This causes frequent programming errors, when developers assume
that holding onto a Surface will keep its associated SurfaceTexture,
ImageReader, etc, also alive.
Bug: 31551063
Test: m offline-sdk-docs, manual viewing of result
Change-Id: I5fb5bb3e3c80c7d5d735417b1697e0fe9a62fc46
When creating an instance of RemoteConnection, the code does not properly
handle the case where the video provider is null. This could cause a
RemoteConnection.VideoProvider to be created for a non-existent binder.
Bug: 31372624
Change-Id: I990084e2d2d68cdf2dda426cb58ad68785b35422
Read and report time cost of uncrypt (in seconds) from last_install.
Bug: 31383361
Change-Id: I283970c33b2a0324f2f51cde328e3e527eff7c58
(cherry picked from commit 3477efce30)
Currently, a container view that doesn’t accept events, but has a
child that accepts events, prevents its parent from receiving
LOCATION/DROP events while the drag is over the container (but not
the child). This is a bug.
With this fix, such a container will prevent the parent from
invoking a (second) LOCATION/DROP event only if the event was really
delivered to any of its descendants.
To know whether the event was delivered, I added
DragEvent.mEventHandlerWasCalled member.
EXITED/ENTERED events are now generated upon delivery of the event
that has coordinates in it.
Current view that has drag focus is now global to reflect the fact
that it’s one per process.
Bug: 31469490
Change-Id: I248e8d1de87b7734853136eb4719f7571cea91d5
We can no longer return the "my_downloads" paths: if those Uris were
shared beyond the app that requested the download, access would be
denied. Instead, we need to switch to using "all_downloads" Uris so
that permission grants can be issued to third-party viewer apps.
Since an app requesting a download doesn't normally have permission
to "all_downloads" paths, DownloadProvider now issues narrow grants
toward the owner of each download, both at device boot and when new
downloads are started.
Bug: 30537115, 30945409
Change-Id: I533125b36444877f54373d88922f2acc777e250b
Currently, every call to avoidBadWifi fetches the current value
of the config variable and setting whenever it is called. This
means that the score of an unvalidated wifi network can
unpredictably change at any time, creating a data race.
Instead, persist the value and only update it when something
changes.
Bug: 31075769
Change-Id: I0f4e0e742c91ef77fabc95d3ebb494338396aca5
getCurrentInputMethodSubtype() acquires InputManagerService.mMethodMap
within its body. There seems to be no reason for holding
InputMethodManager.mH to call getCurrentInputMethodSubtype(). Holding mH
can cause potential deadlock b/w two threads acquiring mH and mMethodMap
in different orders.
Bug: 31247871
Bug: 31273203
Bug: b.android.com/218423
Change-Id: I20cf2c20f49b1b02c0f7a18257b49d4bcc081b5d
Don't use the parole state to deal with the plugged-in trigger.
Otherwise standby apps will only be paroled for a few minutes
after plugging in to charge and not the entire duration.
Use a different intent for charging state, since the CHARGING
intent takes several seconds to be received.
A refactor some time ago changed the charging state behavior
that caused this regression.
Bug: 31399882
Change-Id: Ic036de5e136b3151b225473d0c3f440adb3b48e7
