[Ffmpeg-devel-irc] ffmpeg-devel.log.20120604

[burek]

[00:10] <ubitux> Compn: ping? :)
[00:10] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * rf3fea5bab1 10ffmpeg/libavfilter/vf_lut.c: lavfi/lut: fix component -> rgba mapping
[00:10] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * re10eac910a 10ffmpeg/libavcodec/ (Makefile raw.c): 
[00:10] <CIA-119> ffmpeg: lavc/raw: add test utility for listing fourcc/pix_fmt map
[00:10] <CIA-119> ffmpeg: The test program is useful for checking which raw formats are currently
[00:10] <CIA-119> ffmpeg: identified by a codec tag.
[00:11] <Compn> ubitux : whut
[00:12] <ubitux> Compn: can i ask you to put a few samples on samples.ffmpeg?
[00:12] <ubitux> (http://blackhole.ubitux.fr/sami/)
[00:12] <ubitux> (in the sub directory)
[00:12] <ubitux> i didn't find any SAMI subtitles in it :p
[00:13] <Compn> sure
[00:13] <ubitux> thx :)
[00:16] <Compn> done
[00:16] <Compn> samples/sub/sami
[00:16] <ubitux> awesome, thx :)
[00:17] Action: Compn goes to add samples to sami page in multimediawiki
[00:18] <Compn> oh bah no one made a page for it yet
[00:18] Action: Compn declines
[00:25] <saste> ubitux: any comment on the avfilter_default_filter_name() patch?
[00:26] <ubitux> michaelni didn't comment?
[00:26] <ubitux> saste: is there anything questionable with it?
[00:27] <saste> ubitux: dunno, maybe the name or other bikesheds
[00:27] <ubitux> (btw, isn't it possible to do so with ffprobe as well?)
[00:27] <ubitux> is there any point in exporting the symbol to the user?
[00:28] <ubitux> afaik libav broke the ability for the users to make filters anyway
[00:28] <saste> because it's in my commit queue
[00:28] <saste> ubitux: yes and i honestly don't know why
[00:28] <saste> but it is still possible to create video filters
[00:28] <saste> so the function could be useful
[00:28] <ubitux> most likely because it was incomplete
[00:28] <saste> also what's the point of keeping such a funtion private?
[00:29] <ubitux> keeping functions private allows to change it easily :p
[00:29] <saste> yes but external users can't use them
[00:29] <ohsix> that's what makes them easy to change
[00:29] <saste> and i prefer to avoid that if there is not a technical reason for that
[00:30] <ubitux> bikeshednit++: why the "default"?
[00:30] <ohsix> you package and have an api if you want people using them :]
[00:30] <saste> ubitux: it's framed after av_default_item_name(), just changed av->avfilter and item-> filter
[00:31] <ubitux> ok
[00:31] <ubitux> should be fine then :p
[00:31] <saste> ok thanks i'll push it soon
[00:32] <ubitux> btw, i'm going to push the scene detection patch soon (after i fixed the nonsense description i put while under LSD)
[00:33] <ubitux> so if you have any comment& :p
[00:34] <iive> ubitux: what is this patch about?
[00:35] <ubitux> iive: http://blackhole.ubitux.fr/sami/
[00:35] <ubitux> oups
[00:35] <iive> filter or tunning of the ME s.d.
[00:35] <ubitux> iive: http://imgur.com/a/icpfk sorry
[00:35] <ubitux> filter
[00:35] <ubitux> user purpose
[00:36] <ubitux> not a codec scene cut or anything like that
[00:38] <CIA-119> ffmpeg: 03Luca Barbato 07master * r21e2dc9fb7 10ffmpeg/libavformat/ (flvdec.c flvenc.c): 
[00:38] <CIA-119> ffmpeg: flv: support stream text data as onTextData
[00:38] <CIA-119> ffmpeg: Adobe specifies onTextData as the standard message to use to deliver
[00:38] <CIA-119> ffmpeg: text information.
[00:38] <CIA-119> ffmpeg: Signed-off-by: Luca Barbato <lu_zero at gentoo.org>
[00:38] <CIA-119> ffmpeg: 03Anton Khirnov 07master * rcf3a1948e9 10ffmpeg/libavcodec/avpacket.c: 
[00:38] <CIA-119> ffmpeg: avpacket: fix duplicating side data.
[00:38] <CIA-119> ffmpeg: Use correct side data size instead of just zeroed field.
[00:38] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r007aedeebf 10ffmpeg/libavformat/librtmp.c: 
[00:38] <CIA-119> ffmpeg: librtmp: add rtmp_app and rtmp_playpath private options.
[00:38] <CIA-119> ffmpeg: This makes it easier to switch between native rtmp and librtmp.
[00:38] <CIA-119> ffmpeg: 03Anton Khirnov 07master * ra91943bcef 10ffmpeg/libavformat/librtmp.c: librtmp: don't abuse a variable for two unrelated things.
[00:38] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r7d82020fcb 10ffmpeg/: (log message trimmed)
[00:38] <CIA-119> ffmpeg: Merge remote-tracking branch 'qatar/master'
[00:56] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r4d6a8a2bdb 10ffmpeg/ (26 files in 2 dirs): 
[00:56] <CIA-119> ffmpeg: lavfi: add avfilter_default_filter_name()
[00:56] <CIA-119> ffmpeg: The function is modelled after av_default_item_name(), and will print the
[00:56] <CIA-119> ffmpeg: name of the instance filter if defined, otherwise the name of the filter.
[00:56] <CIA-119> ffmpeg: This allows to show the instance name in the log, which is useful when
[00:56] <CIA-119> ffmpeg: debugging complex filter graphs.
[00:56] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r6a9854e8f4 10ffmpeg/ffmpeg.c: ffmpeg: specify more meaningful names for the lavfi buffer sources
[00:56] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * rf1add01d98 10ffmpeg/ffmpeg.c: 
[00:56] <CIA-119> ffmpeg: ffmpeg: be more verbose when printing an error in transcode()
[00:57] <CIA-119> ffmpeg: Helps debuggability.
