[Ffmpeg-devel-irc] ffmpeg-devel.log.20120629

burek burek021 at gmail.com
Sat Jun 30 02:05:03 CEST 2012


[00:01] <CIA-41> ffmpeg: 03Alex Converse 07master * ra112822597 10ffmpeg/ (5 files in 2 dirs): movenc: Add channel layouts for PCM.
[00:01] <CIA-41> ffmpeg: 03Luca Barbato 07master * r1cb34ea4fe 10ffmpeg/libavformat/flvenc.c: flvenc: K&R formatting cosmetics
[00:01] <CIA-41> ffmpeg: 03Diego Biurrun 07master * r433492ac65 10ffmpeg/doc/git-howto.texi: doc: git: Add checklist with test steps to perform before pushing
[00:01] <CIA-41> ffmpeg: 03Damien Fetis 07master * rb92c7ee662 10ffmpeg/libavformat/ (flv.h flvdec.c flvenc.c): 
[00:01] <CIA-41> ffmpeg: flv: add support for G.711
[00:01] <CIA-41> ffmpeg: Signed-off-by: Luca Barbato <lu_zero at gentoo.org>
[00:01] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r4453f6b861 10ffmpeg/: (log message trimmed)
[00:01] <CIA-41> ffmpeg: Merge remote-tracking branch 'qatar/master'
[00:01] <CIA-41> ffmpeg: * qatar/master:
[00:01] <CIA-41> ffmpeg:  flv: add support for G.711
[00:01] <CIA-41> ffmpeg:  doc: git: Add checklist with test steps to perform before pushing
[00:01] <CIA-41> ffmpeg:  flvenc: K&R formatting cosmetics
[00:01] <CIA-41> ffmpeg:  movenc: Add channel layouts for PCM.
[00:02] Action: Nickname123 is no longer away : Gone for 19 hours 26 minutes 20 seconds
[01:22] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r4e82bdea60 10ffmpeg/doc/git-howto.texi: 
[01:22] <CIA-41> ffmpeg: git-howto: partial rewrite of the push checklist to make it match sanity & reality.
[01:22] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:22] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r7c8b5d1d80 10ffmpeg/Makefile: 
[01:22] <CIA-41> ffmpeg: Makefile: remove checkheaders from the main check target
[01:22] <CIA-41> ffmpeg: checkheaders doesnt pass and noone has even noticed since a very
[01:22] <CIA-41> ffmpeg: long time.
[01:22] <CIA-41> ffmpeg: checkheaders is also unmaintained (please add yourself to MAINTAINERS
[01:22] <CIA-41> ffmpeg: if you want to maintain it)
[01:22] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:22] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rd007f963c2 10ffmpeg/doc/git-howto.texi: 
[01:22] <CIA-41> ffmpeg: git-howto: cleanup, remove unreasonable recommendition.
[01:22] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:22] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r8bff1d7cb0 10ffmpeg/doc/git-howto.texi: 
[01:22] <CIA-41> ffmpeg: git-howto: remove inconvenient and odd 24h limit on rsync
[01:22] <CIA-41> ffmpeg: This was never true for FFmpeg in this form.
[01:22] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:22] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rc103dc0b48 10ffmpeg/doc/git-howto.texi: 
[01:22] <CIA-41> ffmpeg: git-howto: replace confusing and incorrect text about he testsuite by mostly a correct text.
[01:22] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:24] <durandal_1707>  "... especially these days where a lot of things are in dubious state due to the forks issues."
[01:34] <durandal_1707> heh, did anyone tried to fix those valgrind fate errors?, it seems to be in lavfi...
[01:39] <michaelni> i ignored them as i thought they would disappear after some patches from nicolas
[01:40] <michaelni> ill mail nicolas and check if he has something that will fix it or is working on it before ill look 
[02:11] <Zeranoe> I'm wondering if anyone can recommend a good guide for creating git patches? The guide at http://www.mplayerhq.hu/DOCS/tech/patches.txt doesn't really cover using git branches and such. (I'm looking to submit a few patches for FFmpeg and I want to get the right formating).
