[Ffmpeg-devel-irc] ffmpeg-devel.log.20181222

burek burek021 at gmail.com
Sun Dec 23 03:05:04 EET 2018


[00:16:12 CET] <cone-779> ffmpeg 03Carl Eugen Hoyos 07master:6a87729e9da5: lavc/opus_rc: Cast a const pointer to uint8_t *.
[01:26:01 CET] <cone-779> ffmpeg 03James Almer 07master:53d3a1c514bf: APIChanges: add missing entry for SMPTE2094-40 HDR Dynamic metadata API
[08:15:12 CET] <pross> how to describe that feeling you get when your decoder output matches the reference implementation for the first time?
[08:15:40 CET] <BradleyS> i believe the technical term is squee
[08:16:08 CET] <BradleyS> congrats
[11:16:24 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:297e65c676e3: avformat/vividas: check if value from ffio_read_varlen() is too big
[11:16:24 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:14c2f7f65260: avformat/vividas: fix undefined behaviour
[11:16:24 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:e9909fe19418: avformat/vividas: check if size is enough big in xor_block
[13:37:42 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:d62cb29716a5: avformat/vividas: check if viv->sb_pb is not NULL
[15:06:38 CET] <JEEB> I kind of remember us being OK with initializing an iterator within a for loop?
[15:07:22 CET] <JEEB> and if we needed variables within each iteration we could put that in the beginning of the for?
[15:08:57 CET] <atomnuker> yes
[15:09:39 CET] <JEEB> ok, I just took a look at the patch by the hlsenc guy and he weirdly stuck everything in the beginning of the function
[15:11:24 CET] <atomnuker> what's with the sea of dashenc and hlsenc patches and the total lack of dashdec patches?
[15:12:02 CET] <JEEB> well just writing something without an XML parser in the middle is much simpler ;)
[15:12:16 CET] <JEEB> also people generally just want to push to some HTTP end point
[15:12:33 CET] <JEEB> since it seems like some video platforms are starting to support DASH things through HTTP
[15:12:55 CET] <atomnuker> isn't dash http only?
[15:12:58 CET] <JEEB> yes
[15:13:11 CET] <JEEB> also I guess no further comments on https://patchwork.ffmpeg.org/patch/11430/ ?
[16:21:23 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:7ea4b928a264: avfilter/af_sofalizer: fix non-power of 2 IR length filtering in time domain
[17:43:03 CET] <cone-641> ffmpeg 03Michael Niedermayer 07master:4325527e1c4f: postproc/postprocess_template: Avoid using %4 for the threshold compare
[17:43:04 CET] <cone-641> ffmpeg 03Michael Niedermayer 07master:c1cbeb87db4b: postproc/postprocess_template: remove FF_REG_sp from clobber list
[18:47:32 CET] <philipl> what's the best tool to use if I want to look at the SPS/PPS of an h.264 stream?
[18:47:39 CET] <philipl> Will ffprobe do it?
[18:49:39 CET] <kierank> philipl: h264parse
[18:49:41 CET] <kierank> in bmxlib
[18:53:43 CET] <philipl> kierank: thanks.
[18:54:08 CET] <jamrial> philipl: or use the trace_headers bitstream filter
[19:01:58 CET] <cone-641> ffmpeg 03Marton Balint 07master:dfd656632e89: avformat/avio: fix avio_feof documentation
[19:09:15 CET] <philipl> Thanks guys. I was trying to see the differences between two streams to see if I could append them or not. Turns out I didn't need to dig that deep. One was Main and one was High...
[19:27:05 CET] <JEEB> alright, no further comments I see.
[19:27:29 CET] <cone-641> ffmpeg 03Mark Thompson 07master:b97a4b658814: cbs_av1: Fix reading of overlong uvlc codes
[19:30:10 CET] <cone-641> ffmpeg 03Mark Thompson 07release/4.1:b420f2356682: cbs_av1: Fix reading of overlong uvlc codes
[19:39:31 CET] <cone-641> ffmpeg 03Jan Ekström 07master:a1b4f120c031: avformat/mpegts: unset DTS/PTS for subtitle PES packets if PCR not available
[20:52:37 CET] <cone-641> ffmpeg 03Paul B Mahol 07master:7c201e420acc: avfilter/af_headphone: fix filtering of non-power of 2 length IRs in time domain
[20:55:18 CET] <durandal_1707> atomnuker: are you any near finishing non-power of 2 FFT ?
[20:57:23 CET] <atomnuker> yes, the blocker is figuring out how to modify the twiddles, is all
[21:07:49 CET] <atomnuker> I need to bug bofh to tell me again like he promised
[22:34:17 CET] <cone-641> ffmpeg 03Michael Niedermayer 07master:8b53d1322fef: avformat/mxfdec: Do not process zero modified_date timestamp.
[00:00:00 CET] --- Sun Dec 23 2018


More information about the Ffmpeg-devel-irc mailing list