Allows streamlining the way we handle unsupported DRM. In general, either:
* The device does not support the CDM. For example, phones don't support
PlayReady.
* The device does not support the protection scheme. For example, API < 25
devices do not support AES-CBC.
Currently we handle both cases differently. A dummy ExoMediaDrm will allow
the existence of dummy DefaultDrmSessionManagers.
PiperOrigin-RevId: 269315687
Allows shared ownership of ExoMediaDrms. Shared ownership will
allow users to pre-create ExoMediaDrms in their apps, as opposed
to having the DrmSessionManager create the ExoMediaDrm.
Issue:#4721
PiperOrigin-RevId: 269305850
The player may suppress playback when waiting for audio focus even if the
state==Player.READY. There is currently no getter or callback to obtain this
piece of information for UI updates or analytics.
Also, it's a important derived state to know whether the playback position is
advancing. Add isPlaying and the corresponding callback to allow retrieving
this information more easily.
Issue:#6203
PiperOrigin-RevId: 268921721
We currently force render the first frame in all cases. However, when joining
video to an ongoing playback, there is no defined first frame and we should
rather drop frames if we are late until we caught up with the audio playback.
PiperOrigin-RevId: 268698093
When the queue is empty, we may still get multiple requests for a window
sequence number using the same periodUid. We should keep the initially
assigned number because no window change happened.
PiperOrigin-RevId: 268649009
The fLaC prefix is included in the initialization data output
from the MKV extractor, so this is highly likely ot be the
right thing to do.
Issue: #6397
PiperOrigin-RevId: 268244365
This allows to uniquely identify a window within a Timeline. The value is set
correctly for all Window instances, but is not used anywhere yet.
PiperOrigin-RevId: 267556516
We currently don't test if an ad needs to be played in case we are already
playing content.
This is to prevent recreating the current content period when
an ad is marked as skipped. We prefer playing until the designated ad group
position and appending another piece of content. This is less likely to cause
visible discontinuities in case the ad group position is at a key frame
boundary.
However, this means we currently miss updates that require us to play an ad
after a timeline update.
PiperOrigin-RevId: 267553459
Select a track with matching role flags even if someone specifies a preferred role flgs without providing a preferred language
Prefer track without captions or describing music and sound flag when no role flag preference is provided.
Removing this method requires custom MediaSource implementations to switch
to the new version. While this breaks the implmentation, it also avoids problems
with accidentely forwarding to the wrong method or with accidental stack
overflows when none of the two methods gets implemented.
It's unlikely any callers are affected because this is only the extended version
with the defaultProjectionPositionUs parameter that no caller except the
internal player will use.
PiperOrigin-RevId: 267323042
Include Dataspec.httpRequestHeaders in CronetDataSource,
and OkHttpDataSource. Updated documentation of
HttpDataSource.open() to suggest that it should set request
headers (in decreasing priority) from (1) the passed DataSpec,
(2) parameters set with setRequestProperty() and (3) default
parameters set in the HttpDataSource.Factory. No mechanism
has been put in place to enforce this.
PiperOrigin-RevId: 266895574
Sniffing is performed in ProgressiveMediaPeriod even if a single
extractor is provided. Skip it in that case to improve performances.
Issue:#6325
PiperOrigin-RevId: 266766373
The Javadoc for open() won't be read by anyone because it's an overridden method
of the interface and not called directly from application code.
Move doc to class documentation as this is a generic explanation about the
functionality of the class. Also added "all" to clarify that all the parameters
will be added and the order of preference just applies in case of key clashes.
PiperOrigin-RevId: 266753249
This is a no-op interim change to introduce key request error
handling customization. Following changes will allow users to
inject their own LoadErrorHandlingPolicy implementations.
Issue:#6334
PiperOrigin-RevId: 266344399
- Surface information provided by methods isHardwareAccelerated,
isSoftwareOnly and isVendor added in Android Q in MediaCodecInfo
class.
- Estimate this information based on the codec name for earlier API
levels.
Issue:#5839
PiperOrigin-RevId: 266334850
This generalizes our "does file still exist" check to also check
that the file is the expected length. If it's not, we don't trust
it. This avoids infinite recursion in CacheDataSource if a cache
file is truncated underneath the cache.
Issue: #6165
PiperOrigin-RevId: 265707928
Implementations of ForwardingTimeline may override any of the two variants of
this method. We need to ensure that the customized override is always called.
Add back the deprecated method and make it final to forward to the
non-deprecated method in all cases for ForwardingTimelines.
PiperOrigin-RevId: 265419830
The last selection criteria is the audio bitrate to prefer higher-quality
streams. We shouldn't apply this criterium though if the languages of the
tracks are different.
Issue:#6335
PiperOrigin-RevId: 265064756
The implementation can't work properly unless these methods are called by
subclasses, so annotate them to require calling the super implementation when
overriding.
PiperOrigin-RevId: 265017433
The error is closely related to the playback state IDLE and should be updated
in sync with the state to prevent unexpected event ordering and/or keeping the
error after re-preparation.
Issue:#5407
PiperOrigin-RevId: 265014630
Currently the value of FLAG_ALLOW_CACHE_FRAGMENTATION is defined as "1 << 4" but commented as "8". Either the value of FLAG_ALLOW_CACHE_FRAGMENTATION should be "1 << 3", or the comment should be 16. Here I am modifying the comment since it does not affect any current behavior.
PiperOrigin-RevId: 265011839
The app is able to pass a more specialized array type, so the Arrays.copyOf call
produces an array into which it's not valid to store arbitrary AudioProcessors.
Create a new array and copy into it to avoid this problem.
PiperOrigin-RevId: 264779164
This situation happens if the first chunk to load is already behind the end
of the stream. In this case, the preparation never completes because
HlsSampleStreamWrapper gets stuck in a prepared=false and loadingFinished=true
state.
Gracefully handle this situation by attempting to load the last chunk if still
unprepared to ensure that track information is obtained as far as possible.
Otherwise, it wouldn't be possible to play anything even when seeking back.
Issue:#6314
PiperOrigin-RevId: 264599465
Extract supplemental data from block additions in WebM/Matroska.
Allow storing supplemental data alongside samples in the SampleQueue and write
it as a separate field in DecoderInputBuffers.
Handle supplemental data in the VP9 extension by propagating it to the output
buffer.
Handle supplemental data for HDR10+ in MediaCodecVideoRenderer by passing it to
MediaCodec.setParameters, if supported by the component.
PiperOrigin-RevId: 264582805
There is no point in having this parameter as the tag should always be a single
immutable object instantiated at the time the Timeline is created or earlier.
So there is no preformance benefit and it's error-prone in case people
forget to set setTag=true.
PiperOrigin-RevId: 264117041
This solves various issues around event association for buffering and
error throwing around period discontinuities.
The main changes are:
- Logic around being "ready" at the end of a period no longer checks if the
next period is prepared.
- Advancing the playing period no longer checks if the next one is prepared.
- Prepare errors are always thrown for the playing period.
This changes the semantics and assumptions about the "playing" period:
1. The playing period can no longer assumed to be prepared.
2. We no longer have a case where the queue is non-empty and the playing or
reading periods are unassigned (=null).
Most other code changes ensure that these changed assumptions are handled.
Issue:#5407
PiperOrigin-RevId: 263776304
This matches the documentation on MetadataDecoder.decode:
"@return The decoded metadata object, or null if the metadata could not be decoded."
PiperOrigin-RevId: 263767144
Most users are likely to be only interested in the playlist items. So put the
Window definition first and explicitly mention that this is a playlist
item.
PiperOrigin-RevId: 263764515
The current ExoPlayerFactory is growing too big and usage becomes increasingly
complicated because it's not possible to set individual components without
specifying many other defaults.
Adding new builder classes makes building easier and more future-proof.
PiperOrigin-RevId: 263339078
entries are used in .equals(), so it's good to have them printed in toString() too (for test failures) and it makes logging easier too.
PiperOrigin-RevId: 263335503
The current max video buffer is 13MB which is too small for high quality
streams and doesn't allow the DefaultLoadControl to buffer up to its default
max buffer time of 50 seconds.
Also move util method and constants only used by DefaultLoadControl into this
class.
PiperOrigin-RevId: 263328088
We no longer need two modules as AndroidX-Test takes care of the system
abstraction and we no longer have Robolectric Handler/Looper workarounds.
PiperOrigin-RevId: 262363201
I think we need to start clearing the holder as part of the
DRM rework. When we do this, it'll only be valid to read
from the holder immediately after it's been populated.
PiperOrigin-RevId: 262362725
We're no longer tied to the emsg spec, so we can skip unused fields
and assume ms for duration.
Also remove @Nullable annotation from EventMessageEncoder#encode, it
seems the current implementation never returns null
PiperOrigin-RevId: 262135009
We already allow mixed mime type and mixed sample rate adaptation on request,
so for completeness, we can also allow mixed channel count adaptation.
Issue:#6257
PiperOrigin-RevId: 261930046
This also decouples EventMessageEncoder's serialization schema from the emesg spec (it happens to still match the emsg-v0 spec, but this is no longer required).
PiperOrigin-RevId: 261877918
If we keep streams in chunk sources after selecting new tracks, we also keep
a reference to a stale disabled TrackSelection object. Fix this by updating
the TrackSelection object when keeping the stream. The static part of the
selection (i.e. the subset of selected tracks) stays the same in all cases.
Issue:#6256
PiperOrigin-RevId: 261696082
We currently keep the sequence number if we don't reset the position. However,
the sequence number should be kept if we don't reset the state. Otherwise
re-prepare with position reset is counted as new playback although it's still
the same.
PiperOrigin-RevId: 261644924
- When in STATE_SEEK with targetGranule==0, seeking would exit
without checking that the input was positioned at the correct
place.
- Seeking could fail due to trying to read beyond the end of the
stream.
- Seeking was not robust against IO errors during the skip phase
that occurs after the binary search has sufficiently converged.
PiperOrigin-RevId: 261317035
Two tests have very low propability flakiness (1:1000) due to not waiting for
a seek in one case and the chance of already being ended in another case.
Fix these and also adjust wrong comments about state changes.
PiperOrigin-RevId: 261309976
These values can easily default to the empty track group and the empty
selection. As a result we can remove restrictions about not calling
holder.getTrackGroups before the period finished preparation.
PiperOrigin-RevId: 261280927
A previous change switched to calculation of the bitrate based on the
first MPEG audio header in the stream. This had the effect of fixing
seeking to be consistent with playing from the start for streams where
every frame has the same padding value, but broke streams where the
encoder (correctly) modifies the padding value to match the declared
bitrate in the header.
Issue: #6238
PiperOrigin-RevId: 261163904
Android considers ALAC initialization data to consider of the magic
cookie only, where-as FFmpeg requires a full atom. Standardize around
the Android definition, since it makes more sense (the magic cookie
being contained within an atom is container specific, where-as the
decoder shouldn't care what container the media stream is carried in)
Issue: #5938
PiperOrigin-RevId: 261124155
Checking inputPosition == 0 isn't sufficient because the synchronization
at the top of read() may advance the input (i.e. in the case where there's
some garbage prior to the seek frame).
PiperOrigin-RevId: 261086901
This factory was only needed in the past when we didn't have
AnalyticsCollector.setPlayer. Code becomes easier to use without this factory.
