SDK / exoplayer

Issue: androidx/media#213

Without a public constructor, it is not possible to extend this class and override its method.

PiperOrigin-RevId: 491673111

Follow-up CLs will use this duration for silent audio.

PiperOrigin-RevId: 491670359

PiperOrigin-RevId: 491660842

Use the bitrate of the audio format (when available) in
DefaultAudioSink.AudioTrackBufferSizeProvider.getBufferSizeInBytes() to
calculate accurate buffer sizes for direct (passthrough) playbacks.

#minor-release

PiperOrigin-RevId: 491628530

PiperOrigin-RevId: 491623586

To support OPUS offload, we need to provide a few configuration values
that are currently not set due to the lack of devices supporting
OPUS offload.

PiperOrigin-RevId: 491613716

Main change:

- Removed `Codec.EncoderFactory.createForVideoEncoding`'s argument of a list
of allowed MIME types
- Moved the check for whether a video MIME type is supported to VTSP

PiperOrigin-RevId: 491611799

The fix for b/171657375 (internal) has been shipped with 21.1.0 already (see https://developers.google.com/cast/docs/release-notes#august-8,-2022).

PiperOrigin-RevId: 491583727

PiperOrigin-RevId: 491377695

PiperOrigin-RevId: 491336828

The list of charsets is already hard-coded, and using `Charset` types
ensures they will all be present at run-time, hence we will never
encounter an 'unsupported' charset.

PiperOrigin-RevId: 491324466

This should fix how color matrix transforms look when applied on HDR colors

PiperOrigin-RevId: 491323228

This thread just starts the player and handles the player callbacks for
now. Sample pipelines are still run on the playback thread.

PiperOrigin-RevId: 491299671

It's clearer if each test method follows the Arrange/Act/Assert pattern

PiperOrigin-RevId: 491299379

PiperOrigin-RevId: 491289028

FakeClock keeps an internal list of messages to be executed to
ensure deterministic serialization. The next message from the list
is triggered by a separate helper message sent to the real Handler.
However, if the target HandlerThread is no longer alive (e.g. when
it quit itself during the message execution), this helper
message is never executed and the entire message execution chain
is stuck forever.

This can be solved by checking the return values of Hander.post or
Handler.sendMessage, which are false if the message won't be
delivered. If the messages are not delivered, we can unblock the
chain by marking the message as complete and triggering the next
one.

PiperOrigin-RevId: 491275031

Remove self-links, and remove section that is documenting internal
ordering behaviour of
[`SimpleBasePlayer.getCombinedMediaMetadata`](https://github.com/google/ExoPlayer/blob/bb270c62cf2f7a1570fe22f87bb348a2d5e94dcf/library/common/src/main/java/com/google/android/exoplayer2/SimpleBasePlayer.java#L1770)
rather than anything specifically about this method.

#minor-release

PiperOrigin-RevId: 490923719

This is more clear than using Format.NO_VALUE, when we do actually intend for an
output value.

Also, fix @see formatting by using summary fragments instead, and add an error
output for OETF and EOTF transfer functions.

PiperOrigin-RevId: 490910229

Our FakeClock generally makes sure that playback tests are fully
deterministic. However, this fails if the test uses blocking waits
with clock.onThreadBlocked and where relevant Handlers are created
without using the clock.

To fix the flakiness, we can make the following adjustments:
 - Use TestExoPlayerBuilder instead of legacy ExoPlayerTestRunner
   to avoid onThreadBlocked calls. This also makes the tests more
   readable.
 - Use clock to create Handler for FakeVideoRenderer and
   FakeAudioRenderer. Ideally, this should be passed through
   RenderersFactory, but it's too disruptive given this is a
   public API.
 - Use clock for MediaSourceList and MediaPeriodQueue update
   handler.

PiperOrigin-RevId: 490907495

PiperOrigin-RevId: 490836952

Format expects the values of `averageBitrate` and `peakBitrate` in bps and the value fetched from AC3SpecificBox and EC3SpecificBox is in kbps.

PiperOrigin-RevId: 490756581

PiperOrigin-RevId: 490756055

To avoid complicated bit shifting and masking. Also makes the code more readable.

PiperOrigin-RevId: 490749482

Also make sure the demo app doesn't crash when it happens.

