[Ffmpeg-devel-irc] ffmpeg-devel.log.20160417

[burek]

[01:00:14 CEST] <cone-359> ffmpeg 03Rostislav Pehlivanov 07master:77fb7177afd6: vc2enc: don't require interlacing for 1080p50/60 base video formats
[01:00:14 CEST] <cone-359> ffmpeg 03Rostislav Pehlivanov 07master:9e138c481306: vc2enc: use 32x16 slices by default
[01:36:02 CEST] <cone-359> ffmpeg 03Michael Niedermayer 07master:618032da0d05: fate: add test for Ticket4816 (PNG decoding with alpha)
[02:04:49 CEST] <cone-359> ffmpeg 03Michael Niedermayer 07master:487c346d98c0: avcodec/exr: Fix "libavcodec/exr.c:1494:13: warning: ISO C90 forbids mixed declarations and code"
[02:33:18 CEST] <cone-359> ffmpeg 03James Almer 07master:b63ba3a51782: avformat/hashenc: add missing avio_flush to hash_write_trailer
[02:41:41 CEST] <cone-359> ffmpeg 03Ivan 07master:3a727606c474: avcodec/h264: Fix for H.264 configuration parsing
[11:32:37 CEST] <cone-744> ffmpeg 03Paul B Mahol 07master:7a0aee16881d: avcodec/takdec: fix decoding of some sample rates with multichannel coder
[13:19:31 CEST] <cone-925> ffmpeg 03Michael Niedermayer 07master:c8d95e5ceb72: avfilter/colorspacedsp_template: Add argument protecting () to avg()
[13:34:08 CEST] <Daemon404> "Reproduces a bug to remain consistent with libvpx' behaviour." :(
[13:37:52 CEST] <atomnuker> why won't libvpx fix the bug themselves?
[13:37:53 CEST] <wm4> just following the spec
[13:38:27 CEST] <wm4> atomnuker: maybe it'd break some youtube videos
[13:49:48 CEST] Action: Daemon404 wonders if this is the entire OS/2 community on the ML
[14:07:42 CEST] <Compn> haha
[14:07:47 CEST] <Compn> they are active though
[14:08:00 CEST] <Compn> more active than some other archs...
[14:22:30 CEST] <BtbN> how is OS/2 even still alive?
[14:23:02 CEST] <Daemon404> because korea
[14:24:30 CEST] <atomnuker> which korea?
[14:27:23 CEST] <Daemon404> the one who is allowed to post on the internet
[14:29:02 CEST] <JEEB> Daemon404: at least one of the guys is clearly running it in a VM
[14:29:20 CEST] <JEEB> but yes, quite funky that it's still alive there
[15:24:05 CEST] <Compn> best korea has no need for internet
[15:46:52 CEST] <cone-925> ffmpeg 03Ronald S. Bultje 07master:6135aed0d2b2: vf_scale: support bt2020 in in/out_color_matrix properties.
[15:46:53 CEST] <cone-925> ffmpeg 03Ronald S. Bultje 07master:ee729cc0ed0b: vp9: ignore frame_context_index field in key-/intraonly-frames.
[18:23:18 CEST] <cone-795> ffmpeg 03Michael Niedermayer 07master:6d160afab2fa: avfilter/vf_readvitc: Use avpriv_frame_get_metadatap() avoid direct access to AVFrame.metadata (to comply to API)
[19:06:40 CEST] <Daemon404> nevcairiel, https://github.com/dwbuiten/FFmpeg/commit/fef5e35ae84665ff3a948fedec1154d8a110329f
[19:06:46 CEST] <Daemon404> does this look correct to you
[19:06:59 CEST] <Daemon404> (i am uncertain about the last changes from return 1 to 0)
[19:07:18 CEST] <nevcairiel> the return value is just normal API returns now,0 or averror
[19:07:20 CEST] <Daemon404> er, ignore the commented out bit too
[19:07:22 CEST] <Daemon404> im rming that
[19:07:26 CEST] <Daemon404> ok
[19:07:31 CEST] <nevcairiel> EAGAIN in case the bsf eats a packet
[19:07:38 CEST] <nevcairiel> but this one doesnt
[19:07:58 CEST] <Daemon404> yes
[19:10:35 CEST] <Daemon404> nevcairiel, anything look obviously wrong?
