OSDN Git Service

android-x86/frameworks-base.git
10 years agoMerge "Fixed OOM due to visibility transitions." into lmp-dev
George Mount [Mon, 15 Sep 2014 18:12:21 +0000 (18:12 +0000)]
Merge "Fixed OOM due to visibility transitions." into lmp-dev

10 years agoMerge "Fixed enter/exit transitions working against shared elements." into lmp-dev
George Mount [Mon, 15 Sep 2014 18:12:12 +0000 (18:12 +0000)]
Merge "Fixed enter/exit transitions working against shared elements." into lmp-dev

10 years agoMerge "Camera2: Clean up corner case error handling" into lmp-dev
Eino-Ville Talvala [Mon, 15 Sep 2014 18:11:10 +0000 (18:11 +0000)]
Merge "Camera2: Clean up corner case error handling" into lmp-dev

10 years agoFixed enter/exit transitions working against shared elements.
George Mount [Sun, 14 Sep 2014 23:48:21 +0000 (16:48 -0700)]
Fixed enter/exit transitions working against shared elements.

Bug 17499971

The enter and exit transitions were targeting the shared element.
This means you'd get double transitions on the shared element.

Also, the enter transition was not properly initialized to have
an invalid target so it was targeting all views.

Change-Id: I4f3510c70bea90f09be01776bc0ad1ffb42d99d5

10 years agoFixed OOM due to visibility transitions.
George Mount [Fri, 12 Sep 2014 23:39:04 +0000 (16:39 -0700)]
Fixed OOM due to visibility transitions.

Have maximum size for Visibility transition bitmaps.

Also changed Visibility to avoid transitioning out child views
while transitioning out their parents.

Also moved view-to-bitmap code to common utility.

Bug 17469198

Change-Id: I05f069a267539b479da1b39cdad2cc9270c5c381

10 years agoMerge "Remove a minimum wake-up time for a job" into lmp-dev
Matthew Williams [Mon, 15 Sep 2014 17:56:50 +0000 (17:56 +0000)]
Merge "Remove a minimum wake-up time for a job" into lmp-dev

10 years agoMerge "Changed the swipe up search affordance" into lmp-dev
Selim Cinek [Mon, 15 Sep 2014 17:18:39 +0000 (17:18 +0000)]
Merge "Changed the swipe up search affordance" into lmp-dev

10 years agoMerge "Force isolation of clip and matrix within layer" into lmp-dev
Chris Craik [Mon, 15 Sep 2014 15:59:55 +0000 (15:59 +0000)]
Merge "Force isolation of clip and matrix within layer" into lmp-dev

10 years agoChanged the swipe up search affordance
Selim Cinek [Thu, 11 Sep 2014 13:11:00 +0000 (15:11 +0200)]
Changed the swipe up search affordance

The previous card animation is removed and replaced by a animating
circle with a shadow. Also fixes several cases where the card could
either get stuck and the affordance was not launched.

Bug: 17457300
Bug: 17444236
Change-Id: I005313a1dbe63d338490e6100dd3bd01e35687ba

10 years agoMerge "Add hidden API for pausing and resuming NFC polling." into lmp-dev
Andres Morales [Mon, 15 Sep 2014 15:10:43 +0000 (15:10 +0000)]
Merge "Add hidden API for pausing and resuming NFC polling." into lmp-dev

10 years agoMerge "Fix priority for recent task thumbnail loader" into lmp-dev
Jorim Jaggi [Mon, 15 Sep 2014 15:02:40 +0000 (15:02 +0000)]
Merge "Fix priority for recent task thumbnail loader" into lmp-dev

10 years agoFix priority for recent task thumbnail loader
Jorim Jaggi [Mon, 15 Sep 2014 14:07:29 +0000 (16:07 +0200)]
Fix priority for recent task thumbnail loader

Follow up CL for Id9a32f87ca3f9f19e5cad5f115d54b19b26f4f93

Bug: 16513124
Change-Id: I3e95d0ae56a273c482db31b377f66fe7397d01f8

10 years agoMerge "Fix exit transitions hiding views permanently." into lmp-dev
George Mount [Mon, 15 Sep 2014 14:06:50 +0000 (14:06 +0000)]
Merge "Fix exit transitions hiding views permanently." into lmp-dev

10 years agoMerge "Fix cyclic mute" into lmp-dev
Jungshik Jang [Mon, 15 Sep 2014 12:36:24 +0000 (12:36 +0000)]
Merge "Fix cyclic mute" into lmp-dev

