Adds support for fragmented mp4 with multiple sidx boxes

[anthonybajoua]

Summary:

Addresses Github issue #9373
Implements solution 1 proposed by Google maintainer.

• Added test. Generated test file using below command:

ffmpeg -i short_1080p_videoonly_lowbitrate.mp4 -c copy -movflags frag_keyframe+delay_moov+empty_moov+dash sample_fragmented_seekable_multi_sidx.mp4 (Above movflags flags actually result in an unplayable file on this player until this fix.)

Algorithm/Solution

1. Upon encountering multiple sidx atom in track:
2. For each atom in the track:
   1. If sidx atom then aggregate ChunkIndex into master list, else skip
   2. Merge them all into one large chunk index.
   3. Reset the seek pointer to where it was last
      Retry above on I/O failures.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[colinkho]

Hey @tonihei @icbaker , would you guys be able to take a look? This would help us maintain a vanilla fMp4 extractor!

[icbaker]

We can take a look once the CLA check is passing.

[oceanjules]

@anthonybajoua Hey, a reminder to complete the CLA instruction if you want us to look at your code :)

[colinkho]

Seems like the CLA issue is finally resolved. @icbaker Would you be able to take a look?

[Shell1125]

Details

[rohitjoins]

@anthonybajoua Same with this finally block, why do we need to emit seekMap ( which may be incomplete ), when an IO error occurs?

[anthonybajoua]

This is needed in the case where when processing the final sidx atom in a track, and processRemainingSidxAtoms throws an IOException reading the next atom header.

The issue is when an IOException is thrown, there's no clear way to tell that this is the final sidx atom. Does progressively outputting the seek map upon each IOException pose problems?