[Ffmpeg-devel-irc] ffmpeg-devel.log.20191120

[burek]

[00:16:41 CET] <cone-157> ffmpeg 03Michael Niedermayer 07master:d73f06270600: avcodec/dvdec: Use av_clip_uint8 instead of ff_crop_tab
[03:42:28 CET] <cone-636> ffmpeg 03James Almer 07release/4.2:8da31e9eef6d: avcodec/cbs_av1: fix reading reference order hint in skip_mode_params()
[03:42:28 CET] <cone-636> ffmpeg 03James Almer 07release/4.2:252ef2329a53: avcodec/cbs_av1: keep separate reference frame state for reading and writing
[03:42:28 CET] <cone-636> ffmpeg 03Mark Thompson 07release/4.2:82a3a623f012: cbs_h264: Fix missing inferred colour description fields
[03:57:26 CET] <mkver> Is side data of type AV_PKT_NEW_EXTRADATA actually supposed to be padded like normal extradata? I presume it is intended, but it doesn't seem to be stated anywhere.
[03:59:43 CET] <mkver> av_packet_new_side_data() (which is used by av_packet_copy_props() and therefore by av_packet_ref()) at least allocates all side data with padding.
[04:11:20 CET] <BBB> mkver: yes, and should probably be documented explicitly that it needs to be padded identical to normal extradata
[04:11:54 CET] <mkver> Ok. The only place where this is not so seems to be in libavcodec/tests/avpacket.c
[11:26:33 CET] <cone-677> ffmpeg 03Paul B Mahol 07master:e21d4a7ca43e: avfilter/vf_chromakey: add >8 bit support
[13:17:50 CET] <cone-677> ffmpeg 03Paul B Mahol 07master:a16de215c917: avfilter/af_rubberband: fix sample overqueueing
[15:25:34 CET] <cone-677> ffmpeg 03Andreas Rheinhardt 07master:dbc50f8a9350: avformat/matroskadec: Fix default value of BlockAddID
[15:53:07 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:19b8db2908bf: avcodec/atrac9dec: Check precision_fine/coarse
[15:53:08 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:faea5b4462c4: avformat/vividas: Add EOF check in val_1 loop in track_header()
[15:53:09 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:c8c17b8cef77: avcodec/sonic: Fix integer overflow in predictor_calc_error()
[15:53:10 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:b6abdb1faf81: avcodec/ra288: Check block_align for the assumed value
[15:53:11 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:dc3f327e7403: avcodec/apedec: Fixes integer overflow of res+*data in do_apply_filter()
[15:53:12 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:e36ccb5048f0: avcodec/qdm2: The smallest header seems to have 2 bytes so treat 1 as invalid
[15:53:13 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:7fc1baf0ca83: avcodec/vc1_block: Fix undefined behavior in ac prediction rescaling
[15:53:14 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:c54b9fc42fee: avcodec/wmadec: Require previous exponents for reuse
[15:53:15 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:1b7d02642b20: avcodec/ralf: use multiply instead of shift to avoid undefined behavior in decode_block()
[15:53:16 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:f09151fff9c7: avcodec/atrac3: Check for huge block aligns
[15:53:17 CET] <cone-677> ffmpeg 03Michael Niedermayer 07master:a1f8b36cc454: avcodec/iff: Move index use after check in decodeplane8()
[17:48:09 CET] <cone-677> ffmpeg 03Paul B Mahol 07master:176ac987aa25: avfilter/f_graphmonitor: output frames in pts gaps
[18:36:03 CET] <cone-677> ffmpeg 03Alex Mogurenko 07master:b6571777d28d: avcodec/mjpegbdec: Fix yuv444 pix_fmt detection
[19:49:46 CET] <thardin> does VLC use its own demuxers or does it use lavf? or both?
[19:50:19 CET] <JEEB> both
[19:50:28 CET] <JEEB> for a lot of common things it has its own
[19:50:49 CET] <JEEB> but you can also --demuxer=avformat or whatever the parameter was. and less common things will end up in avformat
[19:51:08 CET] <JEEB> mp4, matroska, mpeg-ts, etc at least have vlc's internal demuxers
[19:51:22 CET] <thardin> I'm going to guess it doesn't have its own mxf demuxer
[19:52:39 CET] <JEEB> would initially look like that, yes http://git.videolan.org/?p=vlc.git;a=tree;f=modules/demux;h=7b55ad195c655c58756437b3ed4afcec11c1cee7;hb=HEAD
[22:42:14 CET] <cone-783> ffmpeg 03Paul B Mahol 07master:7ead0daa24f0: avfilter/vf_median: add support for commands
[00:00:00 CET] --- Thu Nov 21 2019