[01:04] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r730d079bf7 10ffmpeg/libavcodec/ffv1.c: 
[01:04] <CIA-119> ffmpeg: ffv1: fix log level of FF_DEBUG_PICT_INFO
[01:04] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:04] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r38c9ebd2a9 10ffmpeg/libavcodec/ffv1.c: 
[01:04] <CIA-119> ffmpeg: ffv1dec: print more information for -debug 1
[01:04] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:41] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r0f0f3bd1e0 10ffmpeg/libavfilter/ (20 files): 
[02:41] <CIA-119> ffmpeg: lavfi: use av_default_item_name() as filter private context logger
[02:41] <CIA-119> ffmpeg: avfilter_default_filter_name() is supposed to access an AVFilterContext
[02:41] <CIA-119> ffmpeg: struct, if used with a private struct it will cause a crash since it will
[02:41] <CIA-119> ffmpeg: access fields which are non defined in the private struct.
[02:56] <michaelni> saste, thx
[07:29] <mancha> is cehoyos around?
[07:47] <CIA-119> ffmpeg: 03Clément BSsch 07master * re00b2d284e 10ffmpeg/cmdutils.c: 
[07:48] <CIA-119> ffmpeg: options: stop after option is found.
[07:48] <CIA-119> ffmpeg: This reverts 78da04384a6c22820518706d84631006d31a85ea; it is not
[07:48] <CIA-119> ffmpeg: required anymore since the previous commit (these extended checks were
[07:48] <CIA-119> ffmpeg: for the -timecode option in both muxers and encoders), and reduces the
[07:48] <CIA-119> ffmpeg: diff with the fork.
[07:48] <CIA-119> ffmpeg: 03Clément BSsch 07master * r6b35f1a2a6 10ffmpeg/ (6 files in 4 dirs): (log message trimmed)
[07:48] <CIA-119> ffmpeg: timecode: move timecode muxer options to metadata.
[07:48] <CIA-119> ffmpeg: Some demuxers set a timecode in the format or streams metadata. The
[07:48] <CIA-119> ffmpeg: muxers now make use of this metadata instead of a duplicated private
[07:48] <CIA-119> ffmpeg: option.
[07:48] <CIA-119> ffmpeg: This makes possible transparent copy of the timecode when transmuxing
[07:48] <CIA-119> ffmpeg: and transcoding.
[09:10] <CIA-119> ffmpeg: 03Clément BSsch 07master * rc790ae0265 10ffmpeg/ffmpeg.c: ffmpeg: add missing return carriage in error message.
[10:37] <ubitux> huh& we can't have multiple tref in a track? oO
[12:20] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * rc02ef07881 10ffmpeg/libavcodec/Makefile: Fix G.723.1 encoder Makefile dependencies.
[12:20] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * r84986b4e61 10ffmpeg/libavcodec/proresdsp.c: 
[12:20] <CIA-119> ffmpeg: Fix compilation condition for some ProRes dsp decoder functions.
[12:20] <CIA-119> ffmpeg: Found, analysed and tested by trac user Jamal.
[12:20] <CIA-119> ffmpeg: Fixes part of Ticket #1404.
[12:20] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * ra4b885d55e 10ffmpeg/libavcodec/Makefile: 
[12:20] <CIA-119> ffmpeg: Fix ProRes decoder Makefile dependencies.
[12:20] <CIA-119> ffmpeg: Found, analysed and tested by trac user Jamal.
[12:20] <CIA-119> ffmpeg: Fixes part of Ticket #1404.
[12:20] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * r568a592418 10ffmpeg/libavcodec/proresdsp.c: 
[12:20] <CIA-119> ffmpeg: Fix compilation condition for some ProRes dsp encoder functions.
[12:20] <CIA-119> ffmpeg: Found, analysed and tested by trac user Jamal.
[12:20] <CIA-119> ffmpeg: Fixes part of Ticket #1404.
[12:34] <CIA-119> ffmpeg: 03Alex Converse 07release/0.5 * r7944a87ba8 10ffmpeg/libavcodec/dpcm.c: (log message trimmed)
[12:34] <CIA-119> ffmpeg: dpcm: ignore extra unpaired bytes in stereo streams.
[12:34] <CIA-119> ffmpeg: Fixes: CVE-2011-3951
[12:34] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:34] <CIA-119> ffmpeg: (cherry picked from commit ce7aee9b733134649a6ce2fa743e51733f33e67e)
[12:34] <CIA-119> ffmpeg: (cherry picked from commit eaeaeb265fe46e1d81452960de918227541873b4)
[12:34] <CIA-119> ffmpeg: Conflicts:
[12:35] <CIA-119> ffmpeg: 03Alexander Strange 07release/0.5 * r0bf8e22cdb 10ffmpeg/libavcodec/h264.c: (log message trimmed)
[12:35] <CIA-119> ffmpeg: h264: Add check for invalid chroma_format_idc
[12:35] <CIA-119> ffmpeg: Fixes a crash when FF_DEBUG_PICT_INFO is used.
[12:35] <CIA-119> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[12:35] <CIA-119> ffmpeg: (cherry picked from commit 6ef4063957aa5025c8d2cd757b6a537e4b6874df)
[12:35] <CIA-119> ffmpeg: Fixes: CVE-2012-0851
[12:35] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[12:35] <CIA-119> ffmpeg: 03Janne Grunau 07release/0.5 * r7867cbaf6c 10ffmpeg/libavcodec/adpcm.c: (log message trimmed)
[12:35] <CIA-119> ffmpeg: adpcm: ADPCM Electronic Arts has always two channels
[12:35] <CIA-119> ffmpeg: Fixes half of http://ffmpeg.org/trac/ffmpeg/ticket/794
[12:35] <CIA-119> ffmpeg: Adresses CVE-2012-0852
[12:35] <CIA-119> ffmpeg: (cherry picked from commit bb5b3940b08d8dad5b7e948e8f3b02cd2eb70716)
[12:35] <CIA-119> ffmpeg: Conflicts:
[12:35] <CIA-119> ffmpeg:  libavcodec/adpcm.c
[12:35] <CIA-119> ffmpeg: 03Alex Converse 07release/0.5 * r5629c39101 10ffmpeg/libavcodec/kmvc.c: (log message trimmed)
[12:35] <CIA-119> ffmpeg: kmvc: Check palsize.
