Add a remote call addBluetoothDevice() using AIDL.
This was needed because onBind() is only called once.
Bug: 23219556
Bug: 23760886
Change-Id: Id7554ca55d596352d11dbd6ae3e403138a29c864
Signed-off-by: Phil Burk <philburk@google.com>
It was found that createByteBufferFromABuffer was recieving a NULL buffer in bug
22726278. Based on returning NULL when buffer->base() == NULL, returning NULL
when buffer == NULL. This does not fix bug 22726278 - but address a problem found
while looking into it.
Bug: 22726278
Change-Id: I933b6a4a535d803576a4cdd7a1ffaf4a428c3201
Add a new getTimestamp() method that returns an error code.
Bug: 21886109.
Bug: 22202349.
Bug: 17548098.
Change-Id: Ib072dd3442de79bab9aca77c3c33b0790b492154
(cherry picked from commit 4e1ccd370fc268f3207c5baee08ac8c943c3fdf9)
The HashMap is being used but devices never added to the map.
Bug: 23429459
Change-Id: I1da5305a56f5bd48e1c5d9345e721ea8dd2eed4e
Signed-off-by: Phil Burk <philburk@google.com>
Stop openBluetoothDevice() from calling its callback twice.
Bug: 23218338
Change-Id: I2e738b22a124c0336c46bef05680a82ee1f59e79
Signed-off-by: Phil Burk <philburk@google.com>
Add a new getTimestamp() method that returns an error code.
Bug: 21886109.
Bug: 22202349.
Bug: 17548098.
Change-Id: Ib072dd3442de79bab9aca77c3c33b0790b492154
This patch does not yet allow to upload files, but uploading (creating)
directories already works.
Bug: 22545670
Change-Id: If4d5a53aa26f791475bb1a783e0ac9540d6760c1
Mostly due to no standard stream configurations being defined,
and for the correct overrides for DEPTH_POINT_CLOUD format.
Bug: 20537722
Change-Id: I8a18f5f68697a09dcc4d7555e51728193fe7f333
The existing APIs required to copy all bytes to memory, which will fail in
case of very large files, like movies.
Bug: 22908937
Change-Id: I23bdcbdbf08b7c1b1017591799bbb94e53249a57
* commit '29e678b9c77fa75232b4b9e147430a920e6f6cc4':
Documenting behavior of returning a 0-length array for formats, channels, rates queries for analog devices.
So that adding and removing device listeners is thread-safe.
Bug: 22909752
Change-Id: I1369ab53c865b54e6e5eb8ea3e8d437fdd1d7197
Signed-off-by: Phil Burk <philburk@google.com>
When connecting/disconnecting and audio device, AudioService
was assuming that the call was always successful through
AudioSystem.setDeviceConnectionState(). In the case of the
connection of a wired headset (mic + headphones), this also
causes A2DP to not be used for media playback.
The connection can fail if the audio device being connected
is not supported by the platform. But if this failure is
not taken into account for DEVICE_OUT_WIRED_HEADSET or
HEADPHONES, A2DP would still be avoided even when an A2DP
device reconnects at boot.
The fix consists in executing the connection logic only when
the connection was deemed successful. Nothing is altered
on the disconnection code path, or the direct connection
of A2DP.
The javadoc is updated in AudioSystem to indicate the return
codes to take into account in setDeviceConnectionState().
Bug 22511833
Change-Id: I22f0d2c7d4ab4fb9ee1be2f248907f721596a16f
