2 receivers, we are parsing the PMTPid so that the standard files should work correctly even for problematic channels which only differ by PMTPid.
For other receivers which don't provide the PMTPid and thus have it set to 0000 (or if PMTPid would not be enough to distinguish), there is an extended format for constcw where you can put the VPID inside (you can obtain it from OSCam log). However, this should only be used for problematic channels and not for the majority.
The extended (OSCam-specific) format is: CAID:PROVIDER:SID:PMTPID:ECMPID:VPID:CW.
* [r9183]
fix wrong upload file
* [r9182]
move lastchannel and add missing info
* [r9181]
rework on protopicons need to change on reader too
* [r9180]
move nostats out of module-webif.c
* [r9179]
fix wrongly used append
* [r9178]
remove Loadbalance statistics for reader from module-webif.c
* [r9177]
add readerstatsrowbit
* [r9176]
simplify & enable picons for readers
* [r9175]
forget one file
* [r9174]
remove LOGO from module-webif.c & move to html
* [r9173]
- tryfix for stapi receivers dynamically registering pids incorrect
* [r9172]
Added documentation for cache HT patch.
* [r9171]
Removed 'gbox_group' from documentation, moved 'gbox_my_password' to 'oscam.conf' documentation.
* [r9170]
Protect against unlikely buffer overflow.
* [r9169]
Fix password check of incoming peer's messages.
* [r9168]
Remove function mgencrypted_fn as nobody was using it and it probably didn't work anyway.
* [r9167]
fix last commit
* [r9166]
fix write emm
http://www.streamboard.tv/wbb2/thread.php?postid=493522#post493522
* [r9165]
Do not use remove() to delete files.
The proper function is unlink().
The bug was spotted by blaky and reported in streamboard forum.
* [r9164]
Cosmetics: In file gbx_card.info use device instead of label. Compatibility reasons. Proposed by felixka.
* [r9163]
- Move Parameter gbox_my_password to [gbox] section in oscam.conf. It belongs there because it may not vary from reader to reader as it is used for routing as well. Can somebody please adjust the wiki?
- Remove Parameter gbox_group from config because it was unused anyway.
* [r9162]
- Move gbox de-/encrypt and (de-)compress functions to a separate file.
- No functionality changes.
* [r9161]
Fix Typo - thanks FilipeAmadeuO
* [r9160]
- fix for timeout on openplipc/enigma2pc if adapter !=adapter0
Tnx milaebrothers for research and providing patch!
* [r9159]
- CW Rate and CASC USERS in USERS to be shown only when build with anticascading module
- Fix typo
Forum: http://www.streamboard.tv/wbb2/thread.php?postid=492696
* [r9158]
- Add cards for betatunnel accrding to entry on gbox account in oscam.user. Suggested by felixka.
- Correct gbx ecmfmt
* [r9157]
- Dynamic streampids for stapi devices
This should fix issues like freezing and pixalization on pmtupdates as seen on "orf bundeslander" and sky channels
* [r9156]
WebIf: Fix typo.
* [r9155]
-fix (again) cw send to legacy camd3
* [r9154]
turn string
i have XXXms by XXX
* [r9153]
short large grouplist
show all on hover
* [r9152]
Fix gbx betatunnel. Thx felixka.
* [r9151]
enable more stats in status page
its needet to read out in html diff
* [r9150]
fix build warning
but why riding a death horse ?
* [r9149]
WebIf: Changelog in forum http://www.streamboard.tv/wbb2/thread.php?postid=492015#post492015
* [r9148]
*this is more safety for commit #9146
* [r9147]
Revert useless cwc improvement (commit 9145) in cache code.
* [r9146]
*changed cwc status share over CE related to this problem: http://www.streamboard.tv/wbb2/thread.php?postid=491977#post491977
-this a importend changeset if one in your sharenetwork uses
cwcycle_usecwcfromce = 1
-before the odd/even Byte is used to mark a cycleturn between CE Partners..
but this is different between providers.. see post above
-now Bit8 of buf[18] mark the cycleturn between CE Partners
* [r9145]
Cosmetic: fix cacheex hits stat.
* [r9144]
Fix problem in LB using check_cache thread (cacheex build). See forum: http://www.streamboard.tv/wbb2/thread.php?postid=491986#post491986
Add cw counter info in log, when getting cw from cache and cw counter>1.
* [r9143]
module-gbox.
Minor.
Just make double sure that there is no buffer overflow because of too many hops.
* [r9142]
Cosmetic: print not found when cacheex and no "normal readers", instead rejected group.
* [r9141]
Fix previous commit.
* [r9140]
Removed paramters "check_cw_count" and "check_cw_count_mode".
Now using new parameter:
cacheex_cw_check = caid[&mask][@provid][$servid]:mode:counter
counter:
set minimum cw counter to allow cw is used.
Default=1, use default behaviour: use cw with highest counter when cache is checked.
mode:
specify behaviour for counter parameter
0: when wait_time expires, serve highest counter's cw got anyway, even if no counter reached.
1: never serve cw (coming from cacheex) stored in cache if its counter not reaches counter. When wait_time expires, requests will go to normal readers! Only when a cw reaches counter, it can be served to clients.
Default 0.
* [r9139]
Do not bloat ecm_request_t with gbox variables.
This is supposed to fix ticket #3555. Thx lattjo.
* [r9138]
fix two compiler warnings
* [r9137]
Just cosmetics.
* [r9136]
- Fix module-gbox for long ecms with length > 255 bytes. Length field is 12 Bit.
- Cleanup some redundant structures in module-gbox. Get rid of gbox_ecm_info, only use ecm_request_t of oscam.
* [r9135]
*backport svn9037
*reduce logs if a bad cw detected from cache
*overhaul cwc light for NDS Caid's
for 0x80 ECM's CW: XXXXXXXXXXXXXXXX0000000000000000 (cwc NDS OK)
for 0x81 ECM's CW: 0000000000000000XXXXXXXXXXXXXXXX (cwc NDS OK)
for 0x8X ECM's CW: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX (cwc NDS NOK)
for a swapped CW the check is pos. passed for a possible backswapping
u can use for cwc light:
cwcycle_check_enable
cwcycle_check_caid
cwcycle_onbad
cwcycle_allowbadfromffb
* [r9134]
a small corection smartreader start sequence + revert the atr iso
insert a smartreader detect find delay for triple and v2 . to avoid the not found delay in first fase.
Revert atr.c to standard iso table for FI=9 cards ass well. In oscam however that iso value does not work for the concerned cards why ... that's another question
* [r9133]
Cards with an FI=9 do not work at speeds below 600 mhz
adapted base atr table to have sfmax off 600 instead off 500
* [r9132]
Sets smartreader mutex lock reader based
* [r9131]
- commit 9093 reverted partly changes to preferlocalcards -> fixed!
* [r9130]
WebIf: Fix for the same behavior of button "debugselect = ALL" in status.html and files.html?file=logfile
-- Andrey Pavlenko <email address hidden> Sun, 12 Jan 2014 14:26:26 +0400