Commit Graph

10601 Commits

Author SHA1 Message Date
Måns Rullgård
c5d6cd5c81 ARM: 1l c&p fix: do not set pred16x16_plane for rv40
Originally committed as revision 20705 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 17:21:43 +00:00
Måns Rullgård
702b5885a1 ARM: NEON optimised H264 16x16, 8x8 pred
Originally committed as revision 20704 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 14:56:45 +00:00
Alex Converse
4fab662769 AAC: Add a global header but output not locked output configuration status.
Originally committed as revision 20701 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 09:12:05 +00:00
Alex Converse
38610d92ff Set the SBR mode to implicit on ADTS frames occurring before the output is locked.
Originally committed as revision 20700 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 09:07:32 +00:00
Mike Melanson
9d8bb0318a Small refactoring: Instead of 4 loops for decoding AC coefficients based
on their grouping, create one loop that indexes into a table of AC VLC
tables.

There is also a small optimization here: Do not call unpack_vlcs()
if there are no fragments in the list with outstanding coefficients.
My profiling indicates that this can save upwards of 1 million
dezicycles per frame throughout the course of unpack_dct_coeffs().

Originally committed as revision 20699 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 04:06:27 +00:00
Mike Melanson
098523eb28 Use a list to track which fragments coded in this frame still have
outstanding coefficients yet to be decoded from the bitstream. Once a
fragment reaches end-of-block, remove it from this new list. This change
makes the VP3/Theora entropy decode process dramatically faster due to
not having to iterate incessantly over fragments which have already been
fully decoded.

Originally committed as revision 20698 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 02:30:15 +00:00
Måns Rullgård
5dad039bf7 ARM: small tweak of NEON H264 IDCT
Originally committed as revision 20697 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 00:37:39 +00:00
Måns Rullgård
1025d19dd7 ARM: NEON 2xN chroma MC
Originally committed as revision 20696 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 00:37:36 +00:00
Måns Rullgård
04e7f6d2d0 ARM: NEON 16x16 and 8x8 avg qpel MC
Originally committed as revision 20695 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-02 00:37:33 +00:00
Baptiste Coudurier
28f2db9670 fix typo, ALT_BITSTREAM_WRITER_LE does not exist
Originally committed as revision 20692 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 22:35:25 +00:00
Michael Niedermayer
94f28061f5 Store original width/height so that rv20 does not get stuck with some
resolution. A sample file to show a difference is at issue833 (http://airfarce.com/video/000211fh.rm)

Originally committed as revision 20691 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 22:30:03 +00:00
Michael Niedermayer
e7e578b73c Make rv1/2 decoder use the next slice if needed.
Fixes issue782.

Originally committed as revision 20690 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 21:10:37 +00:00
Michael Niedermayer
63538a960e Support ASCII pnms.
Implements issue1452.

Originally committed as revision 20687 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 16:56:13 +00:00
Jai Menon
de3553b078 pngdec.c : release allocated buffers.
Originally committed as revision 20680 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 11:38:37 +00:00
Michael Niedermayer
483aad717e slif aka "SoftLab MPEG-2 I-frames Codec" support.
Flipping one bit still is popular it seems as a means to create a new codec.
fixes issue1469.

Originally committed as revision 20679 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 02:07:46 +00:00
Michael Niedermayer
f40a7fd316 4:2:2 4:4:4 bugfix for the error concealment code.
Originally committed as revision 20678 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-12-01 01:01:37 +00:00
Stefano Sabatini
5f1836a7de Rename parameters of put_sbits() to make them consistent with those of
put_bits().

Originally committed as revision 20677 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 23:52:28 +00:00
Michael Niedermayer
52069c4d3c Move "slice below image" check from mpeg_decode_slice to its caller.
Should fix issue1277.

Originally committed as revision 20676 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 23:08:25 +00:00
Michael Niedermayer
d52b4abe8b Move dummy picture allocation code from mpeg1/2 to mpegvideo.
This fixes a infinite loop on a b frame.

Originally committed as revision 20672 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 21:22:01 +00:00
Michael Niedermayer
fd1ef13bb4 Disable error resilience for field pictures, this was never supported,
results where more or less random but should not have crashed.

Originally committed as revision 20671 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 19:15:19 +00:00
Michael Niedermayer
078cdecf9e Set mb_y in mpeg2 field pictures like h264 does.
This fixes -vismv & -debug 16384 with field pictures.

