virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2026-06-19 19:03:00 +02:00
Code Issues Releases Activity
32,949 Commits 39 Branches 425 Tags
777365fe8602d4b92e2664647d0e5fe2418d4bf6
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Justin Ruggles 777365fe86 xa: set correct bit rate 
Also fixes stream duration calculation.
 2012-03-20 14:12:54 -04:00
Justin Ruggles a54bc52265 xa: do not set bit_rate, block_align, or bits_per_coded_sample 
The values in the header refer to decoded data, not compressed data.
 2012-03-20 14:12:53 -04:00
Justin Ruggles 64de57f645 xa: fix end-of-file handling 
Do not output an extra packet when out_size is reached.
Also return AVERROR_EOF instead of AVERROR(EIO).
 2012-03-20 14:12:53 -04:00
Justin Ruggles cd2ffb67ad xa: fix timestamp calculation 
The packet duration is always 28 samples.
 2012-03-20 14:12:53 -04:00
Justin Ruggles 6ca3b248e2 resample: allocate a large enough output buffer 
Fixes invalid writes and crashes when doing conversions such as stereo to 5.1
channels or sample rate conversion on 5.1 channels.
 2012-03-17 13:37:44 -04:00
Justin Ruggles c3da9f5060 wmaenc: remove bit-exact hack 
It may have improved cross-platform stability, but wasn't the only place in
the encoder with bitexact issues. It is no longer needed because we have FATE
tests for float encoders using fuzzy comparison.
 2012-03-17 11:46:15 -04:00
Justin Ruggles 85cf49fab7 FATE: remove WMA acodec tests 2012-03-17 11:46:15 -04:00
Justin Ruggles 3a1e453e54 FATE: add WMAv1 and WMAv2 encode/decode tests with fuzzy comparison 2012-03-17 11:33:35 -04:00
Justin Ruggles a4cf4ef256 FATE: add AC-3 and E-AC-3 encode/decode tests with fuzzy comparison 2012-03-17 11:33:35 -04:00
Justin Ruggles 90e5b58a53 FATE: add capability for audio encode/decode tests with fuzzy psnr comparison 
This allows for testing floating-point audio encoders across different
platforms where exact comparisons are unreliable due to float rounding
differences.
 2012-03-15 17:06:17 -04:00
Justin Ruggles cffd7580bb FATE: allow a tolerance in the size comparison in do_tiny_psnr() 
This will allow for comparing decoded output to the original source when the
decoded size is not exactly the same as the original size.
 2012-03-15 14:40:31 -04:00
Justin Ruggles bb6842966e FATE: use absolute difference from a target value in do_tiny_psnr() 
This will allow comparison to original pre-encoded content instead of
comparing to expected decoded output.
 2012-03-15 14:40:31 -04:00
Justin Ruggles 5ecadc6620 FATE: allow tests to set CMP_SHIFT to pass to tiny_psnr 
This will allow adjusting for any encoder or decoder delay when doing
comparisons.
 2012-03-15 14:40:24 -04:00
Justin Ruggles 0720d263ea FATE: use $fuzz directly in do_tiny_psnr() instead of passing it around 2012-03-15 12:06:56 -04:00
Justin Ruggles 94cf64b81f cosmetics: reindent 2012-03-05 13:08:19 -05:00
Justin Ruggles 8c1d6ac66a avformat: do not require a pixel/sample format if there is no decoder 
Also, do not keep trying to find and open a decoder in try_decode_frame() if
we already tried and failed once.

Fixes always searching until max_analyze_duration in
avformat_find_stream_info() when demuxing codecs without a decoder.
 2012-03-05 13:08:18 -05:00
Justin Ruggles a7fa75684d avformat: do not fill-in audio packet duration in compute_pkt_fields() 
Use the estimated duration only to calculate missing timestamps if needed.
 2012-03-05 13:08:18 -05:00
Justin Ruggles 6c65cf58fd lavf: Use av_get_audio_frame_duration() in get_audio_frame_size() 
Also, do not give AVCodecContext.frame_size priority for muxing.

Updated 2 FATE references:
dxa-feeble - adds 1 audio frame that is still within 2 seconds as specified
             by -t 2 in the FATE test
wmv8-drm-nodec - durations are not needed. previously they were estimated
                 using the packet size and average bit rate.
