Support for loading the pre-compiled binary format for hyphenation
patterns.
Bug: 21562869
Bug: 21826930
Bug: 24570591
Change-Id: Iaeaa9c9ac9dac236af6b0d7894c2e2396bc8447d
(cherry picked from commit 091dba2de1)
Previously, a native pointer was returned. After this change, callers
should call getNativePtr() if they want to get the native pointer instead.
Bug: 21492297
Bug: 24570591
Change-Id: I745bd88e6e8c86078fffb07f90f8d8f24215410c
(cherry picked from commit e7eac6f540)
This makes it so that the socket cannot receive datagrams from
anybody except the DHCP server. This does not improve security,
because we never read from the UDP socket anyway, but it does
make ListeningPortsTest pass.
Bug: 23906864
Bug: 23933386
Change-Id: Ib090273a417f7eb2ac1ee3309260249b72fb8345
When an app doesn't define installLocation, the default behavior
should be to treat it as internal only. This matches all the
published developer documentation.
Without this, apps could be unwittingly be moved to adopted storage
devices.
Bug: 24771264
Change-Id: Iaf38ab45329aad6cb5d6deac81fb1781f680189b
In the case of scaled SurfaceViews with a fixed size.
It is possible for the surface layout size to change
without a change to the surface size or position. In this
case we need to inform the window manager so it can update
frames and scaling. This has presented in Camera and required
workarounds such as changing the fixed surface size to trigger
an update (leading to visual artifacts).
Bug:23974105
Change-Id: I195aa0832907c32874b294101b56e198e853dce5
(cherry picked from commit a8fe43f09c)
When a developer wraps an intent with Intent.createChooser(), they're
indicating that the user should always be prompted, instead of using
any "always use" defaults. A recent CL changed the chooser behavior
to ensure that UI is always shown in the case where there is only one
match.
However, this caused us to start prompting for the GET_CONTENT intent,
for which there is only ever one DocumentsUI system app. Since that
app delivers on the createChooser() contract described above, we're
okay automatically launching it.
Bug: 24464358
Change-Id: I0279d3343479c134a35f41ddf3cb4204d0ae6a90
We cannot pull the "retain loader" state from the Activity; an Activity may
not always be hosting a Fragment. Instead, save the "retain loader" state
inside the individual fragments.
Bug: 23838271
Change-Id: I8358183a7689b5a571ea7be03d769186b2812600
Use the existing PointerEventListener infrastructure to listen for
gestures that look like flings and hint to the power system when they
happen. Since we don't actually have a bound for the fling like a
regular application would, limit them to five seconds and refresh
every time a new fling is seen until the five second time period is
up.
bug 24059298
Change-Id: I5757a1e88f2ab2ef08cccefa8221d809ae71bb6f
getActiveNetworkInfo() and friends already know how to augment their
results to help apps detect when network access is blocked. This
change wires up the new app-idle and device-idle firewall rules to
be reported through these APIs.
This also causes other platform tools like DownloadManager and
SyncManager to respect these new policies.
Bug: 24050462
Change-Id: Id9517b0b70be7e3ca2ab27bed8049db916e4d829
Previously, size of of oat directory was not counted by the getsize command,
because base APK location was passed as apkpath argument.
Bug: 23896047
Change-Id: Ic7b6b725785ff2e2a0cf3887ba68c162b23b1212
Views are not synchronized with adapter state until a layout pass
occurs, which may cause an OOBE if a list item is removed and an
accessibility node is obtained before the next layout pass.
This CL caches the item's enabled state on the bound view's layout
params, which allows us to avoid relying on the adapter to populate
accessibility nodes. It also aborts actions if the target position
is no longer valid.
Updates the documentation on AdapterView to reflect that the result
of getPositionForView() may not be synchronized with the adapter.
Bug: 23943664
Change-Id: I90ca946bead1d87a179eab1c2f482a54bcce9c38
Generalize cpu clusters so we can measure frequency
and power usage across heterogeneous cpu clusters.
This also brings back reading of cpu-times for power calculation.
Bug:22773176
Change-Id: I9c794ae9756c782c0e971c7f5fcebbe70374b269
So the lockout reset exactly fires after 30 seconds, independent
of whether the device was in deep sleep.
Bug: 23668918
Change-Id: Idea1a877c395897c1fd075cf98edf41c8b1f4082
For each runtime permission we have an app op to toggle the
permission for legacy apps as they cannot handle permission
revocations. We were lacking an app op for get_accounts
which prevented the user from controlling access to accounts
regardelss that they change the state of the permission
toggle in the UI. Even worse the permission UI is written
with the assumption that every runtime permission has an
app op and as a result revoking the contacts group (if the
app requests the get_accounts permission) is reset back to
allowed in the UI.
bug:23854618
Change-Id: I9e3f9bfeb320bed561d718db99ee285915d5701b
This is necessary because currently the wifi code just returns
whatever hardware-specific integer it gets back from the HAL,
which is bad because that will be interpreted by the caller as
one of the error codes defined in this class.
In parallel we'll also modify the wifi code to return this new
error code if the hardware returns an error.
Bug: 21405946
Change-Id: Ic9fa1193ced69a4e7ff543e397221c89b10a5a13
Adds the colorTransform field, which defines a vendor-specific color
transform (e.g., wide gamut, sRGB, etc.) to the PhysicalDisplayInfo
class, and populates it from the corresponding field from
ISurfaceComposer.
Bug: 20853317
Change-Id: Ic59ca5142bdaa73c42d9c044d7aae345255f1dad