Originally committed as revision 20670 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 19:14:00 +00:00
Michael Niedermayer
22d4f21331 Make ""MPEG motion vector out of boundary" message more verbose.
Originally committed as revision 20669 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 17:29:21 +00:00
Attila Kinali
76c4a644ee Fix an issue uncovered by commit 20623:
The init functions of mpc7 and mpc8 check whether the vlc has been
initialized already and return early if this is the case (eg by calling
init a second time).
But avctx->sample_fmt and channel_layout is set after the vlc initialization,
causing it not to be set on the second call of init.

Move all manipulations of avctx before the initialization of the vlc,
so that it is always set.

Originally committed as revision 20668 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 10:25:20 +00:00
Michael Niedermayer
ab8c48de8c Start decoding from seq/gop too, not just I frames.
Possibly fixes issue1124.

Originally committed as revision 20667 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 03:32:43 +00:00
Michael Niedermayer
f0ec239438 Fix dual prime motion compensation in field pictures.
Fixes issue1125.

Originally committed as revision 20666 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-30 02:37:03 +00:00
Stefano Sabatini
587edd6af8 Use more consistent / meaningful parameter names for the
ff_copy_bits() function.

Originally committed as revision 20663 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 23:01:29 +00:00
Måns Rullgård
b8cef7be5c nellymoser: use constant seed for dithering RNG
Originally committed as revision 20658 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 15:15:53 +00:00
Måns Rullgård
2d2e72b10e cook: use constant seed for dithering RNG
Originally committed as revision 20657 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 15:15:50 +00:00
Michael Niedermayer
aeaef4ed63 Print errors instead of just ignoring contradictionary values in mpeg2 silently.
Originally committed as revision 20656 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 14:31:40 +00:00
Michael Niedermayer
87f6806069 Override a few values read so as to ensure that things are not inconsistent.
Originally committed as revision 20654 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 03:32:30 +00:00
Michael Niedermayer
d9b99556bf Restructure dummy frame allocation.
New code also allocates a dummy frame for an p field after an i field,
previously that could segfault due to reading from NULL+x.

Originally committed as revision 20653 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 03:05:11 +00:00
Michael Niedermayer
9d13016690 Make sure field_select is not set to nonsensical values even if unused.
Fixed an assert failure.

Originally committed as revision 20646 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-29 01:50:43 +00:00
Michael Niedermayer
1c3260939d Factorize "s->first_slice = 1" out.
Originally committed as revision 20645 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 23:17:46 +00:00
Michael Niedermayer
88eba670fb Also check the location of the picture start code and picture coding
extension a little.

Originally committed as revision 20644 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 23:05:06 +00:00
Michael Niedermayer
fc23d843e1 Check order of startcodes, ignore some obviously wrong ones.
Fixes issue487.

Originally committed as revision 20643 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 22:51:53 +00:00
Michael Niedermayer
bbf266fdb5 Move mpeg_decode_postinit() out of bitstream decoding functions, it does not
belong in there.

Originally committed as revision 20642 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 21:32:33 +00:00
Michael Niedermayer
1206f1d630 Inline mpeg_decode_extension, will simplify future changes.
Originally committed as revision 20641 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 18:57:12 +00:00
Michael Niedermayer
191ad11e91 Simplify mpeg_decode_user_data()
Originally committed as revision 20640 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 14:17:28 +00:00
Stefano Sabatini
8d872e8ac9 Fix put_bits32() doxy.
Originally committed as revision 20639 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 10:58:07 +00:00
Stefano Sabatini
643570d6bb Consistently put braces for function definitions.
Originally committed as revision 20638 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 10:21:10 +00:00
Stefano Sabatini
e74223610b Fix typo, "Write" -> "Writes".
Originally committed as revision 20637 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 10:17:43 +00:00
Stefano Sabatini
106c72faad Use pixel formats names defined in av_pix_fmt_descriptors rather than
in those in pix_fmt_info, and remove the not anymore used
PixFmtInfo.name field.

