SDK / exoplayer

For Live SmoothStreaming, referential equality checking
isn't enough (it breaks once the manifest is updated).
Updated other instances too just for consistency.

Fix DataSourceInputStream.read() for EOF cases

Issue: #72
Issue: #292
Issue: #302
Issue: #425
Issue: #424

The spurious transitions were caused by calling pushTrackSelection
in onRenderers after changing the renderer building state to BUILT
and before calling player.prepare. pushTrackSelection can cause the
ExoPlayer to generate state changes, since it can call setPlayWhenReady.
This change transitions the renderer building state later, so that
when this happens getPlaybackState correctly masks the state and
returns STATE_PREPARING.

And delete things that we're parsing but don't use from TrackElement.

- This makes HLS consistent with Chunk/ExtractorSampleSource.
- It needs to be consistent for WebVTT in HLS.
- Also trim the allocator when done.

- Have TsExtractor report a SeekMap to the output.
- Implement TsExtractor.reset to reset extractor state.
- Add 1x sample.

Play movies has an Allocator that attempts to allocate a single
huge byte[] up front to minimize the risk of GC pauses. This abstraction
will be required to keep that when updating them to the new Exo.

Making Representation and TrackElement extend Format will simplify
this further (TBC whether this is a good idea!).

Issue: #393

This is in preparation for removing bitrate from MediaFormat.

VP8 can be decoded by MediaCodec (since very early versions of android). Now that we want WebmExtractor to be general purpose, adding VP8 makes sense as it is a common use case.

This CL adds support for parsing BlockGroup elements for all
codecs (not just opus). It also adds a test to verify the new
behavior.

This is in preparation for adding support for AC-3 in HLS.

Also parse E-AC-3 boxes.

Refactor the extractor tests to enable testing of multi track Webm support.

Github Issue: #363

This CL adds support for WebM files which have Cues present at the end of the file (i.e.) after Clusters. The file referenced in the bug can now be played back using the demo app. It adds a new flag to WebmExtractor which should be set to true only when being used through ExtractorSampleSource. All others (e.g. DASH) should not set it.

Reference file: http://demos.webmproject.org/exoplayer/glass_vp9_vorbis_cues_at_end.webm

I've found myself doing this a couple of times during local
debugging. It's harmless to have it public, and seems pretty
useful for debugging inside of the mp4 package.

This is in preparation for adding support for AC-3 in HLS.

Also parse E-AC-3 boxes.

fix for issue #403

Seems a little hacky to define this in Track, but there doesn't seem to
be a particularly nice alternative.

Issue: #406

I was hoping not to avoid this, but in the case of a parsing
failure during preparation we keep retrying forever!