PiperOrigin-RevId: 261081860
This new annotation declares everything as non-null by default and can be
used as a package annotation in package-info.java.
In this change the core lib offline package and the mediasession extension is
annotated that way as initial example usage.
PiperOrigin-RevId: 260894548
DefaultMediaClock has currently two non-ideal behaviours:
1. One part of checking if it should use the renderer clock is checking whether
the associated renderer finished reading its stream. This only makes sense
if the renderer isn't already reading ahead into the next period. This can
be solved by forwarding if we are reading ahead to the sync command.
2. When switching from stand-alone to renderer clock we assume they are
exactly at the same position. This is true in theory, but in practise there
may be small differences due to the different natures of these clocks. To
prevent jumping backwards in time, we can temporarily stop the stand-alone
clock and only switch once the renderer clock reached the same position.
PiperOrigin-RevId: 260690468
Both periods are rarely updated in the same iteration. If they are, advancing
the reading period first seems more logical and also more efficient as it may
avoid one extra doSomeWork iteration.
PiperOrigin-RevId: 260463735
When using speed adjustment it was possible for playback to get stuck at a
period transition when the channel count changed: SonicAudioProcessor would be
drained at the point of the period transition in preparation for creating a new
AudioTrack with the new channel count, but during draining the incorrect (new)
channel count was used to calculate the output buffer size for pending data from
Sonic. This meant that, for example, if the channel count changed from stereo to
mono we could have an output buffer size that stored an non-integer number of
audio frames, and in turn this would cause writing to the AudioTrack to get
stuck as the AudioTrack would prevent writing a partial audio frame.
Use Sonic's current channel count when draining output to fix the issue.
PiperOrigin-RevId: 260156541
We keep track of the last publicly known playing period to report it as part
of events happening after the period has been released.
In cases where a period briefly becomes the playing one and is released
immediately afterwards, we currently don't save it as the "last known playing
one". Improve that by saving an explicit reference.
Issue:#5407
PiperOrigin-RevId: 259737218
Creating a period in MaskingMediaSource may result in delayed event reporting
depending on when the actual period gets created. To avoid event reporting
inaccuracies, report the mediaPeriodCreated and mediaPeriodReleased events
directly.
Issue:#5407
PiperOrigin-RevId: 259737170
This changes the logic in the following ways:
- If no preferred language is matched, prefer better scores for the selected
audio language.
- If a preferred language is matched, always prefer the better match
irrespective of default or forced flags.
- If a preferred language score and the isForced flag is the same, prefer
tracks with a better selected audio language match.
PiperOrigin-RevId: 259707430
Some variables were defined although they are the default and other things
were set-up in a non-sensible way, e.g. asserting that audio is selected
although no audio renderer is available, or using unset duration for
everything.
PiperOrigin-RevId: 259532782
PlaybackInfo changes are one of the last ones not masked and reported in the same
way as all other changes.
The main change to support this is to also mask the parameters set in
DefaultAudioSink.
PiperOrigin-RevId: 258787744
1. Using the Locale on API<21 doesn't make any sense because it's a no-op
anyway. Slightly restructured the code to avoid that.
2. API<21 often reports languages with non-standard underscores instead of
dashes. Normalize that too.
3. Some invalid language tags on API>21 get normalized to "und". Use original
tag in such a case.
Issue:#6153
PiperOrigin-RevId: 258773463
2-letter codes (ISO 639-1) are the standard Android normalization and thus we
should prefer them to 3-letter codes (although both are technically allowed
according the BCP47).
This helps in two ways:
1. It simplifies app interaction with our normalized language codes as the
Locale class makes it easy to convert a 2-letter to a 3-letter code but
not the other way round.
2. It better normalizes codes on API<21 where we previously had issues with
language+country codes (see tests).
3. It allows us to normalize both ISO 639-2/T and ISO 639-2/B codes to the same
language.
PiperOrigin-RevId: 258729728
If we use the default start position, we currently resolve it immediately
even if we need to play an ad first, and later try to project forward again
if we believe that the default start position should be used.
This causes problems if a specific start position is set and the later
projection after the preroll ad shouldn't take place.
The problem is solved by keeping the content position as TIME_UNSET (= default
position) if an ad needs to be played first. The content after the ad can
then be resolved to its current default position if needed.
PiperOrigin-RevId: 258583948
These methods helps to indicate that a media source isn't used to create new
periods in the immediate term and thus limited resources can be released.
PiperOrigin-RevId: 258373069
It's easy to forget to forward methods when using basic wrapping. For example,
ExtractorMediaSource.addEventListener is currently a no-op because it's not
forwarded.
PiperOrigin-RevId: 257757556
This is a temporary workaround until we have migrated all MediaSources uses.
This change avoids having to migrate all uses of MediaSources immediately.
PiperOrigin-RevId: 257459138
- Remove manifest argument from callbacks of Player.EventListener and
SourceInfoRefreshListener. Instead make it accessible through
Player.getCurrentManifest() and Timeline.Window.manifest.
- Fix all MediaSource implementation to include the manifest in the
Timeline instead of passing it to the SourceInfoRefreshListener.
- Refactor ExoPlayerTestRunner, FakeTimeline, FakeMediaSource to
reflect these changes and make tests pass.
PiperOrigin-RevId: 257359662
Sending MESSAGE_PREPARE_SOURCE should happen last in the constructor.
It was previously happening before initialization finished (and in
particular before pendingMediaPeriods was instantiated).
Issue: #6146
PiperOrigin-RevId: 257158275
We currently handle most the control code logic after handling special
characters. This includes filtering out repeated control codes and checking
for the correct channel. As the special character sets are control codes as well,
these checks should happen before parsing the characters.
Issue:#6133
PiperOrigin-RevId: 256993672
- Explicitly specify HTTP_METHOD_POST (previously this was
implicit as a result of the body data being non-null)
- Use null when there's no body data (it's converted to
null inside of the DataSpec constructor anyway)
PiperOrigin-RevId: 256573384
Currently, we sometimes apply new playback parameters directly and sometimes
through the list of playbackParameterCheckpoints. Only when using the checkpoints,
we also reset the offset and corresponding position for speedup position
calculation. However, these offsets need to be changed in all cases to prevent
calculation errors during speedup calculation[1].
This change channels all playback parameters changes through the checkpoints to
ensure the offsets get updated accordingly. This fixes an issue introduced in
31911ca54a.
[1] - The speed up is calculated using the ratio of input and output bytes in
SonicAudioProcessor.scaleDurationForSpeedUp. Whenever we set new playback
parameters to the audio processor these two counts are reset. If we don't reset
the offsets too, the scaled timestamp can be a large value compared to the input
and output bytes causing massive inaccuracies (like the +20 seconds in the
linked issue).
Issue:#6117
PiperOrigin-RevId: 256533780
The masking logic for unprepared MediaSources is currently part of
ConcatanatingMediaSource. Moving it to its own class nicely separates the
code responsibilities and allows reuse.
PiperOrigin-RevId: 256360904
1. Only output video starting from a keyframe
2. When calculating the timestamp offset to adjust live streams to start
at t=0, use the timestamp of the first tag from which a sample is actually
output, rather than just the first audio/video tag. The test streams in
the referenced GitHub issue start with a video tag whose packet type is
AVC_PACKET_TYPE_SEQUENCE_HEADER (i.e. does not contain a sample) and whose
timestamp is set to 0 (i.e. isn't set). The timestamp is set correctly on
tags that from which a sample is actually output.
Issue: #6111
PiperOrigin-RevId: 256147747
The total window and period count, as well as the period offset for each holder
are not actually needed and can be removed.
Also added a TODO to remove two other variables if possible.
PiperOrigin-RevId: 255945584
We are currently queuing periods in a way such that the new start position
lines up with the end of the previous period (to ensure continuous playback).
However, if the start position of the new period is larger than the total of
all previously played period durations, we may end up with negative renderer
timestamps when seeking back to the beginning of this new period. Negative
timestamps should be avoided as most decoders have problems handling them
correctly.
This change forces a renderer reset if we detect such a seek to a negative
renderer time and also resets the renderer offset to 0 every time all
renderers are disabled, as this is the only time where we can savely change
the offset of an existing media period.
Also, if playback starts with an ad, we choose the content position as
renderer offset to prevent the whole issue from occurring for the seek-behind-
midroll case.
Issue:#6009
Issue:#5323
PiperOrigin-RevId: 253790054
In some edge cases the renderer position may be slightly ahead of the
buffered position and the total buffered duration is thus negative. We already
filter that in ExoPlayerImpl for the publicly accessible value. However, we
forward the unfiltered value to other components like the LoadControl, which
may be confusing.
Issue:#6015
PiperOrigin-RevId: 253780460
This permission has normal access right and can't be revoked by the user.
However, an app can choose to revoke it when using ExoPlayer, e.g. if
no network is required and the app doesn't want to list this permission.
Support this use case by gracefully catching the exception in the relevant
places.
Issue:#6019
PiperOrigin-RevId: 253759332
We currently report MediaCodec exceptions as unexpected exceptions instead of
as renderer error. All such exceptions are now wrapped in a new DecoderException
to allow adding more details to the exception.
PiperOrigin-RevId: 252054486
It's only thrown in an edge case on API level 20 and below. If it
is thrown it causes playback failure when playback could succeed,
by throwing up through configureCodec.
It seems better just to catch the exception and have the codec be
configured using the format's own width and height.
PiperOrigin-RevId: 251745539
We currently don't display the last frame because the seek time is behind the
last frame's timestamps and it's thus marked as decodeOnly.
This case can be detected by checking whether all data sent to the codec is
marked as decodeOnly at the time we read the end of stream signal. If so, we
can re-enable the last frame. This should work for almost all cases because the
end-of-stream signal is read in the same feedInputBuffer loop as the last
frame and we therefore haven't released the last frame buffer yet.
Issue:#2568
PiperOrigin-RevId: 251425870
Using parallel adaptation for Formats without bitrate information currently
causes an exception. Handle this gracefully and also cases where all formats
have the same bitrate.
Issue:#5971
PiperOrigin-RevId: 250682127
Setting the requested encoding in all cases ensures we receive the relevant
response headers indicating whether gzip was used. Doing that allows to
detect the content length in cases where gzip was requested, but the server
replied with uncompressed content.
PiperOrigin-RevId: 250660890
cache() opens all connections with unset length to avoid position errors.
This makes more data then needed to be downloading by the underlying
network stack.
This fix makes makes it open connections for only required length.
Issue:#5927
PiperOrigin-RevId: 250546175
"Write case, lock not available" was a bit confusing. When the
content is not cached and the lock is held, it's neither a read
or a write case. It's a "can't do anything" case. When blocking,
it may subsequently turn into either a read or a write.
PiperOrigin-RevId: 250530722
The former is deprecated and replaced by the latter in Mockito 2. However, there is a
functional difference: ArgumentMatchers will reject `null` and check the type
if the matcher specified a type (e.g. `any(Class)` or `anyInt()`). `any()` will
remain to accept anything.
PiperOrigin-RevId: 250458607
Removes the need for duplicate calls to SampleQueue#read when
implementing DecryptionResources acquisition in the MediaSources.