10 years agoMerge "Update briefcase icons" into lmp-dev
Alexandra Gherghina [Mon, 15 Sep 2014 11:45:59 +0000 (11:45 +0000)]
Merge "Update briefcase icons" into lmp-dev

10 years agoFix cyclic mute
Jungshik Jang [Mon, 15 Sep 2014 08:48:41 +0000 (17:48 +0900)]
Fix cyclic mute

Bug: 17502271
Change-Id: I837315afce8193ac1948fffed66cbb8c755968e2

10 years agoAggressive tuning of recents memory usage
Jorim Jaggi [Fri, 12 Sep 2014 21:28:58 +0000 (23:28 +0200)]
Aggressive tuning of recents memory usage

Limit the thumbnail count to a maximum of 10 thumbnails and the app
icon count to a maximum of 20 thumbnails, instead of using a
byte-size based limit.

Also prune entries immediately when leaving recents.

Bug: 16513124
Change-Id: Id9a32f87ca3f9f19e5cad5f115d54b19b26f4f93

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 04:19:47 +0000 (04:19 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 04:19:45 +0000 (04:19 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 04:19:43 +0000 (04:19 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 04:05:55 +0000 (21:05 -0700)]
Import translations. DO NOT MERGE

Change-Id: Ib73a5356d8804c8216be0ce59c84637ff609e81a
Auto-generated-cl: translation import

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 04:04:22 +0000 (21:04 -0700)]
Import translations. DO NOT MERGE

Change-Id: Ie385c9b051b20bf52b7c0b266ca4c9147f7fbb90
Auto-generated-cl: translation import

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 04:02:30 +0000 (21:02 -0700)]
Import translations. DO NOT MERGE

Change-Id: Ic6c7d7ca45834f3f0b73967347232991b5c31910
Auto-generated-cl: translation import

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 02:45:02 +0000 (02:45 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 02:44:59 +0000 (02:44 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoMerge "Import translations. DO NOT MERGE" into lmp-dev
Baligh Uddin [Mon, 15 Sep 2014 02:44:56 +0000 (02:44 +0000)]
Merge "Import translations. DO NOT MERGE" into lmp-dev

10 years agoMerge "Remove unnecessary warning." into lmp-dev
Jungshik Jang [Mon, 15 Sep 2014 02:23:06 +0000 (02:23 +0000)]
Merge "Remove unnecessary warning." into lmp-dev

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 02:17:12 +0000 (19:17 -0700)]
Import translations. DO NOT MERGE

Change-Id: I57bb85dcb94f83e095e07ee4956d786849284688
Auto-generated-cl: translation import

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 02:12:57 +0000 (19:12 -0700)]
Import translations. DO NOT MERGE

Change-Id: Idb536e9edfaee903119b8d84188d013e045f5fba
Auto-generated-cl: translation import

10 years agoImport translations. DO NOT MERGE
Baligh Uddin [Mon, 15 Sep 2014 02:11:15 +0000 (19:11 -0700)]
Import translations. DO NOT MERGE

Change-Id: Ifb234cb1ae0c96f393e3b203488f61467d97c267
Auto-generated-cl: translation import

10 years agoNavigationBarView: Fix NPE in Drawable.setVisible()
Dan Sandler [Mon, 15 Sep 2014 01:24:15 +0000 (21:24 -0400)]
NavigationBarView: Fix NPE in Drawable.setVisible()

Bug: 17495735
Change-Id: I9d5ed5a324ae5c08ea0c225104c41b44462aff22

10 years agoRemove unnecessary warning.
Jungshik Jang [Fri, 12 Sep 2014 09:49:08 +0000 (18:49 +0900)]
Remove unnecessary warning.

Bug: 17501260

Change-Id: I5e7edd6d1fc179c00306d745adc99bebdcb65316

10 years agoMerge "Allow creating EAP_TLS network without specifying Phase2.None" into lmp-dev
Prem Kumar [Mon, 15 Sep 2014 00:51:12 +0000 (00:51 +0000)]
Merge "Allow creating EAP_TLS network without specifying Phase2.None" into lmp-dev

10 years agoMerge "A little more work on issue ##16907799." into lmp-dev
Dianne Hackborn [Mon, 15 Sep 2014 00:26:59 +0000 (00:26 +0000)]
Merge "A little more work on issue ##16907799." into lmp-dev

10 years agoA little more work on issue ##16907799.
Dianne Hackborn [Mon, 15 Sep 2014 00:23:54 +0000 (17:23 -0700)]
A little more work on issue ##16907799.

