burek021 at gmail.com
Sat Aug 15 02:05:02 CEST 2015
[00:37:51 CEST] Action: BBB pokes jeeb again
[00:38:00 CEST] <BBB> JEEB: did you have any time to look at the #libav logs?
[01:08:11 CEST] <cone-056> ffmpeg 03Mariusz SzczepaDczyk 07master:affbecb0de8a: lavf/ftp: implement NLST method
[01:43:12 CEST] <philipl> I find it amusing/terrifying that the stagefright vulnerability is specifically regarding handling of mp4 subtitles - my GSoC intern's project. There's lots of room for goofed up size checks in there.
[01:58:21 CEST] <wget> Hi guys. Does anyone know what the FFmtech was about? http://ffmpeg.org/pipermail/ffmtech-board-election/2011-June/000001.html I can only find such emails from MLs, but cannot have a clear definition. WWas that related to libav?
[02:54:44 CEST] <Compn> wget
[02:55:56 CEST] <Compn> wget : it was supposed to be some kind of multimedia funding organization to pay coders to work on the projects
[02:56:08 CEST] <Compn> which they did pay for some coders to work on some ffmpeg/libav code
[02:56:31 CEST] <Compn> it was funded by companies who broke the gpl and settled, i think
[04:23:07 CEST] <BBB> wget: I founded it, its basically what compn said, it was a non-profit to hold money from donations and other things so we could offer reimbursements for expenses and pay in small ways for particular types of work wed like to get done
[04:23:48 CEST] <BBB> wget: I dont know what the current status is, last thing I know, reimar was on the board
[04:23:54 CEST] <BBB> wget: so he might know something
[04:24:07 CEST] <BBB> wget: there were also some libav poeple on the board, e.g. ruggles
[04:24:42 CEST] <BBB> wget: I think in theory it was supposed to be shared between ffmpeg/libav, although in practice Im not sure if anyone agrees that worked out well...
[05:01:13 CEST] <cone-667> ffmpeg 03Michael Niedermayer 07master:02fff499362d: avformat/rmdec: Move MLTI handling out of ff_rm_read_mdpr_codecdata()
[05:01:13 CEST] <cone-667> ffmpeg 03Michael Niedermayer 07master:856452cf63af: avformat/rmdec: MLTI with multiple MDPR support
[05:04:54 CEST] <Compn> lol i didnt know bbb founded it
[10:44:35 CEST] <cone-538> ffmpeg 03Michael Niedermayer 07master:ae2cb9290ac2: ffmpeg_opt: Prefer audio streams with packets
[10:50:50 CEST] <kvz> Hi. I sent an email too root with what I think is a very interesting a hosting proposal for FFmpeg, but didn't hear back yet. Should I assume all is settled and you are not interested, or did something go wrong and do you want to checkout the proposal still? I can put it up in a gist if you want?
[10:55:30 CEST] <michaelni> kvz, there where 2 mails about it, it seems you where not CC-ed, ill forward them to you (2nd contains the first in full quote i think)
[10:56:05 CEST] <michaelni> feel free to reply to root or the ffmpeg-admins list on SF if you like
[10:57:03 CEST] <kvz> ok thanks!
[12:15:26 CEST] <cone-538> ffmpeg 03Stefano Sabatini 07master:819dc7aa0be9: doc/demuxers/mpegts: update documentation
[12:15:27 CEST] <cone-538> ffmpeg 03Stefano Sabatini 07master:90f5668323e6: doc/demuxers/mpegts: apply a minor grammar fix to option description
[12:15:28 CEST] <cone-538> ffmpeg 03Stefano Sabatini 07master:fedc42268429: lavf/mpegts: apply misc option description fixes
[12:30:43 CEST] <lglinskih> kierank: hi! If I want to compare results of my test with ref files from FATE, I need to convert frame data from fltp sample_fmt to s16. Is it ok to use libswresample?
[12:31:02 CEST] <lglinskih> Or should I make it simpler and just use my own ref file with raw data in fltp format?
[13:27:28 CEST] <kierank> lglinskih: hmm I don't know actually
[13:27:33 CEST] <kierank> I think you have to use avresample
[13:36:44 CEST] <nevcairiel> pastebin aint too happy with 5mb of text
[13:55:16 CEST] <cone-538> ffmpeg 03Michael Niedermayer 07master:b2e95e012c3c: avcodec/faxcompr: Factor decode_uncompressed() out
[13:55:17 CEST] <cone-538> ffmpeg 03Michael Niedermayer 07master:012784052a33: avcodec/faxcompr: negate uncompressed runs
[13:59:07 CEST] <michaelni> kierank, lglinskih, why avresample ?
