This changes the current API given app feedback received. In order to use
keycodes as well as chars we also must implement a mapping of Keycodes to
Display labels as only keys with a single char representation are provided
by KeyCharacterMap.
Bug: 27409783
Change-Id: I3da653adc4b8cbc66a4d1aa24a5f9181f71e83c3
Introduce a mapping between dexopt reasons and compiler filters. Use
reasons in package manager and other classes, where possible.
Change PackageDexOptimizer to accept a compilation filter. Adapt for
the split-out profile merging. Pass compilation filter to installd.
Bug: 27689078
Change-Id: I8c0ea6f10fbfdbd096adecc52abfd2466d048fdc
1. Make the overflow show part of the items hidden behind the scroll view
2. Increase the time the scrollbar is on screen so users are less likely to miss it
Bug: 27427333
Change-Id: Ia0bff5fa29cba3103263bc7d063176b0ba7ff7b9
Ensure we have at least 3:1 contrast for the
action buttons. Also ensures that the inline
reply box background has 4.5:1 contrast to white.
Further modifies the color of the inline reply
background to satisfy a 4.5:1 contrast ratio for
the entered text.
Bug: 26831312
Change-Id: If42b1c99d1adee547d0a583c1a69c48ef7287c23
AlertController calls requestWindowFeature() too
later for DialogFragment. Fixed by moving it to the
constructor.
BUG: 27809055
Change-Id: Icef01316b4e857b41ccc29cc56bbcecd7f629342
Before exposing #reportFinish() as a public API, we have to fix an
existing bug that my previous CL [1] for Bug 26945674 forgot to take care.
Currently BaseInputConnection#reportFinish() is always called by using
the root view's Handler. We should move the logic to call
BaseInputConnection#reportFinishInputConnection() from ViewRootImpl to
IInputConnectionWrapper to make sure that the method in question can
always be called on the desired Handler.
To make things simple, instead of explicitly calling #reportFinish()
from IMM, this CL let ControlledInputConnectionWrapper#diactivate()
internally call #reportFinish() as needed. This makes it easier to make
sure that #reportFinish() is called after all the queued method calls
are handled.
[1]: Id9e579bb3e2966986cdcb1c34bc8cacfeca2e1a9
612cce92ad
Bug: 25332806
Change-Id: Ibe94f115e607a198d12ecd3d4e4f91a7d9469c98
Following two fields have basically the same lifetime.
- InputMethodManager#mServedInputConnection
- InputMethodManager#mServedInputConnectionWrapper
Hence we do not need to maintain both of them.
This is a preparation CL for Bug 25332806 and does not change any
user-visible behavior.
Bug: 25332806
Change-Id: I1181e067aa5bedbdf0c7ec1bcec479257aea511c
Helps avoid a continuous launch cycle if a resolver activity resolving
the home intent is the only activity in the system and the lock screen
is up which will put the activity in the stopped state. The activity
then finished itself onStop, but is then relaunched by the activity
manager since there are no other activities in the system and the
home activity should be present.
Bug: 27674536
Change-Id: Iaca67a00c4a37f70aafd18aedcbd8dba2f561203
Add a (configurable) delay between when we start a maintenance
window until the minimum time we will end it.
Also switch to using the alarm manager callback API. (Yay!)
Also fix a little printing problem in the alarm manager dump
so we put the package name and not some class hash in the
summary string of an alarm entry.
Change-Id: I4281e5c80bc8b26ebc1fb6f603ec33ec0e379daa
"Share" didn't make much sense here since SENDTO is used for sending
content to a specific recipient, e.g. emailing a person from their
Contacts page in the dialer.
Bug 23626568
Change-Id: I8ac4f2aee5c41aa6c2d546b39283c6f22127cefe
Zygote process forks every child process for launch the SystemServer and other
applications. When child process is forked, unmount storage inherited
from Zygote process is executed before it gains its own root storage
space.
If Zygote have no storage spaces, unmount operations not needed to
get relevant permission storage space.
Thus unmount is executed only once shortly before the SystemServer is forked.
And the child processes do not unmount its inherited root storage space.
Change-Id: I311ca7ea2c93085f42579f8cf8bdc7ef1d570a04
This patch sets the visibility of a panel to INVISIBLE when it's
alpha is 0. This will ensure that a user cannot interact with a
view that cannot be seen. Not doing this resulted in a bug where
the user will tap on an area of the toolbar and activate an
unintended action.
It also removes the redundant mResetOverflowButtonDrawable. It's
actions are already performed in setPanelsStatesAtRestingPosition.
Bug: 27693433
Change-Id: Iafcde23622c7d1604924c520f6748fdbb8b135e7
am: fb8aa06
* commit 'fb8aa06883faae598512c26d95c5dfeab7f213a5':
Add more @NonNull/@Nullable to TextServicesSettings.
Remove an unnecessary int to String conversion.
Add more @NonNull/@Nullable to InputMethodSettings.
The last caller of the deprecated constructor of InputMethodSettings was
just removed [1]. Since it is not a public API, we can now safely
remove it.
[1]: I09cba4066b95c4a9e89a3e4f83d75b97882502dc
143a6869476a4be5962d4bba3b222d078f46b9a0
Bug: 26279466
Change-Id: I9035c417d6a8166c0ef4e4c7a00c151ffdd5fb49
* changes:
Add more @NonNull/@Nullable to TextServicesSettings.
Remove an unnecessary int to String conversion.
Add more @NonNull/@Nullable to InputMethodSettings.
This follows up to a previous CL [1] for Bug 26279466.
It turns out that we have not clearly defined how nonexistent key should
be handled in InputMethodSettings#getEnabledInputMethodsStr(), e.g. it
returns "" for when mCopyOnWrite is true but returns null when
mCopyOnWrite is false.
Also, since InputMethodSettings now can revert changes made during
mCopyOnWrite is true, the caller may also start receiving null in the
following scenario.
1. call mSettings.switchCurrentUser(userId, true).
2. call mSettings.putEnabledInputMethodsStr(str) where str is non-null.
3. call mSettings.switchCurrentUser(userId, false).
4. call mSettings.getEnabledInputMethodsStr().
If the caller of getEnabledInputMethodsStr() has assumed that it would
never return null, then it would start crashing due to NPE.
With this CL, getEnabledInputMethodsStr() is marked to be @NonNull we
should no longer see such kind of NPE.
[1]: I9c6f9bb3d51174198e5f73588637f87ea0d90e11
68645a638a
Bug: 27687531
Change-Id: I169ad4957e68b65c64251b0849056195b8ca4911
