components by apps which could result in thrashing on the PackageManager. For apps that
do not want to be restarted when such a broadcast is sent, we can just aggregate these broadcasts and
handle them at one go.
Changes include:
New structure to hold pending broadcasts by class name. If a component is enabled or disabled frequently
aggregate component enabled/disabled settings in this structure in a 10 second window and then
send out the accumulated list of broadcasts to the ActivityManager.
A new Handler implementation handles this message
Add new attribute name EXTRA_CHANGED_COMPONENT_NAME in broadcast intent Intent.ACTION_PACKAGE_CHANGED for
additional information for apps like Launcher.
Rename a couple of parameters, the names were too jarring.
Applications can now specify two more aspects of the restore process: whether
they need to run with their own custom Application subclass rather than being
launched in the usual restricted mode during restore, and whether it's okay for
the backup manager to kill the app process once restore has completed. The new
manifest attributes for these are, respectively, android:restoreNeedsApplication
and android:killAfterRestore.
If unspecified in the manifest, restoreNeedsApplication is false, and
killAfterRestore is true.
In order to support kill-after-restore cleanly, this change also adds a new
system-process-only interface to the Activity Manager, which will schedule a
"commit suicide" event on the target app's main thread looper.
The framework backup agents have been given the appropriate new backup
attributes as well.
We need to pass them to WebViewCore as they are per process base. If we drop
them, the counter in the JWebCoreJavaBridge can be out of order.
Part 2 of fixing http://b/issue?id=2087410
Always process class 0 and other unstored SMS (eg, MWI). We were
rejecting all SMS messages in storage full situations, but certain
messages do not require storage.
Also, notify apps when the framework rejects MT SMS, with new
SMS_REJECTED_ACTION intent.
b/2066775
b/2015906
This is necessary partly for general hygiene, but mostly because we need
to be able to determine based on the intent type what style URI to
return in the case of startActivityForResult().
Change-Id: Ib313d830b8646a70d5ac3ded11597af614429262
Rationale: there is no reason to sleep for 4 seconds between
transactions if there is no one using the database. This long
sleep only makes sense if two threads are actively using
the database yielding to each other every several seconds.
Merge commit '56205fea879543a50bb797016832416a8b48cabb' into eclair
* commit '56205fea879543a50bb797016832416a8b48cabb':
Fix an emoji-measuring bug that caused an exception when editing a contact.
Some of this is temporary (in particular the two approaches for getting
process memory, one working but horrible, the other not working but
preferred) until I figure out the best way to do it.
Change-Id: I8c8f25062d481fcea22a47d459b083d2fd8a5040
It was measuring the text to try to determine the size that it needed to
scale the emoji character to. Unfortunately it was accidentally trying
to measure the character under the cursor instead of the emoji character
itself, which is wrong, but more seriously doesn't work at all when the
cursor is at the end of the line.
This was already fixed before in change 144474, but that change never got
merged over to donut. So this merges it now.
Bug 2087915
Change-Id: Ib4804d330a029a966207b3b07271f84e6b2652c0
* changes:
When we get 304 and switch to cacheLoader, need to set mFromCache to be true. So that the cache loading will not create a new mCacheResult which overwrites the old cache data.
Use the fact that the scroller keeps track of the current velocity
to accelerate subsequent flings. The current velocity is added to
the new velocity if the flings are in the same general direction
and if the fling animation hasn't been aborted.
This makes it easier to move to the top or bottom of very large
pages (either in overview mode or reading mode) by taking
consecutive flings as a cue from the user that they want to
page faster. The user can stop the fling with any action as
before.
fixes http://b/issue?id=2066090
This change adds the EventHandler error codes to WebViewClient for the public
api. Previously, the api was ambiguous about the value of the errorCode
parameter and falsely documented that the values are HTTP status codes.
The main consumer of this feature is shortcuts.
The LOOKUP KEY column will contain an encoded concatenation of the contact's row id
as well as sync_ids of all constituent raw contacts. It goes with the "contacts/lookup/*/#" URI.
When we get such a URI, we will first try to load the
contact with the specified _id and lookup_key. If we cannot find the contact
that way, we will go scout for the contact that contains most of the sync_ids
we found in the lookup key.
We will need to make sure that the contact picker returns the lookup-style URIs.
Yet more work on improving the behavior of wallpapers. This fixes a few
problems in their lifecycle (corresponding change in the picker also
required for this), makes their animations better for hardware that supports
alpha fades, adds animations for the wallpapers themselves, eliminates
fixed size wallpapers, and adjusts the API for retrieving a wallpaper
bitmap to take care of scaling the raw wallpaper image to match the current
desired width and height.
Change-Id: If1c0aaceba4ea4e175dcb7a8416ca7ddbb9bfa6f