[02:13] <durandal_1707> Zeranoe: because that is for mplayer and mplayer do not use git but svn
[02:13] <Zeranoe> durandal_1707: Right, but it is still recommend to read here: http://ffmpeg.org/contact.html
[02:14] <Zeranoe> "have a look at the MPlayer patch guidelines most of what is written there applies to FFmpeg as well"
[02:14] <durandal_1707> that is from days when FFmpeg was using svn
[02:14] <llogan> the instrucitons for VLC are somewhat helpful: http://wiki.videolan.org/Git
[02:14] <Zeranoe> Which is why I'm looking for another guide
[02:15] <Zeranoe> llogan: Thank you
[02:15] <j-b> http://wiki.videolan.org/Git#Submitting_patches_to_the_vlc-devel_or_x264-devel
[02:15] <durandal_1707> http://ffmpeg.org/developer.html#Developers-Guide
[02:16] <durandal_1707> doc/git-howto.txt
[02:16] <llogan> Zeranoe: did you get my message a week or two back about libutvideo crashing with your build?
[02:17] <llogan> i haven't tested it since
[02:18] <Zeranoe> llogan: I don't believe I did... Did you file a bug report?
[02:18] <llogan> no, because it worksforme in linux
[02:19] <Zeranoe> What was the error?
[02:19] Action: llogan fires up VM
[02:20] <Zeranoe> I haven't personally used utvideo yet
[02:28] <llogan> Zeranoe: no ffmpeg error. just a windows dialog pops up "ffmpeg.exe has stopped working".
[02:28] <llogan> ffmpeg -i input -an -c:v libutvideo out.mkv
[02:29] <llogan> using latest 64-bit static
[02:30] <llogan> 10.2.4 is from january and i only tested with 11.1.0 on linux.
[02:33] <Zeranoe> llogan: Thanks for the report, I'll dig into it soon. I'm working through a few things before I'll be able to get to it though.
[02:33] <Zeranoe> utvideo seemed to compile fine
[04:00] <CIA-41> ffmpeg: 03Paul B Mahol 07master * r8fbe11e623 10ffmpeg/ (9 files in 2 dirs): 
[04:00] <CIA-41> ffmpeg: Replace Libav with FFmpeg in license headers for files created by me
[04:00] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[07:57] <rik316> elo
[11:37] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r84d73e9d5d 10ffmpeg/libavcodec/dca.c: 
[11:37] <CIA-41> ffmpeg: dca: fix project reference in table name
[11:37] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[11:37] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rc496224374 10ffmpeg/libavutil/pixdesc.h: 
[11:37] <CIA-41> ffmpeg: pixdesc: fix project reference in comment
[11:37] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[11:37] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rbe4ffb28b3 10ffmpeg/libavfilter/vf_ass.c: 
[11:37] <CIA-41> ffmpeg: vf_ass: fix table name to refer to correct lib.
[11:37] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[11:37] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rc83ed18d45 10ffmpeg/libavfilter/formats.h: 
[11:37] <CIA-41> ffmpeg: formats.h: fix project reference in comment
[11:37] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[11:37] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r491846e4be 10ffmpeg/libavutil/avutil.h: 
[11:37] <CIA-41> ffmpeg: avutil: fix project name reference in doxy section
[11:37] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:52] <saste> how it is possible to distinguish packed/planar format in a PCM file?
[12:52] <saste> assuming that the codec used is the same for both "modes" -> av_get_pcm_codec
[12:53] <saste> this is causing a problem to lavfi+aevalsrc, aevalsrc issues planar output but the device thinks it is packed
[12:53] <saste> and fails
[12:56] <saste> or said in other words, why only a few PCM planar variants do exist?
[12:56] <saste> michaelni: ^^
[13:00] <jonsen_> is any magic required to stream .mov/h264 via .asf to windows media player?
[13:00] <jonsen_> can't get it working with ffserver
[13:01] <jonsen_> opening with browser takes that long as the stream isn't finished (header issue?)