PiperOrigin-RevId: 250298175
According to Cue's constructor (for bitmaps) documentation:
+ cuePositionAnchor does horizontal anchoring.
+ cueLineAnchor does vertical anchoring.
Usage is currently inverted.
Issue:#5633
PiperOrigin-RevId: 250253002
The only known use case for contentTypePredicate is to catch
the case when a paywall web page is returned via a DataSource,
rather than the data that was being requested. These days streaming
providers should be using HTTPS, where this problem does not exist.
Devices have also gotten a lot better at showing their own
notifications when paywalls are detected, which largely mitigates
the need for the app to show a more optimal error message or
redirect the user to a browser.
It therefore makes sense to deprioritize this feature. In
particular by removing the arg from constructors, where nearly
all applications are probably passing null.
PiperOrigin-RevId: 249634594
This prevents further unexpected updates if the MediaSource happens to
finish its preparation at a later point.
Issue:#5915
PiperOrigin-RevId: 249439246
The SurfaceListener just sets the surface on the VideoComponent, but
SphericalSurfaceView already accesses the VideoComponent directly so it seems
simpler to update the surface directly.
PiperOrigin-RevId: 249242185
We are currently defaulting to targetSdk=1 as no targetSdk is specified. Only
tests which explicitly ask for another SDK use another test SDK. With the
versioned manifest, all tests run using the targetSDK by default.
PiperOrigin-RevId: 249060796
If a PlayerMessage throws an exception, it is currently not deleted from the
list of pending messages. This may be problematic as the list of pending
messages is kept when the player is retried without reset and the message is
sent again in such a case.
PiperOrigin-RevId: 247414494
Adding an explicit option to clear all downloads prevents repeated database
access in a loop when trying to delete all downloads.
However, we still create an arbitrary number of parallel Task threads for this
and seperate callbacks for each download.
PiperOrigin-RevId: 247234181
1. A content session after an ad has been played was not re-marked as active,
leading to new ad session being marked as active too early.
2. Switching from content to post-roll ended the content session because
the return value of getAdGroupTimeUs of C.TIME_END_OF_SOURCE was not
handled. Using the nextAdGroupIndex instead.
PiperOrigin-RevId: 246977327
As justification for why we should not have this type of logging,
it would scale up to about 13K LOC, 1800 Strings, and 36K (after
pro-guarding - in the case of the demo app) if we did it through
the whole code base*. It makes the code messier to read, and in
most cases doesn't add significant value.
Note: I left the Scheduler logging because it logs interactions
with some awkward library components outside of ExoPlayer, so is
perhaps a bit more justified.
* This is a bit unfair since realistically we wouldn't ever add
lots of logging into trivial classes. But I think it is fair
to say that the deltas would be non-negligible.
PiperOrigin-RevId: 246181421
- Removed DownloadInternal and its sometimes-out-of-sync
duplicate state
- Fixed downloads being in STOPPED rather than QUEUED state
when the manager is paused
- Fixed setMaxParallelDownloads to start/stop downloads if
necessary when the value changes
- Fixed isWaitingForRequirements
PiperOrigin-RevId: 246164845
They behave identically, and the old names are being removed.
Open-source note: The new methods are available in Truth as of version 0.44.
END_PUBLIC
More information:
go/issameas-lsc
Tested:
TAP --sample ran all affected tests and none failed
http://test/OCL:246024032:BASE:246042619:1556672975894:513e7746
PiperOrigin-RevId: 246101315
Can happen if the load position falls behind in every playlist and
when we try to load the next segment, the adaptive selection logic
decides to change variant.
Issue:#5816
PiperOrigin-RevId: 245923006
This resolves some naming confusion that previously existed
as a result of DownloadThread also being used for removals.
Some related variables (e.g. activeDownloadCount) would refer
to both download and removal tasks, whilst others
(e.g. maxParallelDownloads) would refer only to downloads.
This change renames those that refer to both to use "task"
terminology.
This change also includes minor test edits.
PiperOrigin-RevId: 245913671
[Problem]
There are 3 services / modes transported on line 21:
- Captioning
- TEXT (generally not program related)
- XDS (eXtended Data Services)
Bytes belonging to the unsupported modes are interleaved with the
bytes of the captioning mode.
See Chapter 7, Chapter 8.5 and Chapter 8.6 of the CEA608 Standard for
more details.
[Solution]
Drop all bytes belonging to unsupported modes.
[Test]
- All streams containing only captioning services should not be influenced
- Test all 4 CEA 608 channels with live over-the-air content and
using all available TEXT and XDS streams.
There are no logic changes here. It's just moving code around and removing
the "internal" part of names where no longer required.
PiperOrigin-RevId: 245407238
We currently toggle the view in onTouchEvent ACTION_DOWN which is non-standard
and causes problems when used in a ViewGroup intercepting touch events.
Switch to standard Android click handling instead which is also what most
other player apps are doing.
Issue:#5784
PiperOrigin-RevId: 245219728
The AndroidX bundled version (0.42) lags behind the most up-to-date public
release (0.44) making it more difficult to stay close to the actual head
revision which is used internally.
PiperOrigin-RevId: 244848568
This ensures we keep the loading period in sync with the the playing period in
PlybackInfo, when the latter changes to something new.
PiperOrigin-RevId: 244838123
Also re-enable two of them, although note that the fix here is
not related to the flakiness that caused them to be disabled.
I'm re-enabling them since much has changed in DownloadManager,
and the hope is that these tests are no longer flaky.
PiperOrigin-RevId: 244826225
Experiments show this is beneficial for rebuffers with only minor impact
on battery usage.
Configurations which explicitly set a minimum buffer duration are unaffected.
Issue:#2083
PiperOrigin-RevId: 244823642
Before this change we'd release the audio track and create a new one as soon
as audio processors had drained when reconfiguring.
Fix this behavior by stop()ing the AudioTrack to play out all written data.
Issue: #2446
PiperOrigin-RevId: 244812402
- We had buildAddRequest and sendNewDownload. Converged to
buildAddDownload and sendAddDownload.
- Also fixed a few more inconsistencies, and brought the
action constants into line as well.
PiperOrigin-RevId: 244274041
- Listener based reporting of progress allows the content length
to be persisted into the download index (and notified via a
download state change) as soon as it's available.
- Moved contentLength back into Download proper. It should only
ever change once, so I'm not sure it belongs in the mutable part
of Download.
- Made a DownloadProgress class, for naming sanity.
PiperOrigin-RevId: 244242487
1. customCacheKey for DASH/HLS/SS is now asserted against
in DownloadRequest
2. Merging of event delivery in DownloadManager is very
tricky to get right and probably not a good idea
PiperOrigin-RevId: 244048392
Currently SimpleCache will touch cache spans whenever it reads
from them. With legacy SimpleCache setups this involves a potentially
expensive file rename. With new SimpleCache setups it involves
a more efficient but still non-free database write.
For offline use cases, and more generally any use case where the
eviction policy doesn't use last access timestamps, touching is
not useful. This change allows the evictor to specify whether it
needs cache spans to be touched or not. SimpleCache will only touch
spans if the evictor requires it.
Note: There is a potential change in behavior in cases where a
cache uses an evictor that doesn't need cache spans to be touched,
but then later switches to an evictor that does. The new evictor
may temporarily make sub-optimal eviction decisions as a result.
I think this is a very fair trade-off, since this scenario is
unlikely to occur much, if at all, in practice, and even if it
does occur the result isn't that bad.
PiperOrigin-RevId: 244005682
One goal we forgot about a little bit was to allow applications
to provide their own index implementation. This requires the
writable side to also be defined by an interface.
PiperOrigin-RevId: 243979660
ImaAdsLoader gets the player position after the app releases
the player to support resuming ads at their current position
if the same ads loader is reused.
PiperOrigin-RevId: 243969916
ShadowLooper -> ShadowLegacyLooper
ShadowRealisticLooper -> ShadowPausedLooper
ShadowBaseLooper -> ShadowLooper
And all public methods from ShadowLegacyLooper get pushed up to ShadowLooper
Pull Request: https://github.com/robolectric/robolectric/pull/4868
Copybara: OK
Also adjust Google3 tests using custom looper shadows where necessary.
Convert exoplayer to paused looper to eliminate reliance on custom shadows
PiperOrigin-RevId: 243839311
The removed condition only applies when prepare fails and no timeline was or
is known. We should keep a pending seek position in such a case. The internal
player does the same thing already.
PiperOrigin-RevId: 243819466
- Expose constructors that take DatabaseProvider. Deprecate those
that do not.
- Expose Cache.getUid. This will likely be used for naming of the
tables accessed by DefaultDownloadIndex .
PiperOrigin-RevId: 243637786
- Allows enforcing immutability, which in a future CL will allow
avoiding allocating a new array/list on every call.
- Also some left over doc cleanup from the DownloadState rename.
PiperOrigin-RevId: 243627352
The old domain automatically forwards to the new one. For consistency, change
all doc and code references regardless.
Also adds GitHub CNAME config file which configures our page for the custom
domain.
PiperOrigin-RevId: 243592110
handleSourceInfoRefreshed checks for various conditions which may trigger a
change in position. In all cases, we need to resolve the new position for ads,
and if the corresponding MediaPeriods already exist, perform the seek to the new
position.
This change simplified the code by
1. Moving all conditions in a single if-else structure which just determines
the new playback position. This makes it easier to read and to follow.
2. Doing the subsequent seek and position update in one final step to remove
duplicated code.
As a side effect, it also improves some edge cases. For example when skipping
an ad to an already prebuffered period, we now properly update the existing
prebuffered period.
PiperOrigin-RevId: 243588014
- Make a best effort to avoid posting MSG_DOWNLOAD_THREAD_STOPPED
if the internal thread has already quit. Doing so is harmless,
but causes an error in Logcat. We used to generate this warning
quite a bit in ExoPlayerImplInternal as well, and we got quite
a few issues filed about it, so best to avoid it if possible.
- Null out the back reference DownloadThread holds to its manager
on release. This avoids a potential issue where a download thread
can prevent GC of its manager if it doesn't cancel quickly.
There's similar logic (with a similar comment) in Loader.LoadTask.
PiperOrigin-RevId: 243365143
- Rename getAllDownloads to getCurrentDownloads to make it clear
that it doesn't include completed and failed downloads
- Change getDownloadCount to isWaitingForRequirements, which is
what it's used for. Added TODO to make it returns the right thing
PiperOrigin-RevId: 243257856
If the returned retry delay is unset, we should use a DONT_RETRY_FATAL action
instead and dispatch the event with the "canceled" flag.
PiperOrigin-RevId: 243251554
Non-trivial download state transitions are currently split across
DownloadManager and Download. These transitions are part of the
same state machine, so it's clearer if they're all in the same place
(i.e. DownloadManager, since this is the component that transitions
downloads between states).
PiperOrigin-RevId: 243249915
The only state DownloadInternal holds is duplicate of state contained
in Download, but can confusingly be temporarily out of sync. This is
error prone because it's easy to use the wrong variable (e.g.
download.state vs state). DownloadInternals methods are called into
and call out into DownloadManager, which makes some code paths that
are quite hard to follow.
It's possible to simplify DownloadManager quite a lot by removing the
duplicated state in DownloadInternal, at which point DownloadInternal's
methods flatten into DownloadManager, which can just hold an internal
list of Downloads directly.