 2012-03-05 13:08:18 -05:00
Justin Ruggles 11ac796f7a dca_parser: parse the sample rate and frame durations 2012-03-05 13:08:17 -05:00
Justin Ruggles 85469f1c9e libspeexdec: do not set AVCodecContext.frame_size 
It is not necessary
 2012-03-05 13:08:17 -05:00
Justin Ruggles adbf61ff8a libopencore-amr: do not set AVCodecContext.frame_size 
It is not necessary
 2012-03-05 13:08:17 -05:00
Justin Ruggles 0efa4073bf alsdec: do not set AVCodecContext.frame_size 
It is not necessary
 2012-03-05 13:08:17 -05:00
Justin Ruggles f1e73100d9 siff: do not set AVCodecContext.frame_size 
also, properly set AVCodecContext.bits_per_coded_sample, AVStreasm.start_time,
and AVPacket.duration.
 2012-03-05 13:08:17 -05:00
Justin Ruggles ec2e767bf3 amr demuxer: do not set AVCodecContext.frame_size. 
it is not necessary.
 2012-03-05 13:08:17 -05:00
Justin Ruggles 8d1a20aa7c aiffdec: do not set AVCodecContext.frame_size 
It is unnecessary. Also, for some codecs we're reading more than 1 frame per
packet. Instead we use a private context variable to calculate the bit rate,
stream duration, and packet durations.

Updated FATE seek test, which has slightly different timestamps due to a
more accurate bit rate calculation.
 2012-03-05 13:08:17 -05:00
Justin Ruggles 237a855caf mov: do not set AVCodecContext.frame_size 
It is not necessary.
 2012-03-05 13:08:17 -05:00
Justin Ruggles 9727264220 ape: do not set AVCodecContext.frame_size. 
prevents lavf from setting incorrect packet durations.
 2012-03-05 13:08:17 -05:00
Justin Ruggles 2dd18d4435 rdt: remove workaround for infinite loop with aac 
avformat_find_stream_info() no longer hangs while waiting for AAC frame_size
 2012-03-05 13:08:16 -05:00
Justin Ruggles 9c365fe8ae avformat: do not require frame_size in avformat_find_stream_info() for CELT 
In Ogg/CELT, frame_size is found in the same place as the sample_rate and
channels, so we do not need to force the frame_size to be parsed.
 2012-03-05 13:08:16 -05:00
Justin Ruggles fbc8c59679 avformat: do not require frame_size in avformat_find_stream_info() for MP1/2/3 
It was only needed to avoid a bad time base (and thus non-monotone timestamps)
for stream copy to avi.
 2012-03-05 13:08:16 -05:00
Justin Ruggles 84b6ae0808 avformat: do not require frame_size in avformat_find_stream_info() for AAC 
We already will get the needed info because of CODEC_CAP_CHANNEL_CONF
 2012-03-05 13:08:16 -05:00
Justin Ruggles 620b88a302 swfenc: use av_get_audio_frame_duration() instead of AVCodecContext.frame_size 
This way we can do stream copy without having the demuxer wait until
frame_size has been set.
 2012-03-05 13:08:16 -05:00
Justin Ruggles 14aecc50fa rtpenc: use av_get_audio_frame_duration() for max_frames_per_packet 
It is more reliable than AVCodecContext.frame_size for codecs with constant
packet duration.
 2012-03-05 13:08:16 -05:00
Justin Ruggles c019070fda riffenc: use av_get_audio_frame_duration() 
For encoding, frame_size is not a reliable indicator of packet duration.
Also, we don't want to have to force the demuxer to find frame_size for
stream copy to work.
 2012-03-05 13:08:15 -05:00
Justin Ruggles 9524cf79df avcodec: add av_get_audio_frame_duration() function. 
This is a utility function for the user to get the frame duration based on
the codec id, frame size in bytes, and various AVCodecContext parameters.
 2012-03-05 13:08:15 -05:00
Justin Ruggles 6699d07480 avcodec: add av_get_exact_bits_per_sample() function 
This only returns bits per sample when it is exactly correct. That is, the
codec contains only raw samples with no frame headers or padding. This applies
to basically all PCM codecs and a small subset of ADPCM codecs.
