Allow apps to see which of their tasks are affiliated in the
RecentTaskInfo returned by getAppTasks().
Fixes bug 16546323.
Change-Id: I54d75c02c194f58fb6513fa91c312289460f56b4
- Invoke availability listeners when first added
- Handle camera service crashes
- Listen to service death
- Attempt to rebind on every call if service died
- If service is still dead, act as if no cameras are connected
- Fix bug in DeviceImpl that prevented proper handling of errors on open
- For clarity, don't UncheckedThrow for exceptions that are already unchecked
Bug: 16514157
Bug: 16483815
Bug: 16483222
Bug: 16561237
Change-Id: I3d54de1204f5a863882cf675fcee6280d53c4039
Bug: 16239227
Avoid potential issues in views that override invalidate() but
are not prepared to have overridden methods called prior to their
ctor
Change-Id: Ic2574bdfea454451d6881c7e27a6474dc4dbc07c
When a descriptor read/write operation fails with insufficient
authentication/encryption, it will automatically be tried again with
elevated security requirements. The first (failed) attempt should not be
reported to the application requesting the read/write, since it will get
the 2nd callback once the retry succeeds/fails.
Change-Id: Ia27b91afa63b6f9ffdc597d59c24d29f3d8a1008
Starting with startUsingNetworkFeature and stop.
Figure it's easier to code review incremental changes.
Change-Id: I19aee65e740858c3a9a2a1a785663f6fee094334
Add a method to LauncherApps to allow launchers to
show application details for an app in a managed profile.
Bug: 16371359
Change-Id: I23acb4365c09e4a7b9fa742ae6fc7e04434f45aa
This CL changes the relro creation and loading logic.
Instead of using a hardcoded path, now WebViewFactory queries the
package manager to get the lib path, which will be either
/system/priv-app/webview/lib/ or /data/app/... depending on the
presence of an update apk.
BUG:16403706
Change-Id: Icceb95da98e69032a19ac2ae8fe72ba1439e50ad
This makes the code loading the WebView Java code from the .apk
also load the WebView assets into the current AssetManager.
BUG: 11505352
Change-Id: Ib6203f79b98805012ffa20b233037b7da221624c
The intents ACTION_IMAGE_CAPTURE, ACTION_IMAGE_CAPTURE_SECURE and ACTION_VIDEO_CAPTURE are now handled in a way
similar to ACTION_SEND and ACTION_SEND_MULTIPLE.
Migrate the uri in the EXTRA_OUTPUT extra to clipData, and add the flag GRANT_WRITE_URI_PERMISSION.
The userIds are now added to extra uris in the process receiving the intent, (not in the system process), because the
system process may not be able to parcel/unparcel the extras.
BUG: 15534203
Change-Id: I8f79666b726bc6d7745bf777ad3c7518945c5cc3
Also removed different scan methods from IBluetoothGatt and make
BluetoothLeScanner and BluetoothLEAdvertiser singleton.
Change-Id: Ifa2e950b50f100f5507a6dace3bd70db18d7f9ca
This patch removes the @hide annotations and public reference from the
new letterSpacing attribute, so the docs will build. It's still not an
ideal state because a test app can't find the attribute, but at least it
unbreaks the build.
Change-Id: I5425fbe3d63f7a759fabe43bd2136196c7112eed
Add new interface, IHdmiRecordCallback, in order to handle
result of all recording features, One Touch Record and
Timer Recording.
HdmiRecordCallback is clientside interface which uses
RecordSource as return of getOneTouchRecordSource,
instead of raw byte array.
Along with it, add TimerRecordingAction.
Bug: 16160962
Change-Id: If0ed7abc5f45b020b39555cac585becbe8ee7da4
This change makes it convenient to get the HDMI port ID from the device
info without having to look it up through a separate HdmiPortInfo, which
happens in many places.
Also removed port information lookup loop by introducing two maps variable
in Control Service: port path ->port ID, and port ID -> port info.
Bug: 16547583
Change-Id: Ibfff722a16b626230766058efcc819e9c8ca0663
LayoutParams without a specified layout_width value automatically
get a default of 0. This works in most cases, when the view is pinned by
various constraints (such as alignParentLeft and alignParentRight),
but breaks down when the child lies in negative coordinates.
Negative coordinates are supposed to be a signal, in RelativeLayout, that
that coordinate is variable. However, negative margins can cause a negative
value to be calculated, at which point it should be used as a concrete value,
not a behavior trigger.
The fix is to use Integer.MIN_VALUe as the tirgger instead, to distinguish that
impossible value from negative values which could reasonably have been computed
during measurement.
Also, removes dead code from earlier attempt at rtl.
Issue #10606777 Time picker can't display 10, 11, or 12 oclock when font size is huge
Issue #16516577 resolveLayoutDirection doesn't act as intended
Change-Id: Ia564603012f3c45636e69cd3776205fb544fda48
