20166 Commits

Author SHA1 Message Date
tonihei
c79ac5ba21 Remove invalid command comparison
The removed check searched for a player command inside a list of
session commands, which is not allowed by the IntDef definition
and only worked because both types map to a Java int.

PiperOrigin-RevId: 612758442
2024-03-05 02:20:18 -08:00
Copybara-Service
bbdaf2b092 Merge pull request #1025 from v-novaltd:dsparano-exo209
PiperOrigin-RevId: 612485043
2024-03-04 09:55:00 -08:00
claincly
305f24f947 Clarify ExoPlayer.setVideoEffects() javadoc
PiperOrigin-RevId: 612464321
2024-03-04 08:45:38 -08:00
Daniele Sparano
0cdae9af20 Read resolution for debug text view from VideoSize object to reflect changes from video effects 2024-03-04 14:09:54 +00:00
tofunmi
bcfad4b3b4 Move hdrMode from defaultAssetLoaderFactory constructors to create()
Plumbing hdrMode through the default asset loader factory via the constructor is problematic because it breaks API boundaries. It means there is another way to set hdrMode outside of Composition.java and TransformationRequest.java, which is error prone and cause problems if someone an app starts customizing the assetloaderfactory. It also means custom asset loaders can't receive this information without hacking around.

The introduction of the composition-level settings class makes this approach easily extensible for other settings applied on the composition level but use in an individual asset level basis (e.g. ultraHDR support).

PiperOrigin-RevId: 611466920
2024-02-29 07:41:18 -08:00
ibaker
7395c9a159 Remove deprecated CronetDataSourceFactory
PiperOrigin-RevId: 611431225
2024-02-29 04:59:35 -08:00
christosts
1355f4734d Composition preview tests with surfaces
Add abilitiy to use real surfaces in instrumentation tests
using the ActivityScenarioRule and an activity class for testing
purposes.

PiperOrigin-RevId: 611421490
2024-02-29 04:19:44 -08:00
sheenachhabra
bba45b8b35 Rename DefaultMp4Writer to BasicMp4Writer
PiperOrigin-RevId: 611421036
2024-02-29 04:16:15 -08:00
tianyifeng
1bdc58de0b Avoid position jumping back when controller replaces the current item
When the controller replaces the current item, the masking position will be changed to the default position of the new item for a short while, before the correct position comes from the session. This will interrupt the current position fetched from the controller when the playback doesn't interrupted by the item replacing.

Issue: androidx/media#951

#minor-release

PiperOrigin-RevId: 611417539
2024-02-29 04:07:17 -08:00
claincly
27d78e69d7 Clarify setVideoEffect() javadoc
PiperOrigin-RevId: 611147963
2024-02-28 10:31:42 -08:00
Googler
87c4d60c37 Allow CronetDataSource's read buffer size to be configured.
PiperOrigin-RevId: 611146397
2024-02-28 10:27:59 -08:00
sheenachhabra
6809c0a642 Change Mp4Writer to an interface
The Mp4Writer does not have any default implementation,
hence it can be an interface rather than an abstract class.

PiperOrigin-RevId: 611110415
2024-02-28 08:28:59 -08:00
Googler
08993b6fb1 Have VideoSampleExporter output orientation match input
This should enable trim optimization to work correctly in more cases.

PiperOrigin-RevId: 611096958
2024-02-28 07:40:53 -08:00
christosts
55bfe4f95c Composition preview: do not report errors on release
PiperOrigin-RevId: 611096895
2024-02-28 07:37:15 -08:00
ibaker
09ac916119 Remove DefaultTrackSelector.NO_ORDER ordering
This implementation generates lint errors because neither the `first`
nor `second` parameters are used, and that's generally
unexpected/incorrect for a `Comparator` implementation since it should
always consider both its parameters.

PiperOrigin-RevId: 611039632
2024-02-28 02:58:41 -08:00
ibaker
19cd156a8c Add Mp4ExtractorTest case for pixel-motion-photo-2-hevc-tracks.mp4
A follow-up change will add the frame rate to the single-frame track.

