Clearing the exception puts the SimpleDecoder into a silent failure
state - the decoder thread is dead (because decode() has returned
false) but it's still possible to queue buffers to the decoder (they
just never get decoded). This partially reverts
4107375c9d
Also always recreate the decoder when handling an error in TextRenderer
This ensures we can try and decode a later subtitle sample after
encountering a decode error. This behaviour is what nulling out the
exception in SimpleDecoder.flush() was trying to achieve. We need to
ensure we don't start passing data to the new decoder until we've
hit the next key frame, so we throw away any non-keyframe samples
inside TextRenderer#render().
Issue: #7590
PiperOrigin-RevId: 319785908
We currently try to keep track of the playback queue (MediaPeriodQueue)
by listening to onMediaPeriodCreated/onMediaPeriodReleased events.
This approach has some problems:
1. It's easily broken by custom MediaSources that don't report these
events correctly.
2. We need to make some assumptions about what the order of these
events actually means. For example it is currently important that
the playing period gets released last in MediaPeriodQueue.clear()
3. We don't see batched events (like MediaPeriodQueue.clear()), so that
it is impossible to keep the "last reading period" for example. This
information is needed to correctly associate renderer errors to
periods after the queue has been cleared.
All of these problems can be solved by directly tracking the queue.
This also makes the onMediaPeriodCreated/Released/ReadingStarted events
obsolete and they can be removed in a future change.
PiperOrigin-RevId: 319739993
The current code only works if the tts:ruby attributes are defined
directly on the in-line elements. This changes that so we also consider
tts:ruby attributes on `style` nodes referenced by ID.
PiperOrigin-RevId: 319515177
We don't need the renderer immediately after stopping, so the
renderer should not throw a checked exception until it's used again.
This is inline with the not throwing from disable().
Also, none of the known implementation throw an exception at the moment
and all reasonable base classes omit the throws clause already.
PiperOrigin-RevId: 319503643
They are all for Context.getSystemService that is allowed to return
null. In most cases where we need to service, we make an assertion that
it is available.
PiperOrigin-RevId: 319503557
The test was incorrectly assuming that with LooperMode.PAUSE,
HandlerThread instances needed explicit calls to execute
tasks. This commit fixes the test flakiness by manually pausing the
HandlerThead when needed.
PiperOrigin-RevId: 319411552
Without checking if getFramesPerEncodedSample fails,
the frame count becomes negative which leads to
hard to debug errors.
PiperOrigin-RevId: 319247618
Passthrough mode can use a codec or not, but
the code only mentioned "passthrough" in most cases,
making the specific mode confusing.
For example both `MediaCodecRenderer` and
it's derived class `MediaCodecAudioRenderer`
had a private `passthroughEnabled` field,
but they were used for the opposite modes!
This change renames all relevant variables/functions
to explicit `CodecPassthrough` or `Bypass`.
PiperOrigin-RevId: 319225235
The system services may return a null value if the service is
not available. Guard against this by falling back to default values.
PiperOrigin-RevId: 319187882
Issue: #7011
Issue: #6725
Issue: #7066
This also mitigates (but doesn't fix) Issue: #4133 because it
prevents a second key load after a short clear section.
PiperOrigin-RevId: 319184325
SampleStreams should always provide samples starting from a keyframe.
We do not have equivalent logic in any of our extension decoder renderers.
PiperOrigin-RevId: 319012365
The list of MediaSourceHolder in ExoPlayerImpl is only maintained to be able to create a PlaylistTimeline for masking. By keeping only the id and a snapshot of the timeline of the MediaSourceHolder in ExoPlayerImpl, parallel access is prevented and we still have sufficient information to create the masking timeline.
PiperOrigin-RevId: 319003837
We can't restore the previous state of the remaining chunk, so we can't
support discarding from spliced-in chunks. Mark this explicitly instead
of attempting to discard from the previous chunk.
PiperOrigin-RevId: 318983628
This fixes an issue where, even if captioning manager is disabled, the latest used captioning manager preference
related to text size is being applied.
In order to replicate:
1. Go to Captioning Preferences under device Settings and enable it
2. Change the text size to "very large"
3. Observe the selected text size is used for subtitles, for example in Youtube
4. Go to Captioning Preferences under device Settings and disable it
5. Observe the text size used for subtitles does not come back to normal, stays on "very large"
- Deprecate constructors that don't take an executor, to direct
developers toward the new ones. Callers can trivially pass
Runnable::run to one of the new ones if they want old behaviour.
