The previous implementation did the following (after skipping any ID3
headers at the start of the stream):
1. Skip forward byte-by-byte looking for a sync word (0xFFF)
2. Assume this indicates the start of an ADTS frame and read the size
a) If frameSize <= 6 immediately return false
3. Skip forward by frameSize and expect to find another ADTS sync word
(with no further scanning).
b) If we find one, great! Loop from step 2.
a) If we don't find one then assume the **last** sync word we found
wasn't actually one, so loop from step 1 starting one extra byte
into the stream. This means we're looking for a sync word we would
have skipped over in step 3.
The asymmetry here comes from the different handling of frameSize <= 6
(immediately return false) and frameSize being 'wrong because it doesn't
lead to another sync word' (scan the file again from the beginning for
alternative sync words).
With this change both these cases are handled symmetrically (always scan
for alternative sync words). Step 2a) becomes the same as 3b): Loop back
to the beginning of the stream with an incremented offset and scan for
another sync word.
#minor-release
PiperOrigin-RevId: 397285756
This was originally reported on #9390. There was a bug that when HLS
loads failed, the player would endlessly retry and never fail with a
player error.
This change fixes a bug in HlsSampleStreamWrapper.onPlaylistError()
which would return true for a playlist whose load encountered an error
but could not be excluded, whereas the method should return false.
Issue: #9390
PiperOrigin-RevId: 397045802
Previously the released preacquired sessions would start their keepalive
timeout, and so no additional resources would be freed in time for the
manager to retry the session acquisition.
This change adds an additional purge of keepalive sessions *after* the
preacquired sessions are released, which fixes the problem.
PiperOrigin-RevId: 396613352
This is was reported in #9257 where the PlaybackStatsListener may try to
access an emtpy ArrayList.
Issue: #9257
#minor-release
#exofixit
PiperOrigin-RevId: 396329373
This is a candidate fix for #8906. As mentioned in that issue,
negative positions within windows might be (kind of) valid in
live streaming scenarios, where the window starts at some
non-zero position within the period. However, negative positions
within periods are definitely not valid. Neither are positions
that exceed the period duration.
There was already logic in ExoPlayerImplInternal to prevent a
resolved seek position from exceeding the period duration. This
fix adds the equivalent constraint for the start of the period.
It also moves the application of the constraints into Timeline.
This has the advantage that the constraints are applied as part
of state masking in ExoPlayerImpl.seekTo, removing any UI flicker
where the invalid seek position is temporarily visible.
Issue: #8906
PiperOrigin-RevId: 395917413
Instead of selecting the base URL initially or when a load error occurs,
it is now selected when a chunk or initialization chunk is created. The
selected base URL is then assigned to
`RepresentationHolder.lastUsedBaseUrl` that is excluded in case of a
load error. For a next chunk another base URL will be selected by using
the `BaseUrlExclusionList`.
PiperOrigin-RevId: 395721221
Since playlist support was added, it's possible for the player to "have media"
and be in STATE_IDLE. The STATE_IDLE documentation therefore became incorrect.
Issue: #8946
#exofixit
#minor-release
PiperOrigin-RevId: 395653716
The C2 MP3 decoder produces an extra output buffer when draining after
end-of-stream is queued. This output buffer has a later timestamp than the last
queued input buffer so we need to calculate its timestamp to detect a stream
change in the correct position.
Before this CL we used the original input buffer timestamp as the largest
queued timestamp, which caused the stream change to be detected at the correct
position because the original input buffer timestamp was slightly larger than
the actual last output buffer timestamp. After this change we use exact
calculated timestamp as the largest queued timestamp. I manually verified
gapless continues to work on a device using the C2 MP3 decoder by comparing
output of the MP3 gapless and MP3 gapless stripped playlists in the demo app,
and that the last buffer timestamp now matches.
#exofixit
PiperOrigin-RevId: 395428928
The current detection logic checks that the two byte terminator starts
at an even position in the ID3 data, where-as it should check that it
starts at an even position relative to the start of the string.
Issue: #9087
PiperOrigin-RevId: 395274934
This typically happens if there's a discontinuity in the stream.
It's better to say we don't know, than it is to return a negative
position.
Issue: #8346
#exofixit
#minor-release
PiperOrigin-RevId: 395224088
StyledPlayerControlView was checking whether the player is an ExoPlayer
instance to set the track selector. This means that, if apps were
wrapping an ExoPlayer in a ForwardingPlayer (to replace a
ControlDispatcher for example), the track selector wasn't set anymore.
PiperOrigin-RevId: 391776305
Issue: #9254
#minor-release
We used to allow only alphanumerical characters in session id. The spec also
allows "$", "-", "_", ".", "+" (RFC2326 Sections 3.4 and 15.1).
PiperOrigin-RevId: 388873742
This change removes ERROR_CODE_IO_NETWORK_UNAVAILABLE,
ERROR_CODE_IO_NETWORK_CONNECTION_CLOSED, and ERROR_CODE_IO_DNS_FAILED
in favor of keeping only ERROR_CODE_IO_NETWORK_CONNECTION_FAILED.
PiperOrigin-RevId: 388715972
- Fix use of getTimestampOffsetUs in TsExtractor where
getFirstSampleTimestampUs should have been used.
- Don't reset TimestampAdjuster if it's in no-offset mode.
- Improve comment clarity
#minor-release
PiperOrigin-RevId: 388682711
For audio-only playlists, when formats are communicated to the app with
AnalyticsListener.onDownstreamFormatChanged(), the passed MediaLoadData
do not indicate this is an audio track and therefore the
PlaybackStatsListener cannot derive audio format-related information.
This change sets the main SampleStreamWrappers track type to AUDIO, if
the master playlist contains only audio variants.
Issue: #9175
#minor-release
PiperOrigin-RevId: 388676060
The AnalyticsCollector releases listeners lazily so that listener
callbacks triggered on the application looper after
SimpleExoPlayer.release() are still handled. The change in ListenerSet
to post the onEvents callback on the front of the application looper
changed (correctly) how onEvents are propagated, however this made
the AnalyticsCollector deliver onEvents with out-of-order EventTimes.
This change fixes AnalyticsCollector to trigger onPlayerReleased() and
the matching onEvents() event in the correct order.
#minor-release
PiperOrigin-RevId: 388668739
This CL addresses the github issue [#8946](https://github.com/google/ExoPlayer/issues/8964). That issue requests support for `font-size` CSS property in WebVTT subtitle format. This CL:
* Adds support for `font-size` property by extending capabilities of WebVTT `CssParser`. Implementation of `font-size` property value parsing is based on the one in `TtmlDecoder`.
* Adds unit test along with test file containing WebVTT subtitles with all currently supported `font-size` units.
#minor-release
PiperOrigin-RevId: 388423859
- Use timestampOffsetUs == C.TIME_UNSET directly as the way of
determining whether the adjuster has determined the offset,
rather than relying on lastSampleTimestampUs checks for this.
- Remove comment referring to lastSampleTimestampUs as holding
the "adjusted PTS". Its value may not have originated from a PTS
timestamp. It's also confusing to refer to it as "adjusted"
given timestampOffsetUs has not been applied to it.
- Fix PassthroughSectionPayloadReader to make sure it'll never
output a sample with an unset timestamp.
#minor-release
PiperOrigin-RevId: 388226180
Prior to this change, an initalized TimestampAdjuster that's then
reset with DO_NOT_OFFSET would incorrectly continue to apply the
offset.
Also add a test case for this issue, and for some other simple use
cases.
#minor-release
PiperOrigin-RevId: 388182645
