* commit '3f7fd59d151a2773f0e2e93e56b6b13ec6e5334b':
avformat: fix typo in avformat_close_input
mp3enc: write Xing TOC
mp3enc: support MPEG-2 and MPEG-2.5 in Xing header.
mp3enc: downgrade some errors in writing Xing frame to warnings
lavf: flush the output AVIOContext in av_write_trailer().
lavf: cosmetics, reformat av_write_trailer().
avio: flush the internal buffer in avio_close()
Enhance doc on asyncts audiofilter
cmdutils: avoid setting data pointers to invalid values in alloc_buffer()
libavcodec: remove av_destruct_packet_nofree()
Conflicts:
libavcodec/avpacket.c
libavformat/mp3enc.c
libavformat/nutenc.c
libavformat/utils.c
libavformat/version.h
tests/ref/lavf/voc
tests/ref/lavf/voc_s16
Merged-by: Michael Niedermayer <michaelni@gmx.at>
These arguments are either constants or copies of MpegEncContext
fields just as easily accessed within the function.
Signed-off-by: Mans Rullgard <mans@mansr.com>
These functions do not benefit from being inlined. They are large,
and there are no opportunities for constant propagation.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Some implementations of sscanf do not handle a space before a trailing %n
properly.
As an example, MSVC's does this for the second insatnce in this patch, for
an input of "0x3:c0=c1:c1=c0":
1) Match the final "c0" or "c1".
2) Realize it's at the end of the string.
3) Check for %n.
4) There is no %n, but a space instead.
5) Leave 'len' unitilialized.
So, move it out of the sscanf format strings, and call skip_spaces instead.
This bug does not affect skip_spaces since %n is the first and only formatting
string.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
A proper implementation was introduced in
ba53720280 for MSVC, and
MinGW already has vsnprintf.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Previously, object files in, for example, compat/ were left
after a clean or distclean was run.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Remove a bogus warning when using segment_list_type csv.
The LIST_TYPE_EXT constant is only used internally, so it can
be removed when the feature (segment_list_type ext) gets removed.
Signed-off-by: Alexander Strasser <eclipse7@gmx.net>
Should also be faster (though I doubt that hardly ever matters
for the usage here).
Also remove the pointer copy. Since we do not need to reset the
pointer to the start of the string, it is not needed anymore.
Signed-off-by: Alexander Strasser <eclipse7@gmx.net>
This updated version does not deviate from previous behavior on default value of 'buffer_size'
I skipped porting 'sources', 'block' options for now as they're parsed seriously. So i added TODO remarks.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
It is all dead and rotting code.
Also, move yuvcmp.c to tools. Its license is non-existent, however,
and should be clarified.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
The YUV channels of VP6 are encoded in a highly linear fashion which does
not have any slice-like concept to thread. The alpha channel of VP6A is
fairly independent of the YUV and comprises 40% of the work. This patch
uses the THREAD_SLICE capability to split the YUV and A decodes into
separate threads.
Two bugs are fixed by splitting YUV and alpha state:
- qscale_table from VP6A decode was for alpha channel instead of YUV
- alpha channel filtering settings were overwritten by YUV header parse
Signed-off-by: Ben Jackson <ben@ben.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Rather than cleverly managing frame pointers with swaps to avoid
re-using "golden" frames, just do brute-force management of the
4 AVFrames. New strategy is probably no more costly and is easier
to adapt to threaded usage.
Signed-off-by: Ben Jackson <ben@ben.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Makes golden_frame more like other frame data, paves way for threading
alpha channel decode.
Signed-off-by: Ben Jackson <ben@ben.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Based on the code by:
Peter Belkner <pbelkner@snafu.de>,
Michael Niedermayer <michaelni@gmx.at>,
Clément Bœsch <clement.boesch@smartjog.com>,
Reimar Döffinger <Reimar.Doeffinger@gmx.de>, and
Tobias Rapp <t.rapp@noa-audio.com>
Alex Converse <alex.converse@gmail.com>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This is consistent with stdio and is what we want to do in all cases.
Fixes a bug in the voc muxer which didn't flush in write_trailer()
previously. This is the cause of the change in the test results.
According to the sample for trac ticket #1722, PGS subtitles
are decoded from several packets at the same DTS and varying PTS.
The PTS from the presentation segment seem to be the valid one;
in particular, clear subtitles are too early with the other PTS.
If the packet has a PTS and the corresponding time base is known,
set the pts field of the decoded subtitle structure before the
call to the decoder. The decoder is still allowed to change the
PTS if necessary.
Japanese DTV uses some non standard extensions in AAC audio.
One example is 'dual mono', which combines two independent
audio into one stereo stream, storing them in left and right channels
respectively. Historically, dual mono audio has been used for
multi-lingual audio, one for local/native language, and another for english,
and usually the "main" (local language) channel should be output without
any user interactions.
The frames of those dual mono audio are allowed to set
ADTS channel_config field to 0, and just contain two SCE's *WITHOUT* PCE,
which is a non standard extension by Japanese DTV standard.
(ref. ARIB STD-B32 PartII 5.2.3)
This patch adds an AVPacket side data, AV_PKT_DATA_JP_DUALMONO,
which indicates that the AVPacket is likely to contain an audio frame
with the above dual mono extension, and has the parameter to specify
the desired channel selection in that case.
It also makes aacdec to detect dual mono and output just the desired
channel when this side data is attached.
Signed-off-by: Akihiro Tsukada <atsukada@users.sourceforge.net>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