Originally committed as revision 20636 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 01:13:43 +00:00
Stefano Sabatini
f72d40b01a Remove unused PixFmtInfo.is_hwaccel field.
Originally committed as revision 20635 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-28 00:53:15 +00:00
Reimar Döffinger
a7c6e76638 Add support for hardcoding the motionpixels rgb to yuv table.
Originally committed as revision 20627 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 21:20:28 +00:00
Reimar Döffinger
7c7d3abf3c 10l, pix_fmt should be set by the motionpixels decoder, not by the demuxer.
This fixes playback when lavf and lavc are less tightly couples as in e.g. MPlayer.

Originally committed as revision 20626 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 21:15:06 +00:00
Stefano Sabatini
797229a6a2 Make avcodec_get_pix_fmt_name() use av_pix_fmt_descriptors rather than
pix_fmt_info.

Originally committed as revision 20625 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 21:04:33 +00:00
Jai Menon
0bf095a837 Allow decoders to correctly set the sample format by defaulting to
SAMPLE_FMT_NONE in avcodec_get_context_defaults2.

Originally committed as revision 20623 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 20:46:13 +00:00
Stefano Sabatini
14b903f389 Remove x_chroma_shift and y_chroma_shift fields from PixFmtInfo, use
av_pix_fmt_descriptors log2_chroma_w and log2_chroma_h values instead.

Originally committed as revision 20622 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 20:00:03 +00:00
Måns Rullgård
4f898e77fb Revert "Add required header #include to fix 'make checkheaders'."
config.h must not be included in that file.  The table generator runs
on the host system, but config.h describes the target.

Originally committed as revision 20620 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 10:06:41 +00:00
Diego Biurrun
e137934fae cosmetics: K&R style, prettyprinting
Originally committed as revision 20619 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 09:59:10 +00:00
Diego Biurrun
c73ce8e259 Mark apply_mdct() function as static; it is only used within the file.
Originally committed as revision 20618 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 09:55:03 +00:00
Stefano Sabatini
38264abb06 Use av_pix_fmt_descriptors in avcodec_get_chroma_sub_sample(), rather
than the PixFmtInfo x_chroma_shift and y_chroma_shift fields.

Originally committed as revision 20617 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-26 00:33:19 +00:00
Stefano Sabatini
8e861e1b1e Use av_pix_fmt_descriptors in ff_is_hwaccel_pix_fmt() rather than
access the PixFmtInfo.is_hwaccel field which is going to be removed.

Originally committed as revision 20616 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-25 23:52:20 +00:00
Diego Biurrun
ff09a99ef5 -DFRAC_BITS must be added to CPPFLAGS when compiling mpegaudio_tablegen.ho.
Originally committed as revision 20613 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-25 23:29:36 +00:00
Diego Biurrun
154f5b5c9b Add required header #include to fix 'make checkheaders'.
Originally committed as revision 20612 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-25 23:24:09 +00:00
Ramiro Polla
f2526204a4 Fix nv12/nv21 handling. linesize for plane 1 should account for both chroma
planes instead of just doubling the height while computing plane sizes. Also
adjust avpicture_layout() to copy the correct amount of data for plane 1.

Originally committed as revision 20610 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-25 17:14:48 +00:00
David Conrad
800841fd0c Support compiling against libtheora older than 1.1
Originally committed as revision 20609 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-25 16:00:14 +00:00
Vitor Sessak
691a4232ee Make lsp2polyf() function non-static for upcoming usage in SIPR
Originally committed as revision 20602 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-24 16:58:50 +00:00
Jason Garrett-Glaser
096c87f667 Add weightp support in API for libx264.
ffmpeg can encode baseline profile again.

Originally committed as revision 20600 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-24 15:49:34 +00:00
Reimar Döffinger
f3bdc3da15 Call ff_find_hwaccel() after calling avcodec_set_dimensions().
Patch by Reimar

Originally committed as revision 20599 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-24 15:13:47 +00:00
Måns Rullgård
8b4a6d47b2 WMA: extend exponent range to 95
Hopefully this will be enough.  Fixes issue 1565 (again).

Originally committed as revision 20598 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-24 10:10:39 +00:00
Kostya Shishkov
d90aeeaf56 Call avcodec_set_dimensions() instead of simply setting avctx->width/height
when frame dimensions change in RV3/4.

Originally committed as revision 20595 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-24 06:05:41 +00:00
Vitor Sessak
9401357f01 Add missing include. Fix the following compiler warnings:
twinvq.c: In function 'decode_lsp':
twinvq.c:575: warning: implicit declaration of function 'ff_sort_nearly_sorted_floats'

Originally committed as revision 20594 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-23 22:28:45 +00:00
Måns Rullgård
168f92ffaa WMA: extend exponent table up to 75
Fixes issue 1565.

