This method is helpful when iterating the list of track overrides
to figure out which type the override applies to.
Issue: google/ExoPlayer#9665
PiperOrigin-RevId: 409108977
The Javadoc of DefaultTrackSelector can be shortened as it's not the
right place to document detailed options of the Player track selection
parameters.
The documentation page about track selection is updated to the new
APIs and extended with most relevant options and information needed
to work with ExoPlayer's track selection API.
#minor-release
PiperOrigin-RevId: 409088989
MediaSource can be reused with other Player instances after they
have been released, so we need to set the PlayerId when preparing
the source. Access can mostly be handled by the implementation in
BaseMediaSource.
PiperOrigin-RevId: 408878824
Robolectric uses the JRE HttpURLConnection [1], while real Android
devices and emulators use OkHttp to implement HttpURLConnection. This
can lead to important differences in behaviour, so it's better to use
instrumentation tests when specific HTTP behaviour is important.
[1] https://github.com/robolectric/robolectric/issues/6769#issuecomment-943556156
PiperOrigin-RevId: 408840295
The setters in the Builder are already deprecated and using the
old getter is error-prone as they only return the overrides set
with the deprecated setters.
Issue: google/ExoPlayer#9665
PiperOrigin-RevId: 408817640
The `VideoSamplePipeline` handles all steps from decoding to
re-encoding that where previously in
`TransformerTranscodingVideoRenderer`. The renderer is now only
responsible for reading the format, reading input, passing it to the
pipeline and passing the pipeline's output to the muxer.
When no transformations are needed, decoding and re-encoding is
skipped using the `PassthroughPipeline`.
PiperOrigin-RevId: 408619407
We can rename the existing setIndex method to a more generic init
as this method is only called by EPII and implemented by BaseRenderer
anyway.
PiperOrigin-RevId: 408616055
When dropping the remainder, the decoder and encoder timestamps start diverging after a few buffers when no speed changes are supposed to occur. Tracking the remainder keeps them in sync.
PiperOrigin-RevId: 408341074
The platform class is only available from API 31, so we need
a generic wrapper that can be used on all API levels. The wrapper
essentially provides an identifier for a player instance, so naming
it accordingly.
PiperOrigin-RevId: 408292802
And in a couple of related places.
This is for consistency with the rest of the codebase where
we exclusively use indices.
#minor-release
PiperOrigin-RevId: 408273372
The comment "all should be same" is not correct, it is extremely likely they will be the same but not assured. In either case the seek position will work, it just may not land exactly on a sync point in every variant.
When operating the MediaCodec in asynchronous mode, after a
MediaCodec.flush(), we start MediaCodec in the callback thread,
which might trigger errors in some platforms. This change adds an
experimental flag to move the call to MediaCodec.start() back to the
playback thread.
PiperOrigin-RevId: 407801013
When we have multiple overrides for TrackGroups associated with
one renderer, we need to look at all of them to find the non-empty
one. Empty ones should only be used to remove previously selected
tracks for this group and otherwise be ignored.
Currently this is broken because the first override (no matter if
it's empty or not) is used as the final selection for this renderer.
Issue: google/ExoPlayer#9649
#minor-release
PiperOrigin-RevId: 407792330
* Remove GlUtil.Program String[] constructor to unify and just use the
String constructor.
* Add getAttributeArrayLocationAndEnable() to simplify things a tiny bit.
* Increase usage of constant values.
PiperOrigin-RevId: 407570340
`TransformerAudioRenderer` reads input and passes `DecoderInputBuffer`s to the `AudioSamplePipeline`. The `AudioSamplePipeline` handles all steps from decoding to encoding. `TransformerAudioRenderer` receives `DecoderInputBuffer`s from the `AudioSamplePipeline` and passes their data to the muxer.
`AudioSamplePipeline` implements a new interface `SamplePipeline`. A pass-through pipeline will be added in a future cl.
PiperOrigin-RevId: 407555102
If the number of samples changes, the sizes will help us to verify whether they are just split differently or extra data was added.