[13:01] <jonsen_> opening directly via WMP opens but shows nothing
[13:25] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r4674718203 10ffmpeg/ffplay.c: 
[13:25] <CIA-41> ffmpeg: ffplay: fix wrong reference to function in audio_decode_frame() comment
[13:25] <CIA-41> ffmpeg: The comment now references swr_convert(), rather than audio_convert(),
[13:25] <CIA-41> ffmpeg: which was deprecated and/or dropped.
[13:25] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r4fd07b9366 10ffmpeg/ffplay.c: 
[13:25] <CIA-41> ffmpeg: ffplay: avoid useless NULL checks in swr_free()
[13:25] <CIA-41> ffmpeg: swr_free() already checks for nullness, no need to add the check in
[13:25] <CIA-41> ffmpeg: calling code.
[13:32] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * re8e733adcc 10ffmpeg/libavcodec/dca.c: 
[13:32] <CIA-41> ffmpeg: dca: favor native over ffmpeg in table name
[13:32] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[14:51] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r567eb9d344 10ffmpeg/doc/muxers.texi: 
[14:51] <CIA-41> ffmpeg: lavf/segment: fix command with missing -list option
[14:51] <CIA-41> ffmpeg: Replace -list with the correct option -segment_list.
[14:52] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * rd9355a03f2 10ffmpeg/libavformat/hls.c: 
[14:52] <CIA-41> ffmpeg: lavf/applehttp: add log message in case of applehttp_read_header() failure
[14:52] <CIA-41> ffmpeg: Improve error reporting.
[14:52] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r0692d4c890 10ffmpeg/libavformat/segment.c: lavf/segment: add some debugging logs
[14:57] <Compn> The search warrants used by police to raid the New Zealand home of Megaupload founder Kim Dotcom have been ruled illegal. In addition, the data that was sent to the FBI was ruled to be unlawfully obtained.
[14:57] <CIA-41> ffmpeg: 03Lou Logan 07master * r6851130fd6 10ffmpeg/ (25 files in 2 dirs): 
[14:57] <CIA-41> ffmpeg: cosmetics: minor libavcodec spelling errors
[14:57] <CIA-41> ffmpeg: Also update some common misspelled words in patcheck
[14:57] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[14:57] <CIA-41> ffmpeg: 03Martin Sliwka 07master * rdf531b0e10 10ffmpeg/libavformat/file.c: (log message trimmed)
[14:57] <CIA-41> ffmpeg: avformat: disable seeking on FIFOs/named pipes
[14:57] <CIA-41> ffmpeg: Patch is addition to my previous patch
[14:57] <CIA-41> ffmpeg: (https://lists.ffmpeg.org/pipermail/ffmpeg-cvslog/2012-June/051590.html)
[14:57] <CIA-41> ffmpeg: and disables seeking on FIFOs/named pipes by setting
[14:57] <CIA-41> ffmpeg: URLContext::is_streamed (same as pipe: protocol does for stdin/stdout pipes)
[14:57] <CIA-41> ffmpeg: Fixes Ticket986
[15:52] <ubitux> saste: it seems your patch doesn't work if you don't do "-f stream_segment,ssegment"
[15:53] <ubitux> (it looks like -f stream_segment or -f ssegment doesn't work)
[15:54] <ubitux> also, why did you change the -c copy into -codec copy? :P
[15:55] <ubitux> not that it matters much but well...
[16:06] <saste> ubitux: order of application of patches matters
[16:06] <saste> which patch doesn't work?
[16:07] <saste> 1) lavf: allow multiple names in output devices selected by av_guess_format()
[16:07] <saste> 2) lavf/segment: add stream_segment variant of the segment muxer
[16:37] <jonsen_> i troubleshoot the whole day now just to get multicast streaming to work
[16:38] <jonsen_> which container and codec is well suited for that?
[16:39] <kierank> mpeg-ts and h264
[16:39] <jonsen_> and audio?