 2012-03-05 13:08:15 -05:00
Justin Ruggles 1ba08c94f5 vorbisenc: add output buffer overwrite protection 2012-03-04 01:16:54 -05:00
Justin Ruggles fe78470a8b libopencore-amrnbenc: fix end-of-stream handling 
Use CODEC_CAP_DELAY and CODEC_CAP_SMALL_LAST_FRAME to properly pad and flush
the encoder at the end of encoding. This is needed in order to have all input
samples decoded.
 2012-03-04 01:14:53 -05:00
Justin Ruggles b0350c1c30 ra144enc: fix end-of-stream handling 
Use CODEC_CAP_DELAY and CODEC_CAP_SMALL_LAST_FRAME to properly pad and flush
the encoder at the end of encoding. This is needed in order to have all input
samples decoded.
 2012-03-04 01:14:53 -05:00
Justin Ruggles 29e2c85310 nellymoserenc: zero any leftover packet bytes 
fixes writing of uninitialized packet data
 2012-03-04 01:14:52 -05:00
Justin Ruggles 6c7a01621c nellymoserenc: use proper MDCT overlap delay 2012-03-04 01:14:52 -05:00
Justin Ruggles 51ddf35c90 wmaenc: fix m/s stereo encoding for the first frame 
We need to set ms_stereo in encode_init() in order to avoid incorrectly
encoding the first frame as non-m/s while flagging it as m/s. Fixes an
uncomfortable pop in the left channel at the start of playback.

CC:libav-stable@libav.org
 2012-03-03 18:20:10 -05:00
Justin Ruggles 8ed7488ea3 wmaenc: return s->block_align instead of recalculating it 2012-03-03 18:20:10 -05:00
Justin Ruggles 5d652e063b wmaenc: check final frame size against output packet size 
Currently we have an assert() that prevents the frame from being too large,
but it is more user-friendly to give an error message instead of aborting on
assert(). This condition is quite unlikely due to the minimum bit rate check
in encode_init(), but it is still worth having.
 2012-03-03 18:20:10 -05:00
Justin Ruggles dfc4fdedf8 wmaenc: require a large enough output buffer to prevent overwrites 
The maximum theoretical frame size is around 17000 bytes. Although in
practice it will generally be much smaller, we require a larger buffer
just to be safe.

CC: libav-stable@libav.org
 2012-03-03 18:20:10 -05:00
Justin Ruggles 1ec075cfec wmaenc: limit allowed sample rate to 48kHz 
ff_wma_init() allows up to 50kHz, but this generates an exponent band
size table that requires 65 bands. The code assumes 25 bands in many
places, and using sample rates higher than 48kHz will lead to buffer
overwrites.

CC:libav-stable@libav.org
 2012-03-03 18:20:10 -05:00
Justin Ruggles c2b8dea182 wmaenc: limit block_align to MAX_CODED_SUPERFRAME_SIZE 
This is near the theoretical limit for wma frame size and is the most that
our decoder can handle. Allowing higher bit rates will just end up padding
each frame with empty bytes.

Fixes invalid writes for avconv when using very high bit rates.

CC:libav-stable@libav.org
 2012-03-03 18:20:09 -05:00
Justin Ruggles b7beabab4b tiertexseq: set correct block_align for audio 2012-03-03 17:03:27 -05:00
Justin Ruggles f9cf91d822 tiertexseq: set audio stream start time to 0 
Update FATE test to reflect delayed video due to the file having audio-only
frames prior to the first frame with video.
 2012-03-03 17:03:27 -05:00
Justin Ruggles 0883109b27 voc/avs: Do not change the sample rate mid-stream. 
Also, set the time base based on the sample rate.
lavf-voc seek test updated to reflect slightly different seek points.
 2012-03-03 17:03:27 -05:00
Justin Ruggles 4da374f8a9 segafilm: use the sample rate as the time base for audio streams 2012-03-03 17:03:27 -05:00
Justin Ruggles ea289186f0 ea: fix audio pts 
The time base is 1 / sample_rate, not 90000.