I got distracted in the middle of it, and forget to finish
up with the test to not kill processes if they aren't using
an auto create binding.

Change-Id: Ieecfe97fa3208e50cb91ba94be2a8659d128b0de

10 years agoMerge "Clean up view drawable tinting methods, fix default modes" into lmp-dev
Alan Viverette [Sun, 14 Sep 2014 23:36:05 +0000 (23:36 +0000)]
Merge "Clean up view drawable tinting methods, fix default modes" into lmp-dev

10 years agoMerge "Simplify progress bar paths" into lmp-dev
Alan Viverette [Sun, 14 Sep 2014 22:51:42 +0000 (22:51 +0000)]
Merge "Simplify progress bar paths" into lmp-dev

10 years agoMerge "Magic null-background filling for PhoneWindows" into lmp-dev
Adam Powell [Sun, 14 Sep 2014 22:51:28 +0000 (22:51 +0000)]
Merge "Magic null-background filling for PhoneWindows" into lmp-dev

10 years agoForce isolation of clip and matrix within layer
Chris Craik [Sun, 14 Sep 2014 22:49:54 +0000 (15:49 -0700)]
Force isolation of clip and matrix within layer

bug:17379260

Unclipped save layers need to isolate their clips and matrices
so that the save time readout of background content matches clip and
position with the restore/compose step.

Change-Id: I2de474cda76e960d080852f1716d6ddfa8a512d4

10 years agoClean up view drawable tinting methods, fix default modes
Alan Viverette [Sun, 14 Sep 2014 22:48:50 +0000 (15:48 -0700)]
Clean up view drawable tinting methods, fix default modes

Calling setTint now only modifies the tint. It won't force a mode change.

BUG: 17494736
Change-Id: I91392634869ed23981d8e61a403bb2be42aa7a07

10 years agoMagic null-background filling for PhoneWindows
Adam Powell [Thu, 11 Sep 2014 22:09:36 +0000 (15:09 -0700)]
Magic null-background filling for PhoneWindows

In the past it's been a recommended approach to avoiding overdraw for
apps to set their window background to null at runtime if their
content view fully covers their window surface. The problem with this
is the IME.

The IME can force a resize of the window at unexpected times and
unless an app has been configured to fit system windows and manually
cover the padded area that the IME window covers, the asynchronous
nature of the IME-show process can leave surface buffer garbage
visible to the user. In previous platform versions this wasn't an
issue since pre-renderthread we would always animate a crossfade from
the closed to open state. This animation was always a bit of a hack
since it could break the contract of requestLayout/invalidate on the
view hierarchy - it could result in a draw happening into the saved
"before" state of the crossfade before a pending layout.

Now that this has been cleaned up the buffer garbage is sometimes
visible.

To prevent this, PhoneWindow now detects the state of a null window
background and draws solid rects into the area not covered by a
window's content. Which color is determined by the window context's
theme, though this is not a public API available to apps.

Bug 17006497

Change-Id: I714439a1608c4ae135f3d9d49bb165330d9fbe9f

10 years agoMerge "temporarily enable wifi verbose logging in download manager test" into lmp-dev
Guang Zhu [Sun, 14 Sep 2014 22:35:29 +0000 (22:35 +0000)]
Merge "temporarily enable wifi verbose logging in download manager test" into lmp-dev

10 years agotemporarily enable wifi verbose logging in download manager test
Guang Zhu [Sun, 14 Sep 2014 22:30:29 +0000 (15:30 -0700)]
temporarily enable wifi verbose logging in download manager test

also temporarily compile against platfrom, not SDK, since we are
using private API

Bug: 17446070
Change-Id: I6da18aca2d5eaf305940514186ef2434e52f15b9

10 years agoSimplify progress bar paths
Alan Viverette [Sun, 14 Sep 2014 22:10:54 +0000 (15:10 -0700)]
Simplify progress bar paths

BUG: 17366831
Change-Id: I7efa215d399e4bf47b41949d313fba4487f068a0

10 years agoMerge "Fix issue #16907799: Processes containing bound services..." into lmp-dev
Dianne Hackborn [Sun, 14 Sep 2014 22:02:35 +0000 (22:02 +0000)]
Merge "Fix issue #16907799: Processes containing bound services..." into lmp-dev

10 years agoMerge "Route volume commands whenever media stream is in use" into lmp-dev
RoboErik [Sun, 14 Sep 2014 21:45:01 +0000 (21:45 +0000)]
Merge "Route volume commands whenever media stream is in use" into lmp-dev

10 years agoCamera2: Clean up corner case error handling
Eino-Ville Talvala [Sun, 14 Sep 2014 21:07:05 +0000 (14:07 -0700)]
Camera2: Clean up corner case error handling

- If a session is closed, and a new session is created immediately
afterwards, but then fails to be configured, the first session sees an
onUnconfigured call which it wasn't expecting, and throws an
exception on an internal thread, leading to app death.
Add a guard against this case.

