SDK / exoplayer

The ad break time in seconds from IMA was "-1" for postrolls, but this didn't
match C.TIME_END_OF_SOURCE in the ad group times array.

Handle an ad break time of -1 directly by mapping it onto the last ad group,
instead of trying to look it up in the array.

PiperOrigin-RevId: 312064886

The new behaviour:
- If the session is successfully opened, an acquire event is dispatched
  to all attached dispatchers.
- If acquire() is called but the session is already open, the acquire
  event is dispatched only to the dispatcher provided in to acquire()
- If the session is successfully released, a release event is dispatched
  to all attached dispatchers (in theory at most one should ever be
  attached at this point).
- If release() is called but the session isn't released (because
  referenceCount > 0) then a release event is dispatched only to the
  dispatcher provided to release().

PiperOrigin-RevId: 312062422

This is the rename-only part of https://github.com/google/ExoPlayer/pull/7370

PiperOrigin-RevId: 312057896

Issue:#7337
PiperOrigin-RevId: 312042768

This makes test failures with lists of items much easier to read

PiperOrigin-RevId: 312035040

Switch to snapshot Robolectric to pick up the latest version of shadows
required by MediaCodecVideoRendererTest and MediaCodecAudioRendererTest.

PiperOrigin-RevId: 312030332

PiperOrigin-RevId: 311755157

Detect stuck buffering cases in ImaAdsLoader, and discard the ad group after
a timeout. This is intended to make the IMA extension more robust in the case
where an ad group unexpectedly doesn't load.

The timing out behavior is enabled by default but apps can choose to retain
the old behavior by setting an unset timeout on ImaAdsLoader.Builder.

PiperOrigin-RevId: 311729798

EventTime contains information about when an event happened and where
it belongs to. Both places can be fully described using timeline, window
index, media period id and position.

Right now, only the information for where the event belongs to is fully
contained in EventTime, whereas the time when the event happened only has
the position, and none of the other information (timeline, window, period).

This change adds the missing information, so that the EventTime can easily
be used without having access to the Player. This also ensures Event
metadata is self-contained and can be stored and reused later.

issue:#7332
PiperOrigin-RevId: 311727004

Removes the experimental methods to set a timeout when
releasing the player and setting the surface.

PiperOrigin-RevId: 311703988

- Add Widevine AV1 streams
- Remove SD and HD only Widevine streams (we don't need so many!)
- Simplify naming

PiperOrigin-RevId: 311697741

PiperOrigin-RevId: 311628160

PiperOrigin-RevId: 311623784

Some of the CSS font sizes are derived from the current view height, if
this calculation is done before the view has been measured then a zero
view height results in a zero px font size and no visible text.

This can happen when the view type is changed (and so the
WebViewSubtitleOutput has been recently added to the SubtitleView
ViewGroup).

PiperOrigin-RevId: 311552052

ANSI/CTA-608-E R-2014 spec defines exactly 32 columns on the screen,
and limits all lines to this length.

See 3.2.2 definition of 'Column'.

issue:#7341
PiperOrigin-RevId: 311549881

PiperOrigin-RevId: 311531734

PiperOrigin-RevId: 311527440

PiperOrigin-RevId: 311513746

Issue:#7309
PiperOrigin-RevId: 311504497

Some player setup steps that are likely to be only done once
should be moved into the Builder so that player setup can use
a consistent style (builder vs setters).

This also prevents some threading warning issues when the player
is built on a background thread (e.g. for dependency injection
frameworks) and setters can't be used due to threading restrictions.

PiperOrigin-RevId: 311487224

This is the last piece to bring feature parity between
WebViewSubtitleOutput and CanvasSubtitleOutput.

PiperOrigin-RevId: 311325749

Issue:#7309
PiperOrigin-RevId: 311324242

It displays images too, and in fact it's used exclusively to display
images in SubtitleWebView. It also doesn't use a TextView - so all round
a slightly confusing name.

Also rename SubtitleWebView to WebViewSubtitleOutput to match the same
pattern.

PiperOrigin-RevId: 311312758

- Remove proguard-rules.txt symlink
- Guard against classes annotated with @ClosedSource but not excluded

PiperOrigin-RevId: 311312671

This is useful for merging the FFmpeg video support pull request,
since it allows a Decoder base class implementation to catch
DecoderException rather than forcing it to catch Exception (i.e.,
everything). Note that unfortunately, Java doesn't allow catching
of a generic type (i.e., you cannot "catch (E e)") due to type
erasure.

PiperOrigin-RevId: 311300719

PiperOrigin-RevId: 311298802

PiperOrigin-RevId: 311295749

The only reason to require a context for creating an instance of
DefaultMediaSourceFactory is creating a user agent. DownloadHelper
has a bunch of static methods that need to instantiate a
DefaultMediaSourceFactory without a Context. By adding a setter for
the user agent we can remove that restriction and use a default user
agent.

PiperOrigin-RevId: 311291603

We aim for all SinglePeriodTimelines to set window.mediaItem and hence want to eliminate all constructors not passing a media item. The constructor removed in this CL is the easiest one to remove and fix in google3. It is still a breaking change for external ExoPlayer user who provide a custom media source with this timeline.

PiperOrigin-RevId: 311291493

PiperOrigin-RevId: 311290214

PiperOrigin-RevId: 311150195

Two tests had the same code when the were
supposed to test different state.

PiperOrigin-RevId: 311149216

ISSUE: #7114
PiperOrigin-RevId: 311115835

PiperOrigin-RevId: 311106612

Put static methods at the end.

Also add a couple of missing parameter name comments.

PiperOrigin-RevId: 311077684

Log content progress events, as these are helpful to debug triggering of events
based on the content progress.

Don't log AD_PROGRESS events as they occur several times per second while ads
are playing, and the verbosity makes logs difficult to read.

PiperOrigin-RevId: 311077302

It's very unlikely any onQueueInputBuffer implementations are looking
at the subsample encryption data, but from a correctness perspective
we should be passing a buffer object that's self-consistent.

PiperOrigin-RevId: 311011246

PiperOrigin-RevId: 311002702

PiperOrigin-RevId: 310997932

FFmpeg requires input buffers to be sized larger than the size
of the data they contain. This is to allow optimized decoder
implementations that read data in fixed size chunks, without
the risk of such decoders reading beyond the end of the buffer.

Issue: #2159
PiperOrigin-RevId: 310946866