[Ffmpeg-devel-irc] ffmpeg-devel.log.20120210

[burek]

[01:47] <CIA-67> ffmpeg: 03Anton Khirnov 07master * rcd1ad18a65 10ffmpeg/ (72 files in 4 dirs): (log message trimmed)
[01:47] <CIA-67> ffmpeg: rawenc: switch to encode2().
[01:47] <CIA-67> ffmpeg: This changes a number of FATE results, since before this commit, the
[01:47] <CIA-67> ffmpeg: timestamps in all tests using rawenc were made up by lavf.
[01:47] <CIA-67> ffmpeg: In most cases, the previous timestamps were completely bogus.
[01:47] <CIA-67> ffmpeg: In some other cases -- raw formats, mostly h264 -- the new timestamps
[01:47] <CIA-67> ffmpeg: are bogus as well. The only difference is that timestamps invented by
[01:48] <CIA-67> ffmpeg: 03Paul B Mahol 07master * ra16c3a07a6 10ffmpeg/libavformat/apetag.h: 
[01:48] <CIA-67> ffmpeg: apetag: fix comment
[01:48] <CIA-67> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[01:48] <CIA-67> ffmpeg: Signed-off-by: Justin Ruggles <justin.ruggles at gmail.com>
[01:48] <CIA-67> ffmpeg: 03Diego Biurrun 07master * r0bba26466f 10ffmpeg/ (27 files in 8 dirs): cosmetics: Delete empty lines at end of file.
[01:48] <CIA-67> ffmpeg: 03Diego Biurrun 07master * ra2e40cd3e2 10ffmpeg/libswscale/ppc/yuv2rgb_altivec.h: swscale: ppc: Add missing header #includes to pass 'make checkheaders'.
[01:48] <CIA-67> ffmpeg: 03Diego Biurrun 07master * raa06d65693 10ffmpeg/libavcodec/arm/vp8.h: arm: Add missing #include to vp8.h to fix a make checkheaders warning.
[01:48] <CIA-67> ffmpeg: 03Tomas Härdin 07master * ra3923482f2 10ffmpeg/libavformat/mxfdec.c: 
[01:48] <CIA-67> ffmpeg: mxfdec: Make sure x < index_table->nb_ptses
[01:48] <CIA-67> ffmpeg: Avoids a SIGSEGV on files with IndexEntryCount < IndexDuration.
[01:48] <CIA-67> ffmpeg: Signed-off-by: Diego Biurrun <diego at biurrun.de>
[01:48] <CIA-67> ffmpeg: 03Tomas Härdin 07master * reffe3b9575 10ffmpeg/libavformat/mxfdec.c: 
[01:48] <CIA-67> ffmpeg: mxfdec: Check for NULL component
[01:48] <CIA-67> ffmpeg: This fixes a SIGSEGV with zzuf1.mxf.
[01:48] <CIA-67> ffmpeg: Signed-off-by: Diego Biurrun <diego at biurrun.de>
[01:48] <CIA-67> ffmpeg: 03Tomas Härdin 07master * r0373ec635d 10ffmpeg/libavformat/mxfdec.c: 
[01:48] <CIA-67> ffmpeg: mxfdec: Check eof_reached in mxf_read_local_tags()
[01:48] <CIA-67> ffmpeg: This fixes an infinite loop with zzuf2.mxf.
[01:48] <CIA-67> ffmpeg: Signed-off-by: Diego Biurrun <diego at biurrun.de>
[01:48] <CIA-67> ffmpeg: 03Diego Biurrun 07master * rf769cfedd8 10ffmpeg/ (libavcodec/Makefile libavutil/Makefile): build: Add missing directories to DIRS declarations.
[01:48] <CIA-67> ffmpeg: 03Tomas Härdin 07master * r11800c8c89 10ffmpeg/libavformat/mxfdec.c: 
[01:48] <CIA-67> ffmpeg: mxfdec: Never seek back in local sets and KLVs
[01:48] <CIA-67> ffmpeg: Specially crafted files can lead the parsing code to take too long.
[01:48] <CIA-67> ffmpeg: We fix a lot of these problems by not allowing local tags to extend
[01:48] <CIA-67> ffmpeg: past the end of the set and not allowing other KLVs to be read past
[02:41] <Akira^^> is this bug ? RTMP_CHANNELS is 64+0x10000 not 65599 on rtmppkt.h
[02:51] <Compn> sounds like bug
[02:51] <Compn> i think kostya maintains that stuff 
[06:30] <CIA-67> ffmpeg: 03Michael Niedermayer 07master * r29034e6503 10ffmpeg/ffmpeg.c: (log message trimmed)
[06:30] <CIA-67> ffmpeg: ffmpeg: Add threshold to discard crazy/damaged timestamps.
