targetLiveOffsetInMedia_unknownWindowStartTime_doesNotAdjustLiveOffset
is flaky on the release branch. It's not flaky on the dev-v2 branch
due to recent changes to FakeClock's behaviour which aren't yet present
on release-v2.

r2.13.2

PiperOrigin-RevId: 359486817

Latest LTS version is better than latest stable version because it will
be supported for longer.

#minor-release
Issue:#8581
PiperOrigin-RevId: 359467482

PiperOrigin-RevId: 359244236

- Ensure consistency between (Styled)PlayerControlView,
  PlayerNotificationManager, TimelineQueueNavigator and
  DefaultControlDispatcher.
- Handle the case where a live stream has ended when enabling previous
  and next actions (window.isLive() is true and window.isDynamic is
  false in that case)

PiperOrigin-RevId: 359063793

The configuration to publish to Maven seems to be incompatible with the
configuration for bintray, so only enable one at once. Once the GMaven
publishing flow is completely set up we can remove the exoplayerPublishEnabled
constant and the first branch entirely.

Issue: #5246
PiperOrigin-RevId: 359056610

The release AAR and Android sources are included for each published artifact.

Issue: #5246
Issue: #8353

#minor-release

PiperOrigin-RevId: 358146555

This behavior is consistent with DefaultControlDispatcher#dispatchNext().

#minor-release

PiperOrigin-RevId: 357145076

#minor-release

PiperOrigin-RevId: 359041513

PiperOrigin-RevId: 359000734

#minor-release

PiperOrigin-RevId: 358998449

#minor-release

PiperOrigin-RevId: 358847933

The SampleStream.readData contract is that when reading a sample
with a flags-only buffer, the buffer timestamp and flags should
be set and the read position should not be advanced.

#minor-release

PiperOrigin-RevId: 357842130

`ImaAdsLoader` will preload the first ad of a subsequent media item, but the
preloaded ad might not actually play because the user could seek to a non-zero
position in that media item (which could trigger playback of a midroll, not the
preroll). In this case, playback would get stuck because the midroll ad
expected to play after the seek would never load, because the IMA SDK expected
the preroll to play first.

Fix this behavior by discarding the preloaded ad break. If there isn't a seek,
the transition to the next media item is still seamless.

#minor-release

PiperOrigin-RevId: 357682510

Previously it was safe to query the first period in the timeline, but
now we support using the ads loader in concatenations we need to use
the current period index instead.

PiperOrigin-RevId: 357578003

Issue:#8581
PiperOrigin-RevId: 357563419

Issue: #8585
PiperOrigin-RevId: 357553237

The source can be used in compositions (in fact, every source is
automatically used in an internal composition when constructing the
playlist), and there is not really a concept of top-level media source
any more since the Player supports playlists.

The actual restriction is that the content media source needs to have
exactly one period to be able to create a SinglePeriodAdTimeline.

#minor-release

PiperOrigin-RevId: 357544191

The previous logic was changed under the assumption that the first box
inside a meta box was not always an hdlr box, but this is not true.

#minor-release

PiperOrigin-RevId: 357200713

PiperOrigin-RevId: 357160215

r2.13.1

#minor-release

PiperOrigin-RevId: 357273184

PiperOrigin-RevId: 357219132

The available end time was accidentally substracted by the start time
of the last period.

To avoid similar time reference confusion in the future, also renaming
many variables and methods to clearly reflect the time reference point.
And to avoid constant conversion, the processManifest method also
attempts to converge to time relative to the start of the window as
quickly as possible.

Issue: #8537
PiperOrigin-RevId: 357001624

If keepalive is disabled the existing code over-eagerly releases
DrmSession instances. This is arguably OK since a (Default)DrmSession
should be released before its (Default)Manager is released
(since the underlying MediaDrm instance might be released when the
manager is released). And if all sessions are released before the
manager is released then `sessions` is empty, so the loop is a no-op.

Issue: #8576
PiperOrigin-RevId: 356955308

The `DrmConfiguration.sessionForClearTypes` property is often used
to ensure a secure decoder is used for clear ads played in encrypted
content. This is because some devices show black frames when switching
decoders.

Before this change the DRM config isn't propagated down when
constructing the ad media source, meaning
`DrmSessionManager.DRM_UNSUPPORTED` is always used, which will
cause playback to switch from secure to clear decoder when transitioning
to an ad break (ignoring the MediaItem `sessionForClearTypes` option.

Issue: #8568

PiperOrigin-RevId: 356951124

Issue: #4803

PiperOrigin-RevId: 356923345

Issue: #8565

PiperOrigin-RevId: 356691251

Without this a new manager is instantiated for every item in a playlist,
meaning the impact of caching improvements to DefaultDrmSessionManager
are reduced (since the cache doesn't persist across playlist items).

With this change, playlists of items with identical DRM config will use
the same manager instance (and thus share existing sessions).

Issue: #8523
PiperOrigin-RevId: 356690852

Also allow the player's prepared ad media period durations array to exceed the
length of the loaded ad URIs array, as it's possible for the player to buffer
an ad media period fully at the point where it's known that an ad is coming up
but its URI is still unknown.

PiperOrigin-RevId: 356249284

This may happen for HLS live streams without program date time
information.

Issue: #8560

PiperOrigin-RevId: 356227729

#minor-release

PiperOrigin-RevId: 356221487

r2.13.0

#minor-release

PiperOrigin-RevId: 355652144

PiperOrigin-RevId: 355621265

Before, the level was set to null in this case.
MediaCodecUtil.getCodecProfileAndLevel() was therefore returning null
and the fallback to AVC/HEVC was not enabled in MediaCodecVideoRenderer.

Issue:#8530
PiperOrigin-RevId: 355574499

In both cases it's deliberate that all excluded constants should use
the default branch. Furthermore, there are quite a lot of excluded
constants missing, so it's probably better to suppress the warning
than to include them all.

#minor-release

PiperOrigin-RevId: 355426749