[Ffmpeg-devel-irc] ffmpeg-devel.log.20150101

[burek]

[00:41] <thardin> gott nytt år!
[00:58] <kierank> Happy new year
[02:24] <michaelni> Happy new year to you and everyone else as well !
[03:15] <cone-138> ffmpeg.git 03Michael Niedermayer 07master:18802bc81ce8: avfilter/vf_idet: Use frame_requested instead of prev
[06:17] <cone-386> ffmpeg.git 03Michael Niedermayer 07master:d25d92960034: avfilter/vf_idet: Add 9, 12 and 14 bit pixel formats
[12:31] <j-b> Happy New Year!
[12:36] Action: compn wonders when the cone hat goes away
[12:38] <j-b> compn: today
[18:27] <cone-211> ffmpeg.git 03Michael Niedermayer 07master:5c37ffca59e7: avformat/flvdec: add several error messages to error conditions
[18:27] <cone-211> ffmpeg.git 03Michael Niedermayer 07master:322f0f5960a7: avformat/flvdec: do not inject dts=0 metadata packets which failed to be parsed into a new data stream
[18:27] <cone-211> ffmpeg.git 03Michael Niedermayer 07master:eb767a276bfd: avformat/flvdec: Increase string array size
[20:19] <cone-211> ffmpeg.git 03Michael Niedermayer 07master:efc4bfc195da: avformat/flvdec: use named identifier instead of literal numbers as return values
[20:34] <cone-211> ffmpeg.git 03Marton Balint 07master:8c1f5fb0ae19: ffplay: remove flushed state from decoder context
[20:34] <cone-211> ffmpeg.git 03Marton Balint 07master:82928656b134: ffplay: remove unused no_background from videostate
[20:34] <cone-211> ffmpeg.git 03Marton Balint 07master:5e7dcb0491cc: ffplay: add support for stream specifiers in -ast, -vst, -sst options
[20:34] <cone-211> ffmpeg.git 03Marton Balint 07master:46aa75eea1bd: ffplay: dump format before selecting streams
[20:34] <cone-211> ffmpeg.git 03Michael Niedermayer 07master:8009a51f7bbd: Merge remote-tracking branch 'cus/stable'
[22:02] <cousin_luigi> Greetings.
[22:04] <cousin_luigi> https://github.com/FFmpeg/FFmpeg/blob/master/libavformat/flvdec.c#L444 <- is this line supposed to be there?
[22:05] <thardin> that doesn't look right
[22:06] <compn> no it does not
[22:07] <wm4> lol merge error?
[22:07] <ubitux> no, regression from a few hours ago (5c37ffca)
[22:08] <ubitux> didn't even notice while reading the log
[22:08] <wm4> well this caught my eyes http://git.videolan.org/?p=ffmpeg.git;a=blobdiff;f=libavformat/flvdec.c;h=c13c0e48a1a8e4f1d553a9ab0cc3f603be5b37a3;hp=dc0edf20d5ea27164f4e6d68646d3ccf4093e6a7;hb=efc4bfc195dab6efa77b444eac9313df4596a922;hpb=eb767a276bfdb9a0493bdb0b38203638230b7ccb
[22:08] <cousin_luigi> I'm trying to see if that fix will work on 2.3.5.
[22:08] <wm4> returns named constants, but doesn't compare them anywhere
[22:09] <cone-211> ffmpeg.git 03Clément BSsch 07master:6a29499b95bb: avformat/flvdec: remove duplicated line error
[22:09] <ubitux> cousin_luigi: thx
[22:10] <cousin_luigi> wm4: hopefully it will in the future?
[22:25] <compn> maybe its a workaround for some braindead compiler
[22:30] <ubitux> compn: what are you talking about?
[22:30] <compn> what wm4 was talking about, named constants
[22:31] <compn> the git hash url he posted 
[22:31] <compn> ubitux : did you contact that libavfilter patch author about including it upstream ? 
[22:31] <compn> i guess you found another patch too, how many are we collecting now? :)
[22:32] <ubitux> no it's not a compiler workaround
[22:32] <ubitux> >that libavfilter patch author
[22:32] <ubitux> are you talking about nlmeans?
[22:34] <compn> yes nlmeans
[22:34] <ubitux> no i'm reading the paper(s) and reimplementing it
[22:34] Action: compn falls into reimplemented well of doom
[22:34] <compn> ok then 
[22:44] <aetasx> great game
[00:00] --- Fri Jan  2 2015