Bug: 27286867
WindowManager has committed to stopped state
controlling the lifecycle of the Surface, so
make that a first-class thing in HWUI as well.
This makes it more resistent to things like
a rogue updateSurface() happening while mStopped=true,
leading to bad things down the line. Instead let
the surface be changed/updated as often as desired,
and just block any attempt to draw on that surface.
Also removes some unnecessary makeCurrent()s, as
EglManager ensures that we *always* have a valid
GL context now (using a pbuffer surface if there is
no window surface set)
Change-Id: Iead78ddebc7997e8fdb0c9534836352f5e54b9bd
Bug: 27709981
This desperately needs a refactor, but to keep
the current (really needed & nice) behavior of
dispatching after sync finishes would be difficult
to handle cleanly without lots of ripping so... #yolo
Change-Id: I831a06c6ae7412a062720d68ecbe3085190f0258
Gestures now operate on the screen as the user sees it, so they are
affected by magnification. This makes gesture coordinates consistent
with the node bounds in screen and makes an eye or head tracking service
work much more easily.
Bug: 27747314
Change-Id: Idee60398d49d8c9af7405d974893c796b034c974
This change adds support for the new MessagingStyle, which
supports a fully back-and-forth messaging/chat experience
within the notification display itself.
Bug:28043958
Change-Id: Ibb9c2a0784f028d8cf01bd3cee4fecc5f72d68cd
Instead of relying only on kernel support for sdcardfs, give each
device the ability to quickly toggle between sdcardfs and FUSE. Also
add the ability to users to explicitly enable/disable the behavior
for testing and debugging purposes.
Bug: 27991427
Change-Id: Ia57f12f6ead888ab85101004bdd8c8003537b116
Also block all callbacks if the listener is not connected.
Also update the documentation to make lifecycle restictions
more clear.
Bug: 27582839
Change-Id: I029304e1fb87bd8d754cfb4641ba8cc095fc2b2e
Settings screen should apply both primary and managed maximum
timeout policy, even separate profile challenge is enabled.
Bug: 27493348
Change-Id: Ia1ec1cafc7665c54816833af64e0f446a77a55b2
To make the WebViewUpdateService testable we create a new implementation
class that contains the implementation for the actual service. This
implementation can then be tested.
Bug: 27635535
Change-Id: I45c25c71375cc86a04c649a845016d2e7b105a7a
Callers with CONNECTIVITY_INTERNAL permission can read off the netId
that an application is assigned to by default.
Necessary for making connections as the default network for a client
app eg. when downloading stuff on its behalf.
Bug: 27074270
Change-Id: I8d35e8e99126875f55f3c545090326f3e9be43fb
When the parent instance of DPM is used to call an API that is not
parent-aware, we should throw a security exception.
Bug: 27532279
Change-Id: I90abeb4c8522cb9af798301c1bb2048579596993
By setting width to the ListView, the width of the containing text view
won't match with parent.
I locally verified this doesn't revive issue 27341560.
Bug: 28034210
Change-Id: I67886a752ef110d3433dddd6aa8447a9027f8e19
There are a handful of looper threads in the system_process that
are shared by dozens of different internal services. To help track
down what these operations are, tag the processing of each message
with a string that tries describing where it originated from: the
class name of the Handler, and the message number or class name of the
Runnable.
Bug: 28046299
Change-Id: I409ec43fea8daaae4fd70df05d4fed929a7249ae
During package installation, we were parsing the APK twice; once
in the context of the PackageInstaller and once in the context
of the PackageManager. Instead, the installer should just pass
the certificates to be used further in the process.
If the PackageManager doesn't receive certificates [or, if there's
an error using them], it will fallback to re-parsing the APK.
Bug: 27502465
Change-Id: I94ce551af54eaa9916228e933134debe50867d21
