dcpomatic/test/player_test.cc, branch content-changed DCP-o-matic DCP tools https://git.carlh.net/cgit/dcpomatic/atom?h=content-changed 2023-02-27T11:52:28Z Remove debug message. 2023-02-27T11:52:28Z Carl Hetherington cth@carlh.net 2023-02-27T11:52:28Z urn:sha1:8ff6586d568c4a2b0a2ac24e690d172f4c01e3c4 Handle 2D/3D mismatches in the player (#2409). 2023-02-01T00:12:38Z Carl Hetherington cth@carlh.net 2023-01-11T22:00:25Z urn:sha1:2f12058c535045cecc226fe47b3d60da8851a862 Previously if there was 2D content in a 3D project (or vice versa) the player would emit the video anyway and the encoder would have to deal with it. Since previously the FFmpeg encoder did not deal with this quite right, it seems to make sense to move the logic into the player so that neither encoder has to worry about it. Noisy change to get film into AudioContent::modify_trim_start(). 2022-11-25T23:09:27Z Carl Hetherington cth@carlh.net 2022-11-22T01:16:16Z urn:sha1:5799460dc38bafa1da1ce9f7bf43621d3fadf442 Add another test, including a frame rate change. 2022-11-25T23:09:27Z Carl Hetherington cth@carlh.net 2022-11-22T00:29:15Z urn:sha1:5334f1c120ed248f22a5751d47555cf94bf9b4a9 Add failing test for #2373. 2022-11-25T23:09:27Z Carl Hetherington cth@carlh.net 2022-11-21T13:00:17Z urn:sha1:26a8f0c778a4abbe12a5a461dc457aa513773de3 Fix sound corruption with multiple sound files (#2368). 2022-11-25T22:42:55Z Carl Hetherington cth@carlh.net 2022-11-19T16:16:38Z urn:sha1:7d0cd227fa8cf13b04504af22301521dc231c0c7 last_push_end would be initialised to the position of the content, meaning that content at (for example) 60s into the project would be treated as the leader, and content much earlier would then be considered finished by the code which checks ignore_streams_behind (as its last push was more then 5s before the leader). This caused audio corruption, shown up by the test here. Remove some more unnecessary use of shared_ptr. 2022-10-16T22:10:59Z Carl Hetherington cth@carlh.net 2022-10-14T22:28:32Z urn:sha1:7d8d78c183656191ff44c9464e06b843bfadc54d Use a vector rather than a list when returning from content_factory(). 2022-07-11T10:22:11Z Carl Hetherington cth@carlh.net 2022-07-09T18:41:10Z urn:sha1:b1e22dff981fb86ae04b12bb5a064e61030793a8 Set up TextDecoder position based on the time that the next thing will 2022-06-12T13:40:33Z Carl Hetherington cth@carlh.net 2022-06-08T07:58:16Z urn:sha1:c8a036eb727ceddc64a0304d781c916eb952c001 be emitted, instead of the time that the last thing was (#2268). This is to avoid problems with the example shown in the test, where just because a subtitle in source A comes before a subtitle in source B, source A is pass()ed next and may then emit a subtitle which should be after the next one in B. Fix race between the Butler thread starting and audio (perhaps) being disabled. 2022-05-26T18:04:33Z Carl Hetherington cth@carlh.net 2022-05-26T18:04:33Z urn:sha1:8987f133295e352c44e05ef338eacc801c61a629 This could cause Butler::audio to be called with _audio_channels = 0 and _disable_audio = false, causing an exception in AudioBuffers when remap() tried to make an AudioBuffers object with a channel count of 0.