3a5d1ff3ca
Hack in missing bits for ubuntu trusty/libav/avresample
2019-09-13 15:51:35 -04:00
5ddd20ed1a
refactor code. Handle resample buffering during encoding instead of when closing file
2019-09-13 14:34:53 -04:00
701aa8d924
add flushing the resample buffer to get the remaining samples encoded
2019-09-13 10:55:30 -04:00
788b5b480b
hack to make sync work on resample
2019-09-12 17:59:08 -04:00
d9d4c6b4ca
fix build
2019-09-12 17:26:19 -04:00
0a8aba2b76
rework how pts/dts is handled. We still use video_first_pts, but if the result is negative, normalize it to 0. Also now use audio_next_pts/dts to calculate pts when resampling and re-encoding. I believe this resolves the issues with wrong duration in the audio stream.
2019-09-12 16:52:59 -04:00
097ce5e778
add const to cast
2019-09-09 17:09:01 -04:00
792af28c54
return a value from write_packet
2019-09-09 16:55:22 -04:00
d5aa95e45f
cpplint fixes
2019-09-09 16:13:32 -04:00
d5d4d9d19b
ret should be a heap var, not a member of the object. Somehow this also seems to fix crashing on debian buster. I assume the compiler is confused
2019-09-08 16:10:57 -04:00
bd3395ac98
introduce VideoStore::write_packet to refactor out duplicated code. Add the ability to receive multiple packets from audio encoder per input packet
2019-08-28 12:19:54 -04:00
39620c01ff
Add rescaling of pkt pts/dts from audio_out_ctx to audio_out_stream
2019-08-26 16:37:32 -04:00
b26ed9bf62
Yet more debugging. Also, move pts adjustment to before audio encoding
2019-08-26 16:17:45 -04:00
2fe9442257
Fix time_base conversion when flushing audio queue. Add more debugging. COrrect pkt_Duration when resampling audio
2019-08-23 18:17:26 -04:00
396be10d6f
Merge branch 'master' into zma_to_thread
2019-08-22 12:57:00 -04:00
6a9464044b
Demote warnings about fixing non-monotonic dts to debug
2019-08-02 08:03:02 -04:00
1e0c39c632
mostly spacing cleanups. Don't bother setting pkt_duration on resampled frame
2019-07-19 16:28:18 -04:00
13c91bdf60
Add pts adjustment to the delayed flushed encoder packets
2019-07-11 17:56:53 -04:00
94cc85aa36
Sorta fix pts on encoded audio packets. Sync is off, but at least it is close
2019-07-07 08:57:22 -04:00
5ea5bd9bde
Only do audio_first_pts if we have audio_in_stream
2019-07-02 12:26:33 -04:00
4e8ac47060
Use the video_first_pts to set the audio_first_pts
2019-07-02 11:58:22 -04:00
dd57fd95ce
Clean up cruft from videostore api. Fix packetqueue clear_unwanted_packets to take a pre_event_count and take it into consideration when finding the right spot in the queue to start recording.
2019-06-27 11:11:14 -04:00
46f3aabc25
Merge branch 'hwdecode' into storageareas
2019-06-25 15:35:56 -04:00
86c4051c44
handle zm_receive_frame returning AVERROR instead of boolean
2019-06-25 15:34:36 -04:00
9a353d0b6d
don't set frame size
2019-06-24 11:44:21 -04:00
44e4d93445
Merge branch 'master' into storageareas
2019-06-24 11:35:36 -04:00
dab9544845
Revert audio sync fixes ( #2646 )
...
* Reference /run/zm instead of /var/run/zm because systemd-tmpfiles complains about it
* Spacing, google code style and tabs to spaces
* Revert the work I did to srt packets by dts after changing their time_base. It doesn't work. Instead since dts doesn't actually matter to playback, only ffmpeg storing, munge it to whatever is neccessary to make ffmpeg happy.
2019-06-24 11:27:47 -04:00
55d98e4a9f
fix crash
2019-06-20 15:56:25 -04:00
470da03322
Merge sync fixes from storageareas
2019-06-20 15:14:20 -04:00
4c8b8d0c16
Only check pts vs dts if pts isn't AV_NOPTS_VALUE. Also set frame_size in output codec when doing passthrough
2019-06-20 15:09:30 -04:00
16b035f76c
Use common first_dts/pts instead of separate video and audio first pts/dts
2019-06-19 09:22:21 -04:00
acb95709e6
Fix issues with too much audio in events by storing packets in the queue with their timestamps converted to AV_TIME_BASE_Q, so that we can sort video and audio packets together.
