[Ffmpeg-devel-irc] ffmpeg-devel.log.20120123

[burek]

[01:01] <durandal_1707> there is no way to disable writing of ID3 tags?
[01:08] <michaelni> AFAIK, no
[01:08] <michaelni> but it shouldnt be hard to add one 
[01:08] <bcoudu> duh
[01:10] <durandal_1707> there is no way to disable writing of lavf tag?
[01:18] <CIA-31> ffmpeg: 03Dmitry Volyntsev 07master * rbc495bad3d 10ffmpeg/libavformat/rtsp.c: 
[01:18] <CIA-31> ffmpeg: rtsp: Remove a leftover, currently pointless check
[01:18] <CIA-31> ffmpeg: This check isn't relevant in the way the code currently works.
[01:18] <CIA-31> ffmpeg: Also change a case of if (x == 0) into if (!x).
[01:18] <CIA-31> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[01:18] <CIA-31> ffmpeg: 03Martin Storsjö 07master * rdbb06b8c0d 10ffmpeg/libavformat/ (rtsp.c rtsp.h): 
[01:18] <CIA-31> ffmpeg: rtsp: Allow specifying the UDP port range via AVOptions
[01:18] <CIA-31> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[01:18] <CIA-31> ffmpeg: 03Dmitry Volyntsev 07master * r58f0978581 10ffmpeg/libavformat/rtsp.c: (log message trimmed)
[01:18] <CIA-31> ffmpeg: rtsp: Use a random offset for trying to open UDP ports for RTP
[01:18] <CIA-31> ffmpeg: This avoids (for all practical cases) the issue of reusing
[01:18] <CIA-31> ffmpeg: the same UDP port as for an earlier connection. If the remote
[01:18] <CIA-31> ffmpeg: doesn't know the previous session was closed, he might keep
[01:18] <CIA-31> ffmpeg: on sending packets to that port. If we always start off trying
[01:18] <CIA-31> ffmpeg: to open the same UDP port, we might get those packets intermixed
[01:18] <CIA-31> ffmpeg: 03Tomas Härdin 07master * rd32b3f40c7 10ffmpeg/libavformat/mxfdec.c: mxfdec: parse MXF partitions
[01:18] <CIA-31> ffmpeg: 03Philip de Nier 07master * r0662eea6b0 10ffmpeg/libavformat/mxfdec.c: 
[01:18] <CIA-31> ffmpeg: mxfdec: skip to end of structural sets
[01:18] <CIA-31> ffmpeg: This fixes reading of partition packs. The code stops reading after the
[01:18] <CIA-31> ffmpeg: operational pattern and should skip the array of essence container
[01:18] <CIA-31> ffmpeg: labels that follow.
[01:18] <CIA-31> ffmpeg: Signed-off-by: Janne Grunau <janne-libav at jannau.net>
[01:18] <CIA-31> ffmpeg: 03Joakim Plate 07master * r3359246d9a 10ffmpeg/libavformat/mxfdec.c: 
[01:18] <CIA-31> ffmpeg: mxfdec: check return value of avio_seek
[01:18] <CIA-31> ffmpeg: Avoid modifying state if avio_seek fails.
[01:18] <CIA-31> ffmpeg: Signed-off-by: Janne Grunau <janne-libav at jannau.net>
[01:18] <CIA-31> ffmpeg: 03Tomas Härdin 07master * r83ab10f6e9 10ffmpeg/libavformat/mxfdec.c: 
[01:19] <CIA-31> ffmpeg: mxfdec: Move FooterPartition to MXFContext and make sure it is never zero.
[01:19] <CIA-31> ffmpeg: Signed-off-by: Janne Grunau <janne-libav at jannau.net>
[01:19] <CIA-31> ffmpeg: 03Janne Grunau 07master * r682b6db706 10ffmpeg/libavformat/mxfdec.c: 
[01:19] <CIA-31> ffmpeg: mxfdec: Parse IndexTableSegments and convert them into AVIndexEntry arrays.
[01:19] <CIA-31> (77 lines omitted)
[01:51] <CIA-31> ffmpeg: 03Michael Niedermayer 07master * rb9db728a1f 10ffmpeg/libavformat/libavformat.v: 
[01:51] <CIA-31> ffmpeg: libavformat/libavformat.v: fix application name in comment
[01:51] <CIA-31> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[07:31] <ubitux> does anyone know how to run a build only fate?
[09:24] <Tjoppen> michaelni: mail received
[13:46] <ubitux> http://www.raspberrypi.org/archives/571
[14:20] <av500> I wonder if they got Hollywood approval to show these clips
[17:13] <durandal_1707> trac3 and AC-3 decoders, QTRLE encoder, NUT and Matroska muxers. 
