[Ffmpeg-devel-irc] ffmpeg-devel.log.20190512

burek burek021 at gmail.com
Mon May 13 03:05:04 EEST 2019


[01:37:15 CEST] <cone-453> ffmpeg 03Carl Eugen Hoyos 07master:010c0efada6d: lavf/isom: Allow reading mp3-in-isom as written by vlc.
[08:03:37 CEST] <pross> hi. who can add files to the fate samples directory?
[08:53:59 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:014b0e5092a2: lavc/aacenc_ltp: remove unnecessary condition check.
[08:53:59 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:00555f4b8b40: lavc/avpacket: check NULL before using the pointer
[08:53:59 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:3af73f2c8588: lavc/bink: Remove the dead code block
[08:53:59 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:f82a02aa89ce: lavc/libvpxenc: remove redundant condition check
[08:53:59 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:64e610b5f4b7: lavc/mlpenc: remove the redundant condition check
[08:54:00 CEST] <cone-369> ffmpeg 03Jun Zhao 07master:1cc7e2630fec: lavc/pngenc: check malloc fail before using the pointer
[15:26:03 CEST] <thardin> .. vp4 huh?
[16:01:47 CEST] <durandal11707> thardin: vp0
[16:02:34 CEST] <thardin> vp
[16:35:32 CEST] <cone-536> ffmpeg 03Michael Niedermayer 07master:e5ce10514918: avcodec/wmv2dec: Check input bits vs. coded MBs
[16:35:33 CEST] <cone-536> ffmpeg 03Michael Niedermayer 07master:407e7c34ca8a: avcodec/hq_hqa: Check available space before reading slice offsets
[16:35:34 CEST] <cone-536> ffmpeg 03Michael Niedermayer 07master:7c6f2bfdb25c: avcodec/h264_ps: Check offset_for_non_ref_pic, offset_for_top_to_bottom_field and offset_for_ref_frame
[17:07:27 CEST] <cone-536> ffmpeg 03Philip Langdale 07master:5de4f1d871d6: avutil: Add NV24 and NV42 pixel formats
[17:07:28 CEST] <cone-536> ffmpeg 03Philip Langdale 07master:cd483180356c: swscale: Add support for NV24 and NV42
[17:07:29 CEST] <cone-536> ffmpeg 03Philip Langdale 07master:4fa4f1d7a949: swscale: Add test for isSemiPlanarYUV to pixdesc_query
[17:26:08 CEST] <cone-536> ffmpeg 03Paul B Mahol 07master:2a4a364f6a00: avfilter/af_astats: fix initial values of variables
[17:26:09 CEST] <cone-536> ffmpeg 03Paul B Mahol 07master:bd6942b4c031: avfilter/af_astats: fix calculations involving last sample
[19:08:59 CEST] <jkqxz> cehoyos:  I don't think the -pix_fmt option change works.  Avoiding the libavdevice cases is fine, but for normal files you don't necessarily know whether the input is going to be rawvideo before the open, which needs the pixfmt to be set.
[19:27:21 CEST] <cehoyos> jkqxz: Sorry, but doesn't pix_fmt work already for rawvideo files? Or do I misunderstand?
[19:30:21 CEST] <jkqxz> At the point of opening the file, I don't see a way to distinguish between (a) this is a rawvideo file, and -pix_fmt was set to specify the format in the file, and (b) this is a compressed file, and -pix_fmt was set to specify the decode format.
[19:30:42 CEST] <jkqxz> (In case (b) passing the pixel_format option will cause the libavformat open to fail.)
[19:35:28 CEST] <cehoyos> Why must the rawvideo demuxer know the pix_fmt?
[19:35:38 CEST] <cehoyos> (Couldn't this be a new lavc option?)
[19:38:56 CEST] <jkqxz> Because <http://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavformat/rawvideodec.c;h=8bde22c4e93de5aed7094e454d5ad842c2503482;hb=HEAD#l52>.
[19:39:10 CEST] <durandal11707> anybody agains applying asr filter?
[19:46:33 CEST] <cehoyos> But in general, lavc is allowed to set the pix_fmt before read_packet() is called for the first time, no?
[19:50:30 CEST] <jkqxz> Before init, so avcodec_open2(), but yes.
[20:06:17 CEST] <durandal11707> can i get answer - reply or something?
[20:26:47 CEST] <durandal11707> ping
[20:44:32 CEST] <durandal11707> can i get answer?
[20:52:37 CEST] <Lynne> yes, I think you should wait for subtitles in lavfi for the filter to make sense
[20:54:12 CEST] <durandal11707> that will happen in next 1000 years
[20:55:23 CEST] <durandal11707> and filter already makes sense because it exports stuff via frame metadata
[20:57:17 CEST] <cehoyos> Lynne: From which files in lavc do the used functions come from?
[20:58:00 CEST] <jkqxz> durandal11707:  What does the output actually look like?  (How is text split across frames?)
[20:58:11 CEST] <Lynne> cehoyos: libavcodec/fft_template.c
[21:01:05 CEST] <durandal11707> jkqxz: it is output when it is received
[21:48:01 CEST] <beastd> j-b: has the ssh host key for ffmpeg git changed since april 2017?
[22:45:14 CEST] <cone-536> ffmpeg 03Rick Kern 07master:247bda442f51: lavc/videotoolboxenc: add require_sw option to force software encoding.
[22:45:15 CEST] <cone-536> ffmpeg 03Rick Kern 07master:714c8b076624: lavc/videotoolboxenc: simplify frame creation logic between OS X and iOS.
[22:45:16 CEST] <cone-536> ffmpeg 03Rick Kern 07master:98478c05edb6: lavc/videotoolboxenc: Add support for 10-bit HEVC encoding.
[23:00:06 CEST] <j-b> beastd: maybe. I don't run the FFmpeg infrastructure, unfortunately.
[23:01:40 CEST] <beastd> ok thx. thought you might know. maybe anyone else here remembers.
[23:05:16 CEST] <cehoyos> j-b: The question was about the videolan infrastructure, not the FFmpeg infrastructure
[23:05:33 CEST] <cehoyos> git.videolan.org to be exact
[23:12:01 CEST] <jdarnley> beastd: I think I recall it changing within that period
[23:12:23 CEST] <jdarnley> but if you're that concerned about security you need a better source than me
[23:22:09 CEST] <beastd> jdarnley: thx for the info. i am just being a bit paranoid.
[00:00:00 CEST] --- Mon May 13 2019


More information about the Ffmpeg-devel-irc mailing list