This is a first step, which makes it clear that DownloadThready only
needs its immutable DownloadAction + an isRemove flag.
PiperOrigin-RevId: 243245288
Put methods into a more sensible order (the same order as
in the switch statement, which is a more natural order with
e.g. initialize coming first and release coming last).
PiperOrigin-RevId: 242891742
- Changed startDownloads/stopDownloads back to their previous behavior
of starting and stopping all downloads at the manager level.
- Made setManualStopReason methods for the new case of setting a manual
stop reason or some or all downloads.
- Added plumbing to specify an initial manual stop reason when adding a
new download, without also overwriting the manual stop reasons of all
other preexisting downloads. Using the value is left as a TODO pending
a bit of further cleanup that'll make it easier.
PiperOrigin-RevId: 242891688
- Remove completed downloads from progress notification.
- Treat restarting as a downloading, When there are both
remove and download tasks ongoing, we prefer to show a
download notification to a remove one. A restarting task
encapsulates is basically a (remove+download) task. Given
our preference, we should treat it as a download rather
than as a removal.
PiperOrigin-RevId: 242867228
Both manifest loads currently use the default retry delay although a
LoadErrorHandlingPolicy is provided to the MediaSource.
PiperOrigin-RevId: 242852672
- Remove NETWORK_TYPE_NOT_ROAMING and NETWORK_TYPE_METERED
because JobScheduler doesn't support them, and they're
probably not useful to many people (data when roaming is
normally enabled/disabled at the OS level, and restricting
to *only* metered networks seems niche)
- Convert network requirements proper flags
- Stop persisting requirements in DownloadIndex. The direction
we're headed to solve the manager start/stop problem is
going to involve state in DownloadManager determining whether
downloads actually start, and if we're doing that then it's
no worse to do it for this as well
PiperOrigin-RevId: 242713196
Robolectric calls messages posted to the main thread immediately inline. This
will change in the upcoming release and will break tests implicitly relying
on that behavior.
Update ConcatenatingMediaSourceTest to actually wait for the posted message.
PiperOrigin-RevId: 242631531
HashMap.values does not guarantee a stable order, but our tests
assert that the order is stable. It's probably a nice property
that the order is the same as that in which downloads are added,
so this change makes the order stable, with that order.
A future change will likely improve the way this works, by
changing the internal thread "downloads" variable to have type
ArrayList<Download>, at which point the internal thread can just
pass a read-only copy of its list to the main thread whenever a
state changes.
PiperOrigin-RevId: 242460725
- Fix ID being dropped when DownloadAction is serialized and
de-serialized as it's passed to DownloadService
- Properly set DownloadAction.id when building an action from
a DownloadState
- Make ID a required constructor argument. Else it's too easy
to not propagate it by accident.
PiperOrigin-RevId: 242457831
Remove MediaCodecSelector.DEFAULT_WITH_FALLBACK, as codec selectors
will need to be able to return a list of decoder infos even when not
using fallback in a later change. Instead signal that fallback should
be used via a renderer constructor.
Fallback is always disabled for audio.
PiperOrigin-RevId: 242454172
Adding a protected method for that allows to supply a customized
TrackOutputProvider to the ExtractorWrapper used by the ContainerMediaChunk
(e.g. for logging purposes).
The alternative would be to provide a TrackOutputProvider through the
ChunkSampleStream constructor, but the extra initialization taking place
in the ChunkSampleStream constructor would need to move somewhere else and
some methods of BaseMediaChunkOutput would need to move to the
TrackOutputProvider interface. This seems too much effort for niche
customization case.
PiperOrigin-RevId: 242448776
The position should be subtracted from the total content length
retrieved from the cache in this case, both for iteration and for
filling out counters.contentLength (for the latter, note not
doing it this way is inconsistent with what happens when dataSpec
length is set to a known value).
PiperOrigin-RevId: 242445034
It's a transient state whose only use is when passing a removed
download to onDownloadStateChanged. It's a bit strange to have it
as a proper state, since we end up asserting that it's an invalid
value in other places.
This change adds an explicit onDownloadRemoved, which allows
removal of the transient STATE_REMOVED, related assertions, and
uncertainty when dealing with an @State variable whether it's
necessary to handle it being STATE_REMOVED.
PiperOrigin-RevId: 242444128
This is the last cl of a series of cls that will convert DownloadManager
threading model as public methods post actions on an internal thread
which then do the work.
PiperOrigin-RevId: 242437229
The priority task manager only needs to listen to loading state changes and
is independent of the rest of DefaultLoadControl. This also fixes problems
where the player stops loading without consultin the LoadControl.
PiperOrigin-RevId: 242098374
This is the second cl of a series of cls that will convert DownloadManager
threading model as public methods post actions on an internal thread
which then do the work.
PiperOrigin-RevId: 241773765
Necessary for implementing scheme-independant resource management.
Example: A media source needs to have ownership of a crypto resource
while samples are in buffer (but not yet in the renderer).
PiperOrigin-RevId: 241698575
Most of our components that have Handler instances for calling
back to application code instantiate the Handler using Util.getLooper
in the constructor. This makes the two components that do
something else consistent with that model.
PiperOrigin-RevId: 241545575
If no track matches the preferred language or no preferred language is provided,
use the system Locale language as the next decision criterion.
PiperOrigin-RevId: 241322703
- This removes the need for Variant and Rendition to have a common
base class, allowing the url field to be marked as @Nullable in
Rendition but not in Variant.
- The addition of mediaPlaylistUrls is needed for the new StreamKey
indexing for HLS. It's also convenient in a couple of places (e.g.
HlsDownloader), where a list of all media playlist URLs is needed.
- Lots of places where HlsUrl was passed only needed the actual
URL (not the Format, which is the other piece of HlsUrl). Passing
just the URL is a little simpler, and resolves some of the naming
confusion.
Issue: #5596
Issue: #2600
PiperOrigin-RevId: 240970466
This is to make it possible to use equality of HlsUrl.url fields
to determine whether two HlsUrls point at the same media playlist.
This doesn't work currently because it's possible to mix absolute
and relative urls, which will not be equal until after the relative
url is resolved against the playlist baseUrl.
Issue: #5596
Issue: #2600
PiperOrigin-RevId: 240966503
Also make combination of language and role flags the default over the label,
as it's more readable and auto-translated into other languages.
PiperOrigin-RevId: 240801070
Custom factories currently need to copy most of the code in the
AdaptiveTrackSelection.Factory to make use of the same functionality.
Simplify this by allowing to overwrite the factory.
PiperOrigin-RevId: 240773271
This is the first of a series of cls that will convert DownloadManager
threading model as public methods post actions on an internal thread
which then do the work.
PiperOrigin-RevId: 240743400
Still skipping any renditions that have a null URL, which means that
closedCaptions in particular is always empty. This restriction will
be removed in a subsequent change, which will also remove muxedAudioFormat
and muxedAudioCaptions from HlsMasterPlaylist and instead derive them
from the Variant and Rendition information in HlsMediaPeriod.
Issue: #5596
Issue: #2600
PiperOrigin-RevId: 240623500
The one functional change here is to require subclasses to implement getForegroundNotification. It seems easier and less error prone to
force developers to implement it than it does to document the cases
in which they need to override it (which is most cases).
PiperOrigin-RevId: 240566737
Issue: #5596
This arguably made a worse problem than it solved, and also I think subtly broken HLS
downloads by changing which renditions HlsMasterPlaylist.copy would retain in some
cases. Reverting as a mini step toward removing smart logic from the HLS playlist parser.
*** Original commit ***
Expand check for muxed audio media tags to include uris that match variants
Issue:#5313
***
PiperOrigin-RevId: 240348969
If enabled, DefaultTrackSelector returns multiple groups with more than one
track. AdaptiveTrackSelection then decides on the order in which the tracks
are changed such that the BandwidthProvider for each selection can figure out
the allocated bandwidth.
PiperOrigin-RevId: 240150206
It's currently very difficult to actually set a reason for track selection
triggered by a SelectionOverride. It would require creating a new custom
TrackSelection.Factory where the reason gets injected somehow before the
override is specified on the player.
Simplify this whole procedure by allowing to set the reason directly and
forward this reason to all fixed selections created based on this override.
PiperOrigin-RevId: 240114942
3 ("11") is the reserved code for fscod in AC3. The spec says that "If the
reserved code is indicated, the decoder should not attempt to decode audio and
should mute."
We currently throw an exception as we try to access an array with an invalid
index. Now ignoring the track instead by invalidating the format.
Issue:#5638
PiperOrigin-RevId: 240106255
This allows to distinguish between regional variants and scripts.
We still need to normalize the language code itself to make track selection
independent of the whether 2 or 3 letter codes are used.
PiperOrigin-RevId: 239783115
Any seek map with non-zero offsets breaks playback with ICY metadata as
the metadata is no longer read from the correct position.
Issue:#5658
PiperOrigin-RevId: 239605839
*** Reason for rollback ***
Trying to forward fix instead.
*** Original change description ***
Automated g4 rollback of changelist 237051112.
*** Original change description ***
Add SimpleCache deletion functionality
This is needed now that index data may be stored outside
of the cache directory.
***
***
PiperOrigin-RevId: 239429430
*** Original change description ***
Add SimpleCache deletion functionality
This is needed now that index data may be stored outside
of the cache directory.
***
PiperOrigin-RevId: 239294713
When an error occurs we call stopInternal, and this clears the MediaPeriodQueue,
which in turn releases media period holders and notifies that media periods have
been released. AnalyticsCollector updates its information about media periods
using the media period release events, which means that if we post the source
error after stopInternal posts its events we can't determine what media period
the source error corresponds to.
Move error notifications before calling stopInternal, so that
AnalyticsCollector's model of the media period queue contains the loading period
at the point when it handles the error.
For consistency also move the other (non-source) error notifications to match
the new ordering.
PiperOrigin-RevId: 238559324
This simplifies the DownloadHelper code in the demo app and is generally
useful for more specific updates with SelectionOverrides.
PiperOrigin-RevId: 238228276
This replaces the deprecated usages of RobolectricTestRunner and
RuntimeEnvironent and fully migrates the tests to androidx.
PiperOrigin-RevId: 238011667
We may currently throw if the load completes before we finished preparation.
Don't throw and instead check for this condition in maybeThrowPrepareError.
PiperOrigin-RevId: 237237000
These are thrown for non-IOException encountered during loading. Also, they
are thrown from an unexpected position and retrying is likely to fail anyway
because the Extractor is left in an unrecoverable state.
PiperOrigin-RevId: 237043948
SQLiteException is a runtime exception, which makes it easy to
forget to handle it. This change converts SQLiteExceptions into
a checked exception, which is then handled appropriately.
PiperOrigin-RevId: 237038793
Currently a flags-only DecoderInputBuffer may hold only the end-of-stream
flag. This change makes flags-only buffer read any kind of flag set in
the next sample. If no sample is available RESULT_NOTHING_READ is
returned.
PiperOrigin-RevId: 237027581
nalPrefix.readUnsignedIntToInt() will throw if the unsigned output doesn't fit
in an int. Before this change, in the error case we'd retry reading from after
the NAL prefix position. Throw a ParserException so that reading an obviously
invalid NAL prefix will be treated as a fatal exception by the default load
error handling policy.