[17:13] <durandal_1707> July 14, 2007
[17:13] <durandal_1707> crap
[17:14] <durandal_1707> FF_API_READ_SEEK is no more deprecated so /home/fate/src/libavformat/sbgdec.c:1477:5: warning: 'FF_API_READ_SEEK' is not defined, evaluates to 0 [-Wundef] could be resolved
[17:18] <durandal_1707> ubitux: you mean in same patch?
[17:19] <ubitux> you are Paul?
[17:19] <ubitux> if so, yes i meant in the same patch, changing the message of course
[17:19] <ubitux> "replace remaining use of deprecated [...]"
[17:20] <ubitux> with "lavf:" prefix
[17:23] <CIA-31> ffmpeg: 03Paul B Mahol 07master * ra6ef7b3a77 10ffmpeg/libavformat/cafenc.c: cafenc: replace deprecated av_dbl2int()
[17:23] <CIA-31> ffmpeg: 03Paul B Mahol 07master * rc7579ad8e8 10ffmpeg/libavformat/cafdec.c: cafdec: replace deprecated get_strz()
[17:26] <ubitux> haha too late :)
[17:26] <durandal_1707> why my signed-of-by have been removed? And one of actual commiter is not present?
[17:27] <ubitux> ask carl
[17:27] <durandal_1707> yea found how is responsible
[17:27] <durandal_1707> s/how/who
[17:45] <ubitux> michaelni: do you want me to wait for 0.10 to push the af pan channel mapping and af_aresample move to libswr?
[18:13] <michaelni> ubitux, if your code works & is tested no need to wait
[19:06] <ubitux> michaelni: ok&
[19:06] <ubitux> durandal_1707: thank you for taking care of the others get_strz
[19:50] <CIA-31> ffmpeg: 03Michael Niedermayer 07master * r575d494de5 10ffmpeg/libavcodec/pthread.c: 
[19:50] <CIA-31> ffmpeg: pthreads: reset got_frames on flush.
[19:50] <CIA-31> ffmpeg: This fixes memory corruption when seeking in broken streams.
[19:50] <CIA-31> ffmpeg: a random mpeg4 in nut file was used to debug.
[19:50] <CIA-31> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:50] <CIA-31> ffmpeg: 03Michael Niedermayer 07master * rcf7c7f13cd 10ffmpeg/libavcodec/pthread.c: 
[19:50] <CIA-31> ffmpeg: pthreads: Generic progress lubrication support.
[19:50] <CIA-31> ffmpeg: Fixes bug118, bug120 and bug125 at least
[19:50] <CIA-31> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:50] <CIA-31> ffmpeg: 03David Mitchell 07master * rd150a147da 10ffmpeg/ (Changelog libavcodec/pgssubdec.c): (log message trimmed)
[19:50] <CIA-31> ffmpeg: Improve support for PGS subtitles.
[19:50] <CIA-31> ffmpeg: The previous implementation assumed that a new picture would always
[19:50] <CIA-31> ffmpeg: supersede the previous picture. Similarly, presentation segments
[19:50] <CIA-31> ffmpeg: were assumed to pertain to the most-recently-read picture.
[21:34] <ubitux> michaelni: what's going to happen with avconv/ffmpeg?
[21:35] <ubitux> is there some stuff in avconv not present in ffmpeg?
[21:37] <michaelni> ffmpeg supports all things avconv supports
[21:40] <mcbaine1> http://goo.gl/nATCX#
[21:40] <ubitux> michaelni: i remember you were asking a "long" while ago if anyone was willing to merge a chunk from avconv into ffmpeg
[21:41] <ubitux> do you remember this?
[21:41] <michaelni> no
[21:41] <ubitux> mmh ok i'll check
[21:41] <michaelni> wait i remember 
[21:41] <michaelni> it was some of our code i think 
[21:41] <michaelni> that i droped to merge some update
[21:41] <ubitux> oh
[21:41] <ubitux> ok
[21:42] <ubitux> so basically we could just drop avconv?
[21:42] <michaelni> yes
[21:42] <ubitux> ok
[23:44] <ubitux> anyone wants to comment on the yadif openmp patch?
[23:45] <ubitux> i wonder if we could make use of it in several places; though, with threading enabled, i wonder if it won't be a problem
[23:54] <michaelni> ubitux, i thought of just applying it as its so simple :)
[23:54] <ubitux> yes but don't you think it will actually slow down things if threading is enabled?
[23:57] <michaelni> someone should benchmark it
[00:00] --- Tue Jan 24 2012