Commit Graph

489140 Commits

Author SHA1 Message Date
Thomas Nguyen
bfa85d562c Bug 1296751 - Skip observe browser.safebrowsing.provider.*.(nextupdatetime|lastupdatetime). r=francois
MozReview-Commit-ID: 6yNWVJ3LYBE

--HG--
extra : rebase_source : cb60de6f916fdacc9bbb627b1724a68fad708ff1
2016-08-24 17:06:48 +08:00
Nathan Froyd
76bc75d243 Bug 1298171 - store result of BitmapProperties1 in a temporary to avoid warnings; r=jrmuizel
MozReview-Commit-ID: ANy1vKc5p07

--HG--
extra : rebase_source : d2b43299ec7244bc5bccb21c96f6604f50b083e8
2016-08-29 10:31:25 -04:00
Mike de Boer
d826244682 Bug 1295759 - make sure selected ranges in iframes are cleared when the findbar is hidden as well. Adds a test to guard against regressions. r=jaws
MozReview-Commit-ID: 5rudNSNK8GK

--HG--
extra : rebase_source : 03dc19bc9ba83f7601d4d57d2bd9b78d312fda1b
2016-08-29 16:22:29 +02:00
Ryan VanderMeulen
e7d85d7c46 Merge m-c to autoland. a=merge 2016-08-29 12:52:13 -04:00
Ryan VanderMeulen
416afe2f23 Merge inbound to m-c. a=merge 2016-08-29 12:48:10 -04:00
Ryan VanderMeulen
8b0f132c86 Merge fx-team to m-c. a=merge 2016-08-29 12:31:32 -04:00
Ryan VanderMeulen
590377a51f Merge autoland to m-c. a=merge 2016-08-29 12:22:29 -04:00
Henrik Skupin
0dd403224a Bug 1298772 - Bustage fix for firefox-ui-update tests for billboard removal. r=me, a=testonly
MozReview-Commit-ID: 1XQBuCpFFkx
2016-08-29 12:13:39 +02:00
Sebastian Hengst
45f10692c2 Backed out changeset 125b052639cb (bug 1241867) for assertion in test_contentViewer_overrideDPPX.html. r=backout 2016-08-28 00:46:39 +02:00
ffxbld
a2812fa126 No bug, Automated blocklist update from host bld-linux64-spot-561 - a=blocklist-update 2016-08-27 05:56:56 -07:00
ffxbld
21ac721516 No bug, Automated HPKP preload list update from host bld-linux64-spot-561 - a=hpkp-update 2016-08-27 05:56:54 -07:00
ffxbld
a2da90fae1 No bug, Automated HSTS preload list update from host bld-linux64-spot-561 - a=hsts-update 2016-08-27 05:56:51 -07:00
Kai Engert
02e277f76d Bug 1293329, land NSPR_4_13_BETA2, r=ted 2016-08-27 11:06:04 +02:00
Takashi Kawanaka
491499da43 Bug 1291061 - Pass const StyleAnimationValue& to CreateStyleContextForAnimationValue r=birtles
MozReview-Commit-ID: 6PJzBBWSgwQ
2016-08-27 15:27:38 +09:00
Kohei Okuda
5acbf0adda Bug 1293567 - Remove unnecessary call to MaybeUpdateAnimationRule in nsAnimationManager. r=birtles
MozReview-Commit-ID: 2Qdx7nk5fjL
2016-08-27 14:28:24 +09:00
Yusuke Endho
0f4c1508cb Bug 1297000 - Modify comparison of zero TimeDuration to use TimeDuration object rather than converting to milliseconds; r=birtles
MozReview-Commit-ID: DIccqtDb91u
2016-08-27 16:53:33 +09:00
Yuri Kazuna
3a366477e0 Bug 1291481 - Represent direction enum as uint8_t in LayersMessages.ipdlh; r=birtles
MozReview-Commit-ID: 9RcyYqsdsaB
2016-08-27 16:50:29 +09:00
Saheda Reza Antora
91d723340e Bug 1298016 - Identity popup's security view footer color changed to inherit instead of using ButtonText; r=dao 2016-08-27 09:17:34 +02:00
George Wright
b04020febc Bug 1293908 - Specify the LayersBackend to be used when creating Textures r=nical 2016-08-26 20:30:58 -04:00
Seth Fowler
a5b5245612 Bug 1298551 - Respect CachedSurface::SetLocked() even if the underlying ISurfaceProvider is always locked. r=dholbert 2016-08-26 17:03:25 -07:00
Seth Fowler
f8818933fa Bug 1296828 (Part 3) - Update SurfaceCache API to rely on ImageKeys and SurfaceKeys stored on ISurfaceProviders. r=dholbert 2016-08-26 17:03:19 -07:00
Seth Fowler
f9e85bafb2 Bug 1296828 (Part 2) - Store ImageKeys and SurfaceKeys directly on ISurfaceProviders. r=dholbert,edwin 2016-08-26 17:03:17 -07:00
Seth Fowler
8f4ebfcf90 Bug 1296828 (Part 1) - Use ImageKey consistently in SurfaceCache. r=dholbert 2016-08-26 17:03:15 -07:00
Drew Willcoxon
d14c6b15aa Bug 1292345 - Downloads panel didn't shrink to the fit height after all items are downloaded. r=me
MozReview-Commit-ID: AV0S8GM0ztj

