* Introduced `useHdr` for `GlEffect#toGlTextureProcessor`, so
`TextureProcessor` implementations can decide how to handle HDR.
* Creating FP16 color textures for HDR input.
Tested via manual testing, adding a no-op GlEffectWrapper to the transformation to
force use of intermediate textures, adding a linear ramp to the fragment shader,
and trying to ascertain that there's a real reduction in posterization when
switching from 4-bit to 8-bit unsigned bytes, and again from 8-bit unsigned bytes
to 16-bit floating point.
PiperOrigin-RevId: 461613117
(cherry picked from commit f67c1a73f4)
This may happen when a containers' color transfer incorrectly does not match
the video's color transfer.
An example of a file with such a mismatch is the current Transformer demo HDR10
sample file.
Manually tested by confirming that no errors are emitted for SDR and HLG sample
files, and that errors are emitted for our incorrect HDR10 sample file.
PiperOrigin-RevId: 461583532
(cherry picked from commit fd046bd2f6)
- Update profile selection logic to pick an HDR-compatible profile when doing HDR editing on H.264/AVC videos.
- Handle doing the capabilities check for all MIME types that support HDR (not just H.265/HEVC).
- Fix a bug where we would pass an HDR input color format to the encoder when using tone-mapping.
- Tweak how `EncoderWrapper` works so decisions at made at construction time.
Manually tested cases:
- Transformation of an SDR video.
- Transformation of an HDR video to AVC (which triggers fallback/tone-mapping on a device that doesn't support HDR editing for AVC).
- Transformation of an HDR video with HDR editing.
PiperOrigin-RevId: 461572973
(cherry picked from commit 0db07c6791)
Also remove VideoEncoderSettings.colorProfile as there are no
concrete use cases for customizing this and it clashes with picking
the color format automatically based on SDR vs. HDR.
PiperOrigin-RevId: 460746987
(cherry picked from commit db78a87fdc)
We currently start a simple Thread to release AudioTracks
asynchronously. If many AudioTracks are released at the same
time, this may lead to OOM situations because we attempt to
create multiple new threads.
This can be improved by using a shared SingleThreadExecutor.
In the simple case of one simmultaneous release, it's exactly
the same behavior as before: create a thread and release it
as soon as it's done. For multiple simultanous releases we
get the advantage of sharing a single thread to avoid creating
more than one at the same time.
Issue: google/ExoPlayer#10057
PiperOrigin-RevId: 460698942
(cherry picked from commit 1e8d163107)
`HevcConfig.parse` misreads reserved bit to determine NAL unit type. This is currently meant to be always set to 0, but could be given some kind of meaning in a future revision.
Issue: google/ExoPlayer#10366
PiperOrigin-RevId: 460487613
(cherry picked from commit 0f665fbeb8)
Some calls to handleBuffer return false while a previous
flush is still handled in the background.
Fix this by either asserting the method returns true if
we don't expect any delay, or calling it repeatedly until
it returns true (within a timeout).
PiperOrigin-RevId: 460474419
(cherry picked from commit 4289708323)
Pass the color info and HDR static metadata when configuring the decoder
using MediaFormatUtil.maybeSetColorInfo.
PiperOrigin-RevId: 460424985
(cherry picked from commit 46cd92ac1d)
This extension is needed for editing HDR input with OpenGL, as the
ExternalTextureProcessor samples raw YUV values from the
external texture for HDR and converts them to RGB itself rather than
relying on the OpenGL driver to do this automatically as for SDR.
PiperOrigin-RevId: 460424154
(cherry picked from commit 413016f7ef)
This saves an intermediate texture copy step for use-cases
where matrix transformations are the first or only effects
in the chain.
PiperOrigin-RevId: 460239403
(cherry picked from commit 0615922cb6)
- Added setter to disable this feature.
- Added accompanying tests.
- Plan to run tests on the same set of settings on H265.
PiperOrigin-RevId: 460238673
(cherry picked from commit 6e12600099)
The SDR constant also specified a color space and range, in addition to
C.COLOR_TRANSFER_SDR. However, it turns out that SDR videos may use different color
space and range values, so following prior ExoPlayer conventions to have `null`
mean "generic SDR" is preferable here.
PiperOrigin-RevId: 459296746
(cherry picked from commit 4230ea94e9)
If the input is HDR (HLG), check encoder capabilities for HDR support
and request tone-mapping to SDR during decoder configuration otherwise.
Capabilities are only checked for API 31 and above, as HDR editing is
not supported before.