Issue: androidx/media#1051
PiperOrigin-RevId: 611018319
2024-02-28 01:23:01 -08:00
ibaker
c7e00b12b4 Add fps-awareness to DefaultTrackSelector
This change aims to prioritise tracks that have a 'smooth enough for
video' frame rate, without always selecting the track with the highest
frame rate.

In particular MP4 files extracted from motion photos sometimes have two
HEVC tracks, with the higher-res one having a very low frame rate (not
intended for use in video playback). Before this change
`DefaultTrackSelector` would pick the low-fps, high-res track.

This change adds a somewhat arbitrary 10fps threshold for "smooth video
playback", meaning any tracks above this threshold are selected in
preference to tracks below it. Within the tracks above the threshold
other attributes are used to select the preferred track. We deliberately
don't pick the highest-fps track (over pixel count and bitrate), because
most users would prefer to see a 30fps 4k track over a 60fps 720p track.

This change also includes a test MP4 file, extracted from the existing
`jpeg/pixel-motion-photo-2-hevc-tracks.jpg` file by logging
`mp4StartPosition` in
[`MotionPhotoDescription.getMotionPhotoMetadata`](b930b40a16/libraries/extractor/src/main/java/androidx/media3/extractor/jpeg/MotionPhotoDescription.java (L123))
and then using `dd`:

```
mp4StartPosition=2603594

$ dd if=jpeg/pixel-motion-photo-2-hevc-tracks.jpg \
    of=mp4/pixel-motion-photo-2-hevc-tracks.mp4 \
    bs=1 \
    skip=2603594
```

----

This solution is in addition to the `JpegMotionPhotoExtractor` change
made specifically for these two-track motion photos in
5266c71b3a.
We will keep both changes, even though that change is not strictly
needed after this one, because adding the role flags helps to
communicate more clearly the intended usage of these tracks. This
change to consider FPS seems like a generally useful improvement to
`DefaultTrackSelector`, since it seems unlikely we would prefer a 5fps
video track over a 30fps one.

Issue: androidx/media#1051
PiperOrigin-RevId: 611015459
2024-02-28 01:09:50 -08:00
claincly
626a8adfd8 Test effect-enabled playback
This new test is for `ExoPlayer.setVideoEffects()`. It plays the
one-second-long video, applies an overlay that prints the video frame timestamp
onto the frame, captures the output frame and compares the captured output
frame with golden.

PiperOrigin-RevId: 610781590
2024-02-27 09:44:24 -08:00
Googler
6f28eeff31 Pick max H.264 level when trim optimizing
Instead of always starting with the transcoded H.264 level, take the maximum
from transcoded and transmuxed levels

PiperOrigin-RevId: 610759438
2024-02-27 08:27:38 -08:00
Googler
c3aec4f19d Comments for AudioGraph constructor
Add /*effects=*/ at callsites of AudioGraph constructor.

PiperOrigin-RevId: 610741692
2024-02-27 07:12:53 -08:00
sheenachhabra
94e0a27a81 Add unified addMetadata() method in Mp4Muxer
PiperOrigin-RevId: 610710011
2024-02-27 04:34:45 -08:00
Googler
0e0e1c4f1a Tests for composition audio effects
Add tests for the changes to AudioGraph that support composition audio effects

PiperOrigin-RevId: 610691768
2024-02-27 03:04:20 -08:00
Googler
477ace1be9 Composition audio effects
Implement composition-level audio effects in AudioGraph.

PiperOrigin-RevId: 610689632
2024-02-27 02:54:08 -08:00
tofunmi
0480bc31a8 Rename media/bitmap out of media and rename
PiperOrigin-RevId: 610418418
2024-02-26 08:40:19 -08:00
sheenachhabra
63a8fff69e Use ByteBuffer.remaining() instead of ByteBuffer.limit() in BoxUtil
PiperOrigin-RevId: 610414628
2024-02-26 08:24:35 -08:00
huangdarwin
e98858424a Test: Expand GL tone-map pixel difference threshold.
PiperOrigin-RevId: 610374358
2024-02-26 05:13:17 -08:00
michaelkatz
a45e734bb0 Create audio offload failure recovery playback tests
These audio offload failure recovery tests model the DefaultAudioSink failing at audio track init and write operations in offload mode. Playback should recover and try again as DefaultAudioSink will disable offload mode.

