cbc2808761f8a7c04847de236d94be0d0759a321: Bug 1263063 - Part 6: Use TimingParams::EndTime() instead of re-calculation ComputedTiming each time. r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:39:44 +0900 - rev 350877
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 6: Use TimingParams::EndTime() instead of re-calculation ComputedTiming each time. r?dholbert MozReview-Commit-ID: 9pT06LXoU4V
7b7765b71955e8219dc3c7444e3642881a2fda6f: Bug 1263063 - Part 5: Introduce TimingParams::EndTime(). r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:39:42 +0900 - rev 350876
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 5: Introduce TimingParams::EndTime(). r?dholbert MozReview-Commit-ID: 6dpnb65vndl
56e52a836fb15aa3cac9d292dc4c140ac7214bd5: Bug 1263063 - Part 4: Move ActiveDuration() into TimingParams. r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:39:39 +0900 - rev 350875
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 4: Move ActiveDuration() into TimingParams. r?dholbert MozReview-Commit-ID: 4HS8vTJYtDe
e5d320d51e9d78ca63868739022a7b46e32a2dae: Bug 1263063 - Part 3: Change logic in KeyFrameEffect to assume that TimingParams' iteration-count has already been validated as nonnegative & finite. r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:36:03 +0900 - rev 350874
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 3: Change logic in KeyFrameEffect to assume that TimingParams' iteration-count has already been validated as nonnegative & finite. r?dholbert MozReview-Commit-ID: 9mbBHvyjNZg
b0e7493f14c8a0ef35d432b67f7c87ebfad2e5c8: Bug 1263063 - Part 2: Add an assertion to ensure that iteration count is nonnegative and finite. r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:36:01 +0900 - rev 350873
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 2: Add an assertion to ensure that iteration count is nonnegative and finite. r?dholbert MozReview-Commit-ID: JvVaCVQAxW4
52f53557b752eee773cf4069136f281b37564bc5: Bug 1263063 - Part 1: Remove unnecessary clamping of TimingParams::mIterationStart, since it's guaranteed to be nonnegative. r?dholbert draft
Hiroyuki Ikezoe <hiikezoe@mozilla-japan.org> - Thu, 14 Apr 2016 19:35:59 +0900 - rev 350872
Push 15434 by bmo:hiikezoe@mozilla-japan.org at Thu, 14 Apr 2016 10:40:04 +0000
Bug 1263063 - Part 1: Remove unnecessary clamping of TimingParams::mIterationStart, since it's guaranteed to be nonnegative. r?dholbert MozReview-Commit-ID: ApQVt39jasa
b71883aa1ea200d813834aae0e38499816c62411: Bug 1067769 - Part 2: Avoid doing RequestRestyle and mutation batch for null target draft
Boris Chiou <boris.chiou@gmail.com> - Thu, 14 Apr 2016 17:00:55 +0800 - rev 350871
Push 15433 by bmo:boris.chiou@gmail.com at Thu, 14 Apr 2016 10:37:52 +0000
Bug 1067769 - Part 2: Avoid doing RequestRestyle and mutation batch for null target We don't need to restyle if mTarget is nullptr, so skip it. For AnimationMutationBatch, we also skip it if mTarget is nullptr. Note: AnimationColletion, nsAnimationManager and nsTransitionManger are for CSS Animations and CSS Transtions, so the effect should have an effective target. Therefore, I think we don't have to revise them. MozReview-Commit-ID: Egzr1QVEa0v
4df476b34119ef32ab842b89105344bdb90874cf: Bug 1067769 - Part 2: Avoid doing RequestRestyle and mutation batch for null target draft
Boris Chiou <boris.chiou@gmail.com> - Thu, 14 Apr 2016 17:00:55 +0800 - rev 350870
Push 15432 by bmo:boris.chiou@gmail.com at Thu, 14 Apr 2016 10:35:51 +0000
Bug 1067769 - Part 2: Avoid doing RequestRestyle and mutation batch for null target We don't need to restyle if mTarget is nullptr, so skip it. For AnimationMutationBatch, we also skip it if mTarget is nullptr. Note: nsAnimationManager and nsTransitionManger is for CSS Animations and Transtions, so their effect should have effective target, so I think we don't have to revise them. MozReview-Commit-ID: Egzr1QVEa0v
22b1625e127b3242cb57ebfb486bc839f8b386a2: Bug 1067769 - Part 1: Support nullable target in KeyframeEffect(ReadOnly) constructor draft
Boris Chiou <boris.chiou@gmail.com> - Mon, 11 Apr 2016 16:43:00 +0800 - rev 350869
Push 15432 by bmo:boris.chiou@gmail.com at Thu, 14 Apr 2016 10:35:51 +0000
Bug 1067769 - Part 1: Support nullable target in KeyframeEffect(ReadOnly) constructor We set a null target only from Web Animations API, so make sure KeyframeEffectReadOnly::ConstructKeyframeEffect() can handle null target properly. MozReview-Commit-ID: D6PoV7PGFj3
b65dec2d4e1a3f5489884f6d75632ab0f329be37: Bug 1264171 - seperate the fast seek's logic out from accurate seek's logic; r?jwwang draft
Kaku Kuo <tkuo@mozilla.com> - Tue, 12 Apr 2016 17:52:51 +0800 - rev 350868
Push 15431 by tkuo@mozilla.com at Thu, 14 Apr 2016 10:27:50 +0000
Bug 1264171 - seperate the fast seek's logic out from accurate seek's logic; r?jwwang MozReview-Commit-ID: DgM0CuCX58s
18bb3f0681e62a1e0a3c568e0cc5009d8b70eb23: Bug 1261020 - part 2 - modify MDSM to adopt SeekTask; r=jwwang draft
Kaku Kuo <tkuo@mozilla.com> - Thu, 07 Apr 2016 14:30:39 +0800 - rev 350867
Push 15431 by tkuo@mozilla.com at Thu, 14 Apr 2016 10:27:50 +0000