Originally committed as revision 20593 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-23 22:22:16 +00:00
Michael Niedermayer
4c66e8849b Use avcodec_set_dimensions()
Originally committed as revision 20591 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-23 19:45:40 +00:00
Michael Niedermayer
e026902a62 Correct order of arguments for avcodec_check_dimensions().
Originally committed as revision 20590 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-23 19:24:17 +00:00
Michael Niedermayer
4687f908c6 Make sure avcodec_set_dimensions() is used when rv20 changes resolution.
This should fix lowres resolution changes.

Originally committed as revision 20589 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-23 19:22:19 +00:00
Vitor Sessak
419b2be813 Make sorting function used in TwinVQ a shared function
Originally committed as revision 20584 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 22:25:58 +00:00
David Conrad
6c2289b52e Set colorspace for libtheora encoding
Originally committed as revision 20583 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 21:08:46 +00:00
David Conrad
af89b1781f Support 4:2:2 and 4:4:4 subsampling in libtheora encoding
Originally committed as revision 20582 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 21:08:43 +00:00
David Conrad
b0a6d9ca86 Add support for two pass encoding in libtheora
Originally committed as revision 20581 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 21:08:40 +00:00
David Conrad
bdc8c48893 Update libtheora wrapper to use the 1.0 API
Originally committed as revision 20580 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 21:08:37 +00:00
Kostya Shishkov
ec10d2d539 Update dimensions in AVCodecContext when RV3/4 frame dimensions change
Originally committed as revision 20572 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-22 07:48:35 +00:00
Alex Converse
4e878a1898 10l: Fix inverted if-condition from r20448. Fixes issue 1562.
Originally committed as revision 20570 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-21 20:56:46 +00:00
Robert Krüger
ce34ff6b45 Factorize av_get_bit_rate (for future use outside of libavcodec).
Patch by Robert Krüger, krueger signal7 de

Originally committed as revision 20569 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-21 09:58:15 +00:00
Michael Niedermayer
98c82d6910 Allocate pictures with enough padding for jpeg.
Ensure that jpeg does not use mbs that could require larger padding.
This might have been exploitable.

Originally committed as revision 20566 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 22:14:37 +00:00
Reimar Döffinger
9fe0894210 In win32 thread implementation do not access jobnr if the thread is asked
to terminate, jobnr it does not point to a valid location in that case.

Originally committed as revision 20565 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 21:08:26 +00:00
Daniel Verkamp
0aa6a518ad Add codec id for Flash screen codec v2 and hook it up in FLV demuxer
Originally committed as revision 20563 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 19:16:42 +00:00
Ramiro Polla
b16569d256 Indent
Originally committed as revision 20562 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 11:38:37 +00:00
Kostya Shishkov
76267e4e90 Implement missing case for decoding samples with large pivot value in APE
decoder.
This fixes issue 1555

Originally committed as revision 20560 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 07:49:53 +00:00
Kostya Shishkov
e4de5b0fb5 cosmetics: break some long lines and insert few spaces
Originally committed as revision 20559 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 07:26:17 +00:00
Kostya Shishkov
dd562761d9 10l trocadero: ZMBV encoder used zero score to detect whether block should be
XORed with source one or not, which went wrong with new block comparing code.
So track this condition explicitly.

Originally committed as revision 20558 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 07:22:41 +00:00
Baptiste Coudurier
cb893cf386 avoid integer overflow in dnxhd encoder, fixes #1557
Originally committed as revision 20557 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-20 00:13:34 +00:00
Baptiste Coudurier
a4fcd9966b clarify why the encoder is failing
Originally committed as revision 20556 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-19 23:28:21 +00:00
Stefano Sabatini
bf72597ef4 Fix name for PIX_FMT_NV21 PixFmtInfo entry.
Originally committed as revision 20555 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-19 22:56:57 +00:00
Reimar Döffinger
b711aaa2d8 Disable image flipping during JPEG decoding if CODEC_FLAG_EMU_EDGE is set
instead of either aborting on the assert or crashing due to writing beyond
the array due to insufficient padding.

Originally committed as revision 20552 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-19 11:49:03 +00:00
Baptiste Coudurier
8a724ed84d Revert r20548, it was a dream. It seems faad does not ouput the last frame
in any case.

