Re: [Bug 28557] New: Coded frame removal algorithm may introduce discontinuity at end of current coded frame group

I'm building a player implementation using MSE, testing in Chrome 43. 
Seeking back and forth works well, creating multiple TimeRanges in the 
buffer. But this doesn't remain the case after the frame eviction 
algorithm has run. The SourceBuffer.buffered shows that partial segment 
data has been evicted apparently leaving these areas of the buffer  
damaged (ie. once I have moved the playhead such that the eviction 
algorithm focuses on another TimeRange, I can't append to the first area 
of the buffer that had data evicted), these damaged areas consistently 
appear at every eviction point. When the playhead gets to a damaged 
area, in some cases audio will fail.

Not being fully accustomed to the internals, it would be good to know if 
the problem I describe might be a symptom of the issues you describe 
here, that way I can better track progress.

Received on Thursday, 11 June 2015 18:00:52 UTC