[Ffmpeg-devel-irc] ffmpeg-devel.log.20150225

burek burek021 at gmail.com
Thu Feb 26 02:05:02 CET 2015


[00:12] <cone-631> ffmpeg.git 03Martin Storsjö 07master:78791c086bca: rtpdec: Use .init instead of .alloc to set default values
[00:12] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:d52bf89d21b7: Merge commit '78791c086bcaf9eb084c27555b31fea8bbb7624a'
[00:24] <cone-631> ffmpeg.git 03Martin Storsjö 07master:f4b59334bd89: rtpdec: Remove the now unused .alloc field
[00:24] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:f966ac2b6e3f: avformat/rtpdec_vp9: remove alloc/free functions
[00:24] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:64f58d20b200: Merge commit 'f4b59334bd898c47c9ac30999a346176548630a8'
[00:32] <cone-631> ffmpeg.git 03Martin Storsjö 07master:bb4a310bb85f: rtpdec: Don't free the payload context in the .free function
[00:32] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:fa80b76b110c: Merge commit 'bb4a310bb85f43e62240145a656b1e5285b14239'
[00:40] <cone-631> ffmpeg.git 03Martin Storsjö 07master:078d43e23a7a: rtpdec: Free depacketizers if the init function failed
[00:40] <cone-631> ffmpeg.git 03Martin Storsjö 07master:8a273a746061: avio: Add an internal utility function for freeing dynamic buffers
[00:40] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:79052cd4ce4a: Merge commit '078d43e23a7a3d64aafee8a58b380d3e139b3020'
[00:40] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:6e9bbaca6c67: Merge commit '8a273a746061a112e5e35066a8fd8e146d821a62'
[00:49] <cone-631> ffmpeg.git 03Martin Storsjö 07master:199fb4027814: rtpdec: Use ffio_free_dyn_buf
[00:49] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:7d9b06eb179f: Merge commit '199fb40278146c5bb162990c66ad3cd561abc780'
[00:56] <cone-631> ffmpeg.git 03Martin Storsjö 07master:8e32b1f0963d: libavformat: Use ffio_free_dyn_buf where applicable
[00:56] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:7f03f75ce72f: Merge commit '8e32b1f0963d01d4f5d4803eb721f162e0d58d9a'
[01:05] <cone-631> ffmpeg.git 03Martin Storsjö 07master:d594dbecce5f: rtpdec: Rename the free method to close
[01:05] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:3130e5be9285: Merge commit 'd594dbecce5f3af7f80bbf5bb1b516d740b69ccd'
[01:19] <cone-631> ffmpeg.git 03John Stebbins 07master:da7e561964ee: matroskaenc: Allow writing track "forced" flag
[01:19] <cone-631> ffmpeg.git 03Luca Barbato 07master:b31328d00898: xcbgrab: Provide better names for the y and x option
[01:19] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:53dd75e91154: Merge commit 'da7e561964ee833678dc24fae0b0265358126d1e'
[01:19] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:2fbdf30efe99: Merge commit 'b31328d008985f87f0a7c83c700847cef1a4f08c'
[01:51] <cone-631> ffmpeg.git 03Luca Barbato 07master:85b3b1c4ba7a: xcbgrab: Unbreak parsing filename options
[01:51] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:a1b5cf018168: Merge commit '85b3b1c4ba7af9c2658442b0aafd27d613e1854b'
[02:08] <jamrial> http://pastebin.com/WaTVBbpp popcnt
[02:24] <nevcairiel> popcnt sure is nice
[03:33] <cone-631> ffmpeg.git 03Michael Niedermayer 07master:7296716e355c: avcodec/h264: Clear last_pic_for_ec on seeks and reinits
[12:43] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:1f5c7781e63d: avcodec/zmbv: Check len before reading in decode_frame()
[13:13] <JoshX> good day..
[13:13] <JoshX> Hello, Is it possible to get info about an mp4 file which should countain 1800 frames (60 seconds, 30fps) but only contains 1560 frames (240 frames missing, 8 seconds) to find out which frames are missing?
[13:13] <JoshX> or to put it in other words, can i get a list of all the frames in the file with their start and end time and frame number for example?