[12:35] <CIA-119> ffmpeg: Fixes: CVE-2011-3952
[12:35] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:35] <CIA-119> ffmpeg: Based on fix by Michael Niedermayer
[12:35] <CIA-119> ffmpeg: (cherry picked from commit 386741f887714d3e46c9e8fe577e326a7964037b)
[12:35] <CIA-119> ffmpeg: (cherry picked from commit 416849f2e06227b1b4a451c392f100db1d709a0c)
[12:35] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.5 * rae6c57859c 10ffmpeg/libavcodec/qdm2.c: (log message trimmed)
[12:35] <CIA-119> ffmpeg: qdm2: clip array indices returned by qdm2_get_vlc().
[12:35] <CIA-119> ffmpeg: Prevents subsequent overreads when these numbers are used as indices
[12:35] <CIA-119> ffmpeg: in arrays.
[12:35] <CIA-119> (68 lines omitted)
[12:51] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.6 * r1edf848a81 10ffmpeg/libavformat/nsvdec.c: (log message trimmed)
[12:51] <CIA-119> ffmpeg: nsvdec: Fix use of uninitialized streams.
[12:51] <CIA-119> ffmpeg: Fixes CVE-2011-3940 (Out of bounds read resulting in out of bounds write)
[12:51] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:51] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:51] <CIA-119> ffmpeg: (cherry picked from commit 5c011706bc752d34bc6ada31d7df2ca0c9af7c6b)
[12:51] <CIA-119> ffmpeg: Signed-off-by: Alex Converse <alex.converse at gmail.com>
[12:51] <CIA-119> ffmpeg: 03Alex Converse 07release/0.6 * r87007519c8 10ffmpeg/libavformat/nsvdec.c: (log message trimmed)
[12:51] <CIA-119> ffmpeg: nsvdec: Be more careful with av_malloc().
[12:51] <CIA-119> ffmpeg: Check results for av_malloc() and fix an overflow in one call.
[12:51] <CIA-119> ffmpeg: Related to CVE-2011-3940.
[12:51] <CIA-119> ffmpeg: Based in part on work from Michael Niedermayer.
[12:51] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:51] <CIA-119> ffmpeg: (cherry picked from commit 8fd8a48263ff1437f9d02d7e78dc63efb9b5ed3a)
[12:51] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.6 * rb46141b0d1 10ffmpeg/libavformat/dv.c: (log message trimmed)
[12:51] <CIA-119> ffmpeg: dv: Fix null pointer dereference due to ach=0
[12:51] Last message repeated 1 time(s).
[12:51] <CIA-119> ffmpeg: Fixes part2 of CVE-2011-3929
[12:51] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:51] <CIA-119> ffmpeg: Reviewed-by: Roman Shaposhnik <roman at shaposhnik.org>
[12:51] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:51] <CIA-119> ffmpeg: 03Alex Converse 07release/0.6 * ra8f4db0acd 10ffmpeg/libavformat/dv.c: (log message trimmed)
[12:52] <CIA-119> ffmpeg: dv: Fix small stack overread related to CVE-2011-3929 and CVE-2011-3936.
[12:52] <CIA-119> ffmpeg: Found with asan.
[12:52] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:52] <CIA-119> ffmpeg: Signed-off-by: Alex Converse <alex.converse at gmail.com>
[12:52] <CIA-119> ffmpeg: (cherry picked from commit 2d1c0dea5f6b91bec7f5fa53ec050913d851e366)
[12:52] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[12:52] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.6 * r224025d852 10ffmpeg/libavcodec/atrac3.c: (log message trimmed)
[12:52] <CIA-119> ffmpeg: atrac3: Fix crash in tonal component decoding.
[12:52] <CIA-119> ffmpeg: Add a check to avoid writing past the end of the channel_unit.components[]
[12:52] <CIA-119> (136 lines omitted)
[12:52] <CIA-119> ffmpeg: (cherry picked from commit d2205d6543881f2e6fa18c8a354bbcf91a1235f7)
[12:52] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.6 * r83fa442b71 10ffmpeg/libavcodec/eatqi.c: (log message trimmed)
[12:52] <CIA-119> ffmpeg: tqi: Pass errors from the MB decoder
[12:52] <CIA-119> ffmpeg: This silences some valgrind warnings.
[12:52] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[12:52] <CIA-119> ffmpeg: Fixes second half of http://ffmpeg.org/trac/ffmpeg/ticket/794
[12:52] <CIA-119> ffmpeg: Bug found by: Oana Stratulat
[12:52] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:52] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.6 * r598eb973a7 10ffmpeg/: (log message trimmed)
[12:52] <CIA-119> ffmpeg: Merge remote-tracking branch 'qatar/release/0.6' into release/0.6
[12:52] <CIA-119> ffmpeg: * qatar/release/0.6: (32 commits)
[12:53] <CIA-119> ffmpeg:  Bump version number for 0.6.6 release.
[12:53] <CIA-119> ffmpeg:  tqi: Pass errors from the MB decoder
[12:53] <CIA-119> ffmpeg:  ea: check chunk_size for validity.
[12:53] <CIA-119> ffmpeg:  png: check bit depth for PAL8/Y400A pixel formats.