PiperOrigin-RevId: 490725959

*** Original commit ***

Rollback of https://github.com/androidx/media/commit/76df06a7a364c580dfe07d9f069237cd77c5174c

*** Original commit ***

Parse and set `peakBitrate` for Dolby TrueHD(AC-3) and (E-)AC-3

#minor-release

***

***

PiperOrigin-RevId: 490707234

*** Original commit ***

Parse and set `peakBitrate` for Dolby TrueHD(AC-3) and (E-)AC-3

#minor-release

***

PiperOrigin-RevId: 490570517

This reduces the priority to best effort (from the default that seems to be best effort), and allows us to run SSIM even on 8k24fps video. Without this CL,
start()'ing a second codec may result in a MediaCodec.CodecException.

Tested to confirm that transformation8k24():
* fails deterministically without this CL, or with KEY_PRIORITY set to 0.
* succeeds deterministically after this CL (~18s on Samsung Z Fold 4)

PiperOrigin-RevId: 490570416

We cannot check this in code, due to DEVICE_INITIAL_SDK_INT being a @SystemApi, and
reflection being a bit risky/unstable.

PiperOrigin-RevId: 490537916

#minor-release

PiperOrigin-RevId: 490527831

Before this change, the bitmap loading result with mainHandler, in which we set the metadata to `MediaSessionCompat`. However, the `MediaSessionCompat` is not thread safe, all calls should be made from the same thread. In the other calls to `MediaSessionCompat`, we ensure that they are on the application thread (which may be or may not be main thread), so we should do the same for `setMetadata` when bitmap arrives.

Also removes a comment in `DefaultMediaNotificationProvider` as bitmap request caching is already moved to CacheBitmapLoader.

PiperOrigin-RevId: 490524209

The bug has since been fixed.

The values still could change, as the API is labelled as @UnstableApi, so it's
probably fine to leave the <p> tag mostly as is.

PiperOrigin-RevId: 490509205

Issue: androidx/media#209

#minor-release

PiperOrigin-RevId: 490492223

This change includes a change in the `IMediaController.aidl` file and needs
to provide backwards compatibility for when a client connects that is of an older or
newer version of the current service implementation.

This CL proposes to create a new AIDL method `onPlayerInfoChangedWithExtensions`
that is easier to extend in the future because it does use an `Bundle` rather than
primitives. A `Bundle` can be changed in a backward/forwards compatible way
in case we need further changes.

The compatibility handling is provided in `MediaSessionStub` and `MediaControllerStub`. The approach is not based on specific AIDL/Binder features but implemented fully in application code.

Issue: androidx/media#102
#minor-release
PiperOrigin-RevId: 490483068

If there is output data available (outputBuffer.hasRemaining()), then
there is no need to move other data between the underlying processors.
It will not change the buffer being returned by that call to getOutput.

If there is no output data readily available, it's necessary to go to
the AudioProcessors and pass buffers between them, as this may produce
data for output.

PiperOrigin-RevId: 490482653

The direct executor is not the proper way to determine on what thread to run the
`Future.Listener` and the `MediaControllerCreationListener` because the listener
may call the controller passed as argument which must happen on the same thread
that built the controller. This change makes sure this is the case.

PiperOrigin-RevId: 490478587

PiperOrigin-RevId: 490465182

This allows us to release both codecs used in SSIM when one fails to
configure() or start().

Tested and confirmed that on Samsung Galaxy Z Flip 4, running all
TransformationTest.java tests, tests after transform8k24() fails to start the
2nd codec:
* Before this CL, all fail.
* After this CL, all pass.

PiperOrigin-RevId: 490461560

This avoids that apps have to depend on the legacy compat support
library when they want to make this conversion.

Also add a version to both helper methods that takes a Looper to
give apps the option to use an existing Looper, which should be
much faster than spinning up a new thread for every method call.

Issue: androidx/media#171
PiperOrigin-RevId: 490441913

PiperOrigin-RevId: 490376734

When receiving the `onTimelineChanged` callback, we convert the timeline to the list of `QueueItem`s, where decoding a bitmap is needed for building each of the `QueueItem`s. The strategy is similar to what we did in <unknown commit> for list of `MediaBrowserCompat.MediaItem` - set the queue item list until the bitmaps decoding for all the `MediaItem`s are completed.

PiperOrigin-RevId: 490283587