[16:39] <ubitux> saste: oh, right ok
[16:39] <jonsen_> kierank: it's really tricky to do all this right
[16:39] <kierank> aac
[16:39] <kierank> jonsen_: what are you trying to do
[16:40] <ubitux> saste: you should send them as patchset :)
[16:40] <kierank> i run a number of multicast streams
[16:41] <saste> jonsen_: don't forget to update http://ffmpeg.org/trac/ffmpeg/wiki/StreamingGuide when you're finished ;-)
[16:48] <jonsen_> kierank: trying to get ffserver stable to work. Should stream via Multicast. Would love to see it working on the Windows-7 clients in my company. I tried RTP with many different options (maybe not the right ones), tried several containers and codecs, but stream was either not stable or it didn't work at all. Windows Media Client never worked with RTP/Multicast, just with Unicast/HTTP and that never stable. Also the SDP extension for the multicast add
[16:48] <Compn> jonsen_ : unfortunately, no one really has worked on getting ffmpeg to work with wmp
[16:49] <Compn> lu_zero is the ffserver maintainer, you may want to ask him...
[16:49] <Compn> or wait for michaelni 
[16:49] <Compn> also i'm a multicast newbie
[16:50] <Compn> i know all other stream stuff somewhat :)
[16:50] <Compn> jonsen_ : did you try gstreamer ?
[16:50] <jonsen_> from Transport Layer perspective it's the same as Streaming via RTP/UDP Unicast
[16:50] <jonsen_> just unidirectional
[16:51] <jonsen_> just opening URLs to get the Multicast address is something special
[16:51] <jonsen_> the whole stuff is really painful
[16:52] <jonsen_> Compn: no, never tried
[16:52] <michaelni> Compn, lu_zero is not ffserver maintainer
[16:52] <ubitux> he is just a avserver mentor for gsoc afaict
[16:53] <jonsen_> Compn: is it worth to have a look?
[16:53] <Compn> michaelni : oh, i am mistaken then :\
[16:54] <Compn> jonsen_ : well, what are you trying to accomplish end goal ?
[16:54] <michaelni> Compn, np :)
[16:54] <jonsen_> I could also just use VLC to stream to the network
[16:54] <Compn> just streaming to wmp ?
[16:54] <Compn> vlc is also good
[16:54] <jonsen_> Compn: yup, but via Multicast
[16:54] Action: Compn never learned multicast :\
[16:55] <jonsen_> the Multicast is no problem for me
[16:55] <michaelni> btw, for the record lu_zero has 2 commits in ffserver.c out of 467 commits total in ffserver.c
[16:55] <jonsen_> and for the software neither
[16:56] <jonsen_> Transport Layer is important
[16:56] <jonsen_> udp/rtp tcp/http and so on
[16:56] <jonsen_> in the first place without having any knowledge about ffmpeg I thought that it would be right tool for professional streaming purposes
[16:57] <Compn> it is. wmp is the non-professional tool ;)
[16:57] <Compn> ehe
[16:57] <jonsen_> unfortunately there is no way around wmp
[16:58] <jonsen_> i want to stream to 20k users, 19900 of them have WMP :-(
[16:59] <Compn> ah
[17:11] <jonsen_> argh isnt working again with VLC
[17:11] <jonsen_> ffmpeg -i Prototype1920.ts -f mpegts rtp://239.1.2.3:5000
[17:11] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * ra9a7e215e5 10ffmpeg/libavformat/ (allformats.c segment.c version.h): 
[17:11] <CIA-41> ffmpeg: lavf/segment: add stream_segment variant of the segment muxer
[17:11] <CIA-41> ffmpeg: This simplifies usage for segment streaming formats with no global
[17:11] <CIA-41> ffmpeg: headers, tipically MPEG 2 transport stream "ts" files.
[17:11] <CIA-41> ffmpeg: The seg class duplication is required in order to avoid an infinite loop
[17:11] <CIA-41> ffmpeg: in libavformat/utils.c:format_child_next_class().
[17:11] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r3cd4f9fd76 10ffmpeg/libavformat/ (utils.c version.h): 
[17:11] <CIA-41> ffmpeg: lavf: allow multiple names in output devices selected by av_guess_format()
[17:11] <CIA-41> ffmpeg: Consistent with av_find_input_format().
[17:11] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * rc1abfbc47c 10ffmpeg/libavformat/segment.c: 
[17:11] <CIA-41> ffmpeg: lavf/segment: rename segment private context from "c" to "seg" in segment_start()
[17:11] <CIA-41> ffmpeg: Consistent with the rest of the file, less confusing.