2019-06-11 10:19:42 -04:00
8b4fddadfb
out_frame->pts is calculated in resample_audio
2019-05-27 12:29:29 -04:00
c257fb46f6
Merge branch 'storageareas' of github.com:connortechnology/zoneminder into storageareas
2019-05-13 10:34:45 -04:00
22c5d46c65
rescale audio packet duration and pts before feeding to codec after resample
2019-05-12 12:14:03 -04:00
d7daf8550f
Merge branch 'master' into storageareas
2019-05-12 09:37:41 -04:00
453bc2afd8
more frame dumping in resample
2019-05-12 09:36:26 -04:00
67c20aa976
fix frame->duration to frame->pkt_duration
2019-05-10 12:58:54 -04:00
d3a680aaa3
Set out_frame duration when resampling. Better error message if failed to write to fifo
2019-05-10 12:31:10 -04:00
50070a9a7c
Merge branch 'storageareas' of github.com:connortechnology/ZoneMinder into storageareas
2019-05-03 14:59:14 -04:00
1e08b333b4
choose cur_dts instead of 0 for dts
2019-05-03 14:59:09 -04:00
18285e1b94
fix using in_frame->nb_samples instead of out_frame->nb_samples in resample fifo.
2019-04-25 18:45:43 -04:00
8bbcf0f631
merge storageareas
2019-04-25 14:49:16 -04:00
077ad75516
Merge branch 'storageareas' into origin/zma_to_thread
2019-04-25 12:41:48 -04:00
3664084862
Handle out of order dts in writeAudioFrame
2019-04-15 13:27:13 -04:00
96b7d30d7d
Merge branch 'master' into zma_to_thread
2019-04-15 10:47:26 -04:00
439a7b3ab3
Don't set bitrate, etc when h264. Causing crashes in ffmpeg 4.11.
2019-04-14 13:50:10 -04:00
fab548b119
Merge branch 'master' of github.com:ZoneMinder/ZoneMinder
2019-04-08 16:46:28 -04:00
1b98f3529f
allow non-increasing dts in muxer
2019-04-08 14:18:59 -04:00
7f2440177b
remove duplicate dump_codec
2019-04-08 13:51:55 -04:00
3c57d79669
Go back to using video_first_dts
2019-04-08 10:04:26 -04:00
8d3fa807e2
compilation fixes for old ffmpeg
2019-04-03 16:51:41 -04:00
94479fc850
copy pkt flags as well. When pts is NOPTS_VALUE set it to zero
2019-04-03 16:03:21 -04:00
055efdc950
Merge branch 'master' of github.com:ConnorTechnology/ZoneMinder
2019-04-03 14:56:18 -04:00
bd9b5afe46
old ffmpeg doesn't have av_frame_unref and it doesn't seem to be needed
2019-04-03 14:56:14 -04:00
b53e4aa803
fix segfault
2019-04-03 14:51:10 -04:00
b6f35db4de
put back codec closing in destructure, testing with bionic
2019-04-03 14:25:18 -04:00
aece64049f
We cannot avcodec_open2 a context that has been duped. This code currently works without crashes and leaks on ffmpeg 2.8
2019-04-03 13:23:00 -04:00
bd412fa9ca
Merge branch 'master' of github.com:ZoneMinder/ZoneMinder
2019-04-02 14:25:50 -04:00
cd28559c03
when encoding audio, pts and dts have to come from the encoder.
2019-04-02 14:08:46 -04:00
87cb716500
Merge branch 'master' of github.com:ZoneMinder/ZoneMinder
2019-04-01 14:02:46 -04:00
cb33318c4a
fix invalid memory access due to incorrect opening/closing/copying of contexts
2019-04-01 14:02:39 -04:00
ed5a448a1d
Correct pts on audio frame before sending it to the encoder so that when we flush the encoder the resulting packets have the right pts.
