Fix a NPE, update the native methods for NinePatch and update an ICU
method to return the right value as required by DatePicker widget.
Bug: 11140701
Change-Id: Id92fc5359acf6dde1bcdbc781aaf637fdb6eecbf
While under heavy system load, keyguard was able to create widgets before
before ActivityManagerService was ready. The result was a race
between keyguard adding widgets and ActivityManagerService being
ready to send broadcasts.
This fix provides keyguard with an additional signal to know when
the system is booted and widgets are safe to load.
Fixes bug b/11217169
Change-Id: I7a714d65b068678f961e52bdde4e1c20f9c287f0
1. There was leftover code in the printers adapter that was
hard coding the item width with no reason to do so.
2. Changed the icon view visibility to invisible to allow
the text of all list items including save to PDF and all
printers to be left aligned.
bug:11225340
Change-Id: I5b51d0d30577699f5cd73c7c58521358653d135b
Make sure that the BroadcastReceiver registered by ViewFlipper
enforces that messages come in on the correct (UI) thread.
Bug 11228463
Change-Id: I97f2a1c392f9c58a99afebc60170c9656bba86ad
vold now parse out UUID and label for inserted physical devices,
and reports them to framework. Add these to hidden StorageVolume
class for use by DocumentsUI and MediaProvider.
Remove last JNI method in FileUtils!
Bug: 11175082
Change-Id: I1cfcd1ade61767b103f693319ea2600008ee2e3c
This is the API part needed to implement support for custom print
options by a print service. Some printers have quite fancy options
and we want users to benefit using them.
bug:11241800
Change-Id: I3a9df771c0ded559b61c597c686795840d702b27
Do not add the non-standard Java classes. Updates the references to all
non-standard classes to new classes in
com.android.tools.layoulib.create package. This also treats
java.lang.AutoCloseable which is part of Java 7 similarly so that we can
still run on Java 6.
Change-Id: Iac5b272652e2780c9bb72d19f415d150948ca589
Two problems addressed here:
- If a call to startActivity() comes in on an activity that is finishing, we can
end up putting the new activity in a stack that isn't actually in use any more
(if the finishing activity is the last one on that stack). This is a bad case,
anyway, so if this happen the treat it as not being called on an existing
activity and switch to NEW_TASK to find a task for it.
- There was a bug in handling PACKAGE_CHANGE broadcasts that would result in the
app's processes being killed, even though the cleanup through the activities
was done. This could leave the activity stack in a bad state. Fix this to
correctly provide an app id for the changing package so that its processes are
killed.
Change-Id: Iece04e0cf95025c3d30353d68bf3d14fd39d44c3
The window had BOTTOM gravity, but the root view was using the
LinearLayout default (START | TOP), meaning an IME with a tall
inputArea could effectively push the navigation bar guard view
off screen.
This change simply uses BOTTOM instead of TOP for the root view.
Bug:11237795
Change-Id: I681c2b1eddac53589e0d168f09a183d67b1e5322
Prevents the dump actions from being blocked by traversals.
Removed a stale comment on queueOrSendMessage and renamed it to
reflect its current implementation since it no longer has a secondary
queue. Removed an unnecessary lock as well.
Bug: 10948648
Change-Id: Ic5fcc567212fb9cc07566c12f5f419af13c2b0b1
This fixes an issue where the mUpdateSeekBars runnable was being
repeatedly posted to the handler while keeping a reference to
the transport control (and hence keyguard). Even after the view
was detached from keyguard it would continue to post events that
would ultimately keep the instance of keyguard from being GC'd and
cause keyguard to slow down.
Fixes 11169793
Change-Id: If2bb238adf6d78101589926b60a61d9f187c6d1c
Make sure that caches are cleared at the right times and views have
their data populated correctly. Also fix a caching issue in
RemoteController.
Bug 11218218
Change-Id: Ieb833b5dc440ccd1b82050f12eb7059a4e0a412f
If a layout has already been requested when we want to show the
correct widget pane, post the event for later and immediately return.
This works because PagedView has some interesting ideas about
maintaining a consistent state. It will attempt to recalculate the
current page position during measurement, but thanks to a reliance on
this measurement to establish scrolling boundaries that scrollTo uses
for clamping plus other code that depends on the idea that it can add
a new widget pane view and immediately scroll to it, this attempt to
scroll fails if the target page was added and measurement is not yet
complete. This patch ensures we don't hit these edge cases in a few
known ways.
While this ends things in the correct state, it predictably has a few
visual artifacts. Situations where things would completely fail
before, positioning pages incorrectly or fading out wrong pages now
will briefly flash and fade the widget frames or in some cases will
visibly animate from the starting page to the correct one. We should
address these issues in a future release.
Bug 11217368
Bug 9760604
Change-Id: I989d03b5bfdd5b6403a6b67a8aae7585ca0c8ae0
This fixes an issue where keyguard isn't showing but tells
PhoneWindowManager to wait anyway. The fix is to send a null token
when keyguard isn't showing.
Fixes bug 11216030
Change-Id: I4985fe50c63819415d9d5a1de0b741c8aa6391d8
