A few changes to demonstrate a slightly cleaner way to write
JNI code in the framework especially when multiple types are
involved. We use this pattern in many other places outside
of the media stack.
Added more detail to the UnsupportedOperationException that
occurs when buffer formats don't match.
Change-Id: Ic894dc1bd71b387f1be6ea1798fa59e533e9574f
Define specific failure cases so apps have the information they
need to deal with these conditions.
Also adds a new ResourceBusyException to MediaDrm
Change-Id: Iaecf269d58108f28179974b05671bf29b9fe4b7d
related-to-bug: 10157154
related-to-bug: 9695816
Remove dead assignments to member fields in constructor prior to throw.
Execution doesn't continue execution, so no need to initialize fields.
throw doesn't need parentheses.
Fix indentation.
Change-Id: I2bf1b8bfa2c836e53a41eea67552cba40dc6fd43
The stream type is used in audio focus to better deal with
RemoteControlClient/Display in the case of notifications
during media playback. This comment is therefore outdated.
Change-Id: Ifa33bc4c0fbf065d8288ff5fdfaf1bc67b2a337e
Unhide android.media.AudioManager.AUDIOFOCUS_GAIN_TRANSIENT_EXCLUSIVE
Update android.media.AudioManager.requestAudioFocus() javadoc for the
new focus type.
Bug 8251963
Change-Id: I7cd69faaf5f0169ee462b91d0aaa75dc6b092107
Summary of feature:
Do not mute notifications when speech recognition recording is
active, but when an app has requested audio focus with
AudioManager.AUDIOFOCUS_GAIN_TRANSIENT_EXCLUSIVE.
Implementation:
Move definition of AUDIOFOCUS_NONE to AudioManager where other
audio focus codes are defined.
Add support for querying the current audio focus type.
When audio focus is requested as GAIN_TRANSIENT_EXCLUSIVE, make
the corresponding loss by LOSS_TRANSIENT.
Before playing a notification, check whether GAIN_TRANSIENT_EXCLUSIVE
has been requested.
Bug 8251963
Change-Id: I41edc77326b70639d2fdcb4642c53109995b72a8
Context.deleteDatabase() has to be called instead of File.delete()
to guarantee all db related files
like journal, wal, shm, etc are deleted together.
Signed-off-by: jangwon.lee<jangwon.lee@lge.com>
Change-Id: I3c3d7dfa13ffbe709ab7d8a5c7cfe6cfa2864fea
When a new focus owner lands on the stack, don't just make the
previous top of the stack lose focus, propagate the loss throughout
the stack.
Only dispatch focus loss on focus loss state change
Remove canDispatchFocus() method as it now doesn't need to be known
(and shouldn't be known) outside of the implementation.
Fix error where the focus code for a focus gain request should
always be a focus loss code.
Bug 8315302
Change-Id: I92c8f51fdcc090851d34d00fefed916e25da40c1
Implement a state machine to represent how to transition to the
next audio focus loss state, given the current focus loss (state)
and the given external focus gain (transition).
Bug 8315302
Change-Id: I4a3d5a4a53c842caceaa876206a993c4e5446681
Move all audio focus request functionality under a new class/file.
Clean up encapsulation of data related to the request.
Change-Id: I989796e1ee1a5fc99799a64e1612294e0e40fa6d
Move more audio focus-specific functionality into the class
representing each audio focus owner, FocusStackEntry.
Keep track of how each FocusStackEntry instance lost focus.
Change-Id: I35df0717765a26ec747cb0110e2e951d155d1525
According to the API documentation and the ICS implementation
the setDataSource(String) method throws an
IllegalArgumentException if path is null. In JB a
NullpointerException is thrown instead. This fix restores
the earlier behaviour.
Change-Id: Ic47baadf91076acc227d92d84f6b8d1d6ecd0c03
If reading the decoded pixel data (instead of letting the system
render it), this is critical for interpreting the data correctly.
The video MediaCodec CTS tests use this MediaFormat key in this
way as well.
Therefore, make sure the documentation reflects the actual use of
this field - claiming that it is encoder-only is misleading.
Change-Id: I9bfd295a95dd3e7d1fa3189886fd243b07bb82cc
Add fields in an audio focus stack entry to track what type of audio
focus was requested, and why it was lost.
Make audio focus dump easier to read.
Change-Id: I61f6e43c13a32328a07125a2ba0dce1053872c86
This clarifies that it's a channel mask, as opposed to a count.
Also no need to initialize the field in the declaration,
as it's initialized in the constructor.
Include channel mask in error logs.
Change-Id: Ifc02668cf922dc8826a9fb042b3ca52fad377ba6
