Commit Graph

6161 Commits

Author SHA1 Message Date
Chris Craik
363f0d9fd9 am 81defc6a: Merge "Sync ViewOverlay size init with RenderNode" into lmp-mr1-dev
* commit '81defc6a840367dcf9591d14b827f2d8d756ef00':
  Sync ViewOverlay size init with RenderNode
2014-12-02 17:04:34 +00:00
Chris Craik
81defc6a84 Merge "Sync ViewOverlay size init with RenderNode" into lmp-mr1-dev 2014-12-02 16:59:21 +00:00
Chris Craik
2180ba7abd Sync ViewOverlay size init with RenderNode
bug:18381202

Setting mRight/mBottom must be accompanied by updating the RenderNode.

Change-Id: I4dae06a5ca43247b0b3642d85808e2724e486692
2014-12-01 17:55:37 -08:00
Alan Viverette
23dc05634d am e6bd61d5: Merge "Fix visible rect computation for views with padding" into lmp-mr1-dev
* commit 'e6bd61d5bbea012b6b80c8bff915042f615630bc':
  Fix visible rect computation for views with padding
2014-12-01 22:12:44 +00:00
Alan Viverette
e6bd61d5bb Merge "Fix visible rect computation for views with padding" into lmp-mr1-dev 2014-12-01 22:06:13 +00:00
Wale Ogunwale
e05fbd6901 am 997e632f: Merge "Added unique id to display devices and their settings." into lmp-mr1-dev
* commit '997e632f59b070dabf71d9e807847c0643da92ac':
  Added unique id to display devices and their settings.
2014-12-01 20:30:33 +00:00
Alan Viverette
ab2cf6dafe Fix visible rect computation for views with padding
Previously, the computation was double-counting the left and top padding
because it was treating the third and fourth Rect.intersect() arguments
as width and height rather than right and bottom.

BUG: 18418091
Change-Id: I2ab669ee5060372ae11cfe804dcc05c7426be1ec
2014-11-26 14:34:36 -08:00
Wale Ogunwale
361ca21acc Added unique id to display devices and their settings.
The display setting saved to disk were using a localized name for
the key. This is an issue if the user changes languages after the
display settings have been saved. We now use the non-localized
name for the display to access the settings if it is available,
else we fall back on the localized name.

Bug: 18190800
Change-Id: I837c06a8935df10727229a1aa2bb6eeb3953707f
2014-11-26 13:40:52 -08:00
John Reck
08372601ab am 418c8549: Merge "It\'s super critical to call nStart" into lmp-mr1-dev
* commit '418c8549dca8bff8c38c67ed7f7d7fbbaccd4e71':
  It's super critical to call nStart
2014-11-21 23:37:52 +00:00
John Reck
418c8549dc Merge "It's super critical to call nStart" into lmp-mr1-dev 2014-11-21 23:32:08 +00:00
John Reck
72d6e4facb It's super critical to call nStart
Bug: 18204974

Even if we are canceling or ending an animation
nStart() *must* be called, otherwise the native-side listener
is not attached (lazy-attached for JNI cyclic reference reasons),
and then Animator::callOnFinishedListener() no-ops as there's
no listener set

Add a lifecycle verifier to ensure that nStart is always
called on animators that are attached and get finished

Change-Id: Ibc345b5be97b6d3f95a11c361ebe020d030fd3b6
2014-11-21 14:27:21 -08:00
Chris Craik
1ae8a98e8a am 29dc496a: Merge "Revert "Add a way to override Xfermode DO NOT MERGE"" into lmp-mr1-dev
* commit '29dc496a42d49a37dcd99c0465f3cec18a47e6ff':
  Revert "Add a way to override Xfermode DO NOT MERGE"
2014-11-21 22:13:30 +00:00
Chris Craik
29dc496a42 Merge "Revert "Add a way to override Xfermode DO NOT MERGE"" into lmp-mr1-dev 2014-11-21 22:11:13 +00:00
Chris Craik
69e65019b0 Revert "Add a way to override Xfermode DO NOT MERGE"
Feature no longer needed.

bug:18448377

This reverts commit 4678dcc552.