[12:53] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.6 * r679de9b5aa 10ffmpeg/libavformat/electronicarts.c: (log message trimmed)
[12:53] <CIA-119> ffmpeg: ea: check chunk_size for validity.
[12:53] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[12:53] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[12:53] <CIA-119> ffmpeg: (cherry picked from commit 273e6af47b38391f2bcc157cca0423fe7fcbf55c)
[12:53] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[12:53] <CIA-119> (28 lines omitted)
[13:07] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.8 * r26ac878cc2 10ffmpeg/libavcodec/ (h264.c h264_ps.c): (log message trimmed)
[13:07] <CIA-119> ffmpeg: h264: additional protection against unsupported size/bitdepth changes.
[13:07] <CIA-119> ffmpeg: Fixes crashes in codepaths not covered by original checks.
[13:07] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:07] <CIA-119> ffmpeg: (cherry picked from commit 732f9fcfe54fc9a0a7bbce53fe86b38744c2d301)
[13:07] <CIA-119> ffmpeg: Conflicts:
[13:07] <CIA-119> ffmpeg:  libavcodec/h264.c
[13:07] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.8 * r3fc967f6c7 10ffmpeg/libavcodec/h263dec.c: (log message trimmed)
[13:07] <CIA-119> ffmpeg: h263: more strictly forbid frame size changes with frame-mt.
[13:07] <CIA-119> ffmpeg: Prevents crashes because the old check was incomplete.
[13:07] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:07] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:07] <CIA-119> ffmpeg: (cherry picked from commit 2d22d4307dcc1461f39a2ffb9c8db6c6b23fd080)
[13:07] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:07] <CIA-119> ffmpeg: 03Alex Converse 07release/0.8 * r08c81f7365 10ffmpeg/libavcodec/celp_filters.c: 
[13:07] <CIA-119> ffmpeg: celp filters: Do not read earlier than the start of the 'out' vector.
[13:08] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:08] <CIA-119> ffmpeg: (cherry picked from commit 37ddd3833219fa7b913fff3f5cccc6878b047e6b)
[13:08] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:08] <CIA-119> ffmpeg: (cherry picked from commit 9ea94c44b1b414ab3bc6e9220ebb77621423ca38)
[13:08] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:08] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.8 * r3313f31f01 10ffmpeg/libavcodec/h263dec.c: (log message trimmed)
[13:08] <CIA-119> ffmpeg: h263dec: Disallow width/height changing with frame threads.
[13:08] <CIA-119> ffmpeg: Fixes CVE-2011-3937
[13:08] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:08] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:08] <CIA-119> ffmpeg: (cherry picked from commit 71db86d53b5c6872cea31bf714a1a38ec78feaba)
[13:08] <CIA-119> ffmpeg: Conflicts:
[13:08] <CIA-119> ffmpeg: 03Mans Rullgard 07release/0.8 * rc71c77e56f 10ffmpeg/libavcodec/vqavideo.c: (log message trimmed)
[13:08] <CIA-119> ffmpeg: vqavideo: return error if image size is not a multiple of block size
[13:08] <CIA-119> ffmpeg: The decoder assumes in various places that the image size
[13:08] <CIA-119> (76 lines omitted)
[13:08] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.8 * rb15e85d820 10ffmpeg/libavformat/rtpdec_asf.c: (log message trimmed)
[13:08] <CIA-119> ffmpeg: rtpdec_asf: Fix integer underflow that could allow remote code execution
[13:08] <CIA-119> ffmpeg: Fixes MSVR-11-0088
[13:09] <CIA-119> ffmpeg: change properly but this is still a good sanity check.
[13:09] <CIA-119> ffmpeg: Based on change by Michael Niedermayer.
[13:09] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.8 * r628b82294a 10ffmpeg/libavcodec/qdm2.c: (log message trimmed)
[13:09] <CIA-119> ffmpeg: qdm2: clip array indices returned by qdm2_get_vlc().
[13:09] <CIA-119> ffmpeg: Prevents subsequent overreads when these numbers are used as indices
[13:09] <CIA-119> ffmpeg: in arrays.
[13:09] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:09] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:09] <CIA-119> ffmpeg: Signed-off-by: Justin Ruggles <justin.ruggles at gmail.com>
[13:13] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.7 * r4169912f39 10ffmpeg/ (Doxyfile RELEASE VERSION): 
[13:13] <CIA-119> ffmpeg: Update for 0.8.11
[13:13] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:13] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.7 * r26ac878cc2 10ffmpeg/libavcodec/ (h264.c h264_ps.c): (log message trimmed)
[13:13] <CIA-119> ffmpeg: h264: additional protection against unsupported size/bitdepth changes.
[13:13] <CIA-119> ffmpeg: Fixes crashes in codepaths not covered by original checks.
[13:13] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:13] <CIA-119> ffmpeg: (cherry picked from commit 732f9fcfe54fc9a0a7bbce53fe86b38744c2d301)
[13:13] <CIA-119> ffmpeg: Conflicts:
[13:13] <CIA-119> ffmpeg:  libavcodec/h264.c
[13:13] <CIA-119> ffmpeg: 03Ronald S. Bultje 07release/0.7 * r3fc967f6c7 10ffmpeg/libavcodec/h263dec.c: (log message trimmed)
[13:13] <CIA-119> ffmpeg: h263: more strictly forbid frame size changes with frame-mt.
[13:13] <CIA-119> ffmpeg: Prevents crashes because the old check was incomplete.
[13:13] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:13] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:13] <CIA-119> ffmpeg: (cherry picked from commit 2d22d4307dcc1461f39a2ffb9c8db6c6b23fd080)
[13:14] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:14] <CIA-119> ffmpeg: 03Alex Converse 07release/0.7 * r50073e2395 10ffmpeg/libavcodec/motionpixels.c: (log message trimmed)
[13:14] <CIA-119> ffmpeg: motionpixels: Clip YUV values after applying a gradient.