PiperOrigin-RevId: 236839690
There are 6 different situations which are now more explicitly covered to
ensure we handle them correctly:
1. Radio button (non-adaptive, single group only)
Selected -> Do nothing.
Unselected -> Replace override with selected track.
2. Check box (non-adaptive, but multi-group allowed)
Selected -> Remove track (and override). If last one, set isDisabled flag.
Unselected -> Replace override with selected track.
3. Check box (adaptive allowed, single or multi group)
Selected -> Remove track. If last, remove override and maybe set isDisabled.
Unselected -> Add track to override.
PiperOrigin-RevId: 236668642
The release needs to post to the MediaSource thread and also needs to post
back to the DownloadHelper thread to prevent sending updates after release has
been called. The point where we release the downloadHelper also needs to
dismiss the dialog in case the creation has already been queued before the
release.
PiperOrigin-RevId: 236652309
This is more in line with "forgetting all previous state" as documented for
resetState=True. Also only apply the current position as initial seek position
in this case.
PiperOrigin-RevId: 236623630
We currently throw if a pending seek position was valid when the user issued
it on the app thread, but can't be resolved on the playback thread because the
timeline changed in the meantime. Throwing in this case seems wrong as the
user could not have known about the issue (and the seek position was actually
valid). Also, in other cases where the currently playing period is no longer
in the new timeline, we gracefully use a subsequent period or transition to
ENDED state instead of throwing. So it seems more consistent to transition to
ENDED state as well.
PiperOrigin-RevId: 236274862
Instead just set the currently clicked item as override. If the already
selected button is clicked, the override is reset with the same value.
PiperOrigin-RevId: 236118429
We currently do not show check boxes if the renderer does not support adaptive
switches between tracks in one group. However, we assume check boxes in the
onClick method.
Use same logic for both methods to be consistent.
PiperOrigin-RevId: 236118372
The flag in ExoPlayer.prepare is documented as keeping the current window
index and window position. We are currently using the current period UID and
period position instead. This causes problems when the media source is changed
but the position is not reset.
Using the initial seek position instead ensures we actually use the window
index and position.
Issue:#5520
PiperOrigin-RevId: 236101024
Adding new error types may cause issues when listeners assume a fixed set of
error types and don't handle arbitrary defaults.
Fixing error handling in one case and improving documentation to make people
aware of the issue.
PiperOrigin-RevId: 236093265
While disabled the renderer does not have non-null stream formats. This means
that setting the operating rate could cause a NullPointerException if there was
a codec for reuse.
Check for being enabled/started before trying to set the operating rate. After
the renderer is enabled it should receive a new input format which will update
the operating rate as needed.
PiperOrigin-RevId: 235494384
If the DefaultAudioSink was reconfigured in a way that was compatible with the
previous configuration just after having been reset, we would try to drain audio
processors despite not having an AudioTrack. This could result in a
NullPointerException if speed adjustment was active.
Fix this behavior by only trying to drain audio processors if we actually have
an AudioTrack.
PiperOrigin-RevId: 235355466
- Expose constructor (package private for now, for tests only)
- Add some tests for cache initialization
- Add some TODOs for handling initialization failure
PiperOrigin-RevId: 235188386
*** Reason for rollback ***
Breaks ExoPlayer's gradle build
*** Original change description ***
Migrate Mockito Matchers.{hamcrestMethods} to MockitoHamcrest.{hamcrestMethods}
All methods accepting a hamcrest Matcher in org.mockito.Matchers have been changed
to only accept an ArgumentMatcher. The corresponding methods are now available on
MockitoHamcrest.
For more information see http://go/mockito-2-lsc
Tested:
TAP --sample for global presubmit queue
http://test/OCL:235033293:BASE:234998256:1550793421835:c2ccbbda
***
PiperOrigin-RevId: 235186705
All methods accepting a hamcrest Matcher in org.mockito.Matchers have been changed
to only accept an ArgumentMatcher. The corresponding methods are now available on
MockitoHamcrest.
For more information see http://go/mockito-2-lsc
Tested:
TAP --sample for global presubmit queue
http://test/OCL:235033293:BASE:234998256:1550793421835:c2ccbbda
PiperOrigin-RevId: 235153912
*** Reason for rollback ***
Reverting as this may break playback on other Amlogic devices and/or playback of non-interlaced content.
*** Original change description ***
Add max video size workaround for Amlogic decoder.
The Amlogic awesome decoder reduces the video size of interlaced videos by half
if the internal configuration isn't force reset with new maximum input size
values. The product of these new values must exceed 1920x1088 to force the
reset.
Issue:#5003
***
PiperOrigin-RevId: 234967314
Currently IMA determines that the overlay is obstructing, even
if it's empty. Register it as friendly, which means we're
assuming that anything the apps puts in it is necessary for
playback.
PiperOrigin-RevId: 234963065
HlsSampleStream#read should return end of stream when
there is no mapping for the sample stream, instead of
nothing read. This allows the player to transition to
ended.
Issue:#5524
PiperOrigin-RevId: 234764027
- Use Cache UID for CacheContentIndex and CacheFileMetadataIndex. This
enables SD card swapping for a single device.
- I'm hopeful of finding a way to get the Cache UID to DefaultDownloadIndex
so we can do the same there.
PiperOrigin-RevId: 234662753
Horizontal orientation doesn't make sense. I thought about
overriding the setter to throw something, but it's probably
not worth the extra lines.
PiperOrigin-RevId: 234632530
This bug affects any playlist that uses initialization segments. In practice,
almost exclusively fragmented mp4 segmented playlists are affected.
The bottom line is that extractors are chosen for reuse after the
initialization segment connection is open. However, reused extractors
do not need re-parsing the init segment, so loading the initialization is
wasteful.
PiperOrigin-RevId: 234479467
TrackSelectionView requires an initialization with a call to init(...).
That's why we shouldn't let the view retain its view hierarchy automatically
as the views won't be backed by data if restored. Instead add a listener
which lets the containing activity/fragment save and restore the state if
needed.
PiperOrigin-RevId: 234152491
This is a temporary fix to prevent NullPointerException exceptions.
Though writing to cache will still fail if the cache folder isn't a real
folder.
There are a few thing we can try:
* The listing might be failing because the cache folder is just created.
We can wait and try again.
* If the cache folder is a regular file, we can try deleting and creating
a folder with the same name.
PiperOrigin-RevId: 234121925
Previously we would get a new AudioTrack and flush all audio processors if any
AudioProcessor needed to be flushed on reconfiguration. This was problematic for
the case of TrimmingAudioProcessor because it could become active or inactive
due to transitioning to a period with gapless metadata or without it (we don't
keep it active all the time because it is wasteful to populate its end buffer
for content that is not gapless).
This change handles the case where we don't need an AudioTrack but
AudioProcessors do need to be flushed. In this case we drain all the audio
processors when next handling data then switch to the new configuration.
This avoids truncation when period transitions change whether
TrimmingAudioProcessor is active but don't require a new AudioTrack, and is also
a step towards draining the AudioTrack when transitioning between periods if we
do need a new AudioTrack.
To do this, it needs to be possible to drain any pending output data from an
AudioProcessor after it's configured to a new format, so this change makes sure
AudioProcessors allow calling playToEndOfStream and getOutput after
reconfiguration and before flush.
PiperOrigin-RevId: 234033552
This cleanup is in preparation for draining audio processors on reconfiguration
to a format that doesn't require a new AudioTrack to be created.
PiperOrigin-RevId: 233990983
Stop encoding/decoding presentation time as part of the message.
What's actually in emsg boxes is a presentation time delta,
which is why it's only 32 bits, and hence why it doesn't handle
large absolute timestamps. We were using this field to hold
absolute timestamps only for the purpose of passing presentation
times from DashManifestParser.parseEvent back to the calling
method. After this change, we return Pair<Long, EventMessage>
instead.
Issue: #5490
PiperOrigin-RevId: 233561731
builder.mActions is marked as @RestrictTo(LIBRARY_GROUP). We'll
just have to put up with notification flicker on KitKat if the
actions change.
PiperOrigin-RevId: 233420076
Switch from passing an ad UI ViewGroup to passing an object that can also
provide information about controls overlays.
Also switch to using a dedicated overlay for ads instead of the overlay frame
layout, which apps have easy access to.
PiperOrigin-RevId: 233393500
Supporting multiple overrides allows to select tracks from multiple groups, if
enabled. As more options are added, the creation of the dialog is moved to a
separate builder class.
PiperOrigin-RevId: 233366282
Imported from GitHub PR https://github.com/google/ExoPlayer/pull/5438
Plus the following changes:
- Only support profile 5 (handling other profiles requires
backward-compatibility changes in the renderer which are left for a later
change.)
- Only add KEY_PROFILE to the codec configuration MediaFormat for Dolby Vision.
- In MediaCodecUtil support all DV profiles that Android has constants for. This
includes ones that are "not supported for new applications". Since we don't
extract these profiles, this is currently only for the benefit of custom
extractors.
- Misc code style fixes and reordering for consistency.
Merge 37878b975c2bc082b0568e21cbe62bfcef97c10d into 67be9e7783
PiperOrigin-RevId: 233066799
Given the change to require setPlayer on AdsLoaders, it seems like a good
opportunity to clean up deprecated ads-related symbols.
PiperOrigin-RevId: 233020171
When SimpleCache uses a CacheFileMetadataIndex, it will be able to avoid
querying file.length() and renaming files, both of which are expensive
operations on some file systems.
PiperOrigin-RevId: 232664255
Now we're reusing the builder, it's necessary to set maxProgress
back to 0 to avoid the progress bar appearing for end state
notifications.
PiperOrigin-RevId: 232520007
DefaultDownloadIndex is currently just forwarding all calls to its
inner class. getDownloadTable() handles initialization, but this
doesn't really seem different to having an initialization method. It
doesn't guarantee initialization happens, since a bad method
implementation could try and access downloadTable directly, just as
a bad method implementation could forget to call the initialization
method. Hence this change removes the indirection.
PiperOrigin-RevId: 232482228
This combines all the different listeners in ExoVideosPlayerV2 into one and
moves the PlaybackLogger class towards ExoPlayer's AnalyticsCollector with
the same purpose.
In the future this allows two things:
1. Gradually move LogginClient implementations to AnalyticsListener to
prevent custom event forwarding.
2. Using ExoPlayer's QoE extension (which is also an AnalyticsListener).
PiperOrigin-RevId: 232321182
When calling releaseSource(), all pending messages will be removed. That means
that all action-on-completion callbacks which are somewhere in flight are
just dropped without being called. This change adds code to keep track of the
current state of each callback to allow all of them being called when the
source is released.
Issue:#5464
PiperOrigin-RevId: 232312528
This change enables transitioning to/from different Storage
implementations, to allow experimentally enabling (and if
necessary, disabling) SQLiteStorage. All that's left to do
is the final wiring to turn it on
PiperOrigin-RevId: 232304458
If there is data after the last samples in the container, we may request
continue loading after the last samples have been read but before the load has
completed. In this situation the buffered position is returned as
Long.MAX_VALUE, which prevents continuing loading, yet the media period is not
treated as fully buffered because its buffered position is not
C.TIME_END_OF_SOURCE.
