wine/gecko
0
0
Fork 0
mirror of https://gitlab.winehq.org/wine/wine-gecko.git synced 2024-09-13 09:24:08 -07:00
Code Issues Packages Projects Releases Wiki Activity
28,893 Commits 1 Branch 166 Tags 1.2 GiB
2cc8fbe2b4
Commit Graph

252 Commits

Author SHA1 Message Date
Matthew Gregan
2cc8fbe2b4 Bug 494116 - Use system audio timer for a/v sync on Windows. r=chris.double, sr=roc 2009-06-03 10:32:11 +12:00
Chris Pearce
cbd6ab1481 Bug 495507 audio sync - send blank audio when the audio stream has closed, not when there's no audio to present. r+ doublec, sr+roc 2009-06-02 15:05:46 +12:00
Robert O'Callahan
85828511ef Bug 495161. Buffer enough data in the Wave decoder that we never immediately leave BUFFERING after entering it. r=mgregan 2009-06-02 09:55:58 +12:00
Robert O'Callahan
353ec097d1 Bug 495319. Fire 'timeupdate' with the correct final time when we finish playing. r=doublec 2009-06-02 09:54:16 +12:00
Robert O'Callahan
496609122c Backing out bug 487519 due to reftest failures 2009-06-01 00:09:17 +12:00
Robert O'Callahan
7a349ea412 Fixing bustage (bug 495300) 2009-05-31 22:28:57 +12:00
Chris Double
52d8631083 Bug 487519. Work around Ogg bugs by decoding two frames before we seek to find the duration. r=roc 2009-05-31 22:06:30 +12:00
Robert O'Callahan
42b669ed47 Bug 495300. Trigger 'ended' behaviour when seeking to the end of a media resource. r=doublec 2009-05-31 22:02:17 +12:00
Chris Pearce
fb1b735b72 Bug 495156 - Video playback gets stuck forever when pressing End key. r=doublec sr=roc a=blocking1.9.1+ 2009-05-29 00:15:16 -04:00
Chris Pearce
c9cca633a1 Bug 495336 - liboggz oggz_get_next_page() and oggz_read_get_next_page() wrong r=doublec sr=roc a=blocking1.9.1+ 2009-05-29 00:13:23 -04:00
Chris Double
2b08d28a2e Bug 495129. Propagate errors better. r+sr=roc 2009-05-28 16:39:39 +12:00
Robert O'Callahan
6134d7088d Bug 495145. Don't change the state away from ENDED if we try to pause in the ENDED state. r=doublec 2009-05-28 15:49:22 +12:00
Robert O'Callahan
4f81b0342f Since test_load.html waits for load events, make sure it autobuffers and set the cache size to something large so loads aren't suspended. No bug, test-only change. 2009-05-25 12:51:01 +12:00
Chris Double
1ae4c2a0d3 Bug 494035. Never enable tracks we don't use. r=roc 
--HG--
extra : rebase_source : ceffc78e6473273893fd1ad000799b32023f5c76
 2009-05-25 12:45:47 +12:00
Robert O'Callahan
3dc99d8a9f Bug 494336. In DECODER_STATE_DECODING state, if mBufferExhaused is set then stop waiting for the next frame to become available. r=doublec 
--HG--
extra : rebase_source : 8b9a35e65cd0ddde311a493b4762fbd58fa2abf6
 2009-05-25 12:44:45 +12:00
Robert O'Callahan
3d39f15314 Bug 475110. Don't try to handle Wave files loaded via links or <object> using our Wave codec, since there are Wave formats we don't support that are actually used in the wild. r=doublec,a=beltzner 
--HG--
extra : rebase_source : 562c7c7c17157082483e456f33aa8b211fd07aeb
 2009-05-25 12:42:19 +12:00
Chris Pearce
4a9eea3922 Bug 493958. When decoding ahead to the desired seek position, save up the audio data as we go, because some of it might be needed for the destination frame and later frames. Also, if we reach the end of the frames without finding our destination position, keep the last frame and display it. r=doublec,roc 2009-05-21 17:55:12 +12:00
Chris Double
e3232bfecd Fix another FrameData leak (Bug 463358). Fixes intermittent orange. 2009-05-21 03:17:21 +12:00
Robert O'Callahan