[13:14] <CIA-119> ffmpeg: Prevents illegal reads on truncated and malformed input.
[13:14] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:14] <CIA-119> ffmpeg: (cherry picked from commit b5da848facd41169283d7bfe568b83bdfa7fc42e)
[13:14] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:14] <CIA-119> ffmpeg: (cherry picked from commit aaa6a666774eb02c351c84e80622a5c69e9b642e)
[13:14] <CIA-119> ffmpeg: 03Alex Converse 07release/0.7 * r08c81f7365 10ffmpeg/libavcodec/celp_filters.c: 
[13:14] <CIA-119> ffmpeg: celp filters: Do not read earlier than the start of the 'out' vector.
[13:14] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:14] <CIA-119> ffmpeg: (cherry picked from commit 37ddd3833219fa7b913fff3f5cccc6878b047e6b)
[13:14] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:14] <CIA-119> ffmpeg: (cherry picked from commit 9ea94c44b1b414ab3bc6e9220ebb77621423ca38)
[13:14] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:14] <CIA-119> ffmpeg: 03Mans Rullgard 07release/0.7 * rc71c77e56f 10ffmpeg/libavcodec/vqavideo.c: (log message trimmed)
[13:14] <CIA-119> (106 lines omitted)
[13:45] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.10 * r5836110018 10ffmpeg/libavcodec/h263dec.c: (log message trimmed)
[13:45] <CIA-119> ffmpeg: h263dec: Disallow width/height changing with frame threads.
[13:45] <CIA-119> ffmpeg: Fixes CVE-2011-3937
[13:45] <CIA-119> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[13:45] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:45] <CIA-119> ffmpeg: (cherry picked from commit 71db86d53b5c6872cea31bf714a1a38ec78feaba)
[13:45] <CIA-119> ffmpeg: Conflicts:
[13:45] <CIA-119> ffmpeg: 03Alexander Strange 07release/0.10 * r4713234518 10ffmpeg/libavcodec/h264_ps.c: 
[13:45] <CIA-119> ffmpeg: h264: Add check for invalid chroma_format_idc
[13:45] <CIA-119> ffmpeg: Fixes a crash when FF_DEBUG_PICT_INFO is used.
[13:45] <CIA-119> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[13:45] <CIA-119> ffmpeg: (cherry picked from commit 6ef4063957aa5025c8d2cd757b6a537e4b6874df)
[13:45] <CIA-119> ffmpeg: Fixes: CVE-2012-0851
[13:45] <CIA-119> ffmpeg: Signed-off-by: Reinhard Tartler <siretart at tauware.de>
[13:45] <CIA-119> ffmpeg: 03Michael Niedermayer 07release/0.10 * r5872580e65 10ffmpeg/libavcodec/eatqi.c: (log message trimmed)
[13:45] <CIA-119> ffmpeg: tqi: Pass errors from the MB decoder
[13:45] <CIA-119> ffmpeg: This silences some valgrind warnings.
[13:45] <CIA-119> ffmpeg: CC: libav-stable at libav.org
[13:45] <CIA-119> ffmpeg: Fixes second half of http://ffmpeg.org/trac/ffmpeg/ticket/794
[13:45] <CIA-119> ffmpeg: Bug found by: Oana Stratulat
[13:45] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:46] <CIA-119> ffmpeg: png: check bit depth for PAL8/Y400A pixel formats.
[13:46] <CIA-119> ffmpeg: Wrong bit depth can lead to invalid rowsize values, which crashes the
[13:46] <CIA-119> ffmpeg: decoder further down.
[13:46] <CIA-119> (18 lines omitted)
[13:47] <CIA-119> ffmpeg: 03Paul B Mahol 07master * rb0f6df28de 10ffmpeg/libavcodec/ (libopenjpegdec.c libopenjpegenc.c): 
[13:47] <CIA-119> ffmpeg: libopenjpeg: YUV410P and YUV411P support
[13:47] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[14:31] <ubitux> michaelni: i think the aformat should be at the end of the filtergraph
[14:32] <ubitux> right now we have [mapchan] [aformat] [async/disabled] [volume]
[14:32] <ubitux> i think we should have [mapchan] [async] [volume] [aformat]
[14:32] <ubitux> the [aformat] is build according to the final output options the user wants
[14:33] <ubitux> (wouldn't that fix the async issue?)
[14:36] <ubitux> if the user specifies -async and -ar, it will likely do: [async] [aformat], and then auto insert a aresample like [async] [aresample] [aformat] (just a guess)
[14:36] <ubitux> i'm not sure how well it will deal with the auto inserted aresample by async
[15:24] <michaelni> ubitux, the async issue was that there where 2 async filters inserted one by configure_input_audio_filter() and one by configure_output_audio_filter()
[15:24] <michaelni> but maybe i misunderstand you
[15:24] <ubitux> oh, ok
[15:25] <michaelni> not saying the order is corret here
[15:25] <michaelni> if you think its wrong, dont hesitate to change it
[15:25] <ubitux> should we move all the configure_output_audio_filter() auto inserted filter to configure_input_audio_filter()?
[15:25] <ubitux> (isn't the volume also inserted two times etc?)
[15:26] <michaelni> indeed i also just realized that vol is twice :)
[15:29] <michaelni> ubitux, feel free to remove one, or i can do it too if you are busy ?
[15:29] <ubitux> go ahead :p
[15:30] <ubitux> i wonder how this could/need to be factorized
[15:30] <ubitux> where does the auto inserted filters belong to?
[15:31] <ubitux> configure_output_audio_filter() or configure_input_audio_filter()?
[15:31] <ubitux> i guess the two are called with -filter_complex, and only one by default (or with -af)?
[15:35] <michaelni> well, one could allow them on both and send them to the right one depending on where on the command line tehy are specified but it seems overkill
[15:35] <michaelni> the user can manually insert a volume filter when the default doesnt do whast he wants
[15:35] <michaelni> question is input vs output both is kinda bad :)
[15:36] <michaelni> if you add -vol 1 you get 2 volume filters 
[15:37] <michaelni> currently
[15:37] <ubitux> which one do you plan to remove?