[13:59:17 CEST] <kierank> swresample I mean
[13:59:31 CEST] <michaelni> ok
[13:59:55 CEST] <michaelni> lglinskih, if you have questions /need help with swresample use, ping me
[14:15:17 CEST] <thardin> hm, time to push that MXF patch I think
[14:15:22 CEST] <thardin> don't know why I've been putting it off
[14:17:41 CEST] <thardin> can git apply .eml files?
[14:22:41 CEST] <thardin> fatal: remote error: access denied or repository not exported: /ffmpeg.git
[14:23:05 CEST] <BtbN> are you trying to push via git://?
[14:23:31 CEST] <thardin> yes, I should perhaps change to ssh
[14:23:48 CEST] <BtbN> git:// is read-only, as it does not support any form of authentication.
[14:24:27 CEST] <thardin> git at source.ffmpeg.org:ffmpeg.git ?
[14:25:00 CEST] <thardin> fetch works
[14:25:15 CEST] <thardin> --dry-run looks ok, I think
[14:26:54 CEST] <cone-538> ffmpeg 03Marton Balint 07master:95584ddac477: mxfdec: set AVFMT_SEEK_TO_PTS demuxer flag
[14:27:00 CEST] <thardin> there we go
[15:03:25 CEST] <cone-538> ffmpeg 03Sven Dueking 07master:f3fbe790d9d4: avcodec/qsvenc: Set MaxKpbs to rc_max_rate for CBR and VBR (bitrate is equal to rc_max_rate for CBR)
[15:59:00 CEST] <klaxa> i'm having trouble exposing tcp AVOptions through http. i read the part about nesting on https://www.ffmpeg.org/doxygen/2.2/group__avoptions.html but i seem to fail to implement the child_class_next function
[15:59:22 CEST] <klaxa> i'm also not sure if my child_next function is correct, i'll pastebin an excerpt from my http.c
[16:00:36 CEST] <klaxa> https://gist.github.com/klaxa/2c3e25912fd03b22997a
[16:01:33 CEST] <klaxa> grepping through the source i found an implementation in AVFilter, but that has the advantage of keeping a list of AVFilterContexts, from which you can just take the first context and then return its child's avclass
[16:03:28 CEST] <klaxa> another related problem i have is trying to store a socket's address in sockaddr_storage and adding that to the AVOptions of TCPContext, but there is no AVOptionType for that (duh)
[16:10:58 CEST] <michaelni> child_class_next has to iterate over the classes of the childs
[16:11:17 CEST] <michaelni> for sock address, i guess you want to use a char* string
[16:13:45 CEST] <klaxa> so i pass the address just as a string like "126.96.36.199:1234"? or let it point to the binary sockaddr_storage?
[16:15:02 CEST] <michaelni> pass like a string and parse into whatever struct you need, is probably easiest
[16:15:22 CEST] <michaelni> one could also extend AVOptions but thats likely overkill if thats the only place where its needed
[16:16:14 CEST] <klaxa> so after i read it with av_opt_get() i cast it to sockaddr_storage or whatever other sockaddr_* struct i want?
[16:16:29 CEST] <michaelni> for the class ff_tcp_protocol.priv_data_class can probably be used
[16:17:40 CEST] <michaelni> I dont remember off the top of my head how to turn a char* into a sockaddr whatever
[16:18:49 CEST] <klaxa> so after i read it with av_opt_get() i cast it to sockaddr_storage or whatever other sockaddr_* struct i want?
[16:18:54 CEST] <klaxa> whoops
[16:18:56 CEST] <klaxa> ah, i pass a human readable string then, that seems simple enough, thanks!
[17:31:31 CEST] <cone-538> ffmpeg 03Paul B Mahol 07master:4b0e11244287: avfilter/avf_avectorscope: add polar mode
[19:22:23 CEST] <philipl> Does anyone have thoughts on critera for a 2.8 release? My selfish motive is to get a release out with vdpau hevc for kodi 16 to consume.
[19:22:39 CEST] <philipl> Are there specific features that are pending or in development that we want to see in it?
