[Ffmpeg-devel-irc] ffmpeg-devel.log.20190821

[burek]

[00:36:16 CEST] <michaelni_> BtbN, "The build has been scheduled for termination. There may be a delay before a new build can be resubmitted."
[00:36:26 CEST] <BtbN> Thanks.
[00:36:32 CEST] <BtbN> Been stuck for a good 3 months again :/
[00:36:47 CEST] <BtbN> Gonna upgrade the Docker-Image a bit as well
[01:12:23 CEST] <BtbN> https://lists.samba.org/archive/samba-technical/2018-October/130668.html running into this on Debian Stable. Lovely.
[01:13:10 CEST] <BtbN> So, aparently libsmbclient-4.0 header cannot be used when using std=c99
[01:13:14 CEST] <BtbN> even more lovely
[01:23:55 CEST] <cone-901> ffmpeg 03Michael Niedermayer 07master:f51e4d026cc7: avcodec/alsdec: Limit maximum channels to 512
[01:23:55 CEST] <cone-901> ffmpeg 03Michael Niedermayer 07master:d7866bf527a9: avcodec/internal: Bump sane_nb_chanels
[01:23:55 CEST] <cone-901> ffmpeg 03Michael Niedermayer 07master:661a9b274b01: avcodec/alsdec: Fix integer overflow in decode_var_block_data()
[03:21:31 CEST] <nicolas17> .rar /o\
[10:46:53 CEST] <cone-263> ffmpeg 03Steven Liu 07master:d1fe1344ea98: avformat/hlsenc: fix memleak in update_variant_stream_info
[10:46:54 CEST] <cone-263> ffmpeg 03Steven Liu 07master:fa78a6abb379: avformat/hlsenc: simplified code of use_localtime in hls_init
[10:46:55 CEST] <cone-263> ffmpeg 03Steven Liu 07master:c66d468795f7: avformat/hlsenc: avformat/hlsenc: simplified code
[10:52:08 CEST] <durandal_1707> apply my v360 set, it is now urgent
[10:55:56 CEST] <durandal_1707> you complained that its missing Changelog and other stuff
[10:58:58 CEST] <thardin> did anyone submit a patch for the fuzzer yet?
[11:01:07 CEST] <nevcairiel> james did
[11:05:32 CEST] <J_Darnley> I did?  I don't remember that.
[11:05:34 CEST] <J_Darnley> :)
[11:05:53 CEST] <thardin> cool, giving it a spin
[11:40:49 CEST] <thardin> seems to work well
[11:43:00 CEST] <durandal_1707> expert in const-correctness
[11:47:15 CEST] <durandal_1707> who is willing to apply v360 set, get prize in amount of 1000 eur minus currently passed seconds from this moment
[11:56:08 CEST] <durandal_1707> not much time left
[12:10:48 CEST] <thardin> sadly I know nothing of this newfangled 360° stuff
[13:36:14 CEST] <BtbN> "New defects found: 264" oh boy
[13:44:01 CEST] <durandal_1707> thardin: you havent even looked at set
[13:45:50 CEST] <thardin> no I have not
[13:48:12 CEST] <durandal_1707> its trivial changes
[13:48:31 CEST] <kierank> durandal_1707: i will push
[13:49:04 CEST] <kierank> durandal_1707: send patchwork link
[13:51:51 CEST] <durandal_1707> http://ffmpeg.org/pipermail/ffmpeg-devel/2019-August/248607.html
[13:52:18 CEST] <durandal_1707> kierank ^
[13:55:25 CEST] <kierank> done
[13:55:26 CEST] <cone-237> ffmpeg 03Paul B Mahol 07master:33cd5161672a: doc/filters: fix v360 options names
[13:55:26 CEST] <cone-237> ffmpeg 03Paul B Mahol 07master:cc73b849ffb7: add v360 to Changelog
[13:55:26 CEST] <cone-237> ffmpeg 03Paul B Mahol 07master:e6dd9438727d: doc/filters: fix typos
[13:55:26 CEST] <cone-237> ffmpeg 03Paul B Mahol 07master:a8ef41c1f5f0: avfilter/vf_v360: improve comments above headers
[13:55:26 CEST] <cone-237> ffmpeg 03Paul B Mahol 07master:aeae6283a9d5: avfilter/vf_v360: remove unused header
[13:59:37 CEST] <feliwir> Hey, what's up with all the grief on the mailing list?
[14:01:14 CEST] <kierank> FFmpeg is like this
[14:01:22 CEST] <kierank> a bit like the linux kernel used to be
[14:03:22 CEST] <thardin> strong egos abound
[14:08:03 CEST] <durandal_1707> alpha males
[14:09:04 CEST] <thardin> I'd say the opposite. confident people don't act that way
[14:09:44 CEST] <kierank> yes many developers in project are very very paranoid
[14:10:58 CEST] <thardin> paranoia is fine if it's informed and has a valid threat model
[14:12:45 CEST] <durandal_1707> general incompetence
[16:37:10 CEST] <durandal_1707> cehoyos: the configure line is for codecs that are dependent on other code
[16:37:31 CEST] <cehoyos> How do you define "dependent"?
[16:38:06 CEST] <cehoyos> Two solutions are possible, in this case we have two codecs with not a single small function different, why does it make sense to increase the maintenance burden?
[16:38:18 CEST] <durandal_1707> that they cannot be build without other codec
[16:38:50 CEST] <durandal_1707> this is not about maintenace burden
[16:39:55 CEST] <durandal_1707> this is about simple following codecs logic
[16:40:30 CEST] <durandal_1707> look how all adpcm or pcm variants are done
[16:48:16 CEST] <durandal_1707> they certainly do  not depend on some ultimate  pcm/adpcm codec
[17:03:22 CEST] <cehoyos> But the adpcm codecs do not share 100% of their code
[17:15:56 CEST] <durandal_1707> cehoyos: that is irrelevant
[17:16:25 CEST] <durandal_1707> your change is incorrect
[17:23:35 CEST] <durandal_1707> there is no need for dependency
[17:25:13 CEST] <durandal_1707> because decoder does not use explicitly another decoder
[20:59:14 CEST] <durandal_1707> there is startup that uses ffmpeg and claims its 4x  times faster than facebook filter
[23:59:58 CEST] <thardin> so there's already tickets about the cfr -> vfr thing. didn't know that
[00:00:00 CEST] --- Thu Aug 22 2019