Bug 1261020 - part 2 - modify MDSM to adopt SeekTask; r=jwwang MozReview-Commit-ID: IE9quRehVEg
8f191c66d780df0dd06ec8d2582bf5a6b22c4699: Bug 1261020 - part 1 - implement SeekTask; r=jwwang draft
Kaku Kuo <tkuo@mozilla.com> - Thu, 14 Apr 2016 12:01:55 +0800 - rev 350866
Push 15431 by tkuo@mozilla.com at Thu, 14 Apr 2016 10:27:50 +0000
Bug 1261020 - part 1 - implement SeekTask; r=jwwang MozReview-Commit-ID: GTrzI3ZuYeo
b371564155cfc090f247fe82fd846998bb9f8d5e: Bug 1248507 - p11. Detect and report when FFMpeg/Linux fails to load - r=jya draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350865
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p11. Detect and report when FFMpeg/Linux fails to load - r=jya If the FFmpeg decoder module cannot be started, the failure is recorded in the DecoderDoctorDiagnostics structure. In this case, on Linux if there are no suitable decoders for any requested format, a "platform decoder not found" notification is sent to Chrome (a separate bug will implement the actual front-end notification), and logged to the web console. MozReview-Commit-ID: CdaX7QUdWtd
7a5d1a1b0dfe67cc2bcd4d4de4d9fbbea5ec9d50: Bug 1248507 - p10. FFMpeg checks: Notification definition - r=bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350864
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p10. FFMpeg checks: Notification definition - r=bz MozReview-Commit-ID: DRYHBeTv7xA
f257f445256c84cd4d9895ed006399beb6c32085: Bug 1248507 - p9. FFMpeg checks: Console message - r=bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350863
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p9. FFMpeg checks: Console message - r=bz MozReview-Commit-ID: 6YTpH7oa4zF
82fc4e14a55b6268a847d10970294eba9fe7dbd4: Bug 1248507 - p8. Notify decoder-doctor-notification listeners - r?jya,bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350862
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p8. Notify decoder-doctor-notification listeners - r?jya,bz If the Decoder Doctor analysis needs to report something, a notification is sent to listeners of "decoder-doctor-notification", with data identifying the type of notification along with the media format(s) that could not be decoded. In this patch, there are only two notification types: "cannot-play", or "can-play-but-some-missing-decoders" (if pref "media.decoderdoctor.verbose" is true). In a future bug, the Firefox front-end will handle this notification and then optionally display a user notification. Note: "can-play-but-some-missing-decoders" should be useful to help implement the front-end side (as sites like YouTube will probably have some formats we don't handle); it may be removed later on if it has no further use. MozReview-Commit-ID: GL3JRqLxzxL
7599d5dcea0d9aa2aa61762f587293eacb4a6a2c: Bug 1248507 - p7. Minimal notification definition - r=bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350861
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p7. Minimal notification definition - r=bz MozReview-Commit-ID: JWd0BWehttB
228503bdd9a1a2d339e4af8d653d6bebe88d1b1e: Bug 1248507 - p6. Disable window check until bug 1261536 is resolved - r=jya draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:17 +1000 - rev 350860
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p6. Disable window check until bug 1261536 is resolved - r=jya Until bug 1261536 lands, it is possible for a synthetic video document to be created and to start loading (triggering an analysis) *before* that document is attached to a window. So for now we're allowing analyses to be gathered for a document with no window. Best case: The window is attached shortly after, and when the synthesis routine runs 1 second later, it will be able to send its notifications. Worst case: The analyses will linger around for a could of seconds, but nothing more will happen. MozReview-Commit-ID: HYh6b6cCebH
35ca65e34e01b29e1c804143c273bf33159b5ab5: Bug 1248507 - p5. DecoderDoctorDiagnostics implementation - r?jya,bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:16 +1000 - rev 350859
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p5. DecoderDoctorDiagnostics implementation - r?jya,bz Minimal implementation of DecoderDoctorDiagnostics. If the Decoder Doctor analysis needs to report something, a notification is logged to the web console, with the media format(s) that cannot be decoded. In this patch, there are only two notification types: "Cannot play" when no decoders are found for any of the requested formats), or "Can play" (if pref "media.decoderdoctor.verbose" is true) when decoders are missing for only some of the requested formats. MozReview-Commit-ID: 4QkiVvcNSU3
2aff1967fcf65c559ebca901282ecc205a1a26e6: Bug 1248507 - p4. DecoderDoctor base console message - r=bz draft
Gerald Squelart <gsquelart@mozilla.com> - Thu, 14 Apr 2016 20:20:16 +1000 - rev 350858
Push 15430 by gsquelart@mozilla.com at Thu, 14 Apr 2016 10:20:51 +0000
Bug 1248507 - p4. DecoderDoctor base console message - r=bz MozReview-Commit-ID: 6KzDFcwiYjm
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 +300000 tip