As the encoder capabilities check needs to happen before selecting the
encoder to use (as this may depend on the resolution output by the
effects chain), the EncoderWrapper checks all candidate encoders
for the MIME type for HDR capabilities and only requests fallback to
SDR if none of them support it.
When the actual encoder is selected, the wrapper checks that it matches
one of the encoders is checked capabilities for.
PiperOrigin-RevId: 458511599
(cherry picked from commit 474e2f2c77)
Configure the GL shaders and encoder to take in HDR metadata.
This mostly just consists of passing the Format.colorInfo through
the VideoTranscodingSamplePipeline down to the encoder, rather than passing
the PQ-ness down to the GL step.
Due to b/237674316, this will remove HDR10+ support temporarily to introduce
support for HLG10.
Manually tested to confirm that HLG10 operations that don't affect color display
correctly after this CL with "HDR editing" in the demo checked, and continue to display incorrectly (as before this CL) without the option unchecked.
PiperOrigin-RevId: 458490810
(cherry picked from commit 53db305612)
We used "ALL_COOECS" previously, and it is not necessary because "ALL_CODECS"
additionally the codecs that support tunneling/secure decoding, which there
is no use case in Transformer.
PiperOrigin-RevId: 458470278
(cherry picked from commit 8f89eba59c)
Although MediaCodec claims supporting float frame rate, encoder init failed on
API21 Nexus 5. Since it's just a performance hint to the codec, it's OK to
generalize it to other API versions.
PiperOrigin-RevId: 458434650
(cherry picked from commit 821ab26702)
The GlEffectsFrameProcessor that will be part of the effects module
uses the DebugViewProvider. So it does not make sense for it
to be an inner interface of Transformer.
PiperOrigin-RevId: 458014932
(cherry picked from commit f41e6796ce)
The FinalMatrixTransformationProcessorWrapper ensures that the
surface is only replaced when it is not being rendered to and vice
versa.
PiperOrigin-RevId: 458007639
(cherry picked from commit e5527a8add)
The outputHeight in the TransformationRequest is the height of
the frame as it would be displayed (i.e., after applying any
rotation specified in the format). So pass-through should only
be used if the requested outputHeight matches the input
format's height after applying the rotation.
PiperOrigin-RevId: 457934867
(cherry picked from commit c74cf1f1aa)
`MetadataRenderer` is updated to output `Metadata` with its presentation time, in microseconds.
PiperOrigin-RevId: 457444718
(cherry picked from commit 6e9275c13d)
This will be useful for downgrading to a lower resolution during
a slow preview and for processing slide-shows once sequential
multi-asset editing is supported.
PiperOrigin-RevId: 457017255
(cherry picked from commit a230d59f1b)
videoEncoderFormatUnsupported_completesWithError() has recently
been flaky on API 31 emulators on presubmit because a different
exception than the expected exception is thrown.
This disables it on those emulators to reduce testing noise
until the underlying problem is investigated and resolved.
PiperOrigin-RevId: 456765512
(cherry picked from commit d9c63c1e87)
This change is just renaming. There is no functional change intended.
The FrameProcessor interface will be created in a follow-up.
PiperOrigin-RevId: 456741628
(cherry picked from commit 216fefd669)
`TextRenderer` is updated to output `CueGroup`, which contains the presentation time of the cues, in microseconds.
PiperOrigin-RevId: 456531399
(cherry picked from commit bf11a8a831)
After this change GlEffects can use any GlTextureProcessor not just
SingleFrameGlTextureProcessor.
MediaPipeProcessor now implements GlTextureProcessor directly which
allows it to reuse MediaPipe's output texture for its output texture
and avoids an extra copy shader step.
PiperOrigin-RevId: 456530718
(cherry picked from commit e25bf81195)
After this change, FrameProcessorChain chains any GlTextureProcessors
instead of only SingleFrameGlTextureProcessors.
The GlTextureProcessors are chained in a bidirectional manner using
ChainingGlTextureProcessorListener to feed input and output related
events forward and release events backwards.
PiperOrigin-RevId: 456478414
(cherry picked from commit 3a96691654)
In follow-ups the FrameProcessorChain will set an instance of this
listener for each GlTextureProcessor to chain it with its previous
and next GlTextureProcesssor.
PiperOrigin-RevId: 455628942
(cherry picked from commit c92e18ec58)
The wrapper
* catches exceptions for each task and notifies the
listener (this will be used more in follow-ups when processFrame
is split into lots of listeners and callbacks),
* removes finished tasks from the queue and signals any exceptions
that occurred to the listener each time a new task is executed.