- If the lower levels skip a frame (illegal per design), be slightly more
robust to that by accepting any successful result as the latest completed
frame, instead of just incrementing the completed frame count. This will
lead to missing results, but should allow shutdown, etc, to complete
cleanly.

- Convert TIMED_OUT error codes to CAMERA_ERROR CameraAccessExceptions.
This is a common error code returned by waitUntilIdle.

Also, improve debug logging to log a session ID with verbose logging,
and add a few verbose logs.

Bug: 16899526
Change-Id: I7a31f0a12effc2611e1f9c2408224ee82c37c912

10 years agoMerge "Fix jank while unlocking and security is set" into lmp-dev
Jorim Jaggi [Sun, 14 Sep 2014 21:40:13 +0000 (21:40 +0000)]
Merge "Fix jank while unlocking and security is set" into lmp-dev

10 years agoFix issue #16907799: Processes containing bound services...
Dianne Hackborn [Sun, 14 Sep 2014 21:21:18 +0000 (14:21 -0700)]
Fix issue #16907799: Processes containing bound services...

...are killed over eagerly.

When the current foreground activity is moving to the background,
it was briefly going through the CACHED_ACTIVITY state before the
correct LAST_ACTIVITY state, allowing its bound service processes
to be killed (because they went in to the cached list).  To solve
this, as long as a process has stopping activities, it won't go
lower than LAST_ACTIVITY.

Also fixed a problem where we could put a process in CACHED_EMPTY
instead of CACHED_ACTIVITY_CLIENT.  There were a number of cases
in the binding flow and also the client process state transitions
where we would not correctly updateing the bound client activity
state.

And add some sanity code so that if a process hosting a
service is killed, and a client process of that service is in the
cached state, we kill the client process.  This avoids situations
where we can start thrashing around in the cached list because we
are restarting process for no reason -- since they will just
continue to be cached.

Finally, tune the process LRU list to allow twice as many cached
activity processes (from 8 to 16), so we can make better use of
the RAM we have available these days.

Change-Id: Ib0cdf78c321cbb035259fc9dd6ee27b5ba1f90c5

10 years agoMerge "set correct bounds in ActionBar based on visibility of ActionBarView" into...
Chet Haase [Sun, 14 Sep 2014 21:30:20 +0000 (21:30 +0000)]
Merge "set correct bounds in ActionBar based on visibility of ActionBarView" into lmp-dev

10 years agoRoute volume commands whenever media stream is in use
RoboErik [Sun, 14 Sep 2014 20:54:31 +0000 (13:54 -0700)]
Route volume commands whenever media stream is in use

Found a regression in volume handling. Previously we handled
volume commands as long as the media stream was active but we were only
handling them when there was an active session on L. This adds a check to
make sure we handle volume if anything is playing on the media stream.

bug:17498479
Change-Id: Iddd745c8a762cf7ebedb37f1b26fc934db01fba0

10 years agoset correct bounds in ActionBar based on visibility of ActionBarView
Chet Haase [Sat, 13 Sep 2014 01:03:55 +0000 (18:03 -0700)]
set correct bounds in ActionBar based on visibility of ActionBarView

ActionBarContainer was setting the bounds of its background assuming the
visibility of the ActionBarView. But that view becomes GONE when the
ActionBarContextView is visible, causing artifacts such as wrong shadows
when resized (as in custom configuration changes).

Issue #17280341 Quantum: drop shadow on CAB has wrong width after rotation on L, when configuration change is handled by the app

Change-Id: I07e57f00e27b41d5370cb9440b35734a8ec10f3a

10 years agoMerge "Reduce instructions required to obtain and recycle TypedArray" into lmp-dev
Alan Viverette [Sun, 14 Sep 2014 20:04:38 +0000 (20:04 +0000)]
Merge "Reduce instructions required to obtain and recycle TypedArray" into lmp-dev

10 years agoMerge "Correct test data size in keystore signing and verification tests." into lmp-dev
Shawn Willden [Sun, 14 Sep 2014 19:58:05 +0000 (19:58 +0000)]
Merge "Correct test data size in keystore signing and verification tests." into lmp-dev