- Add comment explaining warning suppression added in the CL that
added parallelised download support.
Issue: #5978
PiperOrigin-RevId: 318803296
This enhances readability, particularly as those methods will become
more complex when partially fragmented media will be supported.
PiperOrigin-RevId: 318795536
Previously only pcm encoding were stored in Format,
this was an issue as for audio passthrough and offload
lots of code needs to pass complex format informations
(encoding, sample rate, channel count, gapless metadata)
but could not use Format and each function was taking
each as different parameter.
By allowing Format to contain any encoding, and not only
pcmEncoding, it allows to pass a Format everywhere in ExoPlayer
code that needs a Format.
This patch does not have any functional change. It is only an internal refactor.
PiperOrigin-RevId: 318789444
Gapless offload is fixed in later R beta builds of all Pixels supporting R.
On the firsts R beta builds of Pixel 4, run the following command.
`setprop vendor.audio.offload.gapless.enabled true`.
It can not be enabled on the first Pixel 2 and 3 beta build.
PiperOrigin-RevId: 318436134
This it is enabled only on a list of
manually tested devices.
The list is empty in this CL.
*** Reason for rollforward ***
Fixed dependent cl was rolled forward.
*** Original change description ***
Rollback of 962e08d3be
*** Original commit ***
Add Offload gapless support
Confirmed to work on a Pixel 4 after enabling the feature:
`setprop vendor.audio.offload.gapless.enabled true`
***
***
PiperOrigin-RevId: 318433123
*** Reason for rollforward ***
Fixed dependent cl was rolled forward
*** Original change description ***
Rollback of 2aac0717d7
*** Original commit ***
Propagate format in supportsOutput
It is needed to know if gapless is needed,
as gapless offload might not be supported.
***
***
PiperOrigin-RevId: 318429321
*** Reason for rollforward ***
Rollforward after making sure the handler is created,
and that a test is written preventing a
similar regression.
*** Original change description ***
Rollback of b6f5a263f7
*** Original commit ***
Rollforward of commit 5612ac50a3.
*** Reason for rollforward ***
Rollforward after making sure the handler is created
from the playback thread and not from an app thread.
*** Original change description ***
Rollback of e1beb1d194
*** Original commit ***
PiperOrigin-RevId: 318274400
Ignore two tests in AsynchronousMediaCodecBufferEnqueuerTest
until the ShadowMediaCodec's behavior is updated to apply
input buffer ownership.
PiperOrigin-RevId: 318251859
The existing code moves a multi-line cue box by multiples of the height
of the whole cue box (incorrect), rather than multiples of the first
line of text (correct). These two are equivalent for single-line cues,
which is why I didn't initially spot the problem.
PiperOrigin-RevId: 318036793
Numerical lines conceptually map to a grid of lines in the viewport,
with the Cue text lines being aligned to one of the viewport lines.
It doesn't make sense to position a single-line cue differently based
on lineAnchor when it's expected to 'snap' to a particular line on the
viewport grid. So we redefine the position to be in terms of the cue
lines rather than the bounds of the cue box.
It's also not possible to always handle ANCHOR_TYPE_MIDDLE when
lineType=NUMBER (as it relies on the number of lines in the cue being
odd), so it's easier to ignore lineAnchor completely.
PiperOrigin-RevId: 318034664
This removes a lot of duplication from the module configuration,
avoids divergence, and makes sure that only the important differences
to the default are visible in each module file.
PiperOrigin-RevId: 318024823
Postrolls would be skipped because the period duration wasn't know at
the moment of resuming playback after backgrounding, so the position
wouldn't be resolved to resume the postroll ad.
We have the period duration stored in the AdPlaybackState, so we can
use that directly.
Issue: #7518
PiperOrigin-RevId: 317830418
They currently fall back to the main Looper if the current thread
doesn't have a Looper. All the changed Handlers are guaranteed to
be created on a thread with a Looper (mostly the ExoPlayer playback
Looper) and thus can make this stricter assumption. This makes it
easier to reason about the code as there are no ambiguities as to which
thread the Handler is running on.
PiperOrigin-RevId: 317334503
The setForeground mode method blocks in the same way as release
and should use the same timeout if configured.
In case the method runs into the timeout, a player error is reported.
PiperOrigin-RevId: 317283808
*** Original commit ***
Rename Util methods to clarify which Looper is used.
The method name didn't clarify that either the main or current
Looper is used.