dfae15cade Fix FrameData leak (bug 463358) 2009-05-20 21:11:48 +12:00
Robert O'Callahan
cda3109ffc Bug 493109. Shut down step-decode thread from the main decoder thread, instead of synchronously on the main thread at bad times. r=doublec 
--HG--
extra : rebase_source : 4af72350a51255e7e8748c726a17f9e88e1a540b
 2009-05-19 11:08:08 +12:00
Chris Pearce
b515f90d78 Bug 463358. Seek to any desired frame without visual artifacts by seeking to the previous keyframe and then decoding forward to the desired frame. r=doublec,roc 
--HG--
extra : rebase_source : 602a8f9feb74a1ef88e86424028d8da7bcf57b58
 2009-05-20 15:50:44 +12:00
Robert O'Callahan
efef7a6bc2 Backing out bug 493109 on suspicion of causing crashes in linux tests 2009-05-20 00:24:37 +12:00
Robert O'Callahan
6256de7f9c Backed out changeset 4b86cbae2846 2009-05-20 00:23:51 +12:00
Robert O'Callahan
f843370660 Bug 486922. Clamp out-of-bounds seek requests. r=doublec 
--HG--
extra : rebase_source : deeabc531d4fa6ada6e35740aed44bcc3844f233
 2009-05-19 21:55:04 +12:00
Robert O'Callahan
b4d2d7054a Fix test_media_selection.html to handle IDLE network states, since the cache size limit can suspend its loads 
--HG--
extra : rebase_source : a90486545fbb993c48dba3ee0025418b7bdc915f
 2009-05-19 21:53:38 +12:00
Robert O'Callahan
d848e30d78 Bug 493109. Shut down step-decode thread from the main decoder thread, instead of synchronously on the main thread at bad times. r=doublec 
--HG--
extra : rebase_source : 3a1a053308fc209181f7f23b234ba2f396de2a6e
 2009-05-19 11:08:08 +12:00
Robert O'Callahan
5b5d6baa5f Bug 480058. Handle non-square Theora pixel-aspect-ratio property. r=doublec 
--HG--
extra : rebase_source : 0a43b053b52a9b1e300f7f516517722c44040992
 2009-05-19 11:07:19 +12:00
Chris Double
127a265725 Bug 493673. If a video frame has an audio track but no audio data at all, assume the audio has ended and stuff silence into the audio stream. We need this so that video will keep playing synced to the audio clock. r=roc 
--HG--
extra : rebase_source : a1684fbd769ae48602335d312a33aa87ca17b8aa
 2009-05-19 17:39:40 +12:00
Robert O'Callahan
3d87360eee Bug 493431. When a frame has no video data, don't try to use the video time. r=doublec 
--HG--
extra : rebase_source : 985b41ec426c1cc793634c9f40f6924fbd33ed9b
 2009-05-19 17:30:36 +12:00
Chris Pearce
a0da0c82f6 Bug 487612. Let the start time of all media be 0 even if the internal timestamps start with something nonzero. r=doublec,roc 
--HG--
extra : rebase_source : ee419a06388a5230cddd79c78ddb86ed99fb6400
 2009-05-19 17:25:51 +12:00
Matthew Gregan
f201fe7c68 Bug 454971. Support 'new Audio()' constructor. r+sr=roc 
--HG--
extra : rebase_source : 1738d65cd16a79d69408055a8d9feb5ab5d97923
 2009-05-19 17:18:41 +12:00
Robert O'Callahan
bb3c7fc0f9 Bug 492420. Delay a little before playing a live stream to try to reduce the likelihood we'll have to buffer. r=doublec 
--HG--
extra : rebase_source : 022aa4d89644b3cfef62f95b4349932594de734f
 2009-05-19 17:14:20 +12:00
Robert O'Callahan
acdb5cc508 Bug 480058. Handle non-square Theora pixel-aspect-ratio property. r=doublec 
--HG--
extra : rebase_source : 762e10566ae85d0f73d7080ccb090533d50bfe66
 2009-05-19 11:07:19 +12:00
Robert O'Callahan
0855ba8432 Bug 493509. Don't fire 'suspend' when the resource is actually fully loaded. r=doublec 
--HG--
extra : rebase_source : a7ae1b2f53c7eb4b8a0e270510028d155d79f319
 2009-05-19 11:06:10 +12:00
Robert O'Callahan
f1b439157e Bug 487659. Make sure we never pass a playback rate of 0 to the media cache. r=doublec 