10 years agoMerge "Saver: PowerManager call to set low power mode." into lmp-dev
John Spurlock [Sun, 14 Sep 2014 19:36:11 +0000 (19:36 +0000)]
Merge "Saver: PowerManager call to set low power mode." into lmp-dev

10 years agoMerge "Fix creation issue of stack trace dump directory" into lmp-dev
Ramanan Rajeswaran [Sun, 14 Sep 2014 19:20:35 +0000 (19:20 +0000)]
Merge "Fix creation issue of stack trace dump directory" into lmp-dev

10 years agoMerge "Fix crash in TranserPipe." into lmp-dev
Ramanan Rajeswaran [Sun, 14 Sep 2014 19:20:31 +0000 (19:20 +0000)]
Merge "Fix crash in TranserPipe." into lmp-dev

10 years agoMerge "make wifi cellular params configurable" into lmp-dev
vandwalle [Sun, 14 Sep 2014 19:17:05 +0000 (19:17 +0000)]
Merge "make wifi cellular params configurable" into lmp-dev

10 years agoMerge "Correctly propagate SW_CAMERA_LENS_COVER_BIT" into lmp-dev
Michael Wright [Sun, 14 Sep 2014 18:19:49 +0000 (18:19 +0000)]
Merge "Correctly propagate SW_CAMERA_LENS_COVER_BIT" into lmp-dev

10 years agoSaver: PowerManager call to set low power mode.
John Spurlock [Sun, 14 Sep 2014 15:10:22 +0000 (11:10 -0400)]
Saver: PowerManager call to set low power mode.

- Add an explicit power manager call to set the low power mode state,
  instead of trying manage everything around a single setting.
- When low-power mode is triggered by falling below the configured
  threshold, it does not update the setting.
- The "is-enabled" api returns setting || below configured trigger.
- Move the snooze management into the new api call.
- Callers (sysui + settings) updated to use the api instead of the
  setting.
- Handles the case where the level does an unpowered leap out of the
  low battery level. (Possible if powered in-between while the device
  is off)

Bug:17460535
Change-Id: Ic030504c9cad9868a7137abbe837b170da37852b

10 years agoMerge "Add null checks for quickly disappearing Displays" into lmp-dev
Craig Mautner [Sun, 14 Sep 2014 18:01:39 +0000 (18:01 +0000)]
Merge "Add null checks for quickly disappearing Displays" into lmp-dev

10 years agoMerge "Customized ripple animation for navigation bar" into lmp-dev
Dan Sandler [Sun, 14 Sep 2014 17:44:31 +0000 (17:44 +0000)]
Merge "Customized ripple animation for navigation bar" into lmp-dev

10 years agoCorrectly propagate SW_CAMERA_LENS_COVER_BIT
Michael Wright [Sun, 14 Sep 2014 02:41:20 +0000 (19:41 -0700)]
Correctly propagate SW_CAMERA_LENS_COVER_BIT

Bug: 16034563
Change-Id: I3402d42056cda47fddadd09f838b1082d93fb00d

10 years agoTIF: Turn debugging off
Jae Seo [Sun, 14 Sep 2014 17:30:05 +0000 (10:30 -0700)]
TIF: Turn debugging off

Bug: 17476383
Change-Id: I28ff54a635bf6468c231efe776dbc8e35460d542

10 years agoCorrect test data size in keystore signing and verification tests.
Shawn Willden [Thu, 11 Sep 2014 22:08:44 +0000 (16:08 -0600)]
Correct test data size in keystore signing and verification tests.

The test is sending too much data to be signed, which should actually
fail, and does on Volantis.  Apparently the other keymaster implementors
do something to pass it, because shamu and hammerhead pass, but the test
is wrong.

Change-Id: Ic616a551567d64f5d87d9607ceb08afa7be74f9d

10 years agoMerge "Handle null values in registerService" into lmp-dev
Christopher Lane [Sun, 14 Sep 2014 17:06:45 +0000 (17:06 +0000)]
Merge "Handle null values in registerService" into lmp-dev

10 years agoPipe TelephonyManager.getCallState through TelecommManager
Yorke Lee [Sat, 13 Sep 2014 00:58:48 +0000 (17:58 -0700)]
Pipe TelephonyManager.getCallState through TelecommManager

* Add TelecommManager.getCallState (hidden API)
* Make TelephonyManager.getCallState call through to
TelecommManager, to be consistent with
TelephonyManager.ACTION_PHONE_STATE_CHANGED broadcasts for
overall call state. Telephony continues to manage call states for
individual subscriptions.

