[Ffmpeg-devel-irc] ffmpeg-devel.log.20120714

[burek]

[01:03] <CIA-41> ffmpeg: 03Paul B Mahol 07master * rcd3136e84a 10ffmpeg/libavcodec/exr.c: 
[01:03] <CIA-41> ffmpeg: exr: print channel names not yet supported
[01:03] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[01:08] <durandal_1707> michaelni: what is supposed to call sws_getCoefficients() ?
[01:12] <durandal_1707> exr have chromaticites attribute which define how colors should be interpreted
[01:12] <durandal_1707> so RGB channels becomes XYZ ones
[01:36] <michaelni> durandal_1707, the user application using sws could call it
[01:38] <durandal_1707> michaelni: so i think i know why -colorspace does nothing
[01:38] <durandal_1707> michaelni: given XYZ and ITU.709 values in float could you tell what int32 value should be in that table?
[01:39] <durandal_1707> ff_yuv2rgb_coeffs table
[01:42] <durandal_1707> hmm, numbers are given in CIE x,y coordinates
[01:55] <CIA-41> ffmpeg: 03Paul B Mahol 07master * rbc0ef29cf0 10ffmpeg/libavcodec/exr.c: 
[01:55] <CIA-41> ffmpeg: exr: fix scan_line_size calculation
[01:55] <CIA-41> ffmpeg: This one will also work in cases when file have channels
[01:55] <CIA-41> ffmpeg: we are not going to use.
[01:55] <CIA-41> ffmpeg: Only compressed files where affected by this bug.
[01:55] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[02:58] <CIA-41> ffmpeg: 03Stefan Gehrer 07master * re98330df90 10ffmpeg/libavformat/movenc.c: 
[02:58] <CIA-41> ffmpeg: movenc: write track header transformation matrix depending on "rotate" metadata
[02:58] <CIA-41> ffmpeg: should fix ticket #505
[02:58] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:43] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r2fe186409b 10ffmpeg/libavcodec/ (internal.h utils.c): 
[04:43] <CIA-41> ffmpeg: lavc: skip initial silence when requested
[04:43] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:43] <CIA-41> ffmpeg: 03jamal 07master * r0f03563d7e 10ffmpeg/libavcodec/w32pthreads.h: 
[04:43] <CIA-41> ffmpeg: lavc/w32pthreads: Fix make checkheaders.
[04:43] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:43] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r744bd8d08c 10ffmpeg/ffmpeg.c: 
[04:43] <CIA-41> ffmpeg: ffmpeg: set pkt_timebase
[04:43] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:43] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r825e1e2437 10ffmpeg/ffmpeg.c: 
[04:43] <CIA-41> ffmpeg: ffmpeg: use pkt_pts before pkt->pts for audio
[04:43] <CIA-41> ffmpeg: This fixes the timestamp of the first packet when samples have been skiped.
[04:43] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:43] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * raa32971d2b 10ffmpeg/libavcodec/utils.c: 
[04:43] <CIA-41> ffmpeg: lavc: adjust timestamp in case of skiping initial samples
[04:43] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[09:53] <ubitux> mmh strange the make checkheaders didn't seem to have spotted the problem
[11:03] <saste> michaelni: are you ok with "[PATCH] lavc: add metadata to AVFrame"?
[11:42] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r5a6c7a218c 10ffmpeg/doc/ (Makefile filters.texi libavfilter.texi): 
[11:42] <CIA-41> ffmpeg: doc: merge libavfilter.texi into filters.texi
[11:42] <CIA-41> ffmpeg: There is no much point into splitting documentation for filters. Also
[11:42] <CIA-41> ffmpeg: having a separate document for a library is inconsistent.
[11:42] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * rb65ccce866 10ffmpeg/doc/filters.texi: 
[11:42] <CIA-41> ffmpeg: doc/filters: update documentation
[11:42] <CIA-41> ffmpeg: In particular, assume existence of audio and sink filters.