[19:22:43 CEST] <BtbN> I'm currently working on vaapi hevc for kodi 16.
[19:22:46 CEST] <philipl> The API cleanup, I assume.
[19:23:26 CEST] <philipl> BtbN: You got hardware? It was on my list for after I got a skylake box.
[19:23:44 CEST] <BtbN> Got a Braswell NUC from kodi.
[19:23:53 CEST] <philipl> How kind of them.
[19:24:50 CEST] <BtbN> Haven't yet started, as the system is still setting itself up.
[19:24:59 CEST] <BtbN> But i plan to get stuff done this weekend
[19:25:04 CEST] <philipl> Cool. That's another good thing to wait for, for sure.
[19:25:41 CEST] <philipl> videotoolbox is kind of done? Looks like there's still discussion around async mode going on.
[19:26:28 CEST] <BtbN> No idea
[20:24:09 CEST] <durandal_1707> ubitux: what's with selectivecolor?
[20:26:34 CEST] <ubitux> durandal_1707: can't figure out how the black slider works
[20:26:55 CEST] <ubitux> i need more data but it's really painful to figure out
[20:27:02 CEST] <ubitux> s/figure out/to make some/
[21:17:57 CEST] <durandal_1707> what you guys think about another avf filter abalancemeter
[21:26:34 CEST] <rcombs> explain further?
[21:43:12 CEST] <durandal_1707> shows stereo audio balance
[21:51:19 CEST] <rcombs> I'd like to see something that shows levels on all channels instead of just stereo, along with their names
[21:51:45 CEST] <rcombs> for examining surround files without actually using a surround setup
[21:56:50 CEST] <durandal_1707> rcombs: tried showvolume?
[21:58:15 CEST] <rcombs> oh, huh
[21:58:37 CEST] <rcombs> now I need to work out how to actually use that
[22:02:51 CEST] <rcombs> wm4: is there any way to use multimedia lavfi filters in mpv
[22:03:07 CEST] <rcombs> e.g. showvolume (which does audio->video)
[22:04:35 CEST] <wm4> no
[22:04:57 CEST] <wm4> only the libavdevice lavfi thing
[22:04:59 CEST] <wm4> which sucks
[22:08:56 CEST] <rcombs> wtb
[22:33:41 CEST] <durandal_1707> wm4: you mean complex filtergraph can't be implemented?
[22:33:53 CEST] <durandal_1707> In mpv?
[22:34:19 CEST] <wm4> I guess it could
[22:34:54 CEST] <wm4> but it'd be complicated due to the assumption that audio decoding results in audio output etc.
[23:09:03 CEST] <BBB> JEEB: poke again, did you have time to look at the #libav logs? I really dont mind if you just give me any file you have where I need to do significant effort to parse it myself, thats totally fine
[23:11:02 CEST] <wm4> poor BBB
[23:11:13 CEST] <BBB> ?
[23:11:29 CEST] <wm4> you've been trying to poke JEEB for the logs for months
[23:11:36 CEST] <BBB> less than a week :-p
[23:11:49 CEST] <BBB> if anyone else has logs thats fine also
[23:11:57 CEST] <BBB> but nobody seems to have #libav logs...
[23:12:33 CEST] <wm4> mine actually start in oct 2014
[23:12:38 CEST] <wm4> (not sure what happened)
[23:13:56 CEST] <ubitux> i have them from 2013 at least
[23:14:09 CEST] <ubitux> also 2012 actually
[23:14:28 CEST] <BBB> hurray! were saved at last
[23:14:41 CEST] <BBB> canihavepleaz?
[23:15:07 CEST] <ubitux> i also have from 14 march 2011, but in yet another format
[23:15:19 CEST] <BBB> thats ok, I can convert or count separately
[23:15:22 CEST] <ubitux> basically 3 differents formats
[23:15:25 CEST] Action: wm4 wonders what BBB is going to do with them
[23:15:31 CEST] <ubitux> statistics
[23:15:32 CEST] <BBB> wm4: do some statistics
[23:15:33 CEST] <ubitux> ?
[23:15:35 CEST] <ubitux> :D
[23:15:37 CEST] <ubitux> bingo
[23:15:53 CEST] <ubitux> gimme a moment to archive all of that
[23:16:00 CEST] <BBB> \o\ /o/ \o/ ty
[00:00:00 CEST] --- Sat Aug 15 2015
More information about the Ffmpeg-devel-irc