Several more codecs encode the sample count in the first 4 bytes of the
chunk, so we set the durations accordingly. Also, we can set start_time and
packet duration instead of keeping track of the sample count in the demuxer.
 2012-03-03 17:03:27 -05:00
Justin Ruggles 01be6fa926 psx-str: fix audio pts 
Each packet has 18 sectors with 224/channels samples in each sector.
 2012-03-03 17:03:27 -05:00
Justin Ruggles d0ab585074 vqf: set packet duration 
Fixes timestamp calculation.
The FATE reference is updated because timestamp calculations are now more
accurate. Previous timestamps were based on average bit rate.
 2012-03-03 17:03:26 -05:00
Justin Ruggles 101c369b7c tta demuxer: set packet duration 2012-03-03 17:03:26 -05:00
Justin Ruggles 0b8b7db01b mpegaudio_parser: do not ignore information from the first parsed frame 
Update some demuxing and seeking fate tests.
 2012-03-03 17:03:26 -05:00
Justin Ruggles 5a9b952201 thp: set audio packet durations 2012-03-03 16:58:45 -05:00
Justin Ruggles 5602a464c9 avcodec: add a Vorbis parser to get packet duration 
This also allows for removing some of the Vorbis-related hacks.
 2012-03-03 16:43:11 -05:00
Justin Ruggles 737ca4482b vorbisdec: read the previous window flag for long windows 
When reading sequentially, we are using the actual flag from the previous
frame, but when seeking we do not know what the previous window flag was, so
we need to read it from the bitstream.
 2012-03-03 16:43:11 -05:00
Justin Ruggles f240df6a74 FATE: do not decode audio in the nuv test. 
We already have sufficient coverage for 16-bit pcm.
 2012-02-29 15:45:50 -05:00
Justin Ruggles 929dd8c108 dxa: set audio stream time base using the sample rate 2012-02-29 15:45:50 -05:00
Justin Ruggles aa831c4093 psx-str: do not allow seeking by bytes 2012-02-29 15:45:50 -05:00
Justin Ruggles bdbf1fa405 asfdec: Do not set AVCodecContext.frame_size 2012-02-29 15:45:50 -05:00
Justin Ruggles 4bf6775e9d vqf: set packet parameters after av_new_packet() 
Otherwise the values are overwritten.
 2012-02-29 15:45:50 -05:00
Justin Ruggles 841c17177b FATE: add mp3 test for sample that exhibited false overreads 
related to b716542691
Error messages and audible artifacts were fixed in that commit.
 2012-02-29 15:12:18 -05:00
Justin Ruggles a45a1ea521 libvorbis: add/update error messages 
also use AVERROR codes for some return values instead of -1
 2012-02-29 14:54:25 -05:00
Justin Ruggles 592c4dbc7e libvorbis: use AVFifoBuffer for output packet buffer 
simplifies the code and does less memmove()
 2012-02-29 14:54:24 -05:00
Justin Ruggles 1fe7c1be54 libvorbis: remove unneeded e_o_s check 
vorbis_bitrate_flushpacket() does not return any packets that should not be
output in the bitstream.
 2012-02-29 14:54:24 -05:00
Justin Ruggles 94025d8a99 libvorbis: check return values for functions that can return errors 2012-02-29 14:54:24 -05:00
Justin Ruggles c5063e0348 libvorbis: use float input instead of s16 
libvorbis takes float input, so we can just deinterleave/reorder the input
as-is instead of also converting.
 2012-02-29 14:54:24 -05:00
Justin Ruggles f15c4281dc libvorbis: do not flush libvorbis analysis if dsp state was not initialized 
Fixes a segfault if init() fails before initializing the dsp state
 2012-02-29 14:54:23 -05:00
Justin Ruggles 147ff24a0e libvorbis: use VBR by default, with default quality of 3 2012-02-29 14:54:20 -05:00
Justin Ruggles 182d4f1f38 libvorbis: fix use of minrate/maxrate AVOptions 
- enable the options for audio encoding
- properly check for user-set maxrate
- use correct calling order in vorbis_encode_setup_managed()
 2012-02-29 14:44:15 -05:00
Justin Ruggles eb35ef2932 libvorbis: cosmetics: renaming/pretty-printing/comments/unused code 2012-02-29 14:44:15 -05:00
Justin Ruggles 4e99501f62 (e)ac3enc: select a default bit rate based on the channel layout 2012-02-29 14:36:00 -05:00
Justin Ruggles 6aeea1dfb2 ac3enc: choose the closest bit rate to the one requested instead of failing 2012-02-29 14:36:00 -05:00
Justin Ruggles f234e8a32e movenc: use timestamps instead of frame_size for samples-per-packet 
For encoding, AVCodecContext.frame_size is the number of input samples to
send to the encoder and does not necessarily correspond directly to the
timestamps of the output packets.