[06:30] <CIA-67> ffmpeg: The added tests are limited to the case where timestamp discontinuities
[06:30] <CIA-67> ffmpeg: are not allowed. The default is 30 hours which is arbitrarily picked and
[06:30] <CIA-67> ffmpeg: quite conservative.
[06:30] <CIA-67> ffmpeg: This prevents a out of memory condition due to duplicating a frame
[06:30] <CIA-67> ffmpeg: millions of times.
[06:30] <CIA-67> ffmpeg: 03Michael Niedermayer 07master * r789d6b73cf 10ffmpeg/libavformat/utils.c: 
[06:30] <CIA-67> ffmpeg: lavf: Rewrite metadata printing from dump_metadata().
[06:30] <CIA-67> ffmpeg: This code contained several bugs that mis-formated the output.
[06:30] <CIA-67> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[06:32] <lionaneesh> Hello Guys!
[06:33] <lionaneesh> I was just glancing through the sun rastfile code
[06:33] <lionaneesh> and found that we have a else (1) { if (!1) } condition there
[06:33] <lionaneesh> Should i send a patch?
[06:34] <lionaneesh> michaelni, Here?
[06:36] <michaelni> lionaneesh, iam here
[06:37] <lionaneesh> michaelni,  we have a else (1) { if (!1) } condition in the sunrast encoder
[06:37] <lionaneesh> will send a patch asap
[06:37] <michaelni> ok, thx
[06:39] <lionaneesh> Welcome
[06:40] <michaelni> seems the fix for ticket #865 added it
[06:40] <michaelni> it was depth == 8 before
[06:41] <lionaneesh> michaelni, Exactly
[07:29] <CIA-67> ffmpeg: 03Aneesh Dogra 07master * r1b7ff90107 10ffmpeg/libavcodec/sunrast.c: 
[07:29] <CIA-67> ffmpeg: sunrast: Remove the useless check.
[07:29] <CIA-67> ffmpeg: in , else (1) { if (!1) } the if conditional will never evaluate to be true.
[07:29] <CIA-67> ffmpeg: So as making the check useless.
[07:29] <CIA-67> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:59] <lionaneesh> Hey all :)
[17:40] <CIA-67> ffmpeg: 03Carl Eugen Hoyos 07master * rbaa8ec8d74 10ffmpeg/configure: 
[17:40] <CIA-67> ffmpeg: Add missing swscale dependency for MP-filters.
[17:40] <CIA-67> ffmpeg: Reported-by: Hanspeter Niederstrasser
[17:40] <CIA-67> ffmpeg: Fixes ticket #989.
[17:40] <CIA-67> ffmpeg: 03Hanspeter Niederstrasser 07master * r342afd12de 10ffmpeg/configure: 
[17:40] <CIA-67> ffmpeg: Add missing swresample dependency for ffplay.
[17:40] <CIA-67> ffmpeg: Fixes part of ticket #989.
[17:52] <CIA-67> ffmpeg: 03Carl Eugen Hoyos 07master * r3546f680d5 10ffmpeg/libavcodec/mpeg12.c: Add missing newline to mpeg1/2 intra matrix error message.
[18:16] <Compn> wasnt there a really old commit that made sure every file had a new line at the end ?
[18:16] Action: Compn notices commit that removes all of those new lines
[18:16] <Compn> not really important, just cosmetic bikeshedding
[18:17] <overflow_0f8b> the masters have much freetime
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * r4ab4b3db5a 10ffmpeg/ffmpeg.c: ffmpeg: use exit_program() if configure_video_filters() fails.
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * re22c7b4e25 10ffmpeg/ffmpeg.c: 
[20:35] <CIA-67> ffmpeg: ffmpeg: simplify video frame polling.
[20:35] <CIA-67> ffmpeg: input_video_filter and output_video_filter can't be NULL at this point.
[20:35] <CIA-67> ffmpeg: If they are, the current code would likely crash anyway (since
[20:35] <CIA-67> ffmpeg: filtered_frame would be NULL and sent to do_video_out().
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * raede21f1b6 10ffmpeg/ffmpeg.c: ffmpeg: reindent after video frame polling simplification.
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * rcf31060127 10ffmpeg/ffmpeg.c: ffmpeg: move filtered_frame to video stream processing scope.
