SDK / exoplayer

PiperOrigin-RevId: 277916639

PiperOrigin-RevId: 277916508

PiperOrigin-RevId: 277916279

PiperOrigin-RevId: 277916113

PiperOrigin-RevId: 277911191

PiperOrigin-RevId: 277910909

PiperOrigin-RevId: 277910360

PiperOrigin-RevId: 277896757

PiperOrigin-RevId: 277766372

PiperOrigin-RevId: 277744735

PiperOrigin-RevId: 277732159

PiperOrigin-RevId: 277729887

PiperOrigin-RevId: 277710160

PiperOrigin-RevId: 277709114

Suggested by Android Studio's Inspect Code

SO seems to think there's a performance difference for the | change:
https://stackoverflow.com/questions/22132450/why-is-a-character-class-faster-than-alternation

PiperOrigin-RevId: 277707341

PiperOrigin-RevId: 277706810

The value is stateful and new sessions need to be informed of the current state

PiperOrigin-RevId: 277701955

PiperOrigin-RevId: 277695826

PiperOrigin-RevId: 277691935

This should not introduce any functional changes.

PiperOrigin-RevId: 277691550

E-AC3 with JOC is signaled using the CHANNELS attribute for HLS:
https://developer.apple.com/documentation/http_live_streaming/hls_authoring_specification_for_apple_devices/hls_authoring_specification_for_apple_devices_appendices

PiperOrigin-RevId: 277680300

Some methods marked as throwing DecoderQueryException actually
did not throw it. Switch to marking throws Exception
everywhere.

PiperOrigin-RevId: 277674401

Relevant part of the spec:
https://www.w3.org/TR/webvtt1/#processing-cue-settings

PiperOrigin-RevId: 277523694

This means multiple failures are all logged, instead of the test
stopping when the first assertion fails. Makes tests like this with
lots of independent assertions much easier to work with.

I limited the expect to a single assertCues() call, otherwise
the error message gets very long and is hard to read, but I
still think this is an improvement.

PiperOrigin-RevId: 277523300

Cue's UNSET and null values should be used when the source data
doesn't specify *and* the spec doesn't provide a clear default.
In the WebVTT case, the defaults are clear, so we use them
explicitly.

PiperOrigin-RevId: 277522899

PiperOrigin-RevId: 277519718

PiperOrigin-RevId: 277489763

PiperOrigin-RevId: 277477784

PiperOrigin-RevId: 277462799

The framework opus decoder discards some samples after a call to
flush(). Because we flush a decoder that is being retained across an
input format change, this means that the start of audio gets truncated
when transitioning to a new opus stream. See also
https://android.googlesource.com/platform/frameworks/av/+/refs/heads/android10-release/media/libstagefright/codecs/opus/dec/SoftOpus.cpp.

Avoid this by recreating opus decoders instead of flushing them. It
seems fine to do this for all opus decoders as reinitialization should
be cheap, OEM-provided implementations may also discard samples and
playback shouldn't be interrupted on reinitialization due to the
downstream AudioTrack buffer.

PiperOrigin-RevId: 277458759

The nested ternary is quite hard to read imo, and I believe this is a
no-op change that captures the intent more clearly.

PiperOrigin-RevId: 277320461

The demo app was using this via its dependency on the material library,
but it's preferable to list dependencies explicitly (otherwise the build
would break if we removed the material dependency).

PiperOrigin-RevId: 277316175

PiperOrigin-RevId: 277259448

Which ensures both get updated when the MediaSessionConnector player
changes.

Issue:#6582
PiperOrigin-RevId: 277254889

PiperOrigin-RevId: 277253308

This helps get HlsMediaPeriod off the blacklist.

PiperOrigin-RevId: 277076633

Update it to throw if hasNext() is false, to match Java's Iterator
interface.

This also simplifies the null-checking annotations required

PiperOrigin-RevId: 277059766

PiperOrigin-RevId: 277057328

PiperOrigin-RevId: 277056574

The leanback library doesn't know about non-square pixels. So if
we're playing content that uses non-square pixels, we need to adjust
the video dimensions that we provide to leanback such that it
renders the video with the correct aspect ratio.

PiperOrigin-RevId: 277042560