 2012-02-27 04:33:37 -05:00
Justin Ruggles f3dab5fb6d movenc: use the first cluster duration as the tfhd default duration 2012-02-27 04:33:37 -05:00
Justin Ruggles 681d17264f movenc: factorize calculation of cluster duration into a separate function 2012-02-27 04:33:37 -05:00
Justin Ruggles be8d812c96 vorbisenc: check all allocations for failure 2012-02-25 11:49:43 -05:00
Justin Ruggles f7a2e12f2a roqaudioenc: return AVERROR codes instead of -1 2012-02-25 11:49:43 -05:00
Justin Ruggles 3fa9a999c1 roqaudioenc: set correct bit rate 2012-02-25 11:49:43 -05:00
Justin Ruggles cf57c78b77 roqaudioenc: use AVCodecContext.frame_size correctly. 
It is not allowed to change mid-stream like it does currently. Instead we need
to buffer the first 8 frames before returning them as a single packet, then
only return single frame packets after that.
 2012-02-25 11:49:43 -05:00
Justin Ruggles 56279f1d61 roqaudioenc: remove unneeded sample_fmt check 2012-02-25 11:49:43 -05:00
Justin Ruggles 4b7f8838e8 ra144enc: use int16_t* for input samples rather than void* 2012-02-25 11:49:43 -05:00
Justin Ruggles 03359ebcf9 ra144enc: set AVCodecContext.coded_frame 2012-02-25 11:49:43 -05:00
Justin Ruggles a65f7c96fc ra144enc: remove unneeded sample_fmt check 2012-02-25 11:49:42 -05:00
Justin Ruggles 928672f1f1 nellymoserenc: set AVCodecContext.coded_frame 2012-02-25 11:49:42 -05:00
Justin Ruggles 370b44cda2 nellymoserenc: improve error checking in encode_init() 2012-02-25 11:49:42 -05:00
Justin Ruggles 333506c33f nellymoserenc: return AVERROR codes instead of -1 2012-02-25 11:49:42 -05:00
Justin Ruggles 6f600ab354 libvorbis: improve error checking in oggvorbis_encode_init() 2012-02-25 11:49:42 -05:00
Justin Ruggles be60eec668 mpegaudioenc: return AVERROR codes instead of -1 2012-02-25 11:49:42 -05:00
Justin Ruggles 7a8cbb39f6 libfaac: improve error checking and handling in Faac_encode_init() 2012-02-25 11:49:42 -05:00
Justin Ruggles c9bca80132 avutil: add AVERROR_UNKNOWN 
Useful to return instead of -1 when the cause of the error is unknown,
typically from an external library.
 2012-02-25 11:49:41 -05:00
Justin Ruggles a8bdf2405c check for coded_frame allocation failure in several audio encoders 2012-02-25 11:49:41 -05:00
Justin Ruggles 255ad8881d audio encoders: do not set coded_frame->key_frame. 
it is already set in avcodec_alloc_frame()
 2012-02-25 11:49:41 -05:00
Justin Ruggles 8e2555d3b1 g722enc: check for trellis data allocation error 2012-02-25 11:49:41 -05:00
Justin Ruggles ca300d938b libspeexenc: export encoder delay through AVCodecContext.delay 2012-02-25 11:49:41 -05:00
Justin Ruggles 7600e5c5aa avcodec: document the use of AVCodecContext.delay for audio encoders 2012-02-25 11:49:41 -05:00
Justin Ruggles 19adb0bc2d vorbisdec: add a flush() function 
clear MDCT overlap buffer and reset previous window mode when seeking
 2012-02-24 17:37:25 -05:00