[Matroska-devel] Clarification on CueRelativePosition

wm4 nfxjfg at googlemail.com
Fri Jul 19 16:18:33 CEST 2013

On Fri, 19 Jul 2013 15:58:58 +0200
Moritz Bunkus <moritz at bunkus.org> wrote:

> Hey,
> On Fri, Jul 19, 2013 at 3:51 PM, Moritz Bunkus <moritz at bunkus.org> wrote:
> >> In my experiments, I determined that the final seek position will point
> >> to either a SimpleBlock, or a Block. And never BlockGroup. As
> >> written by mkvmerge v6.2.0. Maybe I am wrong?
> >
> > That's definitely wrong.
> *sigh*
> Unfortunately you're right that the value mkvmerge writes for the
> BlockGroup case actually points to the Block and not the
> BlockGroup. This, however, is a bug in mkvmerge. Thanks for pointing it
> out.
> I'll add a an entry in my bug tracker and will fix it for the next
> release.
> It is SUPPOSED to always point to either a SimpleBlock or a
> BlockGroup. For SimpleBlock it does, and it's OK, for BlockGroup it's
> not OK.

Thanks. If this gets fixed, it would be best to assign a new ID to the
CueRelativePosition element, so that demuxers don't need compatibility
hacks to check whether the element contains a bad value or not.

I still think this two-step seeking is sub-optimal though, although the
correct semantics are much better than what would have been required to
correctly parse the broken CueRelativePosition element.

More information about the Matroska-devel mailing list