2019-04-01 11:13:14 -04:00
e6547953b0
fix segfault using old ffmpeg
2019-03-28 09:24:32 -04:00
5615cafc19
fix compile on old ffmpeg
2019-03-21 17:50:17 -04:00
c159dd8d10
cleanup and fix code that assumed H264
2019-03-21 14:13:56 -04:00
fcb9f76561
Fix out_frame->pts being nothing because we freed in_frame
2019-03-20 16:06:55 -04:00
3931baa2b6
fix build on old ffmpeg
2019-03-20 10:21:43 -04:00
a437c407a7
fix video_out_ctx init so that ffmpeg doesn't complain
2019-03-19 13:17:26 -04:00
9b713a489d
fix sleep time by using a comparison between the frame display time and the distance between now and when we started playing.
...
handle infinite fps
2019-03-06 14:50:36 -05:00
6c8eac1ac8
Merge branch 'storageareas' into h265
2019-03-06 11:10:01 -05:00
73ae3f49ed
Merge branch 'master' into storageareas
2019-03-05 11:35:55 -05:00
b1838de701
test turning off frag_custom
2019-03-05 11:35:16 -05:00
4b3519c84a
Merge branch 'storageareas' into zma_to_thread
2019-02-18 17:04:42 -05:00
5029d7214a
Merge branch 'master' into storageareas
2019-02-18 17:00:45 -05:00
7d6286d517
Merge branch 'storageareas' into zma_to_thread
2019-02-18 16:23:17 -05:00
9996b3e718
Merge branch 'master' into break_videostorage
2019-02-17 12:10:32 -05:00
9656032906
spacing
2019-02-15 17:52:51 -05:00
96906734b8
add an audio_fifo to handle input audio < 1024 samples per frame
2019-02-15 17:26:30 -05:00
4e463c9354
use swr_convert_frame instead of swr_convert
2019-02-14 17:48:58 -05:00
281775cc80
try setting out_frame pts from in_frame
2019-02-14 15:54:47 -05:00
9e07d5cd17
try setting out_frame pts from in_frame
2019-02-14 15:54:00 -05:00
a9e2011727
Merge branch 'storageareas' into zma_to_thread
2019-02-13 12:24:38 -05:00
060751602b
fix missing ;
2019-02-13 10:34:09 -05:00
d1b3b23c09
Merge branch 'master' into storageareas
2019-02-13 09:42:03 -05:00
0bfaf87d27
Mostly code style and whitespacing. However, I do setup more values in the output frame when doing audio resampling
2019-02-12 16:40:27 -05:00
3177764db4
spacing
2019-02-12 14:20:33 -05:00
ed6b22ac06
spacing
2019-02-11 16:29:36 -05:00
a39b92154c
wip
2019-01-17 08:49:48 -05:00
87d1390fed
Merge branch 'storageareas' into h265
2019-01-16 15:20:17 -05:00
5f6bcec4ca
Use start_pts instead of start_dts when calculating output pkt.dts. Because start_dts is often lower than start_pts, we can get into a situation where we calculate a dts that is > pts.
2019-01-15 11:35:37 -05:00
f726666f54
Merge branch 'master' into h265
2019-01-14 12:36:11 -05:00
9f8be169f9
set channels and sample rate in output frame. Use new zm_dump_frame()
2019-01-14 10:48:29 -05:00
c4c6d5b642
switch to swr_convert_frame as a test
2019-01-05 10:15:22 -05:00
28dafd5931
Merge branch 'storageareas' into zma_to_thread
2018-12-26 12:02:00 -05:00
55e0d82f20
spacing
2018-12-24 11:30:22 -05:00
6eb7892fc7
Merge pull request #2357 from connortechnology/libswresample
...
Implement libswresample support as an alternative to libavresample, w…
2018-12-20 16:44:06 -06:00
7a8beffdcc
Merge branch 'master' into storageareas
2018-12-20 15:10:52 -05:00
48865f124b
cap AAC bitrate at 96000 to prevent the aac encoder from complaining
2018-12-20 15:10:04 -05:00
6ee72aa2d9
Implement libswresample support as an alternative to libavresample, which is deprecated
2018-12-20 14:06:56 -05:00
Isaac Connor
Isaac Connor
Isaac Connor
Isaac Connor
Andrew Bauer