Bug: 17378767
Change-Id: Ia5e8b21df801ed3af4f6e14c110a72c92f077f88

10 years agoFix creation issue of stack trace dump directory
Jing Ji [Thu, 7 Aug 2014 18:44:05 +0000 (13:44 -0500)]
Fix creation issue of stack trace dump directory

Use the dirname instead of the filename to create the directory

Change-Id: I16e49303b2ff5e2592ed60eab766db32d02262fe

10 years agoFix crash in TranserPipe.
Xin Guan [Thu, 21 Aug 2014 18:05:34 +0000 (13:05 -0500)]
Fix crash in TranserPipe.

Sometimes the pipe has been closed when it's thread tries to access

E AndroidRuntime: *** FATAL EXCEPTION IN SYSTEM PROCESS: TransferPipe
E AndroidRuntime: java.lang.NullPointerException: Attempt to invoke virtual method 'java.io.FileDescriptor android.os.ParcelFileDescriptor.getFileDescriptor()' on a null object reference
E AndroidRuntime:  at com.android.internal.os.TransferPipe.run(TransferPipe.java:184)
E AndroidRuntime:  at java.lang.Thread.run(Thread.java:818)

Change-Id: I0fcd4a3334b49972903f2cb0edb51323ba3f49e5

10 years agoFix jank while unlocking and security is set
Jorim Jaggi [Sat, 13 Sep 2014 01:57:32 +0000 (03:57 +0200)]
Fix jank while unlocking and security is set

- Delay showing the bouncer and playing the entrance animation for a
  couple of frames.
- Do not disable back button on the normal Keyguard. This makes the
  entrance animation nicer and fixes some jank due to the need to
  draw into both windows.
- Prevent setPadding(...) call when nothing changes. setPadding would
  invalidate the padding for the whole view hierarchy, which results
  in a slower measure() when setting the window to invisible.

Bug: 17419960
Change-Id: I4a239d4af40ad86875e4a0dd08473f19a5c9b961

10 years agoMerge "Minimize the number of default enabled IMEs part 2" into lmp-dev
Yohei Yukawa [Sun, 14 Sep 2014 07:48:26 +0000 (07:48 +0000)]
Merge "Minimize the number of default enabled IMEs part 2" into lmp-dev

10 years agoMerge "Adding null check for invalid recents task list. (Bug 17476926)" into lmp-dev
Winson Chung [Sun, 14 Sep 2014 07:12:37 +0000 (07:12 +0000)]
Merge "Adding null check for invalid recents task list. (Bug 17476926)" into lmp-dev

10 years agoMerge "Fixing issue with stack not being fully centered. (Bug 17015153)" into lmp-dev
Winson Chung [Sun, 14 Sep 2014 07:12:26 +0000 (07:12 +0000)]
Merge "Fixing issue with stack not being fully centered. (Bug 17015153)" into lmp-dev

10 years agoCustomized ripple animation for navigation bar
Christian Robertson [Sun, 14 Sep 2014 06:14:46 +0000 (02:14 -0400)]
Customized ripple animation for navigation bar

To implement the new gel-cap shaped ripple I resurrected the
onDraw and setPressed methods from the KitKat implementation.
The KitKat animation timings were adjusted to match the material
ripple, but the structure of the animators is the same as KitKat.

Since the new ripple is drawn directly to the canvas, the references
to the KitKat background image were removed.

The LogInterpolator is copied from the framework Ripple
to match the material animation curves.

Bug: 17112935
Change-Id: If6a3eb92de794b526338166bf8cb096eb9764cf6

10 years agoMinimize the number of default enabled IMEs part 2
Yohei Yukawa [Sat, 13 Sep 2014 13:03:37 +0000 (22:03 +0900)]
Minimize the number of default enabled IMEs part 2

Previously the system tried to enable at least one auxiliary IME
even when the system is not ready.  However, this doesn't make
much sense because the user should be able to set up their phone
without auxiliary IMEs. Also, IMEs enabled before the system
becomes ready are kept to be enabled after the system becomes
ready. Thus, we should minimize the number of enabled IMEs
until the system becomes ready.

