Builds a MediaTimeline from an HLS manifest.
HLS has a single implicit period spanning the full duration.
Audio is modeled per source: a variant that carries muxed audio (no separate
AUDIO rendition group) contributes an audio quality on its own playlist, while
a variant with a demuxed audio group contributes video only — one audio
quality is created per AUDIO rendition in the group, each carrying that
rendition's language and playlist. Modeling every language as its own audio
quality is what lets the timeline's language filter select among them.
Builds a MediaTimeline from an HLS manifest. HLS has a single implicit period spanning the full duration.
Audio is modeled per source: a variant that carries muxed audio (no separate AUDIO rendition group) contributes an audio quality on its own playlist, while a variant with a demuxed audio group contributes video only — one audio quality is created per AUDIO rendition in the group, each carrying that rendition's language and playlist. Modeling every language as its own audio quality is what lets the timeline's language filter select among them.