Behdad