Closed robUx4 closed 1 year ago
I'm confused. Isn't this the same as #105 which I've merged earlier today?
Ah no, there's an additional commit. Sorry!
I verified that the new shorter values for the upper boundaries are properly read with current libebml, using the code found here https://github.com/ietf-wg-cellar/matroska-specification/issues/726#issuecomment-1465207032.
So this MR should be safe.
Seems to work fine with MKVToolNix. Of course a lot of the checksums in my test suite change, but that's to be expected, and the resulting files seem to be processed further just fine as well.
We should have no problem reading them in older version of libmatroska (rather libebml).
Draft on top of #105.
@mbunkus can you check this is working for you, so we can merge https://github.com/ietf-wg-cellar/matroska-specification/pull/733