Move MMS api to using content provider rather than byte[] to pass MMS message contents.
Rebased and merged into TOT.
Change-Id: I3509b2774b1cb30a1c8100bb25d283140c963b6b
Using compression and decompression for moving bitmap data
acorss processes is slow as compression is expensive. This
change switches to using direct streaming of the bitmap
data.
bug:15938254
Change-Id: I78bc450031ee60ada4c3b66f14586a73c72ce34f
1. Fixed a crash when orientation changes and the content
is scrolled due to wrong size bitmap being requested.
2. Closed a file dscriptior that was being left open.
3. Clearing the bitmap before passing it to the renderer to
ensure it is white for pixels not touched when rendering.
4. Removed debug logs.
5. Switched to the correct layout manager for RecyclerView.
bug:16966145
Change-Id: I8ab9d22635c93cac5ff85c6f4b5d82e58cd8df5c
Bug: 17208461
There's a potential race condition between HardwareRenderer.destroy()
being called (which calls destroyCanvasAndSurface()) and the renderer
being finalized (which is what calls freePrefetchedLayers), during which
time it's possible we get a TRIM_MEMORY_COMPLETE and destroy the EGL
context.
Fix this race condition by moving stopDrawing() and freePrefetchedLayers()
into destroyCanvasAndSurface() where they should have been in the first
place.
Also, if we hit the assertion failure, dump the current state of
Caches to try and provide more context for the failure.
Change-Id: Ife0ba3562041e8b08e87e3e13640472b3004eed6
BUG: 17322886
bindService() to an invalid service might never actually result in
onServiceConnected being called , for e.g. if the client service doesn't
actually implement JobService. This wastes an execution slot as we end
up waiting forever.
Also made the javadocs clearer for the JobScheduler class.
Change-Id: Ie15ebbe18c0b7579f2ab77dd46428d354ef632c3
Ensure background is mutated before changing its alpha.
Ensure resetViews is called when activity options is set.
Bug 16215650
Change-Id: I8482532777652a93b2c0d05c1b902516d281baf0
Bug 17372872
If ChangeTransform was interrupted after its animation completed,
but before the parent's Transition completed, it would think
that its start location was relative to its parent's current
position and not relative to its end position.
Change-Id: Iad25aafc03c41c1e0a7530548fb38acafbaffb8e
Chrome seems to use this method of moving itself to the back on
back presses, causing it to escape lock task mode.
Bug: 17365048
Change-Id: I2e8bc614f771951ed96cfc36f9789d12b1742e70
Rip out the startActivityFromInCall API until we can replace it with
something better.
Bug: 16988478
Change-Id: Ib7b0493353262d9e795d78b78a98dd29e7a3f88e
Recents was depending on using moveTaskToFront() to also move the
task's stack to the front. But that only works if the task is
actually in the stack because moveTaskToFront() doesn't do the
stack movement if the task is no longer in the stack. This change
calls ActivityStack.moveToFront() in addition to moveTaskToFront().
Fixes bug 17361027.
Also logs callingUid when starting an activity.
Change-Id: I023f956bafae00f45c58757ff5707ce520f02b55
(i) is now the system's app notification info panel,
allowing you to block, prioritize, and things like
that.
(gear) is new; it will appear only if your package has a
notification settings activity of its own. If the user
taps this you'll be launched with some useful extras:
notification_id and notification_tag, matching the id
and tag of the notification the user longpressed.
Bug: 17005733
Change-Id: Ib09264c2564152ab27e744c47ed87a0e9223656f
Also added logic to provide nicer animations when falsing.
In addition adapted the clock scale slightly if dragging
in the void.
Bug: 15433087
Change-Id: I4d8eb26cc81d22647ef4d2eca1e69b4994a7f1e2