[11:42] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * rbd58f00ec1 10ffmpeg/doc/filters.texi: doc/filters: prefer "special" over "magic" for special input/output labels
[12:00] <CIA-41> ffmpeg: 03jamal 07master * r3e03f8d8b2 10ffmpeg/ (configure doc/Makefile): build: Fix documentation dependencies and rules
[12:00] <CIA-41> ffmpeg: 03jamal 07master * r2090e40410 10ffmpeg/configure: 
[12:00] <CIA-41> ffmpeg: configure: Check that texi2html supports --init-file
[12:00] <CIA-41> ffmpeg: Fixes ticket #1515
[12:00] <CIA-41> ffmpeg: 03jamal 07master * rd48f014ff1 10ffmpeg/doc/Makefile: doc/Makefile: Add missing $(HOSTEXESUF) to print_options
[15:12] <burek> saste, sorry for pinging you all the time, just one more question please :) what is the diff between "ffmpeg -i ... - > temp1.a" and "ffmpeg -i ... temp1.a"
[15:13] <saste> burek: command > out.lst
[15:13] <saste> will send the stdout output to out.lst
[15:13] <burek> yes
[15:14] <saste> in case of ffmpeg there is not much which is sent to stdout, unless you use the pipe output
[15:14] <burek> but what is the diff between sending the output to stdout and piping it to file instead of just specifying the output file
[15:15] <burek> more precisely, why hasn't this: ffmpeg -i input1.flv -vn -f u16le -acodec pcm_s16le -ac 2 -ar 44100 - > temp1.a < /dev/null &
[15:15] <burek> been written as this: ffmpeg -i input1.flv -vn -f u16le -acodec pcm_s16le -ac 2 -ar 44100 temp1.a
[15:15] <saste> that's equivalent
[15:15] <burek> oh, ok :)
[15:15] <saste> because in the first command you're telling ffmpeg to write to stdout (-)
[15:15] <saste> check the pipe protocol documentation
[15:15] <burek> ok, thanks :)
[15:16] <saste> but usually will use stdout only for textual messages
[15:16] <saste> e.g. ffmpeg -help | grep ...
[15:17] <burek> yes, I know ffmpeg uses stderr mostly, to free up stdout for these kind of piping things, but I wasn't sure if there is actually any difference between specifying the output file name directly
[15:40] <CIA-41> ffmpeg: 03Nicolas George 07master * r7b521c52c4 10ffmpeg/ (Changelog doc/ffmpeg.texi ffmpeg.c): 
[15:40] <CIA-41> ffmpeg: ffmpeg: add -progress option.
[15:40] <CIA-41> ffmpeg: It allows to send program-friendly (easy to parse)
[15:40] <CIA-41> ffmpeg: progress information to a file or an URL.
[17:44] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r57d5a224ce 10ffmpeg/ (3 files in 2 dirs): 
[17:44] <CIA-41> ffmpeg: mp3enc: add lame tag with start padding info
[17:44] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[17:54] <burek> I wrote a general usage script to concat multiple input files and produce an output out of it, based on this faq item: http://ffmpeg.org/faq.html#How-can-I-join-video-files_003f
[17:54] <burek> this is the script, I'll test it later some more and submit it then somewhere (wiki I guess): http://ffmpeg.gusari.org/mcat
[23:50] <CIA-41> ffmpeg: 03Philip Langdale 07master * r2c501ae82b 10ffmpeg/ (8 files in 5 dirs): (log message trimmed)
[23:50] <CIA-41> ffmpeg: movtextdec: 3GPP TS 26.245 Timed Text Decoder.
[23:50] <CIA-41> ffmpeg: This change introduces a basic decoder for 3GPP Timed Text subtitles,
[23:50] <CIA-41> ffmpeg: also known as TX3G, Quicktime subtitles, or "movtext" in the existing
[23:50] <CIA-41> ffmpeg: code.
[23:50] <CIA-41> ffmpeg: This initial change doesn't attempt to parse styling information,
[23:50] <CIA-41> ffmpeg: and just reads the plain text of the subtitles. I intend to add
[00:00] --- Sun Jul 15 2012