[15:38] <michaelni> ill follow the async case and disable the one closer to the disabled async case
[15:38] <michaelni> unless you have a preferrance ?
[15:39] <michaelni> for async my argument was that the timestamps should be cleaned up before the filters as the filters likely prefer clean timestamps
[15:41] <ubitux> no it's ok
[16:03] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r02ff89a1cd 10ffmpeg/ffmpeg.c: 
[16:03] <CIA-119> ffmpeg: ffmpeg: disable audio volume insertion on the output side.
[16:03] <CIA-119> ffmpeg: This fixes -vol leading to 2 inserted volume filters
[16:03] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:55] <cbsrobot> ubitux ping
[16:58] <ubitux> cbsrobot: pong
[16:58] <cbsrobot> nice timecode patch you posted
[16:58] <cbsrobot> one question
[16:58] <cbsrobot> there can be multiple timecode tracks in one file
[16:58] <cbsrobot> is there a reference to which track they belong ?
[16:59] <ubitux> yes
[16:59] <ubitux> but not all the time
[17:00] <ubitux> check the other patches
[17:00] <ubitux> especially the comment in patch 4
[17:00] <cbsrobot> ok
[17:01] <cbsrobot> and I was looking into this some time ago
[17:01] <cbsrobot> so I do not remember very well
[17:01] <cbsrobot> but is there a way to give the timecodetrack a name ?
[17:01] <ubitux> patch 5 handles the case where there is no ref
[17:01] <ubitux> a name? oO
[17:01] <ubitux> what do you mean?
[17:01] <cbsrobot> hmmm - let me dig a bit
[17:01] <ubitux> there is a way to name tracks? :)
[17:02] <cbsrobot> in quicktime I think you see the name of a track
[17:02] <cbsrobot> but let me dig out some samples ...
[17:03] <ubitux> it will likely be a bit more complicated to support such thing anyway
[17:06] <cbsrobot> ah - it's the kgtt tag
[17:07] <cbsrobot> when recording multiple timecode tracks you need to be able to distinguish them in your editing application
[17:07] <ubitux> nothing in the specs
[17:07] <ubitux> (at least the 2001 version)
[17:08] <cbsrobot> some normal values are f.ex: sourceTC, sourceAux1, sourceAux2 
[17:08] <cbsrobot> it might be avendor specific extension
[17:08] <ubitux> fuck that shit? :)
[17:08] <ubitux> do you have use case of this?
[17:09] <cbsrobot> see fcp_export8.mov from http://ffmpeg.org/trac/ffmpeg/ticket/236
[17:09] <cbsrobot> 3 timecode tracks
[17:09] <cbsrobot> do you have a mac available ?
[17:10] <cbsrobot> there is a tool called "Atom inspector" that lets you see and edit the mov atoms
[18:14] <ubitux> cbsrobot: wtf 3 timecode track but only one video stream? oO
[18:14] <cbsrobot> yeah sure
[18:14] <ubitux> what's the point?
[18:14] <cbsrobot> one tc track from the camera
[18:14] <ubitux> and no i have no mac :p
[18:14] <ubitux> never touched one in my whole life
[18:14] <cbsrobot> one timecode track from an external source
[18:15] <ubitux> you won't be able to remux more than one with ffmpeg anyway 
[18:15] <ubitux> ffprobe will give you the timecode, and you would just use -timecode to choose one
[18:16] <cbsrobot> I think it would be useful to add more tc tracks ...
[18:16] <ubitux> you can have one per video track
[18:17] <cbsrobot> need to go afk - sorry
[18:17] <ubitux> cya :)
[19:15] <CIA-119> ffmpeg: 03Clément BSsch 07master * rbae99f7651 10ffmpeg/ffprobe.c: ffprobe/flat: add escape for '`' and '$'.
[19:27] <mancha> I am trying to port the utvideo encoder wrapper to .0.10.x. found i had to edit the parameters of utvideo_decode_frame and utvideo_encode_frame for it to compile (given the change in AVCodec.
[19:28] <mancha> it now compiles but i wonder if i broke something :)  is it as easy as that? rather than have the pointer to AVFramer as a param now i define it in the function as it used to be: AVFrame *pic = avctx->coded_frame;
[19:28] <mancha> since avctx is passed (as pointer to AVCodecContetex)
[19:51] <michaelni> mancha, just look at how other codecs where changed
[19:52] <michaelni> but if it works (encoder & decoder) it likely is correct
[19:52] <michaelni> also if you want it included in the next 0.10 release then seubmit a patch!
[20:02] <mancha> michaelni: OK , i will do this. fortunately utvideodec.cpp does have a before and after but I think i need to ook at actual encoders.
[20:02] <mancha> look*
[20:04] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * r02c2a54bf6 10ffmpeg/libavcodec/jpeglsenc.c: Remove misplaced comment.
[20:04] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * r0a3e15690f 10ffmpeg/libavcodec/ljpegenc.c: 
[20:04] <CIA-119> ffmpeg: Allow colour space autodetection when encoding ljpeg.
[20:04] <CIA-119> ffmpeg: This works as expected if the input stream is not yuv4xxp.
[20:04] <CIA-119> ffmpeg: Reviewed-by: Paul B Mahol
[20:05] <mancha> these wrappers do tend to be very codec dependent thugh - but hopefully enough similairities that I can hack a patch up
[20:42] <CIA-119> ffmpeg: 03Carl Eugen Hoyos 07master * r6b87f0707d 10ffmpeg/libavcodec/dv.c: Do not try to encode DVCPRO HD, it produces broken files.
[21:43] <ubitux> michaelni: https://github.com/ubitux/FFmpeg/compare/master...scenedetect
[21:43] <ubitux> i'm going to push this, any objection?