PiperOrigin-RevId: 231406964
We can keep the reading media period and continue playing if we haven't read
beyond the new duration. Otherwise, we can keep the period, but need to reset
the renderers as we already read too far.
PiperOrigin-RevId: 231406252
It better describes what the class does. More importantly, we've had
inconsistent class names since we added offline support, for which we
added ProgressiveDownloader ("ExtractorDownloader" doesn't make any
sense). We could really do with aligning the names for clarity.
(Sorry)
PiperOrigin-RevId: 231387268
This requires to prepare the media source and the periods in a small helper similar
to the metadata retriever. It also gets rid of the need to have abstract protected
methods to load the manifest, to extract the track groups and to convert to stream keys,
as this can now be handled by the media period.
PiperOrigin-RevId: 231385590
Currently, we remove all variants if none of the stream keys contains any
variants. This causes HlsMediaPeriod to throw exceptions as it expects at least
one variant.
Change it to support master playlists without variants.
PiperOrigin-RevId: 231385547
Converting a single track selection to stream keys is only possible if the output
is independent from other track selections being made.
This is not the case for DASH and HLS embedded track groups which should select the
already selected primary track if possible (and thus needs to know whether a primary
track group is selected).
Also, update the test method to take a period index.
PiperOrigin-RevId: 231385490
The way it is currently, it's very unclear that an operation on the
version table will correctly belong to a transaction in code such as
this, taken from DefaultDownloadIndex:
writableDatabase.beginTransaction();
try {
writableDatabase.execSQL(...);
versionTable.setVersion(...);
writableDatabase.setTransactionSuccessful();
} finally {
writableDatabase.endTransaction();
}
This change explicitly passes the database, to make it obvious that
the operation will really go into the same transaction:
writableDatabase.beginTransaction();
try {
writableDatabase.execSQL(....);
VersionTable.setVersion(writableDatabase, ...);
writableDatabase.setTransactionSuccessful();
} finally {
writableDatabase.endTransaction();
}
PiperOrigin-RevId: 231374933
Float output is only possible from API 21, but the high-res int
to float conversion path was checking for 32-bit PCM not float
output capability.
PiperOrigin-RevId: 231172495
This will allow CachedContentIndex to start using database
storage without us having to change the SimpleCache and
CachedContentIndex constructors to require a Context or a
DatabaseProvider.
PiperOrigin-RevId: 230884501
If are allowing changing durations of periods, we shouldn't use the end position
of clipped content as part of the id as it may change. This change moves the end
position back to MediaPeriodInfo and adds the next ad group index to the id
instead to ensure we still have unique ids for all content parts.
PiperOrigin-RevId: 230878389
The combination of pre-16 API levels accounting for ~0.5% of the device
population, and that the most important components in ExoPlayer (e.g.
the MediaCodec renderers) have always required API level 16, mean it's
very unlikely this will negatively impact on anyone.
PiperOrigin-RevId: 230701808
We need to support serialization to/from an SQLite table. The
model of passing something around for each class to write into
doesn't work well for SQL, and it would be messy to have two
different structural designs for serialization. This change
centralizes the logic in CachedContentIndex, where a centralized
SQL based version can more easily sit alongside it.
PiperOrigin-RevId: 230692291
This method is a generalization of the existing hasReadStreamToEnd. It is
useful to determine whether a renderer already read beyond a new duration of
a period.
PiperOrigin-RevId: 230689165
1. The method kept track of the current period index to check if the next
period is still in the correct period. This is unneccessary since we no longer
use the period index but the actual uid in MediaPeriodId and mismatches are
already detected by canKeepMediaPeriodHolder.
2. We updated the MediaPeriodIndfo twice: once in getFollowingMediaPeriodInfo
and once in getUpdatedMediaPeriodInfo. That's confusing and difficult to
follow. The only difference is that getUpdatedMediaPeriodInfo keeps the
content position while getFollowingMediaPeriodInfo resets it. This is made more
explicit for readability.
3. The durations compatibility check for all following periods was broken as
it compared the same durations (partly due to the confusion caused by 2.)
PiperOrigin-RevId: 230519295
* Add AC-4 MIME type definition
* Add AC-4 format support in Mp4Extractor and TsExtractor
* Add AC-4 Extractor
* Add AC-4 playback support in MPEG-4, MPEG-DASH, TS and HLS
ExoPlaybackExceptions of type SOURCE are always associated with the loading
period and thus we can use the event time for the loading period in
onPlayerError. Renderer and unexpected exceptions are still associated with the
currently playing period.
Issue:#5407
PiperOrigin-RevId: 230216253
The Amlogic awesome decoder reduces the video size of interlaced videos by half
if the internal configuration isn't force reset with new maximum input size
values. The product of these new values must exceed 1920x1088 to force the
reset.
Issue:#5003
PiperOrigin-RevId: 230206675
That's the same position set in MediaPeriod.prepare (where it may be removed
in the future).
Having the position at an earlier point is necessary to fix an
issue with lazy preparation in ConcatenatingMediaSource where the prepare
position was assumed to be known but MediaPeriod.prepare hasn't been called
yet.
Issue:#5350
PiperOrigin-RevId: 229756637
[Problem]
CC1 is shown even when CC2-3-4 is selected
[Solution]
* use incoming channel and frame information
* Misc Codes need to handle use Frame information
* Add checks everywhere
[Test]
* Live channels and already existing test should not show any difference
in behavior when CC1 is selected.
* Live channels might have CC2-CC3-CC4
* Sarnoff test content has specific test cases for channel support
When the extra adaptation set of a switch group isn't defined in the manifest, we
currently assume it's the first adaptation group. This either leads to wrong grouping
or duplicate track groups.
Such a case may easily happen if the manifest is filtered such that only one of the
switch adaptation sets will be present in the manifest.
PiperOrigin-RevId: 229365379
The buffered position is currently based on the mimimum queued timestamp of
all AV tracks. If the tracks have unequal lengths, one track continues loading
without bounds as the "buffered position" will always stay at the shorter
track's duration.
This change adds an optional buffer flag to mark the last sample of the
stream. This is set in the Mp4Extractor only so far. ExtractorMediaSource
uses this flag to ignore AV streams in the buffered duration calculation if
they already finished loading.
Issue:#3670
PiperOrigin-RevId: 229359899
That was previously handled by the player. But since we switched to Handler
messages instead of player messages, we should do that manually.
PiperOrigin-RevId: 229341747
This is not necessary as the track selection needs to be updated with
updateSelectedTrack anyway. It's also error-prone as the selection code
calls into a protected method of a not fully initialized class.
PiperOrigin-RevId: 229331669
DataSpec.FLAG_ALLOW_CACHE_FRAGMENTATION is added to indicate to the
cache when fragmentation is allowed. This flag is set for progressive
requests only.
To avoid breaking changes, CacheDataSink defaults to ignoring the flag
(and enabling fragmentation) for now. Respecting the flag can be
enabled manually. DownloaderConstructorHelper enables respecting of
the flag.
Issue: #4253
PiperOrigin-RevId: 229176835
We currently forget whether a source is seekable at re-preparation. This was
implemented intentionally this way under the assumption that we really can't seek
until we have loaded the seek map again. However, seek operations are only
allowed after a media period is prepared. So there is no harm in remembering
whether a source is seekable.
This problem currently prevents reusing ClippingMediaSources with
ExtractorMediaSource and a non-zero start clip position.
Issue: #5351
PiperOrigin-RevId: 229169441
ExoMediaCrypto.requiresSecureDecoderComponent() is removed, and
FrameworkMediaCrypto.forceAllowInsecureDecoderComponents is made
public to allow determining whether a secure decoder is required
to be implemented in MediaCodecRenderer.
PiperOrigin-RevId: 228909771
There were some edge cases in which we'd forget to release DRM
sessions. For example if we read a format and acquired a
pendingDrmSession (in onInputFormatChanged), then immediately
read another format and overwrote pendingDrmSession, we'd
forget to release the one that's been overwritten.
This change hopefully makes release much clearer. We keep a list
of all drm sessions we're currently holding. Whenever we update
either drmSession or pendingDrmSession, we release any other
sessions that are in the list.
PiperOrigin-RevId: 228905465
The window object is used without being filled with data. This used to work
well for most cases as the same live stream is sending regular updates and the
first update is almost never used if it's not the first item in a playlist.
It causes problems when the first timeline update of a live stream is actually
used for playback (e.g. when the live stream is lazily prepared in a playlist
and played first).
PiperOrigin-RevId: 228530232
Reported in https://github.com/google/ExoPlayer/issues/3860
For failing examples see the github link above.
[Problem]
We drop matching control codes even if they are not received on
consecutive frames.
The specification says
"(4) If the first transmission of a control code pair passes parity,
it is acted upon within one video frame. If the NEXT frame contains
a perfect repeat of the same pair, the redundant code is ignored."
Keyword is the NEXT. The frames must arrive immediately after
each other.
See https://www.law.cornell.edu/cfr/text/47/79.101
[Solution]
Set an additional flag when any data is processed. Control code
duplication checks should be limited only for the first control
byte pairs processed after any control code.
[Test]
Sarnoff tests have equivalent CEA708 and CEA608 Streams.
This is the initialization part of mitigating issue #4253. The
remaining work is on the writing side, and is simply a case of
having startFile return File instances that are sharded into
sub-directories. We still need to decide what scheme we want
to use for doing that.
Issue: #4253
PiperOrigin-RevId: 228306327
Calls to File.length() can be O(N) where N is the number of files
in the containing folder. This is believed to be true for at least
FAT32. Repeated calls for the same file tend to be faster,
presumably due to caching in the file system, however are still
surprisingly expensive. Hence minimizing the number of calls is
preferable.
Issue: #4253
PiperOrigin-RevId: 228179921
They are not longer needed anywhere, are error-prone (because of threading
requirements), and complicate testing and using MediaSources without a player.
PiperOrigin-RevId: 227871157
We currently either use the app thread returned by the player or the thread
the commands are called on depending on whether the media source is already
prepared or not.
This change lets the application decide which callback thread to use. As a
side effect, we also don't longer need access the player instance passed to
MediaSource.prepare.
PiperOrigin-RevId: 227871111
Also configure the FFmpeg context to ignore errors as far as possible (this
appears to have an effect only for certain decoders).
Issue: #5293
PiperOrigin-RevId: 227851397
The top level requirement only tried to ensure that the entire timeline only
has one period. This is already asserted by ImaAdsLoader. AdsMediaSource
itself works fine as long as the wrapped timeline has one period only. This
is now asserted instead.
PiperOrigin-RevId: 227682141
Passing the player through MediaSource.prepare is only needed for the AdsLoader
and complicates other usages of MediaSource. Providing the player directly to
the AdsLoader is also in line with the usage pattern of PlayerView and other
components.
Also rename methods to start/stop to better reflect their usage.
PiperOrigin-RevId: 227682112
Made DownloadState top level class.
Replaced action field DownloadAction fields.
Added removing, removed and restarting states.
Renamed started state to downloading.
PiperOrigin-RevId: 227664735
ExoPlayer methods must not be called from any thread besides the specified
app thread. Therefore we shouldn't use them here. Using a regular Handler
instead is fully equivalent.
