Packages without any resources should not expect to have
a DynamicRefTable.
Bug:16895517
Bug:17056720
Change-Id: Id006f6bdbf08f30505f6ba5982bc9d1b09db0f0a
Use sets backed by ConcurrentHashMaps instead of HashSets, and
CopyOnWriteArrayLists instead of ArrayLists, to prevent concurrent
exceptions if listeners try to remove themselves in callbacks while
iterating over the listeners.
Bug:16325026
Change-Id: I55e081eda6ba19fa466bbf019c648bbdaf833c33
Previously DMAgent would apply these lockdowns before/
after setting the matching user restrictions.
Bug: 16701642
Bug: 16945830
Bug: 16944983
Change-Id: Ib4f7145055687f12408d6ccacd8e6380406a32b2
Ensures that changing the brightness doesn't lead to
another write to settings from SystemUI. Also makes
sure that a BrightnessDialog from a background user
is never showing. While we're in there, fix the obsolete
window flags as well and ensure BrightnessDialog only ever
allocates one BrightnessController.
Bug: 16297345
Change-Id: Id95b598b0fa99680072bc160afb00d2cd49f05ca
The default application icon generally has a different aspect ratio that a
preview image, and cannot transparently be used in place of the image.
Returning null will allow the apps to handle the case as they see fit.
Change-Id: Ie6e8e16dd7b4e58b222187a81c62d27eccab45a9
All reads of those specific settings will go to the primary user.
Inserts to primary also go to managed profiles in order to notify
any observers.
This enables Location settings to be shared by both profiles.
Also some other settings related to IME and Accessibility since
those services are shared across the profiles.
Bug: 16457210
Change-Id: Ib8fd697b5c78027fcbaf245d82dda5e6d6aab4f0
Harden the guarantee that if we're asked about a possible restore,
we always ALWAYS report back to the package manager. This involved
closing "should never happen" edge cases around provisioning/auto-restore
setting that nevertheless were happening.
Also, on the auto-restore setting front, make sure to plumb that
system API through appropriately, since going behind its back and
manipulating the secure setting directly would cause things to get
out of step.
Bug 17060654
Change-Id: I52ca9c1ffbfc0bd6b57196157500d0868bfc2989
- Persist the entire exit condition instead of only the id.
- Make downtime a proper condition provider (similar to the
existing countdown provider for time-based conditions)
- Move all downtime-related items out of ZenModeHelper and
into the new condition provider.
- Reevaluate downtime more often, when any of its inputs change.
- Make sure downtime appears as an available condition in the
condition panel when applicable.
Bug:16296125
Bug:16211189
Bug:17031767
Change-Id: I1d8269a4e6fe170ce776bf932dbbdfb29dd25dd7
This reverts commit 87efe74e092236c372d3b6909009641123aa416a.
This should be fine now with all the dependency CLs +2-ed
Change-Id: I96ad14ad5ff81e6b5391035cb6c5a62339c6cc40
When a String isn't found in the StringPool,
we should not try to construct a String8 object
from the NULL string.
Bug:15163956
Change-Id: I51e701918b10a72c18a860b8a36dce2afd9c0b82
If an unlock handler already exists, we need to try
to update the tech mask for it. Don't permit unlock
handlers with no tech mask.
Bug: 17054331
Change-Id: I54a885d28bdd8ce41d8646d968621c7d6abc9387
This reverts commit 0e475302cd196f45a01a525c49089018b238d4ba.
I'll re-upload this CL with the correct fix in LoadedApk.
Change-Id: I17c6381f8e5907dab9f996f89c5d435b9c08a13a
Cleaned up the code around mMaxPanelHeight of the
PanelView which could lead to flickering during
peeking.
Changed the panel opening logic to account for lag
when we need to wait for a layout, which could lead
to inconsistent animations.
Fixed a bug where holes could appear in the shade
when notifications were updating.
This also improved the general updating behaviour
which is now done in a nicer animation.
Bug: 15942322
Bug: 15861506
Bug: 15168335
Change-Id: Ifd7ce51bea6b5e39c9b76fd0d766a7d2c42bf7a4
We set the system_server classpath in the environment
(like we do with BOOTCLASSPATH). After the zygote forks
the system_server, we dexopt the classpath (if needed)
and then launch the system server with the correct
PathClassLoader. This needed several small / medium
refactorings :
- The logic for connecting to installd is now in a separate
class and belongs in the system_server.
- SystemService / SystemServiceManager have now moved to
classes.jar. They are only used from there, and since they
use Class.forName, we want them to be loaded by the
system_server classloader, and not the bootclassloader.
- BootReceiver now moves to frameworks.jar, because it is
used by ActivityThread and friends.
bug: 16555230
Change-Id: Ic84f0b2baf611eeedff6d123cb7191bb0259e600