[13:40] <Daemon404> JoshX, maybe ffprobe -show_frames file.mp4
[13:41] <Daemon404> but this belongs in #ffmpeg
[13:42] <JoshX> yes and i got my answer.. sorry about that
[13:42] <wm4> does ffprobe still show wrong frame duration for mov?
[13:43] <JoshX> and the show_frames will probably work in my case :) i just need the 'units' for pkt_dts=553
[13:43] <Daemon404> wm4, only the last frame had wrong duration iirc
[13:44] <wm4> JoshX: ffprobe also shows the timestamps in seconds, doesn't it
[13:47] <JoshX> yes it does
[13:47] <JoshX> writing a small script now to find gaps in the timestamps
[13:50] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:bbbc8c618884: avformat/mov: Check ctts_count before use
[14:06] <Daemon404> JoshX, -of json is your pal
[14:06] <Daemon404> easily consumable by most languages
[14:21] <Daemon404> michaelni, supposedly 'colr' for isombff is defined in 14496-12:2012
[14:21] <Daemon404> i cant find any copies floating around though
[14:21] <Daemon404> only the outdated 2005 ver
[14:38] <michaelni> Daemon404, it lists nclx, rICC and prof as allows color_types and refers to  A.7.2 of ISO/IEC 29199-2 for nclx with "but note that the full range flag is here in a different bit position"
[14:39] <Daemon404> right
[14:39] <Daemon404> it differs slightly from the mov colr
[14:42] <michaelni> anyone interrested in isom colr: http://pastebin.com/A3MhtacY
[14:44] <Daemon404> wait you can embedd icc profiles?
[14:44] <Daemon404> ewwwwwwwwww
[14:47] <thardin> yes
[14:47] <Daemon404> time for vlc to link in lcms2?
[14:47] Action: Daemon404 runs
[14:54] <michaelni> http://www.itu.int/rec/T-REC-T.832-201201-I/en <-- jpeg-xr  PTM_COLOR_INFO
[17:26] <cone-666> ffmpeg.git 03Clément BSsch 07master:5c11f5b4a227: avfilter/palettegen: move longest variable to a local scope
[17:47] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:205b2ba3d677: avcodec/hevc_ps: Sanity checks for some log2_* values
[17:53] <JoshX> Daemon404: if i were to 'miss' or 'mangle' an i frame in a vcodec copy stream, would ffmpeg discard all frames untill the next iframe?
[17:53] <JoshX> i seem to miss an iframe, and then miss the next 29 pframes too and then the stream just picks up at the next iframe with a resulting 1 second gap
[17:54] <JoshX> if this ffmpegs way to cope with a video loss and to produce a clean output, preventing mpeg artefacts?
[17:54] <JoshX> or is there something entirely different going on?
[17:57] <cone-666> ffmpeg.git 03Carl Eugen Hoyos 07master:ab3ff19f08b7: lavfi/fade: Do not overread input buffer.
[17:57] <cone-666> ffmpeg.git 03Carl Eugen Hoyos 07master:56ffde3f6190: Allow muxing h264 into ts with startcode "001" instead of "0001".
[17:57] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:d1e2d59f23a9: Merge remote-tracking branch 'cehoyos/master'
[18:41] <cone-666> ffmpeg.git 03Seppo Tomperi 07master:63ca0fe8288d: avcodec/hevcdsp: ARM NEON optimized qpel functions
[21:31] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:6d66e1a1136d: avutil/common: Fix integer overflow in av_clip_int8/16_c
[21:31] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:dd369c9adbdd: avcodec/snow: Initialize spatial_decomposition_count to a valid value
[22:37] <ubitux> wow, the int8 and int16 clipping were broken?
[22:38] <ubitux> fear
[22:45] <wm4> lol
[22:45] <wm4> unmaintenable microoptimized code monmster
[23:09] <cone-666> ffmpeg.git 03Michael Niedermayer 07master:701c965a7606: avcodec/hevc_ps: Check that log2_ctb_size is not smaller than the bounds of all profiles
[23:59] <cone-666> ffmpeg.git 03James Almer 07master:bc65abc8d70d: libavutil: add x86 optimized av_popcount
[00:00] --- Thu Feb 26 2015


More information about the Ffmpeg-devel-irc mailing list