3ba829dddc087cddf76a4591e7746557a3cb1d12: Bug 1377329 - Part 2. Entend the capacity of mFramesMarkedForDisplay only when need. draft
cku <cku@mozilla.com> - Fri, 14 Jul 2017 01:47:34 +0800 - rev 608884
Push 68435 by bmo:cku@mozilla.com at Fri, 14 Jul 2017 08:45:34 +0000
Bug 1377329 - Part 2. Entend the capacity of mFramesMarkedForDisplay only when need. There patch did two things: 1. Enlarge default capacity size of mFramesMarkedForDisplay from 100 to 400. For gmail, 100 is too small, we always need to enlarge capaicty of this array in nsDisplayListBuilder::MarkFramesForDisplayList. 2. Although we cache length in previous patch, nsFrameList::GetLength is still costly in first call. In nsDisplayListBuilder::MarkFramesForDisplayList, we should use it when we have to. MozReview-Commit-ID: 5zSxyakZ5O9
bff909d3dcba24b09d88e815ab79f200fd92e9fa: Bug 1377329 - Part 1. Cache the length of frames in nsFrameList. draft
cku <cku@mozilla.com> - Fri, 14 Jul 2017 12:47:12 +0800 - rev 608883
Push 68435 by bmo:cku@mozilla.com at Fri, 14 Jul 2017 08:45:34 +0000
Bug 1377329 - Part 1. Cache the length of frames in nsFrameList. nsTableRowGroupFrame::GetRowCount() also uses nsFrameList::GetLength(). In this scenario, since using nsFrameList::GetLength() is not preventable, we can cache the length of frames inside nsFrameList to reduce time spend on tracing frame link list. MozReview-Commit-ID: HiDK7mF2AdL
a851d987600580a54ffd322324cae77bc1db06d5: Bug 1362617 - Generalize MOZ_CHROME_MULTILOCALE to work for browser as well. r?ted, r?gps draft
Zibi Braniecki <zbraniecki@mozilla.com> - Tue, 30 May 2017 10:30:26 +0200 - rev 608882
Push 68434 by bmo:gandalf@aviary.pl at Fri, 14 Jul 2017 08:44:51 +0000
Bug 1362617 - Generalize MOZ_CHROME_MULTILOCALE to work for browser as well. r?ted, r?gps In order to get Firefox to work with multiple built-in languages, we want to bundle the same file in desktop as we do in fennec - res/multilocale.json MozReview-Commit-ID: BY1cMYd0q8Q
b38820cde49a2a0fe6bacd6d33b8ce16e34635ee: bug 1370506, update the docs, r?glandium draft
Axel Hecht <axel@pike.org> - Fri, 16 Jun 2017 13:14:28 +0200 - rev 608881
Push 68433 by axel@mozilla.com at Fri, 14 Jul 2017 08:31:20 +0000
bug 1370506, update the docs, r?glandium Mostly removals, but also adding docs on how to create a multilingual package right now. I think I'd like to take another pass at those docs in a follow-up. MozReview-Commit-ID: Dkw4MJ5DLyb
5bb58aafa6eed5f8a1a25003db5279d95fc09aaf: bug 1370506, add l10n-related targets to top-level build.mk, r?glandium draft
Axel Hecht <axel@pike.org> - Fri, 16 Jun 2017 12:13:55 +0200 - rev 608880
Push 68433 by axel@mozilla.com at Fri, 14 Jul 2017 08:31:20 +0000
bug 1370506, add l10n-related targets to top-level build.mk, r?glandium These rules are just convenience, both for developers and automation. Instead of having to hard-code to run make in a particular directory to do l10n-repacks, you can now just do ./mach build installers-de and that's that. MozReview-Commit-ID: C4WKXljjN7n
7e75ab4958a6537ab5466361365744aa16106244: bug 1370506, good default download URL for Desktop l10n repacks, r?glandium draft
Axel Hecht <axel@pike.org> - Fri, 16 Jun 2017 11:59:44 +0200 - rev 608879
Push 68433 by axel@mozilla.com at Fri, 14 Jul 2017 08:31:20 +0000
bug 1370506, good default download URL for Desktop l10n repacks, r?glandium This is just for Nightly, and only for Desktop. Beta and beyond don't have a canonical directory, and Fennec needs the platform in the directory. MozReview-Commit-ID: 9dvJLVlWiQv
735a5d050f02aeb8a9b5d9ed687f583a4c51b7c9: bug 1370506, for Nightly builds, automatically clone l10n repos for localized installers, r?glandium draft
Axel Hecht <axel@pike.org> - Thu, 15 Jun 2017 19:47:28 +0200 - rev 608878
Push 68433 by axel@mozilla.com at Fri, 14 Jul 2017 08:31:20 +0000
bug 1370506, for Nightly builds, automatically clone l10n repos for localized installers, r?glandium Making more decisions on behalf of developers: L10NBASEDIR is always defined, if not specified, it's ~/.mozbuild/l10n-central, or in MOZBUILD_STATE_PATH/l10n-central if the state path in defined in the environment. If a locale isn't checked out, do that. The targets for which that works are merge-%, installers-%, langpack-% But only do that for Nightly builds, as for Beta and beyond, we have explicit revisions to use for the builds, and we don't want to break release builds silently with this. MozReview-Commit-ID: EhGJPLuiyYO