BUG: 17347871
Change-Id: Ife93d909fb8a24471c425c903e2b7048826e17a3

10 years agoMerge "GMSCore and Finsky are fixed. Revert "Revert "Set up final resource identifier...
Qiwen Zhao [Sun, 14 Sep 2014 02:53:31 +0000 (02:53 +0000)]
Merge "GMSCore and Finsky are fixed. Revert "Revert "Set up final resource identifiers.""" into lmp-dev

10 years agoGMSCore and Finsky are fixed. Revert "Revert "Set up final resource identifiers.""
Qiwen Zhao [Sun, 14 Sep 2014 02:53:04 +0000 (02:53 +0000)]
GMSCore and Finsky are fixed. Revert "Revert "Set up final resource identifiers.""

This reverts commit c4bc83b7b5f7911bf750544aa8a2211ae6d11fe2.

Change-Id: I484953c5c3b130b3359ef47aca890fa6903917a6

10 years agoReduce instructions required to obtain and recycle TypedArray
Alan Viverette [Sun, 14 Sep 2014 02:30:11 +0000 (19:30 -0700)]
Reduce instructions required to obtain and recycle TypedArray

Eliminates several field and method accesses by making the TypedArray
pool local to a single Resources object. Removes an unnecessary
synchronization.

ReadXmlAttrs averaged 0.069 ms/op before, 0.064 ms/op after.

BUG: 17484798
Change-Id: If790a5ba6596b47a04f117c8912974798a592052

10 years agoMerge "Fix issue with lifecycle of Activities launched from Lockscreen" into lmp-dev
Adam Lesinski [Sat, 13 Sep 2014 23:11:56 +0000 (23:11 +0000)]
Merge "Fix issue with lifecycle of Activities launched from Lockscreen" into lmp-dev

10 years agoMerge "Give a tablet a chance." into lmp-dev
Craig Mautner [Sat, 13 Sep 2014 22:45:26 +0000 (22:45 +0000)]
Merge "Give a tablet a chance." into lmp-dev

10 years agomake wifi cellular params configurable
vandwalle [Sat, 13 Sep 2014 19:53:26 +0000 (12:53 -0700)]
make wifi cellular params configurable

Change-Id: Ia2a4ac3e11f2b0c7412f33496a655c5d5297ea0c

10 years agoGive a tablet a chance.
Craig Mautner [Sat, 13 Sep 2014 18:17:23 +0000 (11:17 -0700)]
Give a tablet a chance.

500 msec is apparently not enough time to push the pixels around on
a Volantis (or Manta, or occasionally Deb). Increasing the timeout
eliminates turning the screen on before the surface is completely
drawn.

Optimizing the drawing path so that we can return the timeout to
500ms has been filed as b/17496306.

Fixes bug 16849724.

Change-Id: Ic0fca68bc41caa67dba973b5c40915756dc4730f

10 years agoAdd null checks for quickly disappearing Displays
Craig Mautner [Sat, 13 Sep 2014 19:09:31 +0000 (12:09 -0700)]
Add null checks for quickly disappearing Displays

If a display is removed right after it is added we will get null
when trying to retrieve it from the DisplayManager. Check for that
null before commiting to using that display.

Fixes bug 17478906.

Change-Id: I89ae8cc7387729d678b9b1ed5b1797a93cdde3f9

10 years agoMerge "Page content sometimes missing while scrolling." into lmp-dev
Svetoslav [Sat, 13 Sep 2014 18:45:51 +0000 (18:45 +0000)]
Merge "Page content sometimes missing while scrolling." into lmp-dev

10 years agoHandle null values in registerService
Christopher Lane [Sat, 13 Sep 2014 18:13:39 +0000 (11:13 -0700)]
Handle null values in registerService

Bug: 17491971
Change-Id: I0443a9accd86ab70510d16456a03912845718ca9

10 years agoAllow creating EAP_TLS network without specifying Phase2.None
Vinit Deshpande [Sat, 13 Sep 2014 18:10:07 +0000 (11:10 -0700)]
Allow creating EAP_TLS network without specifying Phase2.None

Phase2 isn't required for EAP_TLS networks, and it is possible
for a developer to miss it. We are requiring it for internal
book keeping - so this change just sets Phase2 to 'None' when
EAP_TLS is set on the configuration.

Bug: 17043082
Change-Id: I5a8e0a3a236d0dfd1295584de55d5b9a1b76b483

10 years agoMerge "Renaming Telecomm to Telecom." into lmp-dev
Tyler Gunn [Sat, 13 Sep 2014 17:32:47 +0000 (17:32 +0000)]
Merge "Renaming Telecomm to Telecom." into lmp-dev

10 years agoMerge changes Ife1c7cb0,I398ff797 into lmp-dev
Svet Ganov [Sat, 13 Sep 2014 17:13:31 +0000 (17:13 +0000)]
Merge changes Ife1c7cb0,I398ff797 into lmp-dev

* changes:
  Fixing controls allignment in print UI.
  Fix a crash in print spooler.