--HG--
extra : rebase_source : e8d01256b994368b72c73344232f64cb26d6a357
2016-08-26 16:36:56 -07:00
Alastor Wu
a2979b25f8 Bug 1262053 - part10 : modify tests. r=baku
MozReview-Commit-ID: 98JWlLo7oRG

--HG--
extra : rebase_source : 207b0c68a3218ad260f8b7c97748df7587cb0616
2016-08-29 18:56:43 +08:00
Alastor Wu
b29b15681d Bug 1262053 - part9 : add test case. r=baku,cpearce
MozReview-Commit-ID: DjoIr7jdZeq

--HG--
extra : rebase_source : c4a672cdf2816eaa0462928f42780c260660fb0b
2016-08-29 18:56:40 +08:00
Alastor Wu
56e48dbbf5 Bug 1262053 - part8 : remove function NotifyOwnerDocumentActivityChangedInternal. r=cpearce
MozReview-Commit-ID: DnWgQHGJLU5

--HG--
extra : rebase_source : a6c4ab94fad22b2bef1bdde07bc17246943128f7
2016-08-29 18:56:38 +08:00
Alastor Wu
85c50bb3dd Bug 1262053 - part7 : give audio focus for the non-visited page. r=baku
MozReview-Commit-ID: K1cu8BhEtyV

--HG--
extra : rebase_source : 730746bbee7eed1b3f7568c6e2ace7449933dbb8
2016-08-29 18:56:36 +08:00
Alastor Wu
97b81589f8 Bug 1262053 - part6 : don't need to capture media element without audio. r=baku,cpearce
MozReview-Commit-ID: GO6nXbzYwIy

--HG--
extra : rebase_source : 03aba373fde9133bf373292b2cdded5f04f22781
2016-08-29 18:56:32 +08:00
Alastor Wu
47079de3ad Bug 1262053 - part5 : register audio agent immediately when media element starts playing. r=baku
In ancient degisn, we would only register audio channel after the media element has audio track and enoguh data to playback,
that is because the "audio-playback" event would be dispatched with the registration, and then shows the tab audio indicator.

However, now the event dispatching doesn't follow with the registration, it would be triggered when the media element has
really audible data which would be notified from media decoder.

Therefore, the media element without audio track or without enough data can also register audio channel agent, it won't affect
the display of tab audio indicator. The reason we need to do that is for blocking autoplay media in the non-visited tab.

The autoplay can be adding "autoplay" keyword or playing by the script, and we don't want to dispatch dom event for blocked
media. Therefore, we should register audio channel agent to know whether it needs to be blocked immediately even the media
element doesn't have any enough data which can let us to distinguish it have any audio track or not (this information can
be known from metadata).

First, we must check whether the media is blocked which is notified by audio channel agent, and then we can decide whether
need to dispatch the event. If we don't register audio channel agent, that we can't get blocking information.

MozReview-Commit-ID: HLLkOuecql1

--HG--
extra : rebase_source : 99c34f0185276ecd5b70ae09959b47c584d1564e
2016-08-29 16:34:31 +08:00
Alastor Wu
9021286e9a Bug 1262053 - part4 : don't dispatch dom event for blocked media. r=cpearce
If the media was blocked, we would postpone the dom event and dispatch them after media is resumed.