Issue:#5240
PiperOrigin-RevId: 227650489
- Increase the default cache file size to 5MB
- Recommended a minimum cache file size of 2MB to discourage
applications from specifying values small enough such that
unreasonably large numbers of cache files are generated
- Allow maxCacheFileSize=C.LENGTH_UNSET, equivalent to setting it
to MAX_VALUE. This is just for API consistency with other APIs
we have that accept LENGTH_UNSET
Issue: #4253
PiperOrigin-RevId: 227524233
CacheEvictor.onStartFile currently receives a length, which is the
maximum size of the content that might be written. The LRU cache
evictor will make sure there's sufficient space for the specified
size. If the actual size of the content is unknown, CacheDataSink
passes maxCacheFileSize.
The current behavior isn't ideal. It seems valid for a developer to
specify maxCacheFileSize=Long.MAX_VALUE if they don't want any
file fragmentation in the cache. However if they then attempt to
cache something with unknown length, LRU cache will try and make
room for content of length Long.MAX_VALUE, and end up evicting the
entire cache to do so.
This change alters the logic so a length is only passed if the
actual size of the content is known. In other cases C.LENGTH_UNSET
is now passed. The LRU evictor will not evict until after the file
is committed. Note a custom LRU evictor could still opt to evict to
ensure some application specified amount of space, if that's the
desired behavior.
PiperOrigin-RevId: 227509525
This also applies when seeking or transitioning to unprepared media, which
isn't clear from the current documentation.
Issue:#5267
PiperOrigin-RevId: 226486685
* Add AC-4 MIME type definition
* Add AC-4 format support in Mp4Extractor and TsExtractor
* Add AC-4 Extractor
* Add AC-4 playback support in MPEG-4, MPEG-DASH, TS and HLS
When a Download is created it's set to queued state but doesn't notify
listeners about this state. DownloadManager checks if it can start the
download. After this checks it notifies the listeners.
With this change Download can immediately check if it can be started and
sends correct notification.
PiperOrigin-RevId: 225967129
Problem: There is nothing in the Cache interface that enforces that
these methods are implemented by reading/writing content metadata,
however our own code (CacheDataSource) relies on this property since
it mixes use of [get|set]ContentLength and use of metadata. Using
metadata is the right approach because it minimizes the number of
index writes (e.g. it allows the redirect URI and the content length
to be committed by a single write).
Solution: Remove [get|set]ContentLength, which are redundant anyway,
and use metadata everywhere. Also expose the keys being used through
the public API; there's no particular reason why they shouldn't be.
We previously had an API for getting and setting content length, but
no API for getting and setting redirect uri.
PiperOrigin-RevId: 225383211
This adds the basic track selection capabilties (including tests).
The new capabilities are not exposed yet through the DownloadHelper implementations
and there will also be more helper methods (e.g. to select multiple audio lanuages at
once).
PiperOrigin-RevId: 224518477
We currently have two factory methods where it is completely unclear which one needs
to be overridden.
This change deprecates the old one, adds a Util method to easily map back from the new
to the old behaviour, and updates all implementations of the now deprecated method in
our code.
PiperOrigin-RevId: 224303560
[Problem]
As the alignment of each lineas are calculated individually, the
artificially introduced CENTER and RIGHT alignments can be mixed
with each other and with the default LEFT alignment. The mixed results
look worse than just using the original DEFAULT, and the readability
is radically decreased.
[Solution]
Pre-calculate the alignment for each line, and use the leftmost of them
for all lines. The intention of caption provider could be different, but
the worst case scenario is that we revert back everything to LEFT
alignment that is the only option in the CEA608 standard so we
show everything at the exact location it is set in the content.
The readability is never worse than the one given by the content
provider.
[Test]
This should influence CEA608 pop-up cations only. (Not roll up).
Test various live channels with this change.
Note: Beginning of commercials tend to have PAINT-ON mode as that
is immediately showing incoming characters, not just at the end
of the line like POP-on mode.
This instantiates the renderers and extract the capabilities. None of the known
renderes incurs any overhead during instantiation.
PiperOrigin-RevId: 224118511
- Enable GZIP for media playlist + encryption key chunk requests in
HLS, as we do during playback
- Pass around DataSpecs rather than Uris. This will be needed for if
we add manifest cacheKey support (which seems like a good idea for
completeness, if nothing else)
PiperOrigin-RevId: 224057139
We currently default to not caching data if the content length
cannot be resolved once the DataSource has been open. The
reason for this is to avoid caching progressive live streams.
By doing this we were accidentally not caching in other places
where caching is possible, such as DASH/SS/HLS segments during
playback if the server doesn't include a Content-Length header.
Also HLS encryption key chunks, which were very unlikely to be
cached during playback because we explicitly set FLAG_ALLOW_GZIP
(which normally stops content length from resolving) without
setting FLAG_ALLOW_CACHING_UNKNOWN_LENGTH.
It seems like a good idea to flip the default at this point,
and explicitly disable caching in the one case where we want
that to happen.
PiperOrigin-RevId: 223994110
This is in line with how Player.EventListener and AnalyticsListener methods are
defined and helps to only implement the callbacks needed.
PiperOrigin-RevId: 223991262
Moving most of the logic to the base DownloaderHelper helps to implement track
selection for downloading in a single place instead of multiple places.
PiperOrigin-RevId: 223964869
This option to block bandwidth already exists on the AdaptiveTrackSelection itself
but it's not currently possible to forward the total fixed track bandwidth
automatically.
PiperOrigin-RevId: 223785139
- Asus ZenFone GO (ASUS_X00AD_2)
- Sugar S9 (i9031)
- Redmi Note 3 (kate)
These devices trigger native crashes similar to
https://github.com/google/ExoPlayer/issues/4460
I'm not sure why Asus Zenfone Go (model: ZB500KL, device: ASUS_X00AD_2) was removed here
73af056da3
PiperOrigin-RevId: 223580393
A remove action will eventually just be the unique ID of the download
that should be removed. This is a step toward that goal.
PiperOrigin-RevId: 222832456
[Problem]
PAINT-ON mode is not implemented. From the compliance tests:
* RDC command has no effect except to select paint-on style.
* Next data are written directly to the display upon receipt.
* If other captioning is already on the screen, the four-row
limit is still in effect.
[Solution]
It is a rare use case, we do not support overriding characters in
existing cueBuilders as PAINT-ON would require. But several
compliance tests check if the screen is cleared when the mode
switch happens. We must keep the old captions when switching
to PAINT-ON mode
[Test]
- Live Over-the-Air content, beginning of commercials often uses
PAINT-ON mode
- Add DownloadAction.fromByteArray for symmetry with toByteArray
- Make DownloadService call fromByteArray, for symmetry with calls
to toByteArray
PiperOrigin-RevId: 222801703
We already have tests for comment blocks, and they already
pass because we discard the comment when we fail to parse it
as a cue. We should just skip it directly, however.
The input.getLength() check is invalid because the length may be
unknown (i.e. if the server doesn't include a Content-Length
response header when serving chunks).
Issue: #5063 (tangentially related only)
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222406347
This allows more efficient range removals and is consistent with addition,
which supports adding multiple elements in a single operation.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222396310
This allows to account for bandwidth used by fixed track selections.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222383736
BandwidthProvider allows bandwidth allocation logic to be customized.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222293581
1. Pull up all subclasses of DownloadAction into DownloadAction
2. Add DownloaderFactory for Downloader instantiation, and DefaultDownloaderFactory
to replace the instantiation logic being removed from the DownloadAction
subclasses.
This change will upgrade existing action files gracefully (i.e. it does not
lose compatibility with the existing offline implementation, other than some
minor breaking changes to the API).
TODOs:
1. Move test methods from the XDownloadActionTest classes into DownloadActionTest.
This will be done in a subsequent CL. There's a lot of consolidation that can
be done here, including de-duplicating some of the test code added in this CL.
2. Look at merging DownloaderConstructorHelper into DefaultDownloaderFactory.
3. Use customCacheKey in DASH/HLS/SS Downloaders, for completeness.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222258983
createTrackSelections decides whether to create an adaptive or a fixed track seletion to create.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222231011
Forwarding the listeners to the notification update is meant to ensure we
only notify the listeners which were registered at the time the event happened
However, we currently just copy the reference to the actual list instead of
doing a deep copy of the listeners.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222227735
This ensures that all callbacks use the same notification mechanism.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222226913
The update is actually unrelated and putting it all in one large update
object complicates adding the same update mechanism for other values.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222226825
WindowedTrackBitrateEstimator, uses future and past chunks in a window
of time to estimate track bitrate values.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=222065409
Currently no path may be chosen if
minBufferUs == maxBufferUs == bufferedDurationUs.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221639199
This is needed because the allocator can't be reused for example.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221638233
- Increase the search window size to fix TS seeking for problematic
media we've had provided to us.
- As per my comments on the issue, we should look at doing more here
to better fix the problem. This will solve the worst of the
immediate problem, however.
- The memory usage is non-trivial, particularly with the increased
search window size. I've made the allocations only live whilst
determining duration and seeking to address this. I've done the same
for PS just for consistency.
Issue: #5097
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221449988
The remaining work is to split Window.isDynamic so that it's
possible to represent a window that wont be appended to, but
may still be removed from.
Issue: #4780
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221439220
This is a precursor for fixing the ref'd issue. These classes are
well tested, so the tests passing should give you reasonable
confidence I didn't break anything :).
Issue: #5097
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221435824
LegacyVideoTrackSelection is the previous implementation (unchanged).
VideoTrackSelection is rewritten to extend AdaptiveTrackSelection.
An experiment is added to allow use of the new VideoTrackSelection
implementation. It's envisaged that this transition will be more or
less a no-op change. The experiment flag is only in case of bugs
during roll-out, and will be removed with LegacyVideoTrackSelection
upon successful transition.
VideoTrackSelection also implements discard-on-trick-play-exit, which
was a TODO in the previous implementation.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221153827
- This brings it in line with PlayerControlView. The play action is displayed
instead, and pressing it seeks to the default position of the current window.
- Do the same for MediaSessionConnector
- Add support for PlaybackPreparer consistent with PlayerControlView
Issue: #5072
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221076008
This change also paves the way for splitting out functionality into a utility class.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=221070262
When in REPEAT_MODE_ONE, it's unlikely apps want next/previous methods
on the player to keep them in the same window. Music apps in particular
tend to implement next/previous functionality as though repeat mode were
off when in this mode (i.e. current song loops forever during playback,
but next/previous navigation still navigates to next/previous items).
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=220469655
This fixes an issue where disabling audio focus handling
while audio focus is held would not release audio focus.
A new test was added for this situation.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=220316866
*** Reason for rollback ***
Rolling forward again as [] should fix issue that prompted the rollback
*** Original change description ***
Automated g4 rollback of changelist 219130576.
*** Original change description ***
Re-enable codec re-use
Issue: #2826
***
***
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=220124362
It's no longer guaranteed that onOutputFormatChanged will always be
called when the renderer is re-enabled, since the codec may be kept
and its output format may not change. Hence we need to avoid resetting
state in onDisabled that we need when the renderer is enabled again.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219821519
Currently, MediaCodecRenderer's render loop does not have any time limit. We
always try to drain and feed as much buffers as possible. This may lead to a
pattern of feeding phase takes too much time from previous loop, which makes
the next draining phase drops all buffers, making these buffers available for
feeding all at once, and the pattern keeps repeating.