PiperOrigin-RevId: 455345184
(cherry picked from commit bf9fa12411)
This change adds a SurfaceProvider interface which is necessary to
allow for texture processors whose output size becomes available
asynchronously in follow-ups.
VTSP's implementation of this interface wraps the encoder and provides
its input surface together with the output frame width, height, and
orientation as used for encoder configuration.
The FrameProcessorChain converts the output frames to the provided
orientation and resolution using a ScaleToFitTransformation and
Presentation replacing EncoderCompatibilityTransformation.
PiperOrigin-RevId: 455112598
(cherry picked from commit d20f684986)
to support enabling this as a compile-time error.
More information: go/lsc-extends-object
PiperOrigin-RevId: 454582570
(cherry picked from commit d2f5f212b9)
The native GL errors are in an arbitrary order according to
https://www.khronos.org/registry/OpenGL-Refpages/es2.0/xhtml/glGetError.xml
This means any of them could contain the most useful message
so it's better to use all for the GlException's message rather than
picking the last message and logging all others.
PiperOrigin-RevId: 454130460
(cherry picked from commit 92c2a304e1)
Based on
https://developer.android.com/reference/android/media/MediaCodec#using-an-output-surface,
frame dropping behaviour depends on the target SDK version.
After this change transformer will only use
MediaFormat#KEY_ALLOW_FRAME_DROP if both the target and system SDK
version are at least 29 and default to its pre 29 behaviour where each
decoder output frame must be processed before a new one is rendered
to prevent frame dropping otherwise.
Also remove deprecated Transformer.Builder constructor without a
context and the context setter.
PiperOrigin-RevId: 453971097
(cherry picked from commit 3f718b0d10)
Transformer always enabled glAssertionsEnabled, so there should
be no functional change.
ExoPlayer previously disabled glAssertionsEnabled, so GlUtil logged
GlExceptions instead of throwing them. The GlExceptions are now
caught and logged by the callers so that there should also be no
functional change overall.
This change also replaces EGLSurfaceTexture#GlException with
GlUtil#GlException.
PiperOrigin-RevId: 453963741
(cherry picked from commit dc668f2b59)
This removes the prior restriction of needing to remember not to crop and set aspect ratio in the same Presentation.Builder, and makes each class a bit more targeted.
This is partially made feasible by the past work to merge consecutive
MatrixTransformations into a single MatrixTransformationFrameProcessor, which
ensures that there's no loss in quality between successive MatrixTransformations.
PiperOrigin-RevId: 453660582
(cherry picked from commit b33dc5e57b)
SingleFrameGlTextureProcessor is now an abstract class containing a
default implementation of the more flexible GlTextureProcessor interface
while still exposing the same simple abstract methods for single frame
processing it previously did.
FrameProcessorChain and GlEffect will be changed to use
GlTextureProcessor in follow-ups.
PiperOrigin-RevId: 453633000
(cherry picked from commit 0b37d860d1)
Implementations of this interface will be able to drop or add frames,
change timestamps, accept multiple input frames before producing
output, and process frames on their own background thread.
A default implementation of this interface will be added to SingleFrameGlTextureProcessor in a follow-up.
PiperOrigin-RevId: 453159835
(cherry picked from commit 023d19c8c5)
This internal listener avoids wrapping the TransformationExceptions
in PlaybackExceptions that are handled via the Player.Listener and
is also used for FrameProcessingExceptions which already avoided
the PlaybackException layer previously.
This listener will also be useful in follow-ups for encoder-related
TransformationExceptions that are thrown in the SurfaceProvider that
will be called on the GL thread.
PiperOrigin-RevId: 452074575
(cherry picked from commit 960422e36f)
Once the more advanced GlTextureProcessor interface exists,
it will be possible to change the output size of a GlTextureProcessor
between frames. To keep the re-configuration based on the frame sizes
minimal, things indepedent of the frame size, such as the GlProgram,
can be initialized in the constructor.
PiperOrigin-RevId: 451997584
(cherry picked from commit 54d44d38b6)
The call doesn't currently reset the already loaded suppliers and
factories. Also fix the supplier loading code to use a local copy
of the current dataSourceFactory to avoid leaking an updated
instance to a later invocation.
Issue: androidx/media#116
#minor-release
PiperOrigin-RevId: 460721541
(cherry picked from commit 6be0d6ea25)
I don't think it's useful to keep these in numerical order, it makes
more sense to keep them grouped into a 'logical' ordering.
#minor-release
PiperOrigin-RevId: 460453464
(cherry picked from commit 1b1e0f890d)
We wait until a previous AudioTrack has been released before
creating a new one. This is currently done with a thread
block operation, which may cause ANRs in the extreme case
when someone attempts to release the player while this is
still blocked.