45daf059d4e812dfd509477c8854111f9ed5920a: Bug 1274919 - part5 : add telemetry probe to measure how long the cursor is hovering before opening the tab. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:28:37 +0800 - rev 608877
Push 68432 by alwu@mozilla.com at Fri, 14 Jul 2017 08:28:55 +0000
Bug 1274919 - part5 : add telemetry probe to measure how long the cursor is hovering before opening the tab. Measure the time how long the cursor is hovering before opening the tab which contains suspended playing media. If the tab didn't be opened, the data won't be recorded. MozReview-Commit-ID: A3RBfMyCziv
3ece1d61de4b2b82837a7c79709eb4d876b8341c: Bug 1274919 - part4 : add test. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:28:34 +0800 - rev 608876
Push 68432 by alwu@mozilla.com at Fri, 14 Jul 2017 08:28:55 +0000
Bug 1274919 - part4 : add test. MozReview-Commit-ID: 5wKkE5tphLl
f855331a687b7ce9eb7ebadecaa9966d7056e75f: Bug 1274919 - part3 : only notify the tab with suspened video to resume video decoding. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:27:02 +0800 - rev 608875
Push 68432 by alwu@mozilla.com at Fri, 14 Jul 2017 08:28:55 +0000
Bug 1274919 - part3 : only notify the tab with suspened video to resume video decoding. If the video starts being suspended and its owner tab is in the background, then it might be resumed when the cursor is hovering over the tab. After cursor leaves the tab, we would suspend video again. And we would use a variable to indicate whether we're in this kind of status, it can be checked via "browser.videoDecodingRecoveryMode". If the mode is on, it means we might resume/suspend video. Only do the related operations when this mode is on. MozReview-Commit-ID: CmpE2n4Qtk5
034544d39d6e4e0080f3785d3d26b8cffea34a12: Bug 1380931 - Switch Intl.PluralRules tests to public domain. r?gerv draft
Zibi Braniecki <zbraniecki@mozilla.com> - Fri, 14 Jul 2017 01:24:25 -0700 - rev 608874
Push 68431 by bmo:gandalf@aviary.pl at Fri, 14 Jul 2017 08:24:41 +0000
Bug 1380931 - Switch Intl.PluralRules tests to public domain. r?gerv MozReview-Commit-ID: 33XHMMVhCz
65ed548dee5da29f516a8946f77b6bb327102a60: Bug 1380936 - Add support for 'minimize' command in geckodriver; r?ato,whimboo draft
yangyi_peng@htc.com - Fri, 14 Jul 2017 16:23:34 +0800 - rev 608873
Push 68430 by bmo:yangyi_peng@htc.com at Fri, 14 Jul 2017 08:23:54 +0000
Bug 1380936 - Add support for 'minimize' command in geckodriver; r?ato,whimboo Need to merge after webdriver-rust update to include the commit "Implement the minimize window for webdriver-rust (#105)" MozReview-Commit-ID: F13y0aOn6Ia
f0429765b0bfe8f3fadbedd0f4062e39fa069afb: Bug 1274919 - part5 : add telemetry probe to measure how long the cursor is hovering before opening the tab. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:19:56 +0800 - rev 608872
Push 68429 by alwu@mozilla.com at Fri, 14 Jul 2017 08:20:09 +0000
Bug 1274919 - part5 : add telemetry probe to measure how long the cursor is hovering before opening the tab. Measure the time how long the cursor is hovering before opening the tab which contains suspended playing media. If the tab didn't be opened, the data won't be recorded. MozReview-Commit-ID: HCSwNd3eIP5
9fc5651e8fb518bcb46c51a459812afa3a055e80: Bug 1274919 - part4 : add test. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:19:54 +0800 - rev 608871
Push 68429 by alwu@mozilla.com at Fri, 14 Jul 2017 08:20:09 +0000
Bug 1274919 - part4 : add test. MozReview-Commit-ID: 5wKkE5tphLl
94b24ee236da226f8170660e925cb3e14b007bed: Bug 1274919 - part3 : only notify the tab with suspened video to resume video decoding. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:19:52 +0800 - rev 608870
Push 68429 by alwu@mozilla.com at Fri, 14 Jul 2017 08:20:09 +0000
Bug 1274919 - part3 : only notify the tab with suspened video to resume video decoding. If the video starts being suspended and its owner tab is in the background, then it might be resumed when the cursor is hovering over the tab. After cursor leaves the tab for the specific time, we would suspend video again. And we would use a variable to indicate whether we're in this kind of status, it can be checked via "browser.videoDecodingRecoveryMode". If the mode is on, it means we might resume/suspend video. Only do the related operations when this mode is on. MozReview-Commit-ID: LQthdOmkcqd