***
PiperOrigin-RevId: 317283606
- It seems conceptually simpler for DefaultExtractorsFactory
- It seems unlikely we'll need to diverge the two. In the case of
workaround flags we can just have them be no-ops in the version
that doesn't need them.
PiperOrigin-RevId: 317151955
Looks like this change was introduced in
<unknown commit>
onKeysRemoved is currently not triggered in DefaultDrmSessionManager
as far as I can tell. It seems like it should be called from here.
PiperOrigin-RevId: 317072794
It seems more natural given we always end up instantiating a Matroska extractor,
not one that's specific to the WebM subset of Matroska. There's also no reason
not to support Matroska MIME types in DASH.
PiperOrigin-RevId: 316975451
*** Original commit ***
Rollforward of commit 5612ac50a3.
*** Reason for rollforward ***
Rollforward after making sure the handler is created
from the playback thread and not from an app thread.
*** Original change description ***
Rollback of e1beb1d194
*** Original commit ***
Expose experimental offload scheduling
Add a new scheduling mode that stops ExoPlayer main loop
when the audio offload buffer is full and resume it...
***
PiperOrigin-RevId: 316914147
*** Reason for rollforward ***
Rollforward after making sure the handler is created
from the playback thread and not from an app thread.
*** Original change description ***
Rollback of e1beb1d194
*** Original commit ***
Expose experimental offload scheduling
Add a new scheduling mode that stops ExoPlayer main loop
when the audio offload buffer is full and resume it when
it has been partially played.
This mode needs to be enabled and dissabled manually by the app
for now.
#exo-offload
***
***
PiperOrigin-RevId: 316898804
Ads can appear due to asynchronous ad tag requests completing after
earlier ads in a pod have loaded, so remove the requirement that the
ad count can't change. The MediaPeriodQueue should handling discarding
buffered content if an ad appears before already buffered content, so
I think this case is actually handled correctly by the core player
already.
Also remove the requirement that an ad URI can't change. This is a
defensive measure for now, but it's likely that a later fix in the IMA
SDK for an issue where loadAd is not called after preloading then
seeking before a preloaded ad plays will result in loadAd being called
more than once, and I think it's possible that the second call to
loadAd may have a different URI. Because the ad URI should only change
after an intermediate seek to another MediaPeriod, there shouldn't be
any problems with buffered data not getting discarded.
Issue: #7477
PiperOrigin-RevId: 316871371
The test was trying to synchronize a background decoding thread by
waiting for pending decode calls. However, the background thread needs
to fully queue the newly available output buffer before we can stop
waiting to ensure it's actually fully predictable. So we change the
pending wait to wait until the input buffer is cleared, which only
happens after the decoder is definitely done with it.
Also properly clean-up decoder (including shutting down the background
thread).
PiperOrigin-RevId: 316870659
- Leaving the TODO, since there are still MIME types we're unsure about.
- Removing AAC because xHE-AAC does not have this property. We may re-add
it with an additional profile check to exclude xHE-AAC in the future.
PiperOrigin-RevId: 316715147
We haven't seen it used anywhere in practice. It's a niche feature not
supported by any other extractors, and is one of the very few things
stopping us from simplifying MediaSource implementations to not set the
decodeOnly sample flag. This is a simplification that we want to make,
since the current mechanism doesn't work properly for cases where a
downstream decoder adjusts the buffer presentation timestamps so that
they're different on the output side than on the input side.
PiperOrigin-RevId: 316712302
This flag isn't needed anymore because the waiting for keys happens on
the source side and the source just returns NOTHING_READ under the
same conditions.
PiperOrigin-RevId: 316704214
We started using this method from other tests unrelated to
TestExoPlayer, so the method is better placed inside a generic Util
class.
PiperOrigin-RevId: 316675067
This change adds MediaSource.getMediaItem and deprecates MediaSource.getTag. For backwards compatibility, the tag is made available through the Window with `mediaItem.playbackProperties.tag` as well as in the deprecated `tag` attribute.
PiperOrigin-RevId: 316539752
Using the default offset as a magic constant makes tests hard to
understand. Improve that by looking up the value from the timeline or
setting it explicitly in multiple places, so the relationship becomes
clear.