PiperOrigin-RevId: 407346280
Add protected method DefaultRenderersFactory.getCodecAdapter(), so that
subclasses of DefaultRenderersFactory that override
buildVideoRenderers() or buildAudioRenderers() can access the
DefaultRenderersFactory codec adapter factory and pass it to
MediaCodecRenderer instances they may create.
#minor-release
PiperOrigin-RevId: 407345431
Where this introduced an inconsistency (e.g. assigning to something
called `windowIndex`), I generally renamed the transitive closure of
identifiers to maintain consistency (meaning this change is quite
large). The exception is code that interacts with Timeline and Window
directly, where sometimes I kept the 'window' nomenclature.
#minor-release
PiperOrigin-RevId: 407040052
Where this introduced an inconsistency (e.g. assigning to something
called `windowIndex`), I generally renamed the transitive closure of
identifiers to maintain consistency (meaning this change is quite
large). The exception is code that interacts with Timeline and Window
directly, where sometimes I kept the 'window' nomenclature.
#minor-release
PiperOrigin-RevId: 407040052
The video scaling mode and stream type defines a default constant
that needs to be added to the IntDef definition to be assignable.
PiperOrigin-RevId: 406835696
The return values of AudioManager.getPlaybackOffloadSupport are the same as the values defined in C.AudioManagerOffloadMode.
PiperOrigin-RevId: 406817413
The presentation time in fMP4 is calculated by adding and subtracting
3 values. All 3 values are currently converted to microseconds first
before the calculation, leading to rounding errors. The rounding errors
can be avoided by doing the conversion to microseconds as the last step.
For example:
In timescale 96000: 8008+8008-16016 = 0
Rounding to us first: 83416+83416-166833=-1
#minor-release
PiperOrigin-RevId: 406809844
Currently, clipping errors are never thrown if we already have a
MediaPeriod. This may happen for example for ProgressiveMediaSource
where we need to create a MediaPeriod before knowing whether clipping
is supported. Playback will still fail, but with unrelated assertion
errors that are hard to understand for users.
Fix this by setting the pending error on the ClippingMediaPeriod.
#minor-release
Issue: Issue: google/ExoPlayer#9580
PiperOrigin-RevId: 406809737
These IntDefs are now annotated with TYPE_USE [1], so they can be moved
to directly before the type (int).
[1] Since <unknown commit>
PiperOrigin-RevId: 406803555
This allows the use of the intdef in parameterized types,
e.g. List<@MyIntDef Integer>
For IntDefs that are already released in ExoPlayer 2.15.1 we add
TYPE_USE in addition to all other reasonable targets, to maintain
backwards compatibility with Kotlin code (where an incorrectly
positioned annotation is a compilation failure). 'reasonable targets'
includes FIELD, METHOD, PARAMETER and LOCAL_VARIABLE but not TYPE,
CONSTRUCTOR, ANNOTATION_TYPE, PACKAGE or MODULE. TYPE_PARAMETER is
implied by TYPE_USE.
For not-yet-released IntDefs we just add TYPE_USE.
#minor-release
PiperOrigin-RevId: 406793413
This helps to prevent issues where decoders can't handle negative
timestamps. In particular it avoids issues when the media accidentally
or intentionally starts with small negative timestamps. But it also
helps to prevent other renderer resets at a later point, for example
if a live stream with a large start offset is enqueued in the playlist.
#minor-release
PiperOrigin-RevId: 406786977
The video scaling mode and stream type defines a default constant
that needs to be added to the IntDef definition to be assignable.
PiperOrigin-RevId: 406835696
The return values of AudioManager.getPlaybackOffloadSupport are the same as the values defined in C.AudioManagerOffloadMode.
PiperOrigin-RevId: 406817413
The presentation time in fMP4 is calculated by adding and subtracting
3 values. All 3 values are currently converted to microseconds first
before the calculation, leading to rounding errors. The rounding errors
can be avoided by doing the conversion to microseconds as the last step.
For example:
In timescale 96000: 8008+8008-16016 = 0
Rounding to us first: 83416+83416-166833=-1
#minor-release
PiperOrigin-RevId: 406809844
Currently, clipping errors are never thrown if we already have a
MediaPeriod. This may happen for example for ProgressiveMediaSource
where we need to create a MediaPeriod before knowing whether clipping
is supported. Playback will still fail, but with unrelated assertion
errors that are hard to understand for users.
Fix this by setting the pending error on the ClippingMediaPeriod.
#minor-release
Issue: Issue: google/ExoPlayer#9580
PiperOrigin-RevId: 406809737
These IntDefs are now annotated with TYPE_USE [1], so they can be moved
to directly before the type (int).
[1] Since <unknown commit>
PiperOrigin-RevId: 406803555
This allows the use of the intdef in parameterized types,
e.g. List<@MyIntDef Integer>
For IntDefs that are already released in ExoPlayer 2.15.1 we add
TYPE_USE in addition to all other reasonable targets, to maintain
backwards compatibility with Kotlin code (where an incorrectly
positioned annotation is a compilation failure). 'reasonable targets'
includes FIELD, METHOD, PARAMETER and LOCAL_VARIABLE but not TYPE,
CONSTRUCTOR, ANNOTATION_TYPE, PACKAGE or MODULE. TYPE_PARAMETER is
implied by TYPE_USE.
For not-yet-released IntDefs we just add TYPE_USE.
#minor-release
PiperOrigin-RevId: 406793413
This helps to prevent issues where decoders can't handle negative
timestamps. In particular it avoids issues when the media accidentally
or intentionally starts with small negative timestamps. But it also
helps to prevent other renderer resets at a later point, for example
if a live stream with a large start offset is enqueued in the playlist.
#minor-release
PiperOrigin-RevId: 406786977
Initialize default components lazily in ExoPlayer.Builder to avoid
redundant component instantiations, useful in cases where apps
overwrite default components with ExoPlayer.Builder setters.
The fields in ExoPlayer.Builder are wrapped in a Supplier (rather than
just making then nullable and initializing them in
ExoPlayer.Builder.build()) so that we maintain the proguarding
properties of this class. The exception is
ExoPlayer.Builder.AnalyticsCollector which became nullable and is
initialized in ExoPlayer.Builder.build() in order to use any Clock
that has been set separately with ExoPlayer.Builder.setClock().
#minor-release
PiperOrigin-RevId: 406345976
This makes sure that #EXT-X-RENDITION-REPORT tags can be placed before the list of segments/parts as well. We were previously assuming that these come at the end, which naturally would make sense and is done like this in all examples, but it is not explicitly defined by the spec.
Issue: google/ExoPlayer#9592
PiperOrigin-RevId: 406329684
Only deprecated references remain.
Usages of the deprecated methods will be migrated in a follow-up change.
#minor-release
PiperOrigin-RevId: 405927141
Also add a setRenderersFactory() method, so that all constructor-provided
components can also be passed via setters.
This comment already appears on the constructor that takes all
components, but it applies to these ones as well.
PiperOrigin-RevId: 405917343
`checkNotNull` should be avoided where possible.
This change adds `@EnsuresNonNull` or `@EnsuresNonNullIf` to configuration methods for fields they initialize.
`checkNotNull` is now avoided for the `@MonotonicNonNull` formats by adding `@RequiresNonNull` annotations.
`checkNotNull` is now avoided for the encoder and decoder in `feedMuxerFromEncoder()`, `feedEncoderFromDecoder()`, `feedDecoderFromInput()`, etc. by creating local variables for `encoder` and `decoder` in `render` after the configuration method calls and passing these as non-null parameters.
PiperOrigin-RevId: 405893824
Our package-info.java files are annotated with @NonNullApi which results
in everything being non-null by default, so this annotation is never
needed.
#minor-release
PiperOrigin-RevId: 405864737
Test file produced with:
$ MP4Box -add "sample.mp4#video:colr=nclc,1,1,1" -new sample_18byte_nclx_colr.mp4
And then manually changing the `nclc` bytes to `nclx`.
This produces an 18-byte `colr` box with type `nclx`. The bitstream of
this file does not contain HDR content, so the file itself is invalid
for playback with a real decoder, but adding the box is enough to test
the extractor change in this commit.
(aside: MP4Box will let you pass `nclx`, but it requires 4 parameters, i.e. it
requires the full_range_flag to be set, resulting in a valid 19-byte colr box)
#minor-release
Issue: #9332
PiperOrigin-RevId: 405842520
- This CL does not introduce functional changes.
