[00:56] <BBB> Compn: I assure you I dont have mplayer installed; Id be surprised if I used it in the last 10 years [00:57] <BBB> Compn: so no, not all devs use mplayer as their primary testing tool [00:57] <BBB> if mplayer was user-friendly& [01:00] <wm4> not even I have mplayer installed [01:00] <wm4> too hard to build [01:32] <Daemon404> 23 [01:49] <cone-974> ffmpeg.git 03Michael Niedermayer 07master:2db1bcf1b982: avformat/asfdec: Try to improve skip case [09:56] <ubitux> Timothy_Gu: btw, you do some advertising to Libav features in the release notes; i think you should either credit the project, or remove them from the changelog [09:57] <ubitux> it's not very honest from us to advertise such features [10:51] <cone-816> ffmpeg.git 03Muhammad Faiz 07master:860d12978c3a: doc/filters: fixing typo [11:47] <cone-816> ffmpeg.git 03Timothy Gu 07master:2a9b4c0f05c2: Add a release note for 2.3 [12:00] <michaelni> ubitux, just saw your comment now, but should we credit each author for each feature in the release notes ? [12:01] <ubitux> not each author, it's just that libav definitely didn't do these contributions for ffmpeg, in comparison to other features [12:04] <michaelni> libav is not a person, its the authors who did the work [12:05] <michaelni> also neither project listed a "from project" in the previous changelog [12:05] <michaelni> iam not against it in principle its just very strange [12:09] <michaelni> also i resign from maintaining this just in case its unclear [12:10] <michaelni> i certainly cannot on every merge keep such extra credits up to date when the original authors dont write such credits [12:13] <ubitux> note that i wasn't talking about the list of changes themselves, but more the highlights at the beginning [12:20] <J_Darnley> Regarding that release note, shouldn't this "libavutil xx.yy.1zz" list an actual version? [12:21] <michaelni> yes the versions should be fixed, i mainly pushed it to simplify collaboration, i did expect it to take a few more edits before the release [12:21] <J_Darnley> Oh, that's because 2.3 hasn't been made yet [12:21] <J_Darnley> My bad. [12:32] <michaelni> ubitux, if you want to add some libav references, i guess send a patch that adds them or that removes the parts. But still it feels very strange to me to credit the project instead of the authors [12:35] <michaelni> also for example assuming you talk about opus which says "Copyright (c) 2012 Andrew D'Addesio and Copyright (c) 2013-2014 Mozilla Corporation and @author Andrew D'Addesio, Anton Khirnov" , iam not sure that crediting this as a "from libav" is the proper thing to do [12:36] <michaelni> i dont know which part and how much is from whom but "Mozilla Corporation" certainly isnt part of libav [13:16] <michaelni> ubitux, proper credit for opus might be something like "Initial implementation by Andrew D'Addesio <modchipv12@gmail.com> during GSoC 2012. Completion by Anton Khirnov <anton@khirnov.net>, sponsored by the Mozilla Corporation, additional contributions by several other developers, the code was merged from Libav into FFmpeg" but this feels a bit long for the release notes. i guess ill leave it to you and others to decide wha [13:16] <michaelni> t is the best thing to do about it [14:15] <Compn> 'author copyrights can be found in the source code' ? :) [14:15] <Compn> ehe [14:15] Action: Compn afk [15:08] <cone-773> ffmpeg.git 03Michael Niedermayer 07master:47c84c0bf784: avformat/wavdec: add basic sanity check for the sample count [16:39] <j-b> 'morning [17:15] <michaelni> afternoon [17:22] <cone-773> ffmpeg.git 03Michael Niedermayer 07master:ec24796731d3: RELEASE_NOTES: fill in lib versions [17:22] <cone-773> ffmpeg.git 03Michael Niedermayer 07master:68215724997c: avformat/movenc: dont mark multichannel as mono tracks as containing the center channel [19:00] <michaelni> ubitux, Timothy_Gu (and others), what do you prefer we do about the release_notes author/project credit ? is it ok as it is? should i add the proper full credit as a footnote or something or remove opus or something else ? [19:26] <a141982112> dvbsubdec to overlay fails to produce correct output [19:28] <a141982112> this is might be because of compute_edt set to 1 automatically in ffmpeg [19:30] <a141982112> I wanted to change this behaviour to only when dvbsub is transcoded not overlayed [19:30] <a141982112> is checking filters option in ffmpeg.c is good idea [20:16] <michaelni> a141982112, sounds like a hack [20:18] <a141982112> i need to check that dvbsub subtite stream are mapped to one subtitle stream [20:18] <a141982112> is there any better way [20:41] <a141982112> just now i am using [0:v][0:s]overlay[v] filter to test, is it possible/reasonable when we use filter for subtitle and in outstream we get subtitle stream [21:04] <a141982112> I have sended the hack patch, because code for checking mapped stream was becoming big [21:33] <Timothy_Gu> michaelni, ubitux: how about "This ffmpeg release (also) contains new features from Libav like Opus ..." [21:36] <ubitux> Timothy_Gu: sounds good to me [21:44] <cone-57> ffmpeg.git 03Michael Niedermayer 07master:62227a70f0a4: avformat: add av_stream_get_parser() to access avformat AVParser [21:44] <cone-57> ffmpeg.git 03Michael Niedermayer 07master:8bbadc9b6ec7: ffmpeg: Use av_stream_get_parser() to avoid ABI issues [22:20] <cone-57> ffmpeg.git 03Martin Storsjö 07master:f9900822fc5f: fate: Use the correct, local path to samples for opus reference files [22:20] <cone-57> ffmpeg.git 03Michael Niedermayer 07master:a65d6e17bd04: Merge commit 'f9900822fc5f0b0c7242ca318290c090bd5d0af0' [22:20] <jamrial> Timothy_Gu: for that matter, maybe rewording "We have already ported some 32-bit ARM to AArch64" into "some 32-bit ARM has already been ported to AArch64", since "we" didn't port it. [22:29] <Timothy_Gu> michaelni: do you want to release a beta for 2.3? [22:31] <michaelni> no [22:32] <michaelni> but i need to release a new 2.2, as it has a ABI bug [22:33] <Timothy_Gu> What is your intended codename? [22:35] <cone-57> ffmpeg.git 03Luca Barbato 07release/2.2:6d899d0206e9: lzo: Handle integer overflow [22:35] <cone-57> ffmpeg.git 03Justin Ruggles 07release/2.2:fcbcc561e0fd: Check if an mp3 header is using a reserved sample rate. [22:36] <cone-57> ffmpeg.git 03Justin Ruggles 07release/2.2:46c477c2a14b: Check mp3 header before calling avpriv_mpegaudio_decode_header(). [22:36] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:d832020bd853: Merge commit '6d899d0206e91a3a74680b47ec06ef1dbc151c21' into release/2.2 [22:36] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:00049f193d07: Merge commit 'fcbcc561e0fdc95a7dd48b92db53846726aec27e' into release/2.2 [22:36] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:0d901439729c: Merge commit '46c477c2a14b04a63ab11d31003b48fab6146a96' into release/2.2 [22:37] <michaelni> the remaining names that have been suggested previously where Mandelbrot and Fresnel, havnt really thought about which of the 2 should be next [22:45] <cone-57> ffmpeg.git 03Vittorio Giovara 07release/2.2:74f6df745a05: jpeg2000: fix dereferencing invalid pointers during cleanup [22:45] <cone-57> ffmpeg.git 03Anton Khirnov 07release/2.2:4ec1acc6e4f4: avpacket: fix copying side data in av_packet_copy_props() [22:45] <cone-57> ffmpeg.git 03Martin Storsjö 07release/2.2:16f7cbef5610: oggenc: Set the right AVOption size for the pref_duration option [22:45] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:ebf381168af5: Merge commit '74f6df745a05d3d8b3dcfc28992c69a70ae87957' into release/2.2 [22:45] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:5f1f1868b5ef: Merge commit '4ec1acc6e4f4d0bd28617bbb30ebfb779ce7df92' into release/2.2 [22:45] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:0bbd46c6903b: Merge commit '16f7cbef5610a878317596134607d2a89da66ecf' into release/2.2 [22:53] <Compn> michaelni : einstein ? [22:55] <microchip_> michaelni: Bohr? Von Neumann? :P [22:56] Action: michaelni makes notes [22:57] <cone-57> ffmpeg.git 03Martin Storsjö 07release/2.2:771564945aa9: adpcm: Avoid reading out of bounds in the IMA QT trellis encoder [22:57] <cone-57> ffmpeg.git 03Martin Storsjö 07release/2.2:564c023eba80: adpcm: Write the proper predictor in trellis mode in IMA QT [22:57] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:d47e96090cfa: Merge commit '771564945aa9aebe2f30192b925fcf4909225eca' into release/2.2 [22:57] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:65b839e43a5b: Merge commit '564c023eba807b162527b17af40424de275fc795' into release/2.2 [23:01] <microchip_> lorentz, poincaré, desitter, de broglie, etc - http://phys.kent.edu/~manley/physicists.html [23:04] <cone-57> ffmpeg.git 03Reinhard Tartler 07release/2.2:52dd1a933ed4: Update Changelog for v10.2 [23:04] <cone-57> ffmpeg.git 03Reinhard Tartler 07release/2.2:40dd29653ab8: Prepare for 10.2 Release [23:04] <cone-57> ffmpeg.git 03Andrew Kelley 07release/2.2:95e91aaf335d: af_compand: make sure request_frame always outputs at least one frame [23:04] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:7c81afdba59f: Merge commit '40dd29653ab85812d21fa64e9a665ceb316701ad' into release/2.2 [23:04] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:b2111ad4df29: Merge commit '95e91aaf335dd9c26a9101c38d4926e5271e7e00' into release/2.2 [23:33] <Timothy_Gu> So, 2.3 is Mandelbrot, right? [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:a770a61e6d6e: avfilter/vf_pullup: use ptrdiff_t as stride argument for dsp functions [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:f38c42b913ab: avfilter/x86/vf_pullup: fix old typo [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:b8102ce56d8a: cmdutils_opencl: Use av_malloc_array() [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:e064cce972fe: avcodec/hevc: Use av_malloc(z)_array() [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:64e069efac29: avfilter/f_ebur128: Use av_malloc_array() [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:9752ab6b9e5d: avformat/utils: do not wait for packets from discarded streams for genpts [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:9f8e3e6d1203: avformat: add av_stream_get_parser() to access avformat AVParser [23:38] <cone-57> ffmpeg.git 03Michael Niedermayer 07release/2.2:160e91de89f6: ffmpeg: Use av_stream_get_parser() to avoid ABI issues [23:39] <michaelni> Timothy_Gu, ok [00:00] --- Tue Jul 15 2014