PiperOrigin-RevId: 610372935
2024-02-26 05:06:04 -08:00
sheenachhabra
a5f3db4cfe Fix java doc for setAudioProcessors and setVideoEffects method
PiperOrigin-RevId: 609778651
2024-02-23 11:05:12 -08:00
sheenachhabra
9be7a7ab7f Update media3 version for 1.3.0 stable release
PiperOrigin-RevId: 609745599
2024-02-23 09:10:23 -08:00
tofunmi
5a892509f7 remove media/bitmap/input_images
move the images into the respective places in the file extension directory so they file the pre-existing structure

PiperOrigin-RevId: 609744673
2024-02-23 09:07:03 -08:00
michaelkatz
23a301fc5d Fallback to legacy sizerate check if CDD H264 PerfPoint check fails
Some devices supporting Performance Points for decoder coverage are missing coverage over the CDD requirements for H264. For these cases ExoPlayer should fall back to legacy resolution and frame rate support checks. If there is a stream evaluated as a PerformancePointCoverageResult of COVERAGE_RESULT_NO, then ExoPlayer checks for coverage of the 720p H264 CDD requirement.

Issue: google/ExoPlayer#10898

Issue: androidx/media#693

Issue: androidx/media#966
PiperOrigin-RevId: 609740128
2024-02-23 08:48:09 -08:00
sheenachhabra
284d17cb13 Add release reason argument in the MuxerWrapper.release() method
Also called `muxer.release()` in other tests where it is
appropriate.

PiperOrigin-RevId: 609736956
2024-02-23 08:34:04 -08:00
claincly
68a78b9218 When effects are enabled, opportunistally render to VFP
Even when there's no display surface, MCVR can render frames to VFP, becuase by
the time the frame is processed:

- If there's still no surface, VFP will drop the frame;
- If there's surface, the processed frame would be rendered.

In short, placeholder surface is not needed in effect enabled playback. FWIW,
it is used to swallow frames directly from MediaCodec when there's no output
surface.

PiperOrigin-RevId: 609705222
2024-02-23 06:07:54 -08:00
christosts
0480eff6a1 ExoPlayerImplInternal.releaseInternal(): unblock the app thread
This change makes ExoPlayerImplInternal.releaseInternal() unblock the
app thread if a runtime exception is thrown while releasing components
from the playback thread.

Before this change, if a runtime exception occurred during releasing
components in the playback thread, ExoPlayer.release() would wait for
`releaseTimeoutMs` and then raise a player error. With this change,
the player error is reported only when the playback thread is blocked
but if there is a runtime exception, the application thread is
unblocked.

The impact of this change is potentially fewer ANRs on
ExoPlayer.release() at the expense of less error reporting.

PiperOrigin-RevId: 609702549
2024-02-23 05:54:37 -08:00
huangdarwin
601d6ed587 Update GL thread name, per package name
PiperOrigin-RevId: 609686878
2024-02-23 04:29:31 -08:00
andrewlewis
88c6877418 Slow down dizzy crop demo effect
PiperOrigin-RevId: 609667000
2024-02-23 02:54:43 -08:00
huangdarwin
e4cf72db11 Test: Use TestName testId in TransformerSequenceEffectTest.
Some testcases that got through 0d1b35477d as they were submitted as that CL was in-progress.

PiperOrigin-RevId: 609659607
2024-02-23 02:17:59 -08:00
sheenachhabra
fd8f45b38e Add MdtaMetadataEntry constructor with default locale indicator
PiperOrigin-RevId: 609427529
2024-02-22 10:55:50 -08:00
sheenachhabra
82065e699a Update release notes section for 1.3.0 stable release
PiperOrigin-RevId: 609407248
2024-02-22 10:01:30 -08:00
christosts
440d2ab162 ForwardingAudioSink: add override for release()
We forgot to add it when we added AudioSink.release(). The commit
includes a test that ensures ForwardingAudioSink overrides all the
methods defined in the AudioSink interface.