[17:11] <jonsen_> nor that ffmpeg -i Prototype1920.ts -sn -an -re -f rtp rtp://239.1.2.3:5000
[17:12] <jonsen_> uh, and the rtp muxer can only stream one stream
[17:12] <jonsen_> so this one is showstopper anyway
[17:14] <saste> jonsen_: then use multiple outputs
[17:15] <jonsen_> that works?
[17:15] <jonsen_> i mean how does a client differentiate those streams?
[17:16] <jonsen_> those will come from different source ports
[17:16] <jonsen_> will give it a try
[17:16] <jonsen_> but at the moment just nothing works :D
[17:18] <jonsen_> just for the sake that anyone cares: http://de.pastebin.ca/2165549
[17:19] <jonsen_> for today I give up now
[17:19] <jonsen_> took me 10 hours
[17:46] <saste> michaelni: any hint on the pcm planar -> codecid mapping problem i mentioned this morning?
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * ra19e9f2d5c 10ffmpeg/libavformat/microdvddec.c: lavf/microdvd: rewrite using subtitles queue API.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * rf926d91611 10ffmpeg/ffmpeg.c: ffmpeg: fix a memleak in subtitles decoding.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * r0e7782c08e 10ffmpeg/ (libavcodec/ass.c tests/ref/fate/sub-srt): 
[20:29] <CIA-41> ffmpeg: lavc/ass: honor Default style.
[20:29] <CIA-41> ffmpeg: The "Default" style written in the header is ignored unless you explicit
[20:29] <CIA-41> ffmpeg: it in the Dialogue events (it was valid, just ignored). This requires an
[20:29] <CIA-41> ffmpeg: update of the SubRip test since the ASS output obviously changes.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * r8f4ce626f8 10ffmpeg/libavcodec/microdvddec.c: lavc/microdvddec: support "DEFAULT" properties.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * r04568f8d1a 10ffmpeg/tests/ (Makefile fate/subtitles.mak fate/video.mak): fate: introduce subtitles.mak and move SubRip test in it.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * re301f2f8c6 10ffmpeg/tests/ (3 files in 2 dirs): fate: add JacoSUB and MicroDVD subtitles tests.
[20:29] <CIA-41> ffmpeg: 03Clément BSsch 07master * rd948893dbd 10ffmpeg/libavformat/ (subtitles.c subtitles.h): 
[20:29] <CIA-41> ffmpeg: lavf/subtitles: add some SMIL helpers.
[20:30] <CIA-41> ffmpeg: This is needed for SAMI and RealText demuxers.
[20:30] <CIA-41> ffmpeg: 03Clément BSsch 07master * r53640f42be 10ffmpeg/ (13 files in 6 dirs): SAMI demuxer and decoder.
[20:30] <CIA-41> ffmpeg: 03Clément BSsch 07master * r439e32f9b8 10ffmpeg/ (13 files in 6 dirs): RealText demuxer and decoder.
[20:30] <CIA-41> ffmpeg: 03Clément BSsch 07master * r0ef28e119e 10ffmpeg/libavformat/jacosubdec.c: lavf/jacosubdec: fix FPE in case timeres is badly set.
[20:30] <CIA-41> ffmpeg: 03Clément BSsch 07master * r7c9f9685ae 10ffmpeg/libavformat/ (Makefile subtitles.c subtitles.h): lavf: add internal demuxer helpers for subtitles.
[20:30] <CIA-41> ffmpeg: 03Clément BSsch 07master * r60715511db 10ffmpeg/libavformat/jacosubdec.c: lavf/jacosubdec: use subtitles queue API.
[20:32] <ubitux> hey btw
[20:32] <ubitux> http://irq6.net/haproxy1.5-dev.git/commit/?id=6e0644339f3bcd455693d45219ba80ec23226296
[20:32] <ubitux> it could be nice to do that
[20:33] <ubitux> at the moment we have valgrind with --malloc-fill=0x2a
[20:33] <ubitux> but this is somehow limited
[20:33] <ubitux> it could be nice to have some lrng doing this for all the fate instances
[20:40] <michaelni> nice idea and it should be fairly easy to implement ...