Originally committed as revision 20551 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-19 00:05:30 +00:00
Baptiste Coudurier
bceae2936f faad needs CODEC_CAP_DELAY, otherwise you miss last frame
Originally committed as revision 20548 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-18 21:01:42 +00:00
Diego Biurrun
c173693698 Add functions to return library license and library configuration.
Originally committed as revision 20547 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-18 17:15:17 +00:00
Ronald S. Bultje
6e44ba1550 Use get_bits_left() instead of size_in_bits - get_bits_count().
Originally committed as revision 20543 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-16 17:42:43 +00:00
Carl Eugen Hoyos
830628e11c Cosmetics: Fix indentation.
Originally committed as revision 20542 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-16 01:56:47 +00:00
Alex Converse
b3507d398b Pad the extradata in the AAC ADTS to AudioSpecificConfig bitstream filter.
Originally committed as revision 20541 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-15 19:40:44 +00:00
Vitor Sessak
b2bf3cfe32 Implement ff_decode_pitch_lag() that is used by both AMR and SIPR.
Based on code written by Colin McQuillan during his SoC project.

Originally committed as revision 20540 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-15 10:41:46 +00:00
Thilo Borgmann
25c4fddadc Merge data tables from als_data.h with the decoder source to reduce
code complexity.

Originally committed as revision 20536 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-14 11:49:48 +00:00
Thilo Borgmann
93d38cf6b0 Add long-term prediction to the ALS decoder.
Originally committed as revision 20534 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-14 06:29:19 +00:00
Justin Ruggles
7934cbb3be ac3dec: apply dynamic range compression to correct channels in dual-mono mode.
Originally committed as revision 20533 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-14 00:25:48 +00:00
Mike Scheutzow
eda4ea4e2e Reset rbsp buffer size when freeing its data buffer.
Patch by Mike Scheutzow $(name) AT alcatel 'minus' lucent com

Originally committed as revision 20531 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-13 18:33:28 +00:00
Vitor Sessak
6870aefa40 Bump minor version after MPEG-4 Lossless commit
Originally committed as revision 20528 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-12 20:21:38 +00:00
Carl Eugen Hoyos
c5c2df9423 Bump minor version after applying ALS decoder.
Originally committed as revision 20527 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-12 20:08:17 +00:00
Alex Converse
07580347b9 Mark AOT_ALS as a supported MPEG-4 audio object type.
Originally committed as revision 20524 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-12 05:22:51 +00:00
Alex Converse
9349e5582c Cosmetics: Re-indent
Originally committed as revision 20522 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-12 01:45:04 +00:00
Alex Converse
0bb622ba4d Fix a __warn_memset_zero_len gcc-4.4 warning.
Originally committed as revision 20521 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-12 01:44:38 +00:00
Thilo Borgmann
99971952b0 Add MPEG-4 Audio Lossless Coding (ALS) decoder.
Patch by Thilo Borgmann, thilo D borgmann A googlemail

Originally committed as revision 20517 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 22:36:17 +00:00
Thilo Borgmann
24026a2d49 Add the ability to parse the ALSSpecificConfig from an MPEG-4 AudioSpecificConfig. Patch by Thilo Borgmann
<com googlemail borgmann thilo> as part of GSoC 2009.

Originally committed as revision 20516 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 22:16:54 +00:00
Alex Converse
e92ad83f54 Cosmetics: Remove the implied gap before AOT_USAC in the MPEG-4 AOT enum.
Originally committed as revision 20514 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 20:36:42 +00:00
Alex Converse
01d19fbc9f AAC: Rename sect_len to sect_end. It marks the last sfb of the section, not the
length of the section.

Originally committed as revision 20513 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 20:33:56 +00:00
Alex Converse
5aea268d14 Add a channels field to MPEG4AudioConfig.
Originally committed as revision 20512 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 20:24:23 +00:00
Vitor Sessak
f2b7ce8ae6 Use doubles in ff_set_min_dist_lsf(). After this patch, AMR can use this
function with no change in output.

Originally committed as revision 20511 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 11:00:50 +00:00
Alex Converse
4312b9367a Add Low Delay MPEG Surround to the enum of MPEG-4 Audio Object Types.
Originally committed as revision 20510 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-11 01:04:08 +00:00
Justin Ruggles
3155716677 Add CH_LAYOUT_NATIVE channel mask value to allow user to request native codec
channel order from the decoder rather than the default FFmpeg channel order.

