When users use application specific Bluetooth state changes,
and the bond state changes, we crash.
We need to recheck Bluetooth state in all functions and scenarios
and even maintain a per application pid Bluetooth state, but this fix
is quick fix for a problem.
Change-Id: I905dfb53965907ccb27757ff6ed1816536277f14
Bug: 5143618
Handle imeOptions and inputType attributes in SearchableInfo.
Methods to directly modify those attributes on SearchView.
Change-Id: I340617767f930ecb1b1ad5ab2696bc613982a20b
Provide method to surface progress when building a notification. Also
manages visibility of various items, including hiding the entire line
when unneeded.
Bug: 4022082, 3399877
Change-Id: I90d8fca383c746001aadc6c5f21c68a23bb937e5
Since computing the geocoded location is an expensive operation, add the
geocoded location once and for all when inserting in the call log.
Bug: 5129581
Change-Id: I6017c993a52b461eed395ce5e5a9373018e2087a
While it is valid to request focus while detached from a window,
AbsListView did not handle it gracefully if it had an adapter whose
data changed while the view was detached, since DataSetObservers are
unregistered during this time. Assume that the data could have changed
if we gain focus while detached.
Change-Id: Id135693ffcc48ff5a9d530d335ab449180624c43
getDuration() wasn't returning the correct thing in some cases, and
setStartDelay() had no effect whatsoever. Fixed those.
Change-Id: I89f930535ec1acddf57227937a9ad64986548998
Previously, AnimatorSet incorrectly checked whether child animations were
'running' to figure out what to cancel. If a child animation was started, but
sitting in a startDelay phase, it was not 'running', so the right cancel/end
events would not propagate.
The fix is to add a new isStarted() API to Animator, which returns true when
the animator has started (but not yet ended), regardless of whether the animator
has a startDelay or not. It's basically a superset of the existing isRunning()
method, which only returns true when an animator has actually started setting values.
Change-Id: I126814cb6637b58295b6d18d9b155235671f99be
- do not go to native if this is not needed (empty or null text)
- do expandMetricsFromPaint() in TextLine when needed instead of calling handleText()
with emty string just for doing expandMetricsFromPaint()
Change-Id: I1f566ccef66cbc990ebbb77ef149899119e37e84
The activity manager now take care of plugging the correct settings
into the OOM killer in the kernel. This is a lot cleaner because
it is really central to how the activity manager works, and nobody
else cares about them.
Taking advantage of this, the activity manager computes what it
thinks are appropriate OOM levels based on the RAM and display
size of the device.
Also a small optization to the package manager to keep a binding
to the package install helper for a bit after done using it, to
avoid thrashing on it.
And some new APIs that are now needed by Settings.
Change-Id: I2b2d379194445d8305bde331c19bde91c8f24751
assignContactUri
Also changes:
- Remove handling of mMode, as QuickContact now ignores this anyway
- Don't draw the overlay-triangle if no action can be performed
Bug:5132023
Change-Id: I8d6f8df9caf140fdade8683bd7b22d53c28994f4
Make Switches obey the Grand Unified Theory of Button Ordering.
This also matches their behavior with volume/brightness sliders.
Change-Id: I4e72801e81b10002542a05f61c99a57991f130ff
Clear out saved fragment/view state when a fragment is resumed in case
we didn't need to restore it. This keeps it from hanging around and
preventing proper save/restore behavior later.
Change-Id: Ie3fce5534ee4f8fdb4116281f9ac6a9cfa392561