PiperOrigin-RevId: 316421977
The correct order of initializing the MediaCodec should be (as per
documentation
https://developer.android.com/reference/android/media/MediaCodec#initialization)
"create -> setCallback -> configure -> start"
but the MediaCodecRenderer currently does
"create -> configure -> setCallback -> start"
MediaCodec implementations did not complain about this so far, but the
wrong sequence does not work with the MediaCodec in block mode (new mode
in Android R) and also the ShadowMediaCodec won't operate in
asynchronous mode otherwise. To initialize the MediaCodec in the correct
order, this commit adds configure() in the MediaCodecAdapter so the
MediaCodecRenderer can do:
adapter.configure(); // sets the callback and then configures the codec
adapter.start(); // starts the codec
PiperOrigin-RevId: 316127680
*** Original commit ***
Expose experimental offload scheduling
Add a new scheduling mode that stops ExoPlayer main loop
when the audio offload buffer is full and resume it when
it has been partially played.
This mode needs to be enabled and dissabled manually by the app
for now.
#exo-offload
***
PiperOrigin-RevId: 315948869
*** Original commit ***
Propagate format in supportsOutput
It is needed to know if gapless is needed,
as gapless offload might not be supported.
***
PiperOrigin-RevId: 315947888
*** Original commit ***
Add Offload gapless support
Confirmed to work on a Pixel 4 after enabling the feature:
`setprop vendor.audio.offload.gapless.enabled true`
***
PiperOrigin-RevId: 315946947
The test uses two items with period-in-window offsets and a non-zero default start position. The test also prepares the first item lazily so
that the start position (and thus the renderer offsets) need to change.
This is arguably the most complicated setup that needs to be tested.
PiperOrigin-RevId: 315903958
The order of source info refresh and onPrepared was accidentally
changed by ed88f4f1dd. This changes it back to the correct order
and adds a test
PiperOrigin-RevId: 315885164
Rename the DedicatedThreadAsyncMediaCodecAdapter to
AsynchronousMediaCodecAdapter as it is the only asynchronous adapter
implementation left after the clean-up.
PiperOrigin-RevId: 315873431
Add a new scheduling mode that stops ExoPlayer main loop
when the audio offload buffer is full and resume it when
it has been partially played.
This mode needs to be enabled and dissabled manually by the app
for now.
#exo-offload
PiperOrigin-RevId: 315860373
The option to cancel ongoing loads as part of the queue size evalation
was added recently. This split out the decision to a new method so that
a TrackSelection implementation can independently cancel loads and
discard upstream data. It also clarifies that evaluateQueueSize will
only be called if there is no ongoing load.
Issue: #2848
PiperOrigin-RevId: 315659735
If the previous chunk didn't finish loading, we need to find the appropriate
next chunk based on the current loading position (or the previous chunk's
start time if not independent).
PiperOrigin-RevId: 315658435
This change makes the media item argument in any constructors of the SinglePeriodTimeline non-null. Further a dummy media item is created for deprecated constructors using a tag only.
PiperOrigin-RevId: 315283842
ANSI/CTA-608-E R-2014 Annex B.5 says:
"The concept of a movable base row for a roll-up caption is new."
It means "new" compared to TC1 and TC2 (released in or before 1985).
Issue: #7475
PiperOrigin-RevId: 315258859
Output timestamps are calculated by the codec based on the buffers,
which is offset in Codec2. This adjusts the input timestamps as they
are passed in so they will match the output timestamps produced by
the MediaCodec.
PiperOrigin-RevId: 314963830
FakeMediaSource and FakeTimeline should put a media item to the window just as other media sources and timelines do. This change provides a fake media item for both of them.
Further the MaskingMediaSource needs to provide a media item for when the real timeline of the masked media source is not available. This can be easily done by using mediaSource.getMediaItem() once available. For now a dummy is used to make ExoPlayerTest run green. This can be easily change to use mediaSource.getMediaSource as soon as this method is defined by the MediaSource interface.
PiperOrigin-RevId: 314897474
bear_vorbis_gap.ogg is a copy of bear_vorbis.ogg with 10 garbage bytes
(DE AD BE EF DE AD BE EF DE AD) inserted before the second capture
pattern and 3 garbage bytes inserted at the end (DE AD BE).
Issue: #7230
PiperOrigin-RevId: 314715729
This change makes the media item of Timeline.Window non-null by providing a fallback media item in window.set(...). After having migrated all media sources this should not be needed anymore, but a fallback makes it more safe than just making the mediaItem argument of window.set(...) non-null (which is done in a following CL in this chain of CLs).