Originally committed as revision 20508 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 21:53:22 +00:00
Michael Niedermayer
67e362ca45 indent
Originally committed as revision 20507 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 21:00:46 +00:00
Carl Eugen Hoyos
de9bc64df4 Cosmetics: Fix indentation after r20505.
Originally committed as revision 20506 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 20:55:45 +00:00
Carl Eugen Hoyos
602dd2d359 Add .pix_fmts to H264 VDPAU decoder declaration and remove obsolete test
for codec->capabilities.

Originally committed as revision 20505 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 20:55:00 +00:00
Michael Niedermayer
44be1d6478 Try to fix has_b_frame calculation for open gop files.
Fixes issue1523

Originally committed as revision 20504 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 20:24:41 +00:00
Vitor Sessak
51ae2142d5 Fix palette copying in av_picture_copy(). Previous code worked only if
linesize[1] == 1, which does not make sense.

Originally committed as revision 20503 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 19:36:58 +00:00
NVIDIA Corporation
70e0c871eb Add VDPAU hardware accelerated decoding for MPEG-4 ASP which can be used
by video players.

Original patch by NVIDIA corporation.

Originally committed as revision 20502 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 18:52:39 +00:00
Jai Menon
09bad7e3ca Handle JPEG2000 frames stored in the Quicktime container.
Fixes issue 1525.

Originally committed as revision 20500 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 14:48:32 +00:00
Jai Menon
b70b158833 cosmetics : fix indentation.
Originally committed as revision 20499 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 14:37:42 +00:00
Vitor Sessak
00fa73f052 Avoid variable-length array use in ff_acelp_lspd2lpc()
Originally committed as revision 20496 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 04:17:18 +00:00
Michael Niedermayer
dccda293a0 Add a NULL pointer check to avcodec_close() this should prevent a segfault
when closing without open.

Originally committed as revision 20495 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-10 02:51:47 +00:00
Ronald S. Bultje
c47ca25e74 Make get_bits_left() available for use in libavcodec (was previously held
private in dv.c for some reason). See "[PATCH] get_bits_left()" thread.

Originally committed as revision 20490 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-09 22:10:43 +00:00
Ronald S. Bultje
b1ade3d182 And on the sixth day, God gave us the <= operand, which makes the loop
"for (i = 1; i < num + 1; i++)" look like "for (i = 1; i <= num; i++)".
Programmers worldwide rejoiced and used the operand, e.g. in the thread
"[PATCH] remove ugliness in cel_filters.c", and He saw that it was good.

Originally committed as revision 20489 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-09 22:04:03 +00:00
Vitor Sessak
e26d131845 Do not hardcode filter order in ff_acelp_lspd2lpc()
Originally committed as revision 20485 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-09 12:06:19 +00:00
Francesco Lavra
91cc5d3767 Move ff_reverse in libavcodec to av_reverse in libavutil.
Patch by Francesco Lavra, francescolavra interfree it

Originally committed as revision 20484 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-09 09:11:35 +00:00
Carl Eugen Hoyos
adadf26bb9 Use enum FrameType as parameter to construct_perm_table().
Originally committed as revision 20480 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-08 23:37:59 +00:00
Michael Niedermayer
a16f80f392 Remove unused variable.
1l for me if it wherent sunday and i could buy cola.

Originally committed as revision 20477 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-08 14:54:43 +00:00
Justin Ruggles
187bc061e4 ac3dec: revert r20089. The change was just wrong. 10l for me.
Originally committed as revision 20473 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-08 03:06:05 +00:00
Michael Niedermayer
f3001e0dee Simpliy exp*table init, based on patch by reimar.
Originally committed as revision 20470 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-07 20:10:02 +00:00
Reimar Döffinger
1ffc6e8327 Add ff_init_ff_cos_tabs function and use it in rdft.c to ensure that the
necessary ff_cos_tabs tables are initialized.
Fixes issue 1507 (QDM2 broken since r20237 without hardcoded tables).