[20:42] <michaelni> .... if you arent planing to implement it probably a good idea to open a feature req so its not forgotten ...
[20:44] <ubitux> i'm not :)
[20:45] <ubitux> well, i think the effort of opening an issue is almost equal to the effort of writing the feature& :D
[21:17] <michaelni> hi iive
[21:17] <iive> hi michaelni 
[21:18] <michaelni> do you want to implement filling av_malloc() output by random data ? so use of such data can be detected as difference in the output ?
[21:18] <michaelni> the idea was mentioned by ubitux
[21:18] <michaelni> but we are all to lazy 
[21:19] <iive> how urgent it is?
[21:22] <michaelni> isnt urgent
[21:23] <iive> then if nobody else does it in few days I'll give it a try.
[21:24] <iive> is this functionality to be enabled by cli option?
[21:26] <michaelni> yes cli option is probably the best and thanks
[21:33] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r12863db840 10ffmpeg/libavcodec/wma_common.h: 
[21:33] <CIA-41> ffmpeg: wma_common: Fix license header
[21:33] <CIA-41> ffmpeg: common wma code existed long before Libav
[21:33] <CIA-41> ffmpeg: Reviewed-by: Paul B Mahol <onemda at gmail.com>
[21:33] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[21:33] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rfb7688a83f 10ffmpeg/libavcodec/pngdsp.h: 
[21:33] <CIA-41> ffmpeg: pngdsp: Fix license header
[21:33] <CIA-41> ffmpeg: Libav did not exist in 2008 thus this file cannot have originated from there
[21:33] <CIA-41> ffmpeg: Reviewed-by: Paul B Mahol <onemda at gmail.com>
[21:33] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r7bbb6b38fb 10ffmpeg/libavcodec/ (proresdata.h proresdsp.h): 
[21:33] <CIA-41> ffmpeg: prores: Fix license header
[21:33] <CIA-41> ffmpeg: Libav did not exist in 2010 thus the file cannot originate from there
[21:33] <CIA-41> ffmpeg: Reviewed-by: Paul B Mahol <onemda at gmail.com>
[21:57] <Compn> spend a lot of time on stupid libav/ffmpeg copyright headers
[21:57] <Compn> could we make it 'this file is a part of FFmpeg/Libav ?
[21:57] <Compn> ask libav to do it too
[22:01] <durandal_1707> ^ pointless
[22:02] <bcoudurier> :)
[22:11] <ubitux> iive: the backlog if you're interested in working on the feature: http://pastie.org/private/iiwgwasarxjrk6bqhobg
[22:11] Action: ubitux feels lazy
[22:13] <iive> btw, wasn't avmalloc created to explicitly zero the allocate memory?
[22:13] <ubitux> no, we have av_mallocz for this
[22:14] <iive> hum, my bad.
[22:29] <ubitux> durandal_1707: is your public xpm branch up-to-date?
[22:29] <durandal_1707> what you mean?
[22:30] <ubitux> the patch you submitted on the ml has the encoder & decoder squashed
[22:30] <ubitux> i was wondering of testing it with your remote repo
[22:30] <ubitux> where the enc & dec are separate
[22:30] <ubitux> i was wondering if there was other changes
[22:30] <ubitux> (and which one is the more up to date in that case)
[22:31] <durandal_1707> ahh what i posted is what i have locally
[22:32] <durandal_1707> i usually do not update remote branch after i send patch (unless I fix something....)
[22:33] <ubitux> can i assume the two are mostly similar?
[22:36] <durandal_1707> they are same, except non-xpm code
[22:36] <ubitux> and patch-split :)
[23:18] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r0d3ffde64f 10ffmpeg/ffplay.c: ffplay: split overly long line in audio_decode_frame()
[23:18] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r8179660222 10ffmpeg/ffplay.c: 
[23:18] <CIA-41> ffmpeg: ffplay: vertially align complex if condition in audio_decode_frame()
[23:18] <CIA-41> ffmpeg: Possibly improve readability.