768056fd058dafe99e5dd99abdac98c7c35f5c78: Bug 1274919 - part2 : implement resume/suspend mechanism in MediaDecoder. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:19:45 +0800 - rev 608869
Push 68429 by alwu@mozilla.com at Fri, 14 Jul 2017 08:20:09 +0000
Bug 1274919 - part2 : implement resume/suspend mechanism in MediaDecoder. Create the new class "BackgroundVideoDecodingPermissionObserver" to handle the suspended request sent from the front end side. When the tab goes to background and the video can be suspended, we would start the observer. When the tab goes to foreground and the video won't be suspended anymore, we would stop the observer. MozReview-Commit-ID: AQAog6QRXCj
c6979124cce1cc023ef256581138a999de3bd6d0: Bug 1274919 - part1 : resume video decoding when cursor is hovering over the tab, and suspend it again when cursor leaves the tab. draft
Alastor Wu <alwu@mozilla.com> - Fri, 14 Jul 2017 16:17:35 +0800 - rev 608868
Push 68429 by alwu@mozilla.com at Fri, 14 Jul 2017 08:20:09 +0000
Bug 1274919 - part1 : resume video decoding when cursor is hovering over the tab, and suspend it again when cursor leaves the tab. If the tab is in the background and contains playing video, we would stop the video decoding in order to reduce the power consumption. It would be resumed after the tab goes back to foreground. When resuming the video, we need to seek the video to the specific time. If the video seeking is too slowly, the audio and video would not be synchronized and user would see the black image. To avoid this situation, we need to resume video decoding in advance. When the cursor is hovering over the tab, we assume user would open that tab soon, so we can start to resume the video decoding. MozReview-Commit-ID: 7LaN6UMxlXS
96e18874aa7efc78c20a3b1f4a775c628404b69e: Bug 1380644 - Update Android lint library dependency to 25.3.2. r?grisha draft
Sebastian Kaspari <s.kaspari@gmail.com> - Thu, 13 Jul 2017 15:57:47 +0200 - rev 608867
Push 68428 by s.kaspari@gmail.com at Fri, 14 Jul 2017 08:17:01 +0000
Bug 1380644 - Update Android lint library dependency to 25.3.2. r?grisha This is primarily for local builds that use the latest Android SDK. MozReview-Commit-ID: 5uIjC6EhCho
2e6557e824ccd8aff30dd875aff1444e45efc3c8: Bug 699703 - Need reframe before calling SetSelectionAfterTableEdit. r?masayuki draft
Makoto Kato <m_kato@ga2.so-net.ne.jp> - Fri, 14 Jul 2017 15:48:40 +0900 - rev 608866
Push 68427 by bmo:m_kato@ga2.so-net.ne.jp at Fri, 14 Jul 2017 08:16:24 +0000
Bug 699703 - Need reframe before calling SetSelectionAfterTableEdit. r?masayuki HTMLEditor::TabInTable inserts row element, then select a cell. But when enabling lazy frame construction for editable node, it select invalid cell and table. Because HTMLEditor::SetSelectionAfterTableEdit doesn't select cell correctly on InsertTableRow(). HTMLEditor::SetSelectionAfterTableEdit uses HTMLEditor::GetCellAt, so it depends on frame. So we need flush frame before calling it. Also. a comment of HTMLEditor::InsertTableRow is invalid now because we don't use nsresult version of CreateElementWithDefualts. MozReview-Commit-ID: 698TvmMZgwB
de578a3b78e9ddbeb788b7964ab25fab4e30b04b: Bug 699703 - Need reframe before calling SetSelectionAfterTableEdit. r?masayuki draft
Makoto Kato <m_kato@ga2.so-net.ne.jp> - Fri, 14 Jul 2017 15:48:40 +0900 - rev 608865
Push 68426 by bmo:m_kato@ga2.so-net.ne.jp at Fri, 14 Jul 2017 08:14:02 +0000
Bug 699703 - Need reframe before calling SetSelectionAfterTableEdit. r?masayuki HTMLEditor::TabInTable inserts row element, then select a cell. But when enabling lazy frame construction for editable node, it select invalid cell and table. Because HTMLEditor::SetSelectionAfterTableEdit doesn't select cell correctly on InsertTableRow(). HTMLEditor::SetSelectionAfterTableEdit uses HTMLEditor::GetCellAt, so it depends on frame. So we need flush frame before calling it. Also. a comment of HTMLEditor::InsertTableRow is invalid now because we don't use nsresult version of CreateElementWithDefualts. MozReview-Commit-ID: 698TvmMZgwB
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip