Commit Graph

190 Commits

Author SHA1 Message Date
Baptiste Coudurier e9338da87d fix endianness of time code, and calculate it in fields, according to specs
Originally committed as revision 16730 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-23 20:03:57 +00:00
Baptiste Coudurier 507ff760af Do not use avctx->frame_number which might not be set,
this also fixes a side effect where its value was one too much.

Originally committed as revision 16701 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-20 09:43:58 +00:00
Baptiste Coudurier 15a4e8b85a set correct value for starting line
Originally committed as revision 16698 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-20 09:25:50 +00:00
Baptiste Coudurier fc94187857 fix gxf time code mark out value
Originally committed as revision 16697 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-20 09:14:46 +00:00
David Conrad 4bcafc9a23 Update regression tests after r16651
Originally committed as revision 16656 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-17 18:34:47 +00:00
Baptiste Coudurier 3d0a94f699 calculate pat, pmt and sdt exact size and adjust total bitrate
Originally committed as revision 16610 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-14 23:49:13 +00:00
Baptiste Coudurier 5541759f6f only write dts if they differ from pts, according to specs
Originally committed as revision 16605 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-14 22:17:07 +00:00
Baptiste Coudurier 46911c7ab8 use delay to avoid negative dts
Originally committed as revision 16604 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-14 22:14:35 +00:00
Baptiste Coudurier de34dc39c0 set pts and dts of pes packets exactly according to specs
Originally committed as revision 16603 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-01-14 21:57:10 +00:00
Aurelien Jacobs abce34d979 matroskaenc: require global header
fixes issue696

Originally committed as revision 15667 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-22 21:14:42 +00:00
Michael Niedermayer 3e3c638e69 Switch regression tests to swscale.
Plain C, x86-32 and -64 have been tested and should work, other
archs that had asm optmizations in swscale likely will need some fixes
to either fall back on C if SWS_BITEXACT is set or make the asm match C.
This also disables the PAL8 test as neither swscale nor the old scaler
really support PAL8 output, imgconvert supported a fixed 666 palette
as output and swscale supports fixed 884 and 422.

Originally committed as revision 15305 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-09-13 03:49:54 +00:00
Baptiste Coudurier 9b92ed7666 update swf video frame number when muxing done, fix #439
Originally committed as revision 14293 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-19 03:25:50 +00:00
Michael Niedermayer 4c453ddbbc Make our flac decoder decode all the data at EOF.
Fixes issue524

Originally committed as revision 14223 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-13 21:51:29 +00:00
Michael Niedermayer 7efeb73a5a 10000l for myself, my last change was incomplete.
This one also reverts the libav mpeg ts regression test score change.

Originally committed as revision 14132 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-08 23:25:01 +00:00
Michael Niedermayer 6d77d9acf7 Set cur_dts to 0 only during creating new streams and not every time cur_dts happens
to be AV_NOPTS_VALUE.

Originally committed as revision 14118 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-08 14:07:37 +00:00
Baptiste Coudurier 148c8d8018 store dts in ffm headers
Originally committed as revision 13880 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-22 08:50:57 +00:00
Baptiste Coudurier 5e5f75cf8a remove pts set hack, this is no more correct
Originally committed as revision 13879 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-22 08:44:58 +00:00
Baptiste Coudurier dbedf2aae2 enable feeder threads
Originally committed as revision 13868 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-22 01:35:43 +00:00
Baptiste Coudurier 54e5cf132d do not reset ffm pts, this avoids setting last pts to 0 and screwing up seeking
Originally committed as revision 13856 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-22 00:26:25 +00:00
Baptiste Coudurier b63ba850c1 remove obsolete requirements, quicktime 6 is from 2002, not sure if muxing scheme still fits
Originally committed as revision 13732 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-09 19:56:45 +00:00
Baptiste Coudurier 3c6a9f66b9 pass dts and pts through ffm, should fix streaming b frames
Originally committed as revision 13725 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-09 07:43:14 +00:00
Baptiste Coudurier 2940b38ef2 fix pts handling in ffm
Originally committed as revision 13683 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-07 00:49:03 +00:00
Baptiste Coudurier 995877956f update regression tests for ts due to r13641
Originally committed as revision 13651 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-03 17:59:24 +00:00
Michael Niedermayer bf1310c7b1 Update regression checksums after dts calculation fix (r13470).
Before, all dts where wrong, now just the first is a little odd.