Change-Id: Ib0a19946e966a54857165555827b5fa7b34b6bea
2014-11-21 22:05:33 +00:00
Bryce Lee
deee556641 am d45cedd6: Merge "Make default windows features configurable." into lmp-mr1-dev
* commit 'd45cedd6f5efd141be08ce4ffcecf478fbedb056':
  Make default windows features configurable.
2014-11-21 19:21:25 +00:00
Bryce Lee
d45cedd6f5 Merge "Make default windows features configurable." into lmp-mr1-dev 2014-11-21 19:18:50 +00:00
Bryce Lee
d6e6e7235a Make default windows features configurable.
Bug: 18434078
Change-Id: I0bafce1a201088a8681f4b9e43237e23b49296a2
2014-11-21 11:08:45 -08:00
Filip Gruszczynski
5ea0411f39 am a83c5806: Merge "SwipeDismissLayout makes activity opaque only after entry animation ends." into lmp-sprout-dev automerge: 96c5d37
* commit 'a83c5806ded083f51549ba4bf70e0c1f6a17fc61':
  SwipeDismissLayout makes activity opaque only after entry animation ends.
2014-11-21 10:21:19 +00:00
Filip Gruszczynski
a83c5806de Merge "SwipeDismissLayout makes activity opaque only after entry animation ends." into lmp-sprout-dev
automerge: 96c5d37

* commit '96c5d37fb5139465cdd3d882b5deba830497557a':
  SwipeDismissLayout makes activity opaque only after entry animation ends.
2014-11-21 10:13:59 +00:00
John Reck
d3e5777e79 am 83fb64e8: Merge "If the surface is lost, release it" into lmp-mr1-dev
* commit '83fb64e8ec4c6bc0930f50b24bd1c36a81f6a1c2':
  If the surface is lost, release it
2014-11-21 00:04:18 +00:00
John Reck
83fb64e8ec Merge "If the surface is lost, release it" into lmp-mr1-dev 2014-11-21 00:00:08 +00:00
John Reck
9b5d427fe0 am 0b2dac34: Merge "Rename callDrawGLFunction" into lmp-mr1-dev
* commit '0b2dac3452a58e76be7379b85e7505c531842b57':
  Rename callDrawGLFunction
2014-11-20 23:29:31 +00:00
John Reck
0b2dac3452 Merge "Rename callDrawGLFunction" into lmp-mr1-dev 2014-11-20 23:26:07 +00:00
John Reck
680df8d60e Rename callDrawGLFunction
Bug: 18471503
Change-Id: I00cb356ff2338da7fd07b9ba30b5b3dcbdffdb2e
2014-11-20 14:40:52 -08:00
Alan Viverette
a19f057792 am 8ccf071a: Merge "Move default token handling into WindowManagerImpl" into lmp-mr1-dev
* commit '8ccf071ab83510c8ef7b4d311d894d5c4a352f6c':
  Move default token handling into WindowManagerImpl
2014-11-20 19:21:11 +00:00
Alan Viverette
8ccf071ab8 Merge "Move default token handling into WindowManagerImpl" into lmp-mr1-dev 2014-11-20 19:16:58 +00:00
Chris Craik
57b641a2ed am 3b2ba44b: Merge "Improve logging around performance critical events" into lmp-mr1-dev
* commit '3b2ba44be3699a4e133d7a09cb4ed182405d7ff9':
  Improve logging around performance critical events
2014-11-20 17:50:20 +00:00
Chris Craik
3b2ba44be3 Merge "Improve logging around performance critical events" into lmp-mr1-dev 2014-11-20 17:36:26 +00:00
John Reck
b13de07f0b If the surface is lost, release it
Bug: 18403246

Surface#isValid() is used throughout ViewRootImpl to
determine Surface validity. However, this will return true
until Surface#release() is called, even if the underlying
BufferQueue is abandoned. We have a very strong signal
from EGL that the Surface is now dead, so use that signal
to call Surface#release() so ViewRootImpl's isValid() usage
will be valid.

Change-Id: Ib355560ef4ca7625c3c784174863d2d55267d93d
2014-11-20 07:38:00 -08:00
Chris Craik
70850ea258 Improve logging around performance critical events
bug:17702227

Add details useful to developers (such as layer size/View name), and
switch away from logging implementation names/details, since they
are generally not relevant to developers.

Change-Id: Iee605d182f241450f7e75a6d0c283d51fa1312f5
2014-11-19 17:24:30 -08:00
Alan Viverette
7c9746d4ef Move default token handling into WindowManagerImpl
BUG: 18451795
Change-Id: I1fc6db988ad879fded5318f33d08a4f09da38907
2014-11-19 17:02:16 -08:00
Chet Haase
a79d2e8664 am 85603a7f: Merge "Fix seeking and scaled duration behavior" into lmp-mr1-dev
* commit '85603a7faf0a8067803ae6424bc590f6e2a17e07':
  Fix seeking and scaled duration behavior
2014-11-20 00:07:50 +00:00
Chet Haase
85603a7faf Merge "Fix seeking and scaled duration behavior" into lmp-mr1-dev 2014-11-20 00:02:09 +00:00
Chet Haase
0d1c27a713 Fix seeking and scaled duration behavior
The animation scaled was not being factored in early enough in the
activity lifecycle. Also, setCurrentPlaytTime() was not accounting for
the scaled duration correctly. Finally, added setCurrentFraction() as
a more general-purpose seeking facility.

