[Webkit-unassigned] [Bug 225630] [MSE] QuotaExceededError Exception not thrown even if the sum of totalTrackBufferSize and appendBuffer size exceeds maximumBufferSize.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=225630

--- Comment #10 from Toshio Ogasawara <toshio.ogasawara at access-company.com> ---
(In reply to Jean-Yves Avenard [:jya] from comment #9)
> To add.
> 
> The issue here is a spec issue.
> 
> We are to reject a buffer only once we know that the source buffer is full.
> The first appendBuffer should always complete.
> 
> https://w3c.github.io/media-source/#sourcebuffer-buffer-full-flag
> "The buffer full flag keeps track of whether appendBuffer() is allowed to
> accept more bytes. It is set to false when the SourceBuffer object is
> created and gets updated as data is appended and removed."
> 
> "buffer full flag" only gets modified to true in the 3.5.1 Segment Parser
> Loop algorithm, step 6.3
> https://w3c.github.io/media-source/#sourcebuffer-segment-parser-loop
> "If this SourceBuffer is full and cannot accept more media data, then set
> the buffer full flag to true."
> 
> On the 2nd call to the appendBuffer, in the Prepare Append algorithm, step
> 3.5.4.6:
> https://w3c.github.io/media-source/#sourcebuffer-prepare-append
> 
> "If the buffer full flag equals true, then throw a QuotaExceededError
> exception and abort these steps."
> 
> This change as such, made the WebKit implementation non-compliant. 
> 
> However, I can understand why this change would be desired.
> 
> I will have a follow-up bug to rework how we handle source buffer size and
> checking capacity. It will make it easier to toggle between the two
> behaviours if needed (one before and after this fix)

Thank you for your comment.
I will consider.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20210607/3eaa8bdb/attachment.htm>