PiperOrigin-RevId: 314527983
Before loading a new chunk, the player will call reevaluateBuffer
anyway, so we don't have to do this directly after cancelation.
This simplifies some logic because we can remove the pending queue
size variable.
PiperOrigin-RevId: 314313268
If no deprecated methods on the factory is called, the media item instance that is passed to the createMediaSource method must be passed down to the constructor of the media source.
PiperOrigin-RevId: 314193865
This change applies the same approach of handling tag/streamKeys from factories like in the SmoothStreaming and Hls factories. It is functionally equivalent but improves readability.
PiperOrigin-RevId: 313771318
I was trying to understand why this isn't always checked and didn't
realise there was already a bug tracking this - this way a future
confused reader knows something isn't working quite as intended.
PiperOrigin-RevId: 313765935
If the sample times don't overlap and are independent, splicing makes no
difference because all samples (starting from the first one, which must be
a keyframe) will be appended to the end of the queue anyway.
PiperOrigin-RevId: 313594372
We aim to have the same media item instance published in the timeline that is passed to the createMediaSource(mediaItem) method. This means when the manifest uri is manually replaced by a developer, the media item published in the next timeline update is still the same.
Note: This leads to the fact that the manifest published in that timeline is loaded from a different URI than the URI of the media item in the same timeline.
PiperOrigin-RevId: 313375421
- The new CacheWriter is simplified somewhat
- Blocking on PriorityTaskManager.proceed is moved out of
CacheWriter and into the Downloader tasks. This is because
we want to shift only the caching parts of the Downloaders
onto their Executors, whilst keeping the blocking parts on
the main Downloader threads. Else we can end up "using"
the Executor threads indefinitely whilst they're blocked.
Issue: #5978
PiperOrigin-RevId: 313222923
Add an `_` in long constants.
Eg: 10000 => 10_000.
I'm proposing this change because I have had multiple
missread due to confusing the number of 0 in a long number.
More specifically, added an underscore to all number matching:
`final.*\ [0-9]{2,}000;`
PiperOrigin-RevId: 313186920
The test was flaky because it didn't wait for pending commands to finish
after pausing the test playback.
Also add more debug information to the toString() method because the
expected and actual state only differed in the nextAdGroupIndex in the
flaky case.
PiperOrigin-RevId: 313175919
Sometimes it's useful to be able to block until something on some other thread
"really has finished". This will be needed for moving caching operations onto
the executor in Downloader implementations, since we need to guarantee that
Downloader.download doesn't return until it's no longer modifying the
underlying cache.
One solution to this is of course just to not interrupt the thread that's
blocking on the condition variable, but there are cases where you do need to do
this in case the thread is at some other point in its execution. This is true
for Downloader implementations, where the Download.download thread will also
be blocking on PriorityTaskManager.proceed. Arranging to conditionally
interrupt the thread based on where it's got to is probably possible, but seems
complicated and error prone.
Issue: #5978
PiperOrigin-RevId: 313152413
Every timeline refresh currently assumes that it corresponds to a manifest
refresh and issues the respective load events. However, there are other
timeline updates that don't have a manifest refresh (e.g. ad state updates)and thus shouldn't issue these events.
PiperOrigin-RevId: 313150489
A recent change tried to make this condition clearer by using Integer.MAX_VALUE, but
this only works if the comparison also compares against larger values.
PiperOrigin-RevId: 312697530
The code that uses these variables is already commneted out. Android Studio
complains about unused variables and code and it's better to comment them
out as long as they are not used.
PiperOrigin-RevId: 312660512
This change adds an overloaded createMediaSource method which allows developers to pass in a media item with a in-memory manifest. Without this method the developer would not have a chance to pass in a non-dummy media item when using the factory for creting a DASH media source with an in-memory manifest.
PiperOrigin-RevId: 312660418
Previously if the AudioCapabilities reported that an encoding/channel count was
supported, DefaultAudioSink could try to play it via passthrough. However,
DefaultAudioSink does not support passthrough of every possible format (for
example, it's likely that AAC passthrough does not work given it's never been
tested and recent GitHub issues indicate that trying to use it leads to no
audio being played).
Add additional checks to make sure the encoding is in the list of encodings that
are known to work with passthrough in DefaultAudioSink.
issue:#7404
PiperOrigin-RevId: 312651358
*** Original commit ***
Remove set timeout on release() and setSurface()
Removes the experimental methods to set a timeout when
releasing the player and setting the surface.
***
PiperOrigin-RevId: 312647457