- This change will allow searching for the clli box while
parsing the mdcv box in order to construct the HDR
static info contained in ColorInfo.
#minor-release
PiperOrigin-RevId: 405656499
Also check that the output video MIME type is supported with the given container MIME type in `TranscodingTransformer` and `TransformerBaseRenderer`.
PiperOrigin-RevId: 405645362
This has a few benefits:
* Aligns the Builder constructors with the setters
(setRenderersFactory is missing, but can be easily added in a
follow-up change).
* Allows DefaultMediaSourceFactory to be stripped by R8 and
makes the shrinking dev guide for the cases of providing a custom
MediaSourceFactory or directly instantiating MediaSource instances
less weird too.
#minor-release
PiperOrigin-RevId: 405632981
When no encoder video MIME type is specified, the `TransformerTranscodingVideoRenderer` now uses the video MIME type of the input for the encoder format.
The input format is now read in a new method `ensureInputFormatRead` which is called before the other configuration methods. This removes the logic for reading the input format from `ensureDecoderConfigured`, because it is now needed for both encoder and decoder configuration but the encoder needs to be configured before GL and GL needs to be configured before the decoder, so the decoder can't read the format.
The width and height are now inferred from the input and the frame rate and bit rate are still hard-coded but set by the `MediaCodecAdapterWrapper` instead of `TranscodingTransformer`.
PiperOrigin-RevId: 405631263
The static and dynamic metadata now build up in a list, such that when
the MediaMetadata is built, they are applied in an event order. This
means that newer/fresher values will overwrite older ones. The MediaItem
values are then applied at the end, as they take priority over any other.
#minor-release
PiperOrigin-RevId: 405383177
This introduces a new option `setAudioMimeType` in `TranscodingTransformer.Builder` and a corresponding check whether the selected type is supported. This check is done using `supportsSampleMimeType` which is now part of the `Muxer.Factory` and `MuxerWrapper` rather than `Muxer`.
A new field `audioMimeType` is added to `Transformation` and the `TransformerAudioRenderer` uses this instead of the input MIME type if requested.
PiperOrigin-RevId: 405367817
Decoded video frames can be large and there is no need to retrieve the
corresponding ByteBuffer as we render the decoded frames on a surface
for better performance.
PiperOrigin-RevId: 405364950
The current API exposes an `ImmutableMap` of
`TrackGroup` -> `TrackSelectionOverride`.
This has several disadvantages:
- A difficult to use API for mutation
(`ImmutableMap.Builder` doesn't support key removal).
- There is no track selection specific methods,
how the generic map API mapps to the selection override is not complex
but to obvious for a casual reader.
- The internal data type is exposed, making internal refactor difficult.
This was done to have the API ready as quick as possible.
When transitioning the clients to the map API in <unknown commit>,
it became clear that the map API was too verbose and not mapping
to the clients needs, so utility methods
were added to make operations clearer and more concise.
Nevertheless, having to use utility method to use easily and correctly
an API is not the sign of a good API.
This cl refactors the track selection API for several improvements:
- Add a type `TrackSelectionParameters` that encapsulate the internal
data structure (map currently).
- For iteration, expose as a list.
- Add a `Builder` for easy mutable operations.
- Add track selection specific methods to avoid having utilities functions.
- Those operations are the same as `DefaultTrackSelector.Parameters`
for easier migration. (`setOverride` was renamed to `addOverride`)
- Move `TrackSelection` classes outside of `TrackSelectionParameters`
as their own top level classes.
The migration of the client code is straightforward as most of it
were already using the previously mentioned utility functions
that are now native methods.
The full migration has not been done yet, and is pending on this cl approval.
PiperOrigin-RevId: 405362719
* @Reason is not a TYPE_USE annotation, so should appear before any modifiers and after Javadocs.
(see go/java-style#s4.8.5-annotations)
* Curly braces should be used for inline Javadoc tags: {@code ...}
(see http://go/bugpattern/InvalidInlineTag)
This CL looks good? Just LGTM and Approve it!