[23:18] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r347ecfdc63 10ffmpeg/ffplay.c: 
[23:18] <CIA-41> ffmpeg: ffplay: move assignment in else block in audio_decode_frame()
[23:18] <CIA-41> ffmpeg: Avoid confusing and pointless double assignment of variable
[23:18] <CIA-41> ffmpeg: resampled_data_size.
[23:18] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r94a00ec8af 10ffmpeg/ffplay.c: 
[23:18] <CIA-41> ffmpeg: ffplay: group together and vertically align correlated parameters in log function
[23:18] <CIA-41> ffmpeg: Possibly improve readability.
[23:27] <CIA-41> ffmpeg: 03Ronald S. Bultje 07master * rdfb57fc596 10ffmpeg/libavformat/rtpdec.c: 
[23:27] <CIA-41> ffmpeg: rtpdec: Don't explicitly include unistd.h any longer
[23:27] <CIA-41> ffmpeg: unistd.h used to be required for gethostname. On windows, gethostname
[23:27] <CIA-41> ffmpeg: is provided by winsock2.h. Now network.h includes both unistd.h and
[23:27] <CIA-41> ffmpeg: winsock2.h if they exist.
[23:27] <CIA-41> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[23:27] <CIA-41> ffmpeg: 03Samuel Pitoiset 07master * re312fcde6a 10ffmpeg/doc/general.texi: 
[23:27] <CIA-41> ffmpeg: doc: Indicate that RTMPT is natively implemented in libavformat
[23:27] <CIA-41> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[23:27] <CIA-41> ffmpeg: 03Ronald S. Bultje 07master * rf985113075 10ffmpeg/libavutil/random_seed.c: 
[23:27] <CIA-41> ffmpeg: random_seed: Only read /dev/*random if we have unistd.h
[23:27] <CIA-41> ffmpeg: unistd.h is used for open/read/close, but if this header does not
[23:27] <CIA-41> ffmpeg: exist, there's probably no use in trying to open /dev/*random
[23:27] <CIA-41> ffmpeg: at all.
[23:27] <CIA-41> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[23:27] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r8a85660d3b 10ffmpeg/: (log message trimmed)
[23:28] <CIA-41> ffmpeg: * qatar/master:
[23:28] <CIA-41> ffmpeg:  file: Only include unistd.h if it exists
[23:28] <CIA-41> ffmpeg:  random_seed: Only read /dev/*random if we have unistd.h
[23:28] <CIA-41> ffmpeg:  doc: Indicate that RTMPT is natively implemented in libavformat
[23:28] <CIA-41> ffmpeg:  rtpdec: Don't explicitly include unistd.h any longer
[23:28] <CIA-41> ffmpeg: 03Ronald S. Bultje 07master * r3b1ab197be 10ffmpeg/ (configure libavformat/file.c): (log message trimmed)
[23:28] <CIA-41> ffmpeg: file: Only include unistd.h if it exists
[23:28] <CIA-41> ffmpeg: It is included for the open/read/write/close functions. On
[23:28] <CIA-41> ffmpeg: MSVC, where this header does not exist, the same functions
[23:28] <CIA-41> ffmpeg: are provided by io.h, which is already included.
[23:28] <CIA-41> ffmpeg: On windows, these functions are provided by io.h. Make sure
[23:28] <CIA-41> ffmpeg: io.h is included if it exists, regardless of the setmode
[23:29] <ubitux> michaelni: it looks like the sunos box is missing some samples rsync
[23:34] <michaelni> ubitux, fixed
[23:34] <michaelni> the problem is i didnt find a rsync client that works reliable for sunos so i decided to run it by hand
[23:36] <michaelni> the samples dont change often enough for that to rmatter ,much ,..
[23:37] <ubitux> ok :)
[23:37] <ubitux> thx
[23:55] <durandal_1707> michaelni: what blocksize dsp support?
[23:57] <durandal_1707> the codecs uses blocksize of 4 and there is way to copy pixels horizontally and vertically
[00:00] --- Sat Jun 30 2012


More information about the Ffmpeg-devel-irc mailing list