1. We use a delayed callback to throttle the amount of accessibility
scroll events fired by the view tree. The callback to do so was
not properly reset when removed putting the view tree in a bad
state resulting in no scroll events being fired at all.
bug:6549005
Change-Id: Ibf72d7e009e4545a336c9471f46015910290703e
1. This attribute specifies whether a view can take accessibility
focus. It has three values: 1) auto - the system determines
based on whether the view is actionable and has actionable
predecessor. Accessibility services can put accessibility focus
on such a node at will; 2) yes ; this view always takes access
focus; 3) no - the view cannot takes accessibility focus and
accessibility services cannot put accessibility focus on it.
Change-Id: I2ebf4e7c75bf6b39e1742b6868b37ccdd4cc7d28
1. Iterators were skipping content on reversing direction.
2. The cursor was positioned at the beginning of the next text segment
when moving forward and at end of the previous text segment when moving
backwards. This is incorrect and now the cursor is positioned at the
end of the segment when moving forward and at the beginning when moving
backward.
3. The cursor position was not properly set when reaching the end/start
of the text.
4. The iterators were reporting strictly the next/previous segment even
if the cursor is within such a segment. Thus, when traversing some
content may be skipped. Now moving forward moves the selection to
the next segment end and the start position is either the old index
if it was within a segment or the start of the segment. Same in
reverse.
bug:6575099
Change-Id: Ib48a649cec53910339baf831a75e26440be6e576
...content provider and updating its oom adj
This introduces the concept of an "unstable" reference on a content
provider. When holding such a reference (and no normal stable ref),
the content provider dying will not cause the client process to be
killed.
This is used in ContentResolver.query(), .openAssetFileDescriptor(),
and .openTypedAssetFileDescriptor() to first access the provider
with an unstable reference, and if at the point of calling into the
provider we find it is dead then acquiring a new stable reference
and doing the operation again. Thus if the provider process dies
at any point until we get the result back, our own process will not
be killed and we can safely retry the operation.
Arguably there is still the potential for a race -- if somehow the
provider is killed way late by the OOM killer after the query or
open has returned -- but this should now be *extremely* unlikely.
We also continue to have the issue with the other calls, but these
are much less critical, and the same model can't be used there (we
wouldn't want to execute two insert operations for example).
The implementation of this required some significant changes to the
underlying plumbing of content providers, now keeping track of the
two different reference counts, and managing them appropriately. To
facilitate this, the activity manager now has a formal connection
object for a client reference on a content provider, which hands to
the application when opening the provider.
These changes have allowed a lot of the code to be cleaned up and
subtle issues closed. For example, when a process is crashing, we
now have a much better idea of the state of content provider clients
(olding a stable ref, unstable ref, or waiting for it to launch), so
that we can correctly handle each of these.
The client side code is also a fair amount cleaner, though in the
future there is more than should be done. In particular, the two
ProviderClientRecord and ProviderRefCount classes should be combined
into one, part of which is exposed to the ContentResolver internal
API as a reference on a content provider with methods for updating
reference counts and such. Some day we'll do that.
Change-Id: I87b10d1b67573ab899e09ca428f1b556fd669c8c
Make StateMachine#quit non-conditional and remove the need to
process the SM_QUIT_CMD it is now private.
Rename halting to onHalting.
Add onQuitting
Change the message specific logging to be more generic and change
the xxxProcessedMessagesYyy methods to xxxLogRecXyy names. Also add
addLogRec(String) and addLogRec(String, State) as the generic logging
methods.
bug: 5678189
Change-Id: I22f66d11828bfd70498db625fe1be728b90478b7
Make clearer how the platform is handling key events following some
unfortunate uses by third party applications. Also highlight the
changes in Jelly Bean default keyboard.
Bug: 6566711
Change-Id: Ibcdaf54c6d629fd0733529bfe2fffc82f555f084
This method was only supported with the JSC JavaScript engine.
V8 became the default JavaScript engine in Froyo and this method
has not been invoked since. Support for building JSC was removed
in b/5495373.
Bug: 6295376
Change-Id: I3dbe83f375b88ebbbb713d3c3fa5a2fa323a0d45
Bug 6476578
The latest bug report show a query.length() of 33 while
mQueryTextView.length() is 0 on line 514.
I can see 2 reasons which can explain this discrepancy:
- the mQueryTextView has a filter, which alters the text.
- some asynchronous event (IME?) changes the text in the mean time.
I would favor the second one, which seems to break a lot of single
thread assumptions in the code and generates other IOOB exceptions.
Note that depending on what they are used for, it may be more consistent
to use mQueryTextView.getText() instead of query in the following
assignment.
Change-Id: Ie8a5486b11a80543f8f90980454933c5a74c073e