[19:11:09 CEST] <nevcairiel> i usually preferred to get rid of buf/buf_size and reference in->data directly, but not like I care
[19:11:38 CEST] <Daemon404> i can do that; it's just cosmetics
[19:12:14 CEST] <nevcairiel> posted two comments
[19:14:24 CEST] <Daemon404> nevcairiel, av_packet_copy_props is called directly after new_pacjet, right?
[19:14:32 CEST] <nevcairiel> makes no difference when
[19:14:36 CEST] <Daemon404> ok
[19:14:42 CEST] <nevcairiel> just after new packet
[19:15:22 CEST] <Daemon404> fixes for both comments pushed
[19:15:37 CEST] <nevcairiel> posted a third one =p
[19:15:57 CEST] <nevcairiel> dont think copy props can fail
[19:16:07 CEST] <Daemon404> it can
[19:16:18 CEST] <Daemon404> https://www.ffmpeg.org/doxygen/2.4/avpacket_8c_source.html#l00561
[19:16:33 CEST] <Daemon404> at least we have code that checks if it can
[19:21:00 CEST] <Daemon404> also fixed
[19:21:31 CEST] <Daemon404> im not sure if out must be freed on error or not
[19:21:39 CEST] <nevcairiel> it must
[19:21:42 CEST] <nevcairiel> you have ownership
[19:21:49 CEST] <nevcairiel> well, if you allocated anything
[19:21:55 CEST] <nevcairiel> if its untouched then it d oesnt matter
[19:22:08 CEST] <Daemon404> all right
[19:22:25 CEST] Action: Daemon404 really couldnt care less about learning bsf internals here, just wants to finish it
[19:23:20 CEST] <Daemon404> tehre
[19:23:50 CEST] <Daemon404> i find it odd that the filter must free the in packet, though
[19:30:40 CEST] <nevcairiel> this way it could buffer it if it uses it partially
[19:30:43 CEST] <nevcairiel> it takes ownership
[19:30:58 CEST] <Daemon404> hmm ok
[19:31:02 CEST] <Daemon404> any other issue left?
[19:38:08 CEST] <nevcairiel> didnt see anything
[19:38:12 CEST] <nevcairiel> but no clue what that thing even does
[19:38:12 CEST] <Daemon404> ok
[19:48:57 CEST] <cone-795> ffmpeg 03Anton Khirnov 07master:33d18982fa03: lavc: add a new bitstream filtering API
[19:48:58 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:af9cac1be175: Merge commit '33d18982fa03feb061c8f744a4f0a9175c1f63ab'
[19:49:09 CEST] <Daemon404> rcombs, ^
[19:50:47 CEST] <cone-795> ffmpeg 03Anton Khirnov 07master:4426540f0c3e: avconv: switch to the new BSF API
[19:50:48 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:98e3153fa3a1: Merge commit '4426540f0c3ee516662f79d0a6ab5b95503b6611'
[19:52:51 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:6202e2fede75: indeo4: Rework stream analysis report
[19:52:52 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:1117d6f4b122: Merge commit '6202e2fede75df92cbc374a3f7d6893d0c5ac721'
[19:53:25 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:64250d94b74d: indeo4: Consistently initialize variables
[19:53:26 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:d9357332ae9a: Merge commit '64250d94b74d3fd47cc8b1611f48daf6a6ed804a'
[19:55:01 CEST] <petrurares> Hi there! I want to save a new frame side data, concretely AVMasteringDisplayMetadata. I tried to put some random data in doc/examples/decoding_encoding.c example but I can't retrieve back such data
[19:55:11 CEST] <petrurares> any idea how to do this? I want to make a test but firstly I need to know how to do that. Thanks :)
[20:06:54 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:709c0f79d803: nuv: Use the correct context for av_image_check_size
[20:06:55 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:dff9894acdf8: Merge commit '709c0f79d8032fcf733bfe58e79ca7ff0858c8bc'
[20:12:02 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:0c6a70873fc6: intrax8: Move error resilience out of intrax8
[20:12:03 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:750562549cee: intrax8: Wrap multiline macros in do{}while(0) clauses
[20:12:04 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:42244ce07a1f: intrax8: Move a comment to the place it corresponds
[20:12:05 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:909f5e71001f: Merge commit '0c6a70873fc6e43194b471d112c30823b6c8d0b4'