The problem can be avoided by just returning false from
DefaultAudioSink.handleBuffer to try again until the previous
AudioTrack is released.
Reproduction steps to force the issue:
1. Add Thread.sleep(10000); to the AudioTrack release thread.
2. Add this to the demo app:
private int positionMs = 0;
Handler handler = new Handler();
handler.post(new Runnable() {
@Override
public void run() {
player.seekTo(positionMs++);
if (positionMs == 10) {
player.release();
} else {
handler.postDelayed(this, 1000);
}
}
3. Observe Player release timeout exception.
These steps can't be easily captured in a unit test as we can't
artifically delay the AudioTrack release from the test.
Issue: google/ExoPlayer#10057
PiperOrigin-RevId: 459468912
(cherry picked from commit a80dd60449)
ProgressiveMediaPeriod loads all available tracks into SampleStreams
(because it needs to read the data anyway and it allows easy activation
of tracks without reloading). However, the SampleStreams for disabled
tracks are not read and no one if waiting for them.
The buffered position is used for user-visible state (e.g. in the UI)
and to check how much data is already buffered to decide when to stop
buffering (using LoadControl). Both values benefit from only
using the actually enabled tracks to better reflect what is available
for playback at the moment.
Issue:Issue: google/ExoPlayer#10361
PiperOrigin-RevId: 458475038
(cherry picked from commit 577e19168d)
As per MP4 spec, bitrates in esds boxes can be a 32 bit number which doesn't fits in Java int type, so now reading it as a long value. Our class for holding media format, only allows bitrates value to be an int as we don't expect the bitrates to be greater than or equal to 2^31. So we're limiting the values for bitrates to Integer.MAX_VALUE.
#minor-release
PiperOrigin-RevId: 458423162
(cherry picked from commit 9e10286b5e)
Previously two timelines that differed only in shuffle order were
considered equal, which resulted in no call to
Player.Listener.onTimelineChanged when calling
ExoPlayer.setShuffleOrder. This in turn resulted in no call to
MediaControllerCompat.Callback.onQueueChanged.
Also make a small fix inside ExoPlayerImpl.setShuffleOrder, to ensure
that the new shuffle order is used when constructing the masked
timeline.
Issue: google/ExoPlayer#9889
#minor-release
PiperOrigin-RevId: 457703727
(cherry picked from commit 5c7ec13e85)
NoUidTimeline still exists as a private detail of TestUtil, but it no
longer extends ForwardingTimeline because the interactions are quite
hard to reason about.
#minor-release
PiperOrigin-RevId: 457703593
(cherry picked from commit 292f6de630)
1. The offloadSchedulingEnabled value doesn't need to be in
PlaybackInfo because it's never updated in EPII.
2. The sleepingForOffload value in EPII wasn't updated explicitly
(just via the return value of a method). It was also only
meant to be enabled while the player is actively playing, but
confusingly triggered from a path where the player may
theoretically be buffering as well.
3. The offload sleeping (=not scheduling doSomeWork) was interwoven
into the actual scheduling code making it slightly hard to follow.
This can be improved slightly by keeping the offload sleeping
decision and the scheduling separate.
PiperOrigin-RevId: 457427293
(cherry picked from commit aedde2de39)
The offload sleeping stops as soon as a new DO_SOME_WORK message
is handled (because this indicates an expected change in rendering
and we want to stop sleeping until we know it's safe to do so).
Every exit path from doSomeWork needs to clear other pending
DO_SOME_WORK messages as these requests have already been handled by
the current method invocation. This currently doesn't happen from the
offload sleeping return path and a previously queued DO_SOME_WORK
message can immediately wake up the rendering loop again.
Fix this by moving the message removal to the beginning of the
doSomeWork method (as it prevents forgetting it in one of the
exit paths later).
PiperOrigin-RevId: 456259715
(cherry picked from commit 251389d744)
The API 32 SDK has incorrect versioning metadata for Spatializer. It
reports the whole class has only been present since API 33 (which is
surely impossible given it's present in the API 32 SDK):
https://issuetracker.google.com/234009300
The metadata seems to be correct in the API 33 SDK, so this baseline
will no longer be needed when we bump to `compileSdkVersion = 33`.
Fixing lint errors in the string.xml files makes no sense because these are
overridden with the next automated string import. Adding a lint-baseline.xml
instead for the ui module.
See https://issuetracker.google.com/208178382
#minor-release
PiperOrigin-RevId: 455354304
(cherry picked from commit 495480a66a)