Originally committed as revision 13491 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-28 10:25:20 +00:00
Baptiste Coudurier 0bf017c48b update regression ref due to r13412
Originally committed as revision 13427 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-26 08:22:38 +00:00
Henrik Gulbrandsen 4fa6eac5f1 Correct ASF duration.
Patch by Henrik Gulbrandsen: henrik gulbra net

Originally committed as revision 12916 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-04-21 08:19:16 +00:00
Baptiste Coudurier 3a71577f87 homogenous regression tests output for image formats
Originally committed as revision 12639 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 21:29:38 +00:00
Baptiste Coudurier 3251f24be7 missed one
Originally committed as revision 12638 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 21:28:52 +00:00
Baptiste Coudurier 7261f9bd25 cosmetics, remove leading whitespaces
Originally committed as revision 12637 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 21:21:49 +00:00
Baptiste Coudurier e516d3488a change rle encoder to count up to 127, sgi does not support 128
Originally committed as revision 12631 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 17:27:58 +00:00
Justin Ruggles fe790f89e8 Use 15-bit LPC precision by default. This generally gives better compression
results. Also update Ogg regression test.

Originally committed as revision 12630 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 16:01:10 +00:00
Justin Ruggles 6e330816d4 fix Ogg/FLAC header size
Originally committed as revision 12626 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-30 03:40:03 +00:00
Panagiotis Issaris 781138d005 Update the regression tests reference values after the changes to the Matroska
muxer/demuxer in revision 12358.

Originally committed as revision 12407 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-10 15:19:52 +00:00
Michael Niedermayer e942293db1 It appears last_pts was not set ...
Originally committed as revision 11939 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-15 13:21:00 +00:00
Michael Niedermayer a712d725c0 Header elision muxing support.
This only removes 2 bytes from MP3 and MP2 currently.
Up to 4 could be removed from MP3/MP2 though this might need a 2pass muxer.
Primitive code to remove headers from MPEG-1/2/4 is there too but for the
single file I tried it on (the one in the regression tests), it was a loss
because all video frames were >4096 byte, so that it is disabled ATM.

Originally committed as revision 11936 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-15 02:57:19 +00:00
Benoit Fouet 3f8e1fc80c Update regression test checksum for ffm.
Originally committed as revision 11933 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-14 09:20:53 +00:00
Baptiste Coudurier 37b74c8651 Fix issue 338:
Correctly interleave ogg packets per granule
and set eos correctly, 2 packets buffering is needed.
It duplicates interleave_per_dts a bit,
if someone has a good solution, I'll implement it.

Originally committed as revision 11867 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-05 01:16:47 +00:00
Oded Shimon 4ddcc983b4 back_ptr calculation used wrong timebase when searching for back syncpoint
Originally committed as revision 11856 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-04 10:48:41 +00:00
Oded Shimon 02c986a8d6 Fix intented algo in syncpoint writing before keyframes in NUT muxer
Originally committed as revision 11855 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-04 10:45:32 +00:00
Oded Shimon 82badd27f1 nut->max_distance was not set at all anywhere in the NUT muxer
Originally committed as revision 11854 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-04 10:34:35 +00:00
Oded Shimon 103d74ea5c ff_nut_reset_ts() expected to get 'ts*time_base_count', but muxer only
gave it 'ts'. Fixed by changing ff_nut_reset_ts() and demuxer params

Originally committed as revision 11851 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-04 10:27:32 +00:00
Måns Rullgård 2009721d99 split lavf regression tests
Originally committed as revision 11833 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-03 19:26:29 +00:00
Måns Rullgård 4383ed5cba split regression tests allowing make to run them in parallel
Originally committed as revision 11832 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-03 19:26:26 +00:00
Michael Niedermayer 175e23e875 Consider initial EOR in the back_ptr calculation.
Originally committed as revision 11550 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-01-17 13:59:52 +00:00
Michael Niedermayer 7548783bc5 correct invalid duration
Originally committed as revision 11152 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-12-03 10:23:09 +00:00
Kostya Shishkov ec28509fed Fix regression tests after r11127 and 11129
Patch by kostya

Originally committed as revision 11128 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-12-02 17:46:27 +00:00
Baptiste Coudurier 5118cae694 ogg muxer
Originally committed as revision 10999 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-11-10 18:27:03 +00:00
David Conrad 7a297ceb8f Update mkv regression tests after r10421
Originally committed as revision 10430 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-09-07 15:58:08 +00:00
David Conrad 46491f138c Matroska regression tests
Originally committed as revision 10391 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-09-05 01:05:54 +00:00
Neil Brown 9fcbcca60a If a stream has no start time, but the first packet has a 'pts', use that
pts to set the start_time.
patch by Neil Brown: [neilb suse de]
original thread: [FFmpeg-devel] [patch 3/3] Make timing calculations less
dependant on start_time being defined.
date: 08/16/2007 08:27 AM

Originally committed as revision 10285 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-09-03 07:56:26 +00:00