[20:12:06 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:e445b80c741c: Merge commit '750562549ceef268b29b94f6a887d9cf331a8c78'
[20:12:07 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:2c3e38f44fd2: Merge commit '42244ce07a1f4f5108ae86d50fe72db785d483ed'
[20:14:22 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:6f5ff559dbd0: intrax8: Adjust printf conversion specifier for sizeof expression
[20:14:23 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:7120bff0a326: Merge commit '6f5ff559dbd01fa4ea96bc0a1d2e0c21f2db8d13'
[20:28:14 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:2ade1cdafb96: intrax8: K&R formatting cosmetics
[20:28:15 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:ba5bcf96124a: Merge commit '2ade1cdafb96bf47e77f7ed74731d78a30aae950'
[20:30:17 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:ad8aa8e6c648: intrax8: Move documentation from implementation to header file
[20:30:18 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:c29ac4abb818: Merge commit 'ad8aa8e6c648b61f01b9f106f27b9d4f3d094345'
[20:32:11 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:0372e73f917e: intrax8: Check and propagate errors from ff_intrax8_common_init
[20:32:12 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:f64b53f2bf5f: Merge commit '0372e73f917e72c40b09270f771046fc142be4a7'
[20:32:52 CEST] <cone-795> ffmpeg 03Vittorio Giovara 07master:d909f43b5c77: vc1dec: wmv2dec: Validate ff_intrax8_common_init return value
[20:32:54 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:61400a1dc751: Merge commit 'd909f43b5c773a73c8d526638744547ba4aa8c59'
[20:37:22 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:8dead2aaca4a: Move const qualifier before type name
[20:37:23 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:48362ceadeb2: doc: Update paths to match new examples location
[20:37:24 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:9d4799c247e3: Merge commit '8dead2aaca4aa8b84b77b05745755afb56b7d37a'
[20:37:25 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:b0c75963c87d: Merge commit '48362ceadeb2eb5286ae94ef7f9542d990ff7ec7'
[20:40:21 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:30e9ef21cea0: timefilter-test: Only compile timefilter-test if JACK is enabled
[20:40:22 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:65a802401c6c: build: Add component for the SRTP common code
[20:40:23 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:9b36f446acbf: Merge commit '30e9ef21cea09fa5e880e979c9f5b39edccbb6f4'
[20:40:24 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:8f540b86125a: Merge commit '65a802401c6cc136576bb2e613c0577cbf622aa8'
[20:41:06 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:52385410b7de: des-test: Move a variable declaration to avoid an ifdef
[20:41:07 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:4b23a8137fd3: Merge commit '52385410b7de012c2562fb5ba8f07a4e1112c4e7'
[20:44:52 CEST] <cone-795> ffmpeg 03Carl Eugen Hoyos 07master:1106a2c5dd7e: lavc/webp: Print a warning for unsupported chunks.
[20:46:19 CEST] <cone-795> ffmpeg 03Diego Biurrun 07master:a84713e70d06: parseutils-test: Move some variable declarations to avoid block braces
[20:46:20 CEST] <cone-795> ffmpeg 03Derek Buitenhuis 07master:d97a61a8f132: Merge commit 'a84713e70d06238100cb0f867ad6031c272054b0'
[21:26:25 CEST] <JEEB> wm4: I love how YUV_TO_RGB1_CCIR et al are completely undocumented unless you go search for the values
[21:30:45 CEST] <wm4> they also can't be used without carefully looking what these macros expand to
[21:31:45 CEST] <JEEB> yeah
[21:32:14 CEST] <cone-795> ffmpeg 03Thilo Borgmann 07master:37a4d3383fb9: lavfi/perspective: Add basic timeline editing.
[21:32:31 CEST] <JEEB> which explains why my initial question regarding lack of explanation of those macros 
[21:32:38 CEST] <JEEB> *-lack
[21:32:42 CEST] <JEEB> *-lack of
[21:33:01 CEST] <JEEB> didn't get any real reply :P
[00:00:00 CEST] --- Mon Apr 18 2016