This CL adds an experimental parameter for setting a time limit. If the time
limit is exceeded, the feeding process stops even if more input could be fed.
The default behavior is unchanged
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219820386
This makes extractor selection a bit more efficient for some CMAF files.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219795105
This prevents leaving an inconsistent state after a EOF exception.
Issue:#5039
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219785275
Before this change, the player state would be STATE_ENDED then
STATE_BUFFERING (when the postroll ad was marked as played) then
STATE_ENDED again.
Queueing a final content media period with start point equal to
the content duration after a postroll ensures that the player
state doesn't change to STATE_ENDED transiently.
Switch from using C.TIME_END_OF_SOURCE to C.TIME_UNSET for media
periods that should not have an end point and are not followed
by an ad.
Content media periods before postrolls have end position
C.TIME_END_OF_SOURCE. If the postroll ad loads, its content
position is set to the content duration, which should be known
at the point of loading the postroll, then a final 'empty'
content media period with start position equal to its duration
is queued. If the postroll fails to load, this empty content
media period is queued up directly after the preceding content
media period.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219443683
- Handling EVENT_PROVISION_REQUIRED is unnecessary because we handle
it via NotProvisionedException, which was thrown on all API levels.
- Handling EVENT_KEY_EXPIRED during playback is unnecessary. All it
does is cause an error to be thrown, but an error will be thrown
anyway out of the MediaCodec when it tries to use the expired keys.
It's currently a race condition where the error gets thrown from,
where-as having it always be thrown from one place is preferable.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219172021
This paves the way to cleanly fix the first two issues
listed in [] onDisabled will null inputFormat,
but nulling of codecFormat will remain tied to the codec
being released.
Issue: #2826
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=219125458
ExoCastMessage contains all player management messages that the sender app can
send to the receiver app. ExoCastMessages can be serialized as JSON strings.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=218846977
- Only use bitrate when comparing two tracks if their bitrates are
actually different.
- For audio, prefer to use bitrate over selection flags.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=218660886
- Improve variable naming
- In the edge case that bytes are transferred in a sample that
has 0 elapsed time, carry the bytes forward into the next
sample. If we don't do this then the estimate will be calculated
as though those bytes were never transferred. In practice I
suspect this very rarely occurs.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=218551208
If EventDispatcher.removeListener is called to remove a listener,
and if the call is made from the same thread that said listener
handles events on, then it should be guaranteed that the listener
will not be subsequently invoked on that thread.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=218331427
In ConcatenatingMediaSource, the source may be removed before it started
preparing (this may happen if lazyPreparation=true). In this case, we
shouldn't call releaseSource as the preparation didn't start.
Issue:#4986
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=218141658
There are multiple subtle issues with the current implementation:
1. setOperatingRate can cause a codec initialization even if the
renderer is disabled. This is not supposed to happen.
2. If the codec is released whilst the renderer is disabled, the
renderer can instantiate a new codec using the old format when
it's enabled again, only to immediately have to reconfigure or
release it if the actual format to be played is different.
3. Codec reuse does not take into account renderer configuration.
The specific case where this is problematic is if the video
renderer is re-enabled with a different tunneling session id.
The reused codec is then not configured correctly.
Also moved availableCodecInfos reset into releaseCodec for sanity.
Issue: #2826
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217924592
This is a no-op change as the respective values are not used so far but
the change makes the current state cleaner and is less error-prone.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217892421
- Added ability to override the DefaultBandwidthMeter network type.
- These methods currently only work properly if called before the
meter is used, but the plan is for them to be handled in the same
way as internally detected network changes.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217887019
This method only ensured backward compatibility while the new version was not
public yet. Now ExoPlayer 2.9.0. is public, this workaround can be removed.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217881691
I don't think there's any particular reason for releasing
dummySurface when we release the codec. It seems to make
more sense to decouple the two, and treat a dummySurface as
a resource that should be released on reset.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217711788
setSurface only sets the output surface on the codec if the
renderer is enabled or started. This was fine in the past
because the codec was always null in other states, however
the recent codec reuse work means this is no longer the case.
The output surface should always be set if the codec is
non-null. If setting the surface requires releasing the codec
instance, we should only instantiate the new instance if
enabled or started. This is in line with what onDisabled does
if flushing requires releasing the codec
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217676434
Just not doing it seems simplier and more obviously correct
than suppressing the warnings in our proguard file.
Issue: #4890
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217675527
1. Remove option to pass null context to Util.getNetworkType. IMO it's
clearer for the caller to just not call the method in this case.
2. Stop using deprecated DefaultBandwidthMeter.Builder constructor in
all but one place.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217147091
The assertion was so weak it probably wouldn't detect genuine
misuse of the DefaultAllocator API, so it seems fine just to
remove it.
We don't really know what happens when the player is allowed to
continue on the affected devices, but hopefully it either "just
works" or fails in a more graceful way.
Issue: #4532
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217113060
Robustness fix to make sure we ignore tags with known names, but
which are nested inside of unknown tags. For example we don't want
to parse the third period in:
<Period>...</Period>
<Period>...</Period>
<CustomTag>
<Period>...</Period>
</CustomTag>
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=217101630
For decoder reuse, we want disable() to flush the decoder. However,
if the flush needs to release the decoder for some reason, it seems
non-ideal to immediately re-initialize it. Re-initialization can
also throw ExoPlaybackException, which we don't want for disabling.
This change allows a variant of flush that wont re-initialize the
decoder if it has to be released, which will be used from disable().
Issue: #2826
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=216834862
reloading frmo snapshots. b/117347850. This is a no-op change as the current
default == _fullboot. When we globally flip and make snapshots the default,
these tests would fail and we are moving them over so that when we flip your
tests still pass.
https://groups.google.com[]forum/#!topic/mobile-ninjas/Y69PTAT5GyY
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=216634430
- Always clip to period duration for the last chunk. We previously
did this only when the last chunk explicitly exceeded the period
end time. We now also do it when the chunk claims to end at the
period boundary, but still contains samples that exceed it.
- If pendingResetPositionUs == chunk.startTimeUs == 0 but the
chunk still contains samples with negative timestamps, we now
clip them by setting the decode only flag. Previously we only
clipped such samples if the first chunk explicitly preceeded the
start of the period.
Issue: #4899
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215763467
- It's a no-op for now
- Renderers that want to support retaining resources will move
some functionality from their disable() implementations into
reset()
- ExoPlayerImplInternal will be updated to not always call reset()
immediately after disable(), which is what will enable resources
to actually be retained.
Issue: #2826
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215729783
This makes the following changes to improve consistency among the PlaybackInfo
values:
1. Update buffered position and total buffered duration after loading period
is set as both values are affected by a loading period update.
2. Add copyWithPosition to allow updating the position without resetting the
loading period.
3. Forward the total buffered duration to playing position updates
as it may have changed with the new playing position.
Issue:#4899
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215712328
When SubtitlePainter positions the cues centred in the given box, it must add
the left offset of the box to get the correct position.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215535289
SubtitleView forwards the cue box position to SubtitlePainter. This should be
the position relative to the canvas of the SubtitleView. Currently, however,
we forward the position relative to the parent of SubtitleView. That causes
problems if SubtitleView has a non-zero offset position to its parent.
Issue:#4788
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215535281
This removes the experimental bandwidth meter and uses it as the new default.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215404065
A lot of methods just forward to other methods and there is no conceivable
way a player should implement it another way. Moving these methods to a
base player class allows to remove duplicated code across our player
implementations.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215374192
For example, using ExoPlayer.STATE_IDLE falls back to the Player.STATE_IDLE
automatically and thus there is no need to keep the deprecated members.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215363841
If we prepare a deferred media period before the actual timeline is available,
we either prepare with position zero (= the default) or with a non-zero
initial seek position.
So far, the zero (default) position got replaced by the actual default position
(including any potential non-zero window offset) when the timeline became known.
However, a non-zero initial seek position was not corrected by the non-zero
window offset. This is fixed by this change.
Related to that, we always assumed that the deferred media period will the
first period in the actual timeline. This is not true if we prepare with an
offset (either because of an initial seek position or because of a default
window position). So, we also determine the actual first period when the
timeline becomes known.
Issue:#4873
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215213030
These two methods are meant to indicate to the track selection that it's
started or stopped being used. This is helpful to schedule background tasks
related to track selection (e.g. register network change listeners etc.).
This intention is not clearly stated in the method docs.
Also, all track selections of all prebuffered periods stay enabled in
parallel at the moment. As the whole purpose of these methods is to know
whether dynamic updates via updateSelectedTrack may happen, it's better to
only enable track selections of the current loading media period. That's
similar to how we always forward the loading track selections to the
LoadControl.
This change:
1. Improves the JavaDoc of TrackSelection.
2. Disables track selections if loading moves to another period.
3. Reenables track selection if loading moves back to a previous period.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215199987
Switch to the recommended way of checking whether the app is running on a TV
device.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=215177587
Extend Sony workaround up to and including Oreo.
Due to a platform issue the Display API couldn't report 1080p UI
and 4k SurfaceView support until Oreo. Since Oreo it is still
common for devices to misreport their display sizes via Display,
so this change switches to using system properties up to and
including Pie.
On Pie treble may prevent writing sys.display-size so check for
vendor.display-size instead.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=214987203
More specifically, if the parts of the Format that are used
for decoder configuration are unchanged.
Issue: #2826
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=214791234
If a source is removed from the playlist, the player may still call createPeriod
for a period of the removed source as long as the new timeline hasn't been handled
by the player. These events are stale and can be ignored by using a dummy media
source. The stale media period will be released when the player handles the updated
timeline.
Issue:#4871
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=214787090
This simplifies code skipping items in a playlist programatically.
Issue:#4863
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=214580742
This method needs to be called whenever the track selection of the current
loading period changes, but also when the loading period itself (and thus
the "loading track selection") changes. These are the same situations in which
we update the loading media period id and thus we can move both updates in
a common helper method.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=213959982
From API 23 this uses the timed format queue. Before API 23 the
format is notified as soon as the buffer is queued.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=213830729
Before this change we would reset the start trim to zero after initial
configuration (at the start of playback) and after seeking to any position. The
fact that no trimming was applied at the start of playback meant that after the
first period transition we'd see a mismatch between the next buffer timestamp
(equal to the duration of the period taking into account edits) and the duration
of audio submitted to the sink.
This change modifies the behavior so that we reset the start trim to zero only
if some audio was queued since configuration. This is incorrect in the case of
starting playback at a non-zero position, but fixes the common case of starting
at zero. As before, a later seek to any position is handled via a flush and
resets the trim as required.
Transitions from one period to the next are unaffected by this change.
One way to implement start trimming correctly would be to provide the input
buffer timestamp to the audio processors and only trim when handling audio from
the start of the stream, but that is a larger change so left for later.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=213828511
After a period transition the first buffer queued has the sum of previous period
durations added to its source presentation timestamp. These durations take into
account gapless edits, but the check on the timestamp was based on the submitted
frame count, not the frame count after trimming.
This change fixes an issue where audio/video would gradually drift apart due to
accumulated error in the audio track position, which could lead to freezing due
to the audio renderer stopping being ready and switching to the standalone media
clock.
Issue: #4559
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=213819908