This CL doesn’t look good? This is what you can do:
* Revert this CL, by replying "REVERT: <provide reason>"
* File a bug under go/error-prone-bug for category ErrorProneStyle if the change looks generally problematic.
* Revert this CL and not get a CL that cleans up these paths in the future by
replying "BLOCKLIST: <provide reason>". This is not reversible! We recommend to
opt out the respective paths in your CL Robot configuration instead:
go/clrobot-opt-out.
This CL was generated by CL Robot - a tool that cleans up code findings
(go/clrobot). The affected code paths have been enabled for CL Robot in //depot/google3/java/com/google/android/libraries/media/METADATA which is reachable following include_presubmits from //depot/google3/third_party/java_src/android_libs/media/METADATA.
Anything wrong with the signup? File a bug at go/clrobot-bug.
#codehealth
PiperOrigin-RevId: 404769260
The HLS Parser converts from a string decimal duration in seconds into long
microseconds.
Because the conversion passes through a java double type it can result in
representation errors.
For example:
`#EXTINF:4.004` -> `Segment.durationUs` of 4003999
This matters because the first sample (which is the IDR) for a segment will be discarded following a seek because of the logic in the `SampleQueue`:
````java
buffer.timeUs = timesUs[relativeReadIndex];
if (buffer.timeUs < startTimeUs) {
buffer.addFlag(C.BUFFER_FLAG_DECODE_ONLY);
}
````
The HLS implementation of `getAdjustedSeekPositionUs()` now completely supports `SeekParameters.CLOSEST_SYNC`
and it's brotheran, assuming the HLS stream indicates segments all start with
an IDR (that is EXT-X-INDEPENDENT-SEGMENTS is specified).
This fixes issue #2882 and improves (but does not completely solve #8592
Issue: #9392 reports occasional IllegalStateExceptions from release()
in crashlytics,`with no way to reproduce locally. It seems likely there
is a bug somewhere in DRM handling, and ideally we would find that and
fix it.
However we haven't been able to find the problem, and in the meantime
these exceptions cause the entire app to crash. Although this is
arguably useful from a debugging perspective, it's obviously a poor
experience for developers and users, since all we're actually trying to
do is release the session, so maybe we shouldn't strictly care that it's
already released?
This change replaces the exception with an error log, which might be a
useful debugging hint if we see other DRM unexpected behaviour due to
references to released sessions being held for too long.
PiperOrigin-RevId: 403942546
Builder setters should only accept null when the underlying property can
be null. In this case null is directly converted to an empty map/list.
PiperOrigin-RevId: 403406626
The new forcedSessionTrackTypes field was introduced in
<unknown commit>.
These usages are migrated in a follow-up change to add confidence that
the deprecated field continued to work correctly.
PiperOrigin-RevId: 403342893
The previous name is quite easy to misread because it sounds like it
splits up like "(session) for (clear types)" when it's meant to be
"(session for clear) (types)".
The old field is left deprecated for backwards compatibility. The
DrmConfiguration.Builder methods are directly renamed without
deprecation because they're not yet present in a released version of
the library.
PiperOrigin-RevId: 403338799
The possibilities to set a ControlDispatcher have been removed in
<unknown commit> so that the ControlDispatcher is always a
DefaultControlDispatcher.
PiperOrigin-RevId: 403327092
Common houses DataSource as an interface for reading data,
but most of the concrete implementations are in ExoPlayer.
This means that in practice, if an app wants to use a module
that reads using DataSource (e.g. extractor), they may be
forced to depend on ExoPlayer as well to get a concrete
implementation (e.g. FileDataSource). This change moves the
DataSource implementations into common to resolve this.
PiperOrigin-RevId: 403222081
Also fix a typo where windowIndex was being passed to Objects.hashCode
twice.
The old field is left deprecated for backwards compatibility. Usages
will be migrated in an upcoming change.
PiperOrigin-RevId: 403049260
An upcoming change will modify ExoPlayer.Builder#build() to return
ExoPlayer, so any places that explicitly need a SimpleExoPlayer
instance should be using SimpleExoPlayer.Builder.
PiperOrigin-RevId: 403028312
As suggested in parent change, return a list of
`TrackType` instead of appending to it.