--HG--
extra : rebase_source : af86613bb5283e777acf1ca01d67eaa734cb8a0f
 2009-05-19 11:05:13 +12:00
Robert O'Callahan
0785b005eb Bug 493187. Call nsOggDecoder::UpdateReadyStateForData more often so we don't fail to notice state changes, and make it smarter about examining the frame queue. r=doublec 
--HG--
extra : rebase_source : 9a45e7cf9dc048956971ecf83b0bcc9bb2e45491
 2009-05-19 10:47:08 +12:00
Robert O'Callahan
a3f433b1b3 Bug 493443. Set readyState to HAVE_CURRENT_DATA whenever the Ogg frame queue is empty or we're buffering/seeking, instead of trying to hardcode it based on other events like PlaybackEnded. r=doublec 
--HG--
extra : rebase_source : 0b2f938e5f19cbb21258248da9da09e2a525db81
 2009-05-19 10:44:17 +12:00
Benjamin Smedberg
bb1c33c4ea Bug 488175 - flatten dist/include and provide mechanism to install certain headers in namespaced subdirectories (e.g. #include "mozilla/XPCOM.h") r=ted 
--HG--
rename : xpcom/ds/nsTimeStamp.cpp => xpcom/ds/TimeStamp.cpp
rename : xpcom/ds/nsTimeStamp.h => xpcom/ds/TimeStamp.h
 2009-05-18 14:15:05 -04:00
Robert O'Callahan
ceaf55f434 Fix test_media_selection.html; its video resource needs to be autobuffer so it doesn't stop loading 
--HG--
extra : rebase_source : 937602951c6a8c9831f6a73adc913457e913a7d6
 2009-05-18 20:18:43 +12:00
Robert O'Callahan
2d8939e3ca Fix test_autobuffer2 to handle the case when a suspend happens when all the data is already loaded. 
--HG--
extra : rebase_source : 7923d3fde3e4aef1723e424631f65a64d663f2a3
 2009-05-18 16:37:10 +12:00
Robert O'Callahan
5c3106e300 Relax time constraint in test_buf482461 a bit. 
--HG--
extra : rebase_source : 01fda6b6e4f50a8207208b22811a335117f9518c
 2009-05-18 15:20:40 +12:00
Robert O'Callahan
9ae2cf2709 Bug 479863. AdFire 'suspend' event and do correct networkState transitions when downloads suspend and resume. Also avoid firing 'stalled' while the download is suspended. r=doublec 
--HG--
extra : rebase_source : 0615871c062638d575a66e29b630040821a9f9cf
 2009-05-18 14:03:37 +12:00
Robert O'Callahan
710d28a4e6 Bug 479863. Add ability to close and reopen suspended streams, and recover from server-side disconnect of suspended streams. r=doublec 
--HG--
extra : rebase_source : 5ed7825cd919ad3f997f58dafb61c2d4249b60d4
 2009-05-18 14:02:20 +12:00
Robert O'Callahan
a2a28c523b Bug 479863. Stop buffering media elements by default, implement 'autobuffer' attribute to control buffering. r=doublec 
--HG--
extra : rebase_source : 5fa0b63da981fdfcf7b1555174b6d1122bf2ae97
 2009-05-18 14:00:44 +12:00
Robert O'Callahan
80975dcc56 Fix seeking tests to avoid false assumptions 2009-05-18 12:25:05 +12:00
Robert O'Callahan
ee1a877ccf Fix test_seek5; it should not depend on reading currentTime for the first seek during the asynchronous 'seeking' event, since the second seek may well have already happened. 2009-05-18 12:15:32 +12:00
Chris Pearce
ff942d65f4 Bug 469408. Make seeking in bufferred ranges fast by trying a 'bounded seek' in each buffered data range before falling back to a seek over the whole resource. r=doublec,sr=roc 2009-05-18 10:15:57 +12:00
Robert O'Callahan
c2bc1f2907 Bug 476973. Don't get confused by repeated seeks in the Ogg decoder. r=doublec 2009-05-18 10:03:03 +12:00
Josh Aas
2f6891c68f Disable test "test_videoDocumentTitle.html" due to random failures. b=492821 2009-05-15 23:20:39 -04:00
Robert O'Callahan
9bc46422a9 Disable test_autobuffer2.html for now, it fails a lot (bug 493692 filed) 2009-05-19 17:44:23 +12:00