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * ra78dd826ee 10ffmpeg/ffmpeg.c: 
[20:35] <CIA-67> ffmpeg: ffmpeg: move video stats code to do_video_out().
[20:35] <CIA-67> ffmpeg: This will allow some simplifications in transcode_video().
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * r9303b794b8 10ffmpeg/ffmpeg.c: 
[20:35] <CIA-67> ffmpeg: ffmpeg: move do_video_stats() above do_video_out().
[20:35] <CIA-67> ffmpeg: This avoid a forward declaration.
[20:35] <CIA-67> ffmpeg: 03Clément BSsch 07master * r086fc4d765 10ffmpeg/ffmpeg.c: 
[20:35] <CIA-67> ffmpeg: ffmpeg: move filtered_frame to the CONFIG_AVFILTER scope.
[20:35] <CIA-67> ffmpeg: This simplifies a bit the #ifdefery.
[20:48] <CIA-67> ffmpeg: 03Clément BSsch 07master * ra3c4371f3c 10ffmpeg/ffmpeg.c: ffmpeg: prefer NULL over 0 for av_log context pointer parameter.
[21:05] <ubitux> [FFmpeg-devel] [PATCH] BluRay protocol using libbluray
[21:05] <ubitux> oh :o
[21:05] <ubitux> nice nice nice :)
[21:06] <nevcairiel> like i mentioned in the GSoC discussion, its a days work
[21:06] <nevcairiel> if even that
[21:07] <nevcairiel> although this implementation has timestamp discontinuities on every clip change, but supposedly every mpegts consumer should be able to deal with that, so meh
[21:08] <nevcairiel> anyhow, libblurays basic reading api is refreshing simple
[21:11] <ubitux> ok :)
[21:12] <pasteeater> anyone know why cehoyos removes the CC from bug reports?
[21:24] <Compn> pasteeater : removes himself you mean ?
[21:24] <Compn> i remove myself on bugs too, its easier to read on web interface than having email clog up
[21:27] <pasteeater> Compn: no. the reporters email address. they add themselves for notifications and he removes them.
[21:43] <ubitux> pasteeater: maybe trac automatically notifies the reporter even if in non-cc?
[21:46] <nevcairiel> it does
[21:46] <nevcairiel> although maybe it does that by automatically adding the reporter to cc? :)
[21:48] <Compn> you could just ask carl :P
[22:25] <Plorkyeran> CC is for the people in addition to reporter and owner that want to be notified
[22:27] <vadim> hi 2 all
[22:57] <ubitux> exit(ret); /* not all OS-es handle main() return value */
[22:57] <ubitux> is this still true?...
[23:04] <CIA-67> ffmpeg: 03Michael Niedermayer 07master * raf46ca7356 10ffmpeg/libavcodec/msmpeg4.c: 
[23:04] <CIA-67> ffmpeg: msmpeg4: replace 999999 by INT_MAX and initial by a valid index.
[23:04] <CIA-67> ffmpeg: Fixes Ticket990
[23:04] <CIA-67> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:19] <CIA-67> ffmpeg: 03Reimar Döffinger 07master * r5338566eaf 10ffmpeg/libavcodec/mpegvideo.c: 
[23:19] <CIA-67> ffmpeg: mpegvideodec: minor simplifications.
[23:19] <CIA-67> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[23:19] <CIA-67> ffmpeg: 03Reimar Döffinger 07master * r3fe00cac71 10ffmpeg/libavcodec/mpegvideo.c: 
[23:19] <CIA-67> ffmpeg: Fix bitexact intra mismatch control.
[23:19] <CIA-67> ffmpeg: The DC coefficient should be included, too.
[23:19] <CIA-67> ffmpeg: This probably was missed because DC quantizer is always
[23:19] <CIA-67> ffmpeg: even for MPEG-1/2 but this function is also used for MPEG-4.
[23:19] <CIA-67> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[23:19] <CIA-67> ffmpeg: 03Reimar Döffinger 07master * r394d41ee30 10ffmpeg/libavcodec/ (pngdsp.h x86/pngdsp.asm): 
[23:19] <CIA-67> ffmpeg: Partially revert "Fix png decoding on x86."
[23:19] <CIA-67> ffmpeg: This partially reverts commit 58dabf7bf2fdd08f79173da0df613127ff783028.
[23:19] <CIA-67> ffmpeg: It is no longer necessary to use unaligned mov.
[23:19] <CIA-67> ffmpeg: The swapped mov argument fix remains though.
[00:00] --- Sat Feb 11 2012