dcpomatic/src, branch 2958-3d-subs DCP-o-matic DCP tools https://git.carlh.net/cgit/dcpomatic/atom?h=2958-3d-subs 2025-03-24T09:20:32Z fixup! Basic support for variable-Z 3D subtitles. 2025-03-24T09:20:32Z Carl Hetherington cth@carlh.net 2025-03-24T09:20:32Z urn:sha1:39bf0d30645b3a9f9a6206f459fc62edfe946379 Basic support for variable-Z 3D subtitles. 2025-03-24T00:12:45Z Carl Hetherington cth@carlh.net 2025-03-20T22:43:34Z urn:sha1:86ba8bd464c04d622c28b6461386aedcae44d7c6 Fix crash due to crosss-thread list access. 2025-03-22T12:29:59Z Carl Hetherington cth@carlh.net 2025-03-21T12:37:34Z urn:sha1:e93b66f4236ec25dbd0603bbe74c904b53180e91 I've seen it happen that item points to a different QueueItem after the lock was unlocked here, which I didn't think was possible. Backtraces suggest that when problems happen _queue is being sorted in another thread, and perhaps this is not allowed (I couldn't find any conclusive proof in any documentation). There is also a potential problem with the newly-added zombify() method, which takes the lock and clears the list. Use make_shared. 2025-03-22T12:29:59Z Carl Hetherington cth@carlh.net 2025-03-21T12:34:41Z urn:sha1:3f1cffe8a0afe23577fe7351c05f1aa1dc492032 Simplify conditions. 2025-03-22T12:29:59Z Carl Hetherington cth@carlh.net 2025-03-21T12:34:33Z urn:sha1:fb1bda83c7e03ec27ca5faeaf10ba2c0b8b0d3b6 Only Type::FULL can have non-null encoded, so I don't think there's any point in checking for both things. Fix menu sensitivity. 2025-03-22T12:29:59Z Carl Hetherington cth@carlh.net 2025-03-20T00:36:45Z urn:sha1:e2c32319d1cc3ebab4259e1f6c9f840298819065 Broken in e1c43614ad31ce276b923f9b1d280ad0c102c4b1 Fix progress reporting. 2025-03-22T12:29:59Z Carl Hetherington cth@carlh.net 2025-03-20T00:25:58Z urn:sha1:721b17d3384989f2d378a37bbfb25fac0551b117 earliest_time will be empty when we're at the end of the film, not the beginning. White space. 2025-03-22T12:29:58Z Carl Hetherington cth@carlh.net 2025-03-20T00:07:50Z urn:sha1:f6f1e325b56acce9a1e7349daeb896bc76fddffc Remove before appending to try and reduce flicker on Windows. 2025-03-22T12:29:58Z Carl Hetherington cth@carlh.net 2025-03-19T23:31:42Z urn:sha1:1f1d8897877175903caa561c5495cde0de01c441 Move remote backoff sleep to before we pop a frame from the queue. 2025-03-22T12:29:58Z Carl Hetherington cth@carlh.net 2025-03-19T23:31:22Z urn:sha1:e32e6654b983bf4928b6f27a627b1e6305188cf6 Otherwise in the case of failure it can take a while for the frame to become available again for a functioning thread to encode.