This has the slight disadvantage of iterating twice
over the (short) list, but clarifies the code.
PiperOrigin-RevId: 402844458
Update the UI module, the demos and most other users
to make use of the new player TracksInfo and track
selection override APIs.
PiperOrigin-RevId: 402817857
When ExtractorInput.getLength() returns a defined length, the
SubtitleExtractor will create a buffer of the same length, call
ExtractorInput.read() until it has read the input bytes, plus one more
time where ExtractorInput.read() returns RESULT_END_OF_INPUT. The last
call to ExtractorInput.read() however will make the SubtitleExtractor to
increase its buffer (including a copy) unnecessarily.
This change makes the SubtitleExtractor avoid calling
ExtractorInput.read() if the expected number of bytes have already
been read, so that the internal buffer does not grow.
PiperOrigin-RevId: 402583610
Enable subtitle output in the PlaybackOutput and disable the text
renderer in the MkvPlaybackTest. Add WebvttPlaybackTest to test the
output of side-loaded WebVTT subtitles.
PiperOrigin-RevId: 402526588
Before this change, the SubtitleExtractor did not mark the limit
of the input buffer, thus the SubtitleDecoder attempted to decode
more bytes. If the subtitle file had a new line at the end, this
bug would make the SubtitleDecoder append an line break after the
last subtitle.
PiperOrigin-RevId: 402523039
*** Original commit ***
Migrate callers of ExoPlayer.Builder#build() to buildExoPlayer()
An upcoming change will update build() to return Player.
PiperOrigin-RevId: 401468532
In a future change it will be updated to return ExoPlayer
We no longer need separate methods to build Player and ExoPlayer, so
buildExoPlayer will be removed shortly.
PiperOrigin-RevId: 401441016
Temporary file copy of Transformer, which uses
TransformerTranscodingVideoRenderer instead of
TransformerMuxingVideoRenderer to transform files. This allows devs to test
transcoding more easily using the Transformer demo, while external engineers
continue to see the completely working Muxing-based transformer.
In the future, this will replace the Transformer class.
PiperOrigin-RevId: 401020893
*** Original commit ***
Rollback of 8ed6c9fcf5
*** Original commit ***
Fix capitalization of language in track selector
Issue: #9452
***
***
PiperOrigin-RevId: 400942287
TracksInfo is very similar to
`MappingTrackSelector.MappedTracksInfo` with some
fields removed to simplify the Player API,
notably it doesn't expose the renderer concept.
A significant difference is the addition of a `selected` boolean
field which avoids having a separate `getCurrentTrackSelection`
API.
This cl is a part of the bigger track selection change,
splitted for ease of review.
In particular, the MediaSession implementation and UI usage
have been slitted in child cls.
Find all cls with the tag:
#player-track-selection
PiperOrigin-RevId: 400937124
This change makes asynchronous queueing non-experimental, it enables the
feature by default on devices with API level >= 31 (Android 12+) and
exposes APIs for apps to either fully opt-in or opt-out from the
feature.
The choice to use or not asynchronous queueing is moved out of
MediaCodecRenderer to a new MediaCodecAdapter factory, the
DefaultMediaCodecAdapterFactory. This is because, at the moment,
if an app passes a custom adapter factory to a MediaCodecRenderer and
then enables asynchronous queueing on it, the custom
adapter factory is not used but this is not visible to the user.
The default behavior of DefaultMediaCodecAdapterFactory is to create
asynchronous MediaCodec adapters for devices with API level >= 31
(Android 12+), and synchronous MediaCodec adapters on devices with older
API versions.
DefaultMediaCodecAdapterFactory exposes methods to force enable or force
disable the use of asynchronous adapters so that applications can enable
asynchronous queueing on devices with API versions before 31 (but not
before 23), or fully disable the feature. For applications that build
MediaCodecRenderers directly, they will need to create a
DefaultMediaCodecAdapterFactory and pass it to the renderer constructor.
For applications that rely on the DefaultRenderersFactory, additional
methods have been added on the DefaultRenderersFactory to control
enabling/disabling asynchronous queueing.
Issue: #6348
PiperOrigin-RevId: 400733506