[21:43] <ubitux> (i updated the doc, move to double, and changed a few minor things)
[21:44] <ubitux> ('going to add a Changelog entry too)
[21:57] <teratorn> ubitux: huh, neat. I was just looking at doing some scnene-change detection stuff... is this going to be a new feature of ffmpeg, or are there existing functionality?
[21:58] <ubitux> new feature
[21:58] <teratorn> coolness
[21:58] <ubitux> at least as a user option
[21:58] <ubitux> teratorn: http://imgur.com/a/icpfk
[21:58] <ubitux> all generated with the command given in the documentation
[21:59] <ubitux> it looks like it is a fairly wanted feature
[21:59] <teratorn> we already have our own thing in our product... it's a pretty crummy algorithm though
[22:00] <teratorn> it's on the todo list to refactor/rewrite it :)
[22:00] <ubitux> may i ask who is "our"? :)
[22:00] <ubitux> and for what this scene detection stuff is used?
[22:00] <teratorn> http://catturavideo.com
[22:00] <teratorn> mostly we capture lectures. so powerpoint slide change is the usual use-case :)
[22:00] <ubitux> catsvideo @_@
[22:01] <ubitux> ok
[22:01] <teratorn> cats?
[22:01] <ubitux> i read "cat" in your url
[22:01] <teratorn> oh
[22:01] <ubitux> sorry, bad joke :)
[22:03] <teratorn> could you describe your algorithm ? :)
[22:03] <teratorn> afaict its only looking at luma values?
[22:03] <ubitux> rgb
[22:03] <teratorn> packed rgb?
[22:04] <gnafu> ubitux: What video is that in the second set of thumbnails?
[22:04] <ubitux> teratorn: sum of the diff between two frames (SAD), average it (MAFD), diff = previous MAFD - current MAFD; then check if the diff or mafd are above a certain threshold
[22:05] <ubitux> gnafu: https://www.youtube.com/watch?v=6QFwo57WKwg
[22:05] <ubitux> gnafu: awesome video, don't miss a single second :)
[22:05] <ubitux> (HD!)
[22:06] <teratorn> ubitux: may i ask where that came from? your brain, research paper, etc?
[22:06] <ubitux> shotdetect algorithm
[22:06] <ubitux> slightly changed
[22:07] <ubitux> i tried to implement a better one (the dynamic threshold one) but it didn't give good results
[22:07] <ubitux> maybe i misunderstood sth
[22:07] <ubitux> but this gave results good enough to me
[22:07] <ubitux> and it can be improved later anyway
[22:07] <ubitux> the thing is, the paper is saying "hey one threshold sucks, we're going to compute it dynamically!"
[22:07] <ubitux> so far so good
[22:07] <ubitux> but they introduce 4 constants instead
[22:08] <ubitux> that you have to tweak&
[22:08] <ubitux> so& :)
[22:08] <teratorn> I always thought that taking the image in 2x2 squares of averaged pixels might lead to better results.... depending on compression level individual pixels might vary a bit from frame to frame (even if the source image is unchanging)
[22:08] <gnafu> ubitux: Thanks.  It seemed interesting.  I have downloaded it for later viewing :-).
[22:08] <ubitux> gnafu: no, watch it now or i'll kill a kitten.
[22:09] <gnafu> But I'm at work, and it's preferred not to watch YouTube videos :-o.  Plus, I don't have headphones.
[22:09] <ubitux> ok, granted then.
[22:09] <ubitux> you need the music
[22:09] <ubitux> teratorn: what do you mean by 2x2 SAD?
[22:09] <gnafu> And for an additional reason to wait 'til home: my monitor at home is much nicer and full 1080p./
[22:10] <ubitux> computing SAD by mb or linearly should give the same results :p
[22:10] <teratorn> mmmh.
[22:10] <ubitux> gnafu: i pardon you for this time then
[22:10] <teratorn> with that algo. I guess it would make no difference
[22:10] <ubitux> teratorn: what's interesting by computing sad by MB is that you can deal with error
[22:11] <teratorn> MB?
[22:11] <teratorn> macroblock?
[22:11] <ubitux> for instance someone proposed to detect too bright or too dark MB
[22:11] <ubitux> yes
[22:11] <ubitux> so if in the middle of your frame you have a MB with a too big SAD
[22:11] <ubitux> you can just put a -20% on it
[22:11] <ubitux> (it's likely the MB was badly decoded or sth like this)
[22:12] <teratorn> ah
[22:12] <teratorn> so some sanity filtering ;)
[22:12] <ubitux> at least that's what the document was proposing
[22:12] <ubitux> with a few others improvement
[22:12] <ubitux> but well, whatever :)
[22:12] <teratorn> which doc?
[22:12] <ubitux> just a sec
[22:13] <ubitux> teratorn: http://www.hindawi.com/journals/ijdmb/2010/864123/
[22:13] <ubitux> which actually takes this solution: publik.tuwien.ac.at/files/pub-et_9739.pdf 
[22:13] <ubitux> (the 4 constants thing i was talking about)
[22:14] <ubitux> and then propose 3-4 improvments
[22:14] <ubitux> such as error handling, tweaking the constants, etc
[22:15] <teratorn> the only sane way for us to tweak such an algorithm, I think, is to get together like 10 or 20 sample videos that are representative of our typical input. manually construct scene-change files. then compare against results of algorithm and tweak
[22:16] <teratorn> it sound scarier than tweaking presets :)
[22:21] <michaelni> ubitux, patch should be ok but arent you missing a dependancy on avcodec for configure ?
[22:21] <ubitux> +select_filter_deps="avcodec" ?
[22:21] <ubitux> that's not enough?
[22:22] <ubitux> ah maybe a dsputil.o in the Makefile
[22:22] <ubitux> ?
[22:22] <ubitux> (isn't it a builtin anyway?)