MozReview-Commit-ID: LcdJtH16qQn

--HG--
extra : rebase_source : 209d18925a88a9e0dac82f2c0695ec1b60f54f63
2016-08-29 16:34:28 +08:00
Alastor Wu
c060486878 Bug 1262053 - part3 : modify media element for blocking autoplay media. r=cpearce
MozReview-Commit-ID: 8e13lkYTN46

--HG--
extra : rebase_source : 134d698c6136b8dcf0ef67c6b40df787c42cb886
2016-08-29 16:34:26 +08:00
Chris Pearce
1231058432 Bug 1262053 - part2 : remove old media.block-play-until-visible behaviour. r=cpearce
MozReview-Commit-ID: GujLSVfu2rp

--HG--
extra : rebase_source : 6d229b3817736039e46713c9e029dec8fd8d5913
2016-08-29 10:42:10 +08:00
Dan Glastonbury
b194d26e61 Bug 1262053 - part1 : unblock window's media when the page was first visited. r=baku
MozReview-Commit-ID: 5g5cYdgC8sg

--HG--
extra : rebase_source : 5af18ff846f464b1d3d195d7dc2634e03601cfcc
2016-08-29 10:41:02 +08:00
Ryan VanderMeulen
4460ee8698 Backed out changeset 5f43357fe5b4 (bug 1284909) for bustage on a CLOSED TREE. 2016-08-29 11:25:00 -04:00
Wes Kocher
cbdf3625fa Merge m-c to inbound, a=merge 2016-08-26 16:32:56 -07:00
J. Ryan Stinnett
fff17715fe Bug 1297569 - Disable XUL popup contextmenu handler for all remote targets. r=felipe
This expands the existing escape hatch added by bug 516753 (which disabled this
same handler for content within a remote xul:browser).  After this change, it is
is disabled for any remote target (`EventStateManager::IsRemoteTarget`), which
in practice means remote xul:browser or remote iframe mozbrowser.

MozReview-Commit-ID: Ix7fdtDH54R
2016-08-26 18:32:09 -05:00
Wes Kocher
7f0262e905 Merge fx-team to central, a=merge 2016-08-26 16:30:01 -07:00
Wes Kocher
f09e8fef1a Merge inbound to central, a=merge 2016-08-26 16:20:50 -07:00
Wes Kocher
44a1966767 Merge autoland to central, a=merge 2016-08-26 16:07:49 -07:00
Timothy Nikkel
d4bf99143c Bug 1298100. Remove nsDisplayListBuilder::mHadToIgnoreSuppression because it is unused. r=cjku 2016-08-26 18:02:51 -05:00
Timothy Nikkel
ca29dc5318 Bug 1297934. Enforce the invariant that every display item class has a different display item type for nsDisplayFocusRing. r=mattwoodrow 2016-08-26 18:02:49 -05:00
James Willcox
f55f874788 Bug 1297751 - Allow E10S to be enabled on Fennec via browser.tabs.remote.force-enable r=sebastian 2016-08-26 16:57:27 -05:00
David Keeler
247986c342 bug 1298056 - fix HSTS preload update script so it will continue when requests result in errors r=jcj DONTBUILD NPOTB a=KWierso
MozReview-Commit-ID: b697Aa4iqQ

--HG--
extra : amend_source : 96e9e1c05520fab13f79990a99a4c507fe83fd44
2016-08-26 14:18:39 -07:00
Boris Zbarsky
17194b94cf Bug 1295322 followup: disable the test on Android, where we have no feed: handler and hence get a CLOSED TREE. 2016-08-26 17:15:53 -04:00
Wes Kocher
493b1533ea Backed out changeset e82fc4bb4cc4 (bug 1298324) for windows build bustage a=backout
CLOSED TREE

--HG--
extra : amend_source : b780f41c130ad5be13c451b4bcbc2b8cfd5570a1
2016-08-26 14:07:32 -07:00
Geoff Brown
c686d38861 Bug 1271264 - Populate clipboard earlier to avoid intermittent failure in testAccessibleCarets; r=sebastian 2016-08-26 14:34:55 -06:00
Ryan Hunt
4656974491 Bug 1298324 - Change D3D11DeviceStatus to a union. r=dvander
MozReview-Commit-ID: LCy2NSch39X
2016-08-26 13:32:39 -07:00
Boris Zbarsky
f795d19994 Bug 1258911. Correctly reframe an absolutely positioned frame that goes from having a transform to not having one and has an abs pos kid with a fixed-pos descendant. r=dbaron
Instead of skipping the absolute and fixed child lists, we walk all kids of the
frame.  But before recursing down into things that are absolute containing
blocks we ensure that we're only looking for fixed-pos placeholders, so we don't
reframe if we have a relatively positioned descendant with absolutely positioned
kids, for example.  Note that this part is pure optimization attempt, and it
might be cheaper to not do it: IsAbsoluteContainingBlock is not that cheap and
the situations where we avoid reframing due to this optimization are likely
fairly rare.
2016-08-26 15:38:16 -04:00
Boris Zbarsky
0e5b46b6d6 Bug 1295322 test. r=bholley 2016-08-26 15:38:16 -04:00