It turns out we were not returning the density for anything retrieved from a
TypedArray... which basically means any bitmap references from a layout or style...!!!
This is now fixed.
Also fiddle with the density compatibility mode to turn on smoothing in certain situations,
helping the look of things when they need to scale and we couldn't do the scaling at
load time.
Since the search UI is handled by the system process, and
the service calls to show and hide it are asynchronous,
the tests that checked the visibility of the search UI were
very flaky. This change sclaes the tests back to just check
that nothing crashes when showing and hiding the search UI.
Fixes http://b/issue?id=1993675
This change allows us to use drawables that match the current screen
density even when being loaded in compatibility mode. In this case,
the bitmap is loaded in the screen density, and the bitmap and
nine-patch drawables take care of accounting for the density difference.
This should be safe for existing applications, for the most part, since
they shouldn't really be pulling the bitmap out of the drawable. For
the small rare chance of them breaking, it worth getting the correct
graphics. Also this will only happen when there is actually a resource
of the matching density, and no existing apps should have resources for
anything besides the default density (though of course all of the
framework resources will be available in the native density).
As part of this, the bitmap density API has been changed to a single
integer provider the DPI unit density.
Before, if a column contained null, MatrixCursor.getString()
returned the string "null", and the get<Number>() methods
threw exceptions.
Now, getString() returns null, and get<Number>() return 0.
Fixes http://b/issue?id=1996628
Instead of a list, we now just have a single boolean indicating whether an
application is density aware, and this set set to true by default as of
Donut.
This changes the names of the directories in aapt, to what you see
in the list of DpiTest resources. Also adds a new "long" configuration
for wide screens, which the platform sets appropriate, and introduces
a new kind of resizeability for not large but significantly larger
than normal screens which may have compatibility issues.
The Bitmap functions to get the scaled width/height couldn't actually
do the right thing because they didn't know the destination they would
be drawing to. Now there are two forms of them, taking an explicit
parameter specifying the destination.
Hopefully this gives enough time for more memory recycling. Because it is observed that several seconds after a suite is run, the memory usage could drop by 1-2M.
ServiceTest appears to be causing the whole android unit_tests suite to fail in the
continuous system. Suppressing to stop the lab failure spam while continuing to investigate.
This adds a new API with the activity manager to find out about movement between
activities. For my sanity, the old IActivityWatcher is now renamed to
IActivityController, and the new activity movement interface is named
IActivityWatcher.
This changes the search manager itself to use the new API to manage its state.
Note that there are still problems when going back to the search dialog after
it was hidden -- the suggestions window no longer appears until you explicitly
dismiss and re-show it.
Details:
- Add a new SearchDialogWrapper class that makes sure
all access to the SearchDialog is run one a single thread
other than the main ServerThread.
- Don't save/restore seach dialog state in Activity.
This resulted in lots of calls to the SearchManager
throughout the life cycle of all activities, for
the questionable benefit of restoring the search dialog
in a few cases.
- Remove search UI state save/restore, and the isVisible()
method from SearchManagerService. They are no longer used,
and were tricky to implement since they return values from
the search UI thread to the service.
- Handle configuration changes in searchDialogWrapper instead
of calling through from Activity.
Fixes http://b/issue?id=1938101
TODO:
- Activity.performPause() calls stopSearch(). This call may not happen
until the new activity has been started. If the new activity starts a
search immediately, this search could be cancelled by the old activity's
call top stopSearch().
Add small unit test for NeighboringCellInfo in AndroidTests. The old telephony unit test directory, CoreTest/android, is marked as deprecated in AndroidTestHowto.
new file: tests/AndroidTests/src/com/android/unit_tests/NeighboringCellInfoTest.java
+ Now rechecks the cached IBinder each time the wrapper is used, and if it's
still null (i.e. the BackupManager was constructed before the system service
came up) it's refetched. This lets even system code cache a single
BackupManager instance and just keep making calls through it without worrying
about interactions with the life cycle of the backup service.
+ Added a static dataChanged(packageName) method as a convenience for code that
needs to indicate that some other package needs a backup pass. This is useful
even for third party code in the case of multiple packages in a shared-uid
situation.
Otherwise, the different text measurement methods produce slightly
different results and cause the text to show different results for
Spanned vs. non-Spanned source strings.
Bug 1837826