Issue #18222006 Animator duration scale ignored in some situations
Issue #17951668 add ability to seek fraction, not just time

Change-Id: Idad401f5ff5026d7046c36eee09d78a4793215dc
2014-11-19 15:29:47 -08:00
Filip Gruszczynski
6eafa902cb SwipeDismissLayout makes activity opaque only after entry animation ends.
Bug: 18340863
Change-Id: Ic60fa2463618f86b1ae23fc4a0c06cd348f28334
2014-11-19 11:05:43 -08:00
Chris Craik
df57b22bc6 am 0455174d: Merge "Update clipToPadding docs" into lmp-mr1-dev
* commit '0455174d347bf451e830584bc3f6f35c3c560282':
  Update clipToPadding docs
2014-11-19 00:09:32 +00:00
Chris Craik
0455174d34 Merge "Update clipToPadding docs" into lmp-mr1-dev 2014-11-18 23:33:30 +00:00
John Reck
51ce9cacb6 am 37cdc191: Merge "Don\'t invalidate() on setClipBounds" into lmp-mr1-dev
* commit '37cdc191e380cca8edcc58f116b9b319bea1c140':
  Don't invalidate() on setClipBounds
2014-11-18 21:46:44 +00:00
John Reck
37cdc191e3 Merge "Don't invalidate() on setClipBounds" into lmp-mr1-dev 2014-11-18 21:41:17 +00:00
John Reck
fcd38813ef am 3cf61d4b: Merge "Fix invalidateOutline" into lmp-mr1-dev
* commit '3cf61d4b79a22f89b8fb839ef9ee2146fac4e703':
  Fix invalidateOutline
2014-11-18 21:37:40 +00:00
John Reck
9029e5ecd0 Don't invalidate() on setClipBounds
Bug: 17510133

This is a RenderNode property now, so use
the faster invalidateViewProperty() shortcut
since a re-record isn't necessary

Change-Id: If3999bce9a1fb9b60e42f0ee624bb554361f96ac
2014-11-17 15:28:55 -08:00
John Reck
c99d3c99f8 Fix invalidateOutline
Bug: 18175261

invalidateOutline was switched to a lazy-method, but this doesn't
work because invalidateViewProperty intentionally does not
do a traversal, therefore the invalidate was never consumed.

However it was attempting to be lazy about work that is cheap to
do, so nuke the lazy aspect and restore invalidateOutline's previous,
correct behavior. rebuildOutline is kept to avoid triggering
traversals in places they are not needed

Change-Id: I70f8cbacd54a607c0bf0bc7fe6eea78554cb2ea3
2014-11-17 12:35:53 -08:00
John Reck
ff5978988a am b3ec64e1: Merge "Wire up surface width/height to lockHardwareCanvas" into lmp-mr1-dev
* commit 'b3ec64e1005c82d41b6d1a1d86b6933fd25d87d2':
  Wire up surface width/height to lockHardwareCanvas
2014-11-17 15:23:11 +00:00
John Reck
b3ec64e100 Merge "Wire up surface width/height to lockHardwareCanvas" into lmp-mr1-dev 2014-11-17 15:19:23 +00:00
Chris Craik
b165296a9c Update clipToPadding docs
Clarify that it doesn't clip anything but children,
and only if padding is non-zero.

Change-Id: I3e3ae31f52fc22304eb13d88765f9f4fc6e43e66
2014-11-14 17:05:06 -08:00
John Reck
f1812b17bb am 8f5839be: Merge "Document that circular reveal is async" into lmp-mr1-dev
* commit '8f5839be21f028885b6171e75cba479a020a3a0c':
  Document that circular reveal is async
2014-11-14 22:07:29 +00:00
John Reck
8f5839be21 Merge "Document that circular reveal is async" into lmp-mr1-dev 2014-11-14 22:03:33 +00:00
John Reck
a3b35907de Document that circular reveal is async
Bug: 18058966
Change-Id: I64b35441f12bec433f633580319ec73c1e11e7a2
2014-11-14 14:02:55 -08:00
Raph Levien
58fa003577 am 505ffe36: Merge "Check bounds on CharSequence drawText methods" into lmp-mr1-dev
* commit '505ffe3633f45778e493e23ad8c163f09ed1dbb5':
  Check bounds on CharSequence drawText methods
2014-11-14 20:56:46 +00:00
Raph Levien
505ffe3633 Merge "Check bounds on CharSequence drawText methods" into lmp-mr1-dev 2014-11-14 20:54:29 +00:00