Originally committed as revision 20464 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-05 18:29:06 +00:00
Reynaldo H. Verdejo Pinochet
136086523f Correct double include introduced by r20450
Originally committed as revision 20452 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-05 01:10:27 +00:00
Vitor Sessak
c902340599 Reduce stack memory allocation in MP3 decoder
Originally committed as revision 20451 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-04 23:33:08 +00:00
Reynaldo H. Verdejo Pinochet
0c50f8e6cc Implement ff_scale_vector_to_given_sum_of_squares()
to aid generic gain control routines. Changes
for qcelp are included. Patch Collin McQuillan.

Originally committed as revision 20450 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-04 19:29:29 +00:00
Reimar Döffinger
223217746c Pad ff_cos_tabs and ff_sin_tabs so that index n points to the table for n bits.
While this "wastes" up to 2x32 bytes it makes the code slightly simpler and
less confusing.

Originally committed as revision 20449 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-04 17:30:23 +00:00
Alex Converse
981b8fd777 Don't lock the channel output configuration based on the first value seen for
non extradata formats. Instead lock it only after the successful decoding of a
frame. This fixes issue 999.

Originally committed as revision 20448 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-03 22:50:02 +00:00
Andrew Wason
42022e0362 add yuvs fourcc to raw.c and isom.c, fixes samples/mov/yuvs.mov
patch by Andrew Wason

Originally committed as revision 20439 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-03 16:17:21 +00:00
Justin Ruggles
31f8c10983 Simplify AC-3 critical band end calculation (correctly this time).
Originally committed as revision 20438 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-03 03:26:26 +00:00
Justin Ruggles
94593c9b52 Revert r20050, which caused decoding errors with at least one valid AC-3 file.
Originally committed as revision 20437 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-03 02:59:18 +00:00
Michael Niedermayer
ad9feac9fb Avoid huge array for rgb ljpeg on the stack.
Code tested with an ljpeg i had laying around.

Originally committed as revision 20433 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-02 14:12:25 +00:00
Ben Littler
e47ca4f798 map VYUY fourcc to rawcodec
fixes http://samples.mplayerhq.hu/V-codecs/ATI-VCR12/ATI/y422test.avi

Originally committed as revision 20432 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-02 00:10:10 +00:00
Justin Ruggles
4c1645a458 Set sample format in flac_decode_init() rather than in
ff_flac_parse_streaminfo() since the latter is shared with the raw FLAC
demuxer, which should not be setting the sample format.

Originally committed as revision 20430 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-01 16:48:05 +00:00
Justin Ruggles
dee34af4d6 Set the sample format for Smacker audio in the decoder rather than in the
demuxer.

Originally committed as revision 20429 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-11-01 15:08:49 +00:00
Vitor Sessak
083c5a48ec Implement AMR gain function that is used by both AMR and SIPR.
Based on AMR SoC code by Robert Swain and Colin McQuillan.

Originally committed as revision 20421 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-31 02:02:30 +00:00
Michael Niedermayer
8a08175f3f Change whitespace placement a little to improve readabiliy slightly.
Thats not an attempt to emulate indent -kr behavior down to the finest
fineprint, first it would not be worth the work, second it would be less
readable, third it would  not be K&R but indent -kr then.

Originally committed as revision 20416 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-29 10:58:59 +00:00
Michael Niedermayer
f92e769666 Multiple inclusion guards for synth_filter.h.
Originally committed as revision 20415 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-29 10:46:29 +00:00
Francesco Lavra
e58992e41f Support demuxing teletext in DVB streams.
Original patch by Francesco Lavra, francescolavra interfree it

Originally committed as revision 20414 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-29 10:46:16 +00:00
Michael Niedermayer
d886e4df10 Vertically align function arguments.
Originally committed as revision 20413 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-29 10:43:56 +00:00
Måns Rullgård
de4070b520 Add generic rules for building helpers with HOSTCC
Originally committed as revision 20412 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-29 00:55:35 +00:00
Reimar Döffinger
7a41027cc4 vorbis_dec: factor out duplicated code for index reading, bounds checking and
error message into a shared macro.

Originally committed as revision 20411 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-28 22:41:45 +00:00
Reimar Döffinger
3db0208c42 Change format string from "x" to PRIx32 to hopefully fix a warning with Cygwin gcc 4.2.4.
Originally committed as revision 20407 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-28 20:02:40 +00:00
Måns Rullgård
785d3c97b2 Makefile cosmetics
Originally committed as revision 20406 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-10-28 19:49:19 +00:00