Audio: 1, 2, 3, -, -, 6, -, -, 9
Video: 1, 2, 3, -, -, -, -, -, 9
Should the packager emit 2 discontinuity tags for audio or just discard timestamp 6 and merge them into 1 discontinuity?
Section 8.15 of the HLS authoring spec says:
"All variants and renditions MUST have discontinuities at the same points in time."
Does this mean that for audio/video and text variants all need to align?
However the documentation for EXT-X-DISCONTINUITY-SEQUENCE seems to imply that a/v may not necessarily have the same count of discontinuities. Please clarify which is correct and how the packager should have emitted the discontinuity tags above.
Thanks!