10 years agoMerge "Minimize the number of default enabled IMEs part 1" into lmp-dev
Yohei Yukawa [Sat, 13 Sep 2014 10:33:48 +0000 (10:33 +0000)]
Merge "Minimize the number of default enabled IMEs part 1" into lmp-dev

10 years agoRenaming Telecomm to Telecom.
Tyler Gunn [Sat, 13 Sep 2014 05:16:17 +0000 (22:16 -0700)]
Renaming Telecomm to Telecom.

- Changing package from android.telecomm to android.telecom
- Changing package from com.android.telecomm to
com.android.server.telecomm.
- Renaming TelecommManager to TelecomManager.

Bug: 17364651
Change-Id: I192cb5d189f55db012ea72ee82ccc5aedbc21638

10 years agoFixing controls allignment in print UI.
Svet Ganov [Sat, 13 Sep 2014 02:54:05 +0000 (19:54 -0700)]
Fixing controls allignment in print UI.

Making some allignment tweaks based on UX feedback.

bug:17489187

Change-Id: Ife1c7cb0f55eb0398bca4a5e015ace24437ff2a6

10 years agoFix issue with lifecycle of Activities launched from Lockscreen
Adam Lesinski [Sat, 13 Sep 2014 02:42:24 +0000 (19:42 -0700)]
Fix issue with lifecycle of Activities launched from Lockscreen

ActivityManager still thinks the device is going to sleep, so
any Activity started while the lockscreen is showing is immediately
stopped. When Keyguard calls ActivityManager.keyguardWaitingForActivityDraw(),
then we must make sure that the ActivityManager thinks we are not sleeping.

Bug:17459745
Change-Id: I5c45c88e3a72e1f08e1b32101d0d0b4675f7a5ff

10 years agoFix a crash in print spooler.
Svet Ganov [Sat, 13 Sep 2014 02:25:27 +0000 (19:25 -0700)]
Fix a crash in print spooler.

A recent bug fix exposed another one where we were not updating
the internal state of the remote print document after it finishes
updating. This resulted in a crash on every print.

bug:16966145

Change-Id: I398ff7976533241e3d4cb6cd27f69cdc25e67be9

10 years agoChange the property name to pass selinux.
Yuncheol Heo [Sat, 13 Sep 2014 02:21:46 +0000 (11:21 +0900)]
Change the property name to pass selinux.

- Added the prefix 'persist.sys.' to pass selinux and to make the
  properties persistent after reboot.
- Shortened the name to fit the length to 31.
- Got together the properties in the near place.

Bug: 17491732
Change-Id: I051f1bacc51ee372806077159020600ffeb30f48

10 years agoMerge "Update ambient shadow for TV." into lmp-dev
Tim Kilbourn [Sat, 13 Sep 2014 01:11:42 +0000 (01:11 +0000)]
Merge "Update ambient shadow for TV." into lmp-dev

10 years agoMerge "Work on issue #17477510: Time change notifications sent too often" into lmp-dev
Dianne Hackborn [Sat, 13 Sep 2014 00:56:56 +0000 (00:56 +0000)]
Merge "Work on issue #17477510: Time change notifications sent too often" into lmp-dev

10 years agoMerge "Add support for dozing after screen off." into lmp-dev
Jeff Brown [Sat, 13 Sep 2014 00:53:03 +0000 (00:53 +0000)]
Merge "Add support for dozing after screen off." into lmp-dev

10 years agoUpdate ambient shadow for TV.
Tim Kilbourn [Fri, 12 Sep 2014 18:35:21 +0000 (11:35 -0700)]
Update ambient shadow for TV.

Bug: 17462175
Change-Id: I2fb6e939fc428e95b944e87a3a82a3db6130c762

10 years agoMerge "Correctly initialize animator asynchronousity" into lmp-dev
Chris Craik [Sat, 13 Sep 2014 00:46:10 +0000 (00:46 +0000)]
Merge "Correctly initialize animator asynchronousity" into lmp-dev

10 years agoCorrectly initialize animator asynchronousity
Chris Craik [Sat, 13 Sep 2014 00:40:20 +0000 (17:40 -0700)]
Correctly initialize animator asynchronousity

bug:17483792
Change-Id: Icb6b6e6318d4c79ee94228806cb60b275424b9c9