[22:24] <michaelni> ohh i missed it, must have forgotten once i read the patch to its end :)
[22:24] <michaelni> so yes LGTM
[22:24] <ubitux> ok great :)
[22:24] <ubitux> thx
[22:26] <CIA-119> ffmpeg: 03Clément BSsch 07master * r72868144e5 10ffmpeg/ (Changelog configure doc/filters.texi libavfilter/vf_select.c): 
[22:26] <CIA-119> ffmpeg: lavfi/select: add scene detection.
[22:26] <CIA-119> ffmpeg: Based on the shotdetect algorithm (http://shotdetect.nonutc.fr/ by Johan
[22:26] <CIA-119> ffmpeg: MATHE johan.mathe tremplin-utc.net).
[22:26] <CIA-119> ffmpeg: 03Clément BSsch 07master * r5e8225004f 10ffmpeg/Changelog: Changelog: add ffprobe flat output.
[22:29] <pasteeater> ubitux: interesting
[22:31] <ubitux> feel free to propose any better tweaking
[22:34] <ubitux> michaelni: what is the point of check_recording_time() and why is it disabled?
[22:36] <CIA-119> ffmpeg: 03Clément BSsch 07master * rdf03ae8dd8 10ffmpeg/ffmpeg.c: ffmpeg: remove two unused variables.
[22:42] <RobertNagy> is there anyway to avoid pooling of buffer refs in lavfi?
[22:48] <michaelni> the buffer pool isnt optional currently, why do you ask?
[22:50] <michaelni> ubitux, dont remember exactly why its disabled, it must have caused a bug ...
[22:50] <michaelni> have you found a case where it does something positive ?
[23:05] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r8eb9bf0933 10ffmpeg/libavfilter/ (avfiltergraph.c internal.h): 
[23:05] <CIA-119> ffmpeg: avfiltergraph: make some functions static.
[23:05] <CIA-119> ffmpeg: They are not used outside of avfiltergraph.c
[23:05] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r2f51ec2b94 10ffmpeg/avconv.c: avconv: split checking for active outputs out of transcode().
[23:05] <CIA-119> ffmpeg: 03Anton Khirnov 07master * ra508e7a1ff 10ffmpeg/avconv.c: avconv: split selecting input file out of transcode().
[23:05] <CIA-119> ffmpeg: 03Anton Khirnov 07master * rbb7431f4fc 10ffmpeg/avconv.c: 
[23:05] <CIA-119> ffmpeg: avconv: check output stream recording time before each frame returned from filters
[23:05] <CIA-119> ffmpeg: There may be multiple frames returned, so with just one check we can
[23:05] <CIA-119> ffmpeg: write more than requested to the output.
[23:05] <CIA-119> ffmpeg: 03Anton Khirnov 07master * ra982e5a031 10ffmpeg/libavformat/avidec.c: 
[23:05] <CIA-119> ffmpeg: avidec: make scale and rate unsigned.
[23:05] <CIA-119> ffmpeg: The specs say they are unsigned 32bit integers.
[23:05] <CIA-119> ffmpeg: 03Alex Converse 07master * r41e9682af2 10ffmpeg/ (7 files in 3 dirs): movenc: Write chan atom for all audio tracks in mov mode movies.
[23:05] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r944d049eaa 10ffmpeg/: (log message trimmed)
[23:05] <CIA-119> ffmpeg: Merge remote-tracking branch 'qatar/master'
[23:05] <CIA-119> ffmpeg: * qatar/master:
[23:05] <CIA-119> ffmpeg:  movenc: Write chan atom for all audio tracks in mov mode movies.
[23:05] <CIA-119> ffmpeg:  mpegtsenc: use avio_open_dyn_buf(), zero pointers after freeing
[23:05] <CIA-119> ffmpeg:  doc/avconv: add some details about the transcoding process.
[23:05] <CIA-119> ffmpeg:  avidec: make scale and rate unsigned.
[23:06] <CIA-119> ffmpeg: 03JindYich Makovi
ka 07master * r84e430dd7b 10ffmpeg/libavformat/mpegtsenc.c: 
[23:06] <CIA-119> ffmpeg: mpegtsenc: use avio_open_dyn_buf(), zero pointers after freeing
[23:06] <CIA-119> ffmpeg: Per suggestion by Michael Niedermayer.
[23:06] <CIA-119> ffmpeg: Signed-off-by: JindYich Makovi
ka <makovick at gmail.com>
[23:06] <CIA-119> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[23:06] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r2b1f105f1b 10ffmpeg/doc/avconv.texi: doc/avconv: add some details about the transcoding process.
[23:06] <ubitux> <@michaelni> have you found a case where it does something positive ? // no i just saw a warning with the function being unused and i was curious what this was good for
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * r03095d73a3 10ffmpeg/ffplay.c: 
[23:39] <CIA-119> ffmpeg: ffplay: put audio_open into a seperate function
[23:39] <CIA-119> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * re1248f5c52 10ffmpeg/ffplay.c: 
[23:39] <CIA-119> ffmpeg: ffplay: put audio parameters to their own struct
[23:39] <CIA-119> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * r7342037973 10ffmpeg/ffplay.c: 
[23:39] <CIA-119> ffmpeg: ffplay: remove VideoState from audio_open
[23:39] <CIA-119> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * r7821264c26 10ffmpeg/ffplay.c: 
[23:39] <CIA-119> ffmpeg: ffplay: simplify audio_open, rename parameters to more explanatory names
[23:39] <CIA-119> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * r0dad529201 10ffmpeg/ffplay.c: 
[23:39] <CIA-119> ffmpeg: ffplay: try more channel count combinations for SDL_OpenAudio
[23:39] <CIA-119> ffmpeg: This should fix ticket 1384.
[23:39] <CIA-119> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[23:39] <CIA-119> ffmpeg: 03Marton Balint 07master * r22505c188c 10ffmpeg/ffplay.c: 
[00:00] --- Tue Jun  5 2012