bug:8037003
-Merges replay methods
-Bounds checking for DrawBitmapMesh, DrawRects and DrawDisplayList
-Use clip as bounds for otherwise unbounded draw operations
Clip-as-bounds is correct for drawColor and functor, but other draw
operations (pos text, text on path, and layers) still need true bounds
calculation
Change-Id: I5d5149d2c624f01e3fe46628bf156e835e69b9d5
External bug: http://code.google.com/p/android/issues/detail?id=49379
Drawing was previously wrapped in a try/finally block which was silently
swallowing user code exceptions. For instance, if a View throws a
NullPointerException in its onDraw() method, the previous implementation
would silently crash in native code.
This change extracts the section that builds display lists into a new
method that does not contain any try/finally block.
In addition, this change logs any exception thrown while drawing
display lists.
Change-Id: I0abffa4c9183d41aac8b0f8442813e56b957f08f
The exposed APIs are slightly simpler than the full APIs used internally.
Only APIs useful to applications are exposed.
Change-Id: Ie03014628d40ad5ef63dedbc52ce3def84429d54
The window manager now keeps track of the overscan of
each display, with an API to set it. The overscan impacts
how it positions windows in the display. There is a new set
of APIs for windows to say they would like to go into the
overscan region. There is a call into the window manager to
set the overscan region for a display, and it now has a
concept of display settings that it stores presistently.
Also added a new "wm" command, moving the window manager
specific commands from the "am" command to there and adding
a new now to set the overscan region.
Change-Id: Id2c8092db64fd0a982274fedac7658d82f30f9ff
Eliminate the short period of time between the tap timeout and long
press causing onSingleTapConfirmed not to be dispatched when apps
expect.
Bug 8124095
Change-Id: I9841ab2a8eb3fee7d57e744e1c8e0e42e108d5f6
SurfaceControl is the window manager side; it can
control the attributes of a surface but cannot push buffers
to it. Surface on the other hand is the application (producer)
side and is used to push buffers to the surface.
Change-Id: Ib6754c968924e87e8dd02a2073c7a447f729f4dd
Bug #7146141
This new cache is used in a similar way to LayerCache. It helps
reuse already allocated stencil buffers and thus avoid churning
memory on every frame.
Change-Id: I19551d72da52c40039e65904563600e492c8b193
This reverts commit 6c0307dd0a, reversing
changes made to a2cd828b74.
Conflicts:
packages/SystemUI/res/values-sv/strings.xml
Change-Id: Ia178efe8b14751583d47b2826bfe3d3d5463dd2e
Previous code incorrectly included pivot point in calculation, which
resulted in offset rectangle
Issue #8178031 getHitRect() incorrect when view is transformed
Change-Id: Ie6c4c51614577d208016584cf1bf4eed570f730d
Loading of the resource id name was unnecessarily complex and
was generating spam in the log as a result.
bug:8153518
Change-Id: Ib2c1118fd36b86ba8ce803ba3190fa01ab06e8c3
- add RTL apis to ViewParent
- remove awful and performance bottleneck casts in View during LayoutDirection,
TextDirection and TextAlignment resolutions.
Change-Id: I92a56b16382e13c28c1dad35f6f205620eb01bfe
- Following disable and reenable of the WindowOrientationListener
the state was the same as before. State should be reset to default.
- Provide a Handler to the sensor task to deliver events on the
same Thread that WindowManagerPolicy operates on.
- Expand lock protection to all of WindowOrientationListener.
- Move WindowOrientationListener to policy package.
- Make SensorEventListenerImpl non-static.
Fixes bug 7964531.
Change-Id: I17cecf3d0b6d125cb3e4d7350c3adb3f62b684bd
# By Ki-Hwan Lee
# Via Android Git Automerger (1) and others
* commit '110fe9b303be527c30a8af95b6fb7afac12968b5':
Fix ViewRootImpl to find missing focus using D-pad.
# By Ki-Hwan Lee
# Via Gerrit Code Review (1) and Ki-Hwan Lee (1)
* commit 'fe1f3a1beff9f73f6a04bcc35239038a21bc38ff':
Fix ViewRootImpl to find missing focus using D-pad.
A layer will, by definition, clip its content already. The extra
clip added by the UI toolkit/renderer is wasteful and in some
situations very expensive. If a layered view is rotated, the clip
will be transformed into a non-rectangular region, thus forcing
the renderer to rely on the stencil buffer. The whole point of
using layers is to draw efficiently so let's avoid this case.
Change-Id: I813465f6468cc386b84a4d058c11a258fef8507a
Ensure profileRendering reschedules a previously created profiler callback if
disabled and reenabled
Change-Id: I181ac5043966b99bb8d21e69ce4fdd7494102acb
By using D-pad, no-focus in non touch mode is rare but legal in a case like below.
1. The first request to get focus for a new activity is handled in the first
performTraversals() call when activity is not ready for a complete view hierarchy.
So there might be no focusable yet.
2. If the activity has some menus, ActionMenuView can be attached to the view hierarchy
in the PhoneWindow.preparePanel() soon.
So there can be focusables but still not focused.
Fixed ViewRootImpl.deliverKeyEventPostIme() to handle this case to resurrect a focus
if there are focusables.
How to reproduce:
(1) Open "API Demos" application -> Views -> Search View
(2) Select "Action Bar" item using the D-pad
(3) Try to focus the Search View, using the D-pad.
Change-Id: Ic379774f0307f168f0ed775d0f6a9078ac5c9713