PiperOrigin-RevId: 609402258
2024-02-22 09:59:15 -08:00
michaelkatz
4192924622 Use playing period TrackSelectorResult in track reselection update
If the reading period has already advanced and a track reselection procs that only affects the reading period media, then ExoPlayer may try and apply the reading period's track selection incorrectly unto the playing period. ExoPlayer should apply the playing period's track selection to the playing period instead.

PiperOrigin-RevId: 609375077
2024-02-22 08:08:18 -08:00
huangdarwin
0d1b35477d Test: Use testId throughout media3 tests.
No more instances of this codesearch query after this CL in non-parameterized
tests:
`String\ testId\ \=\ \" f:media3`
PiperOrigin-RevId: 609364413
2024-02-22 07:28:55 -08:00
ibaker
d1ae9ffc52 Add more details about why Extractor.sniff returned false
PiperOrigin-RevId: 609335656
2024-02-22 05:20:16 -08:00
sheenachhabra
41886434ad Remove duplicate Mp4Location class
PiperOrigin-RevId: 609313551
2024-02-22 03:40:35 -08:00
michaelkatz
9046f2edb6 Allow renderer retry for audio track offload initialization failure
If render error occurs due to AudioTrack initialization failure in offload mode, then ExoPlayer should allow retry as subsequent attempt will be with DefaultAudioSink disabling offload.

PiperOrigin-RevId: 609304897
2024-02-22 03:00:10 -08:00
tianyifeng
d952a06214 Fix a bug in retaining streams when preload a PreloadMediaPeriod again
The `PreloadMediaPeriod.selectTracksForPreloading` can be called for multiple times at the preloading stage (before the period is being played). For example, when the period resumes preloading. This change fix the assertion failure in `ProgressiveMediaPeriod.selectTracks` caused by the wrong implementation of `PreloadMediaPeriod.selectTracksForPreloading` when it is trying to retain the previously preloaded streams.

Also the `TrackSelectorResult` parameter is changed to a list of `ExoTrackSelection`. We should compare the selections only rather than considering the `RendererConfiguration` in the `TrackSelectorResult` to decide whether to retain the streams, as for preloading case the renderers haven't consumed the samples yet.

PiperOrigin-RevId: 609126868
2024-02-21 14:41:33 -08:00
huangdarwin
fde6a32156 Transformer: Input decoder output color to VideoSampleExporter.
Previously, the track format was used in VideoSampleExporter. Now, we use a
simulated decoder output format.

As the last expected change for this bug, also adds release notes

PiperOrigin-RevId: 609080629
2024-02-21 12:15:11 -08:00
ibaker
bb5c688543 Update TestPlayerRunHelper to fail on non-fatal errors by default
Also introduce a fluent API that allows callers to ignore non-fatal
errors (while avoiding adding boolean overloads for every method).

**Most** tests want to fail on non-fatal errors (since they likely
indicate user-visible issues like codec errors etc), only tests
explicitly testing fallback in error scenarios should want to ignore
them.

Before this change there were a few `playUntilXXX` methods. These can
now all be triggered via `play(player).untilXXX`, which means
effectively every 'until' condition is available in a 'play until'
variant that calls `play` just before waiting for the condition.

PiperOrigin-RevId: 608988234
2024-02-21 07:39:07 -08:00
huangdarwin
2e9cc2784f Test: Add sequence effect test with bt601 image and bt709 video.
The lack of this test was the *real* root cause for our partner's regression :P

PiperOrigin-RevId: 608956530
2024-02-21 05:22:11 -08:00
Googler
dca7bae416 Allow mismatching H.264/AVC level for trimming or resuming
MediaCodec docs already allude to potentially mismatching H.264 level
between container and bitstream. Relax the initialization data check to
reflect this.

PiperOrigin-RevId: 608942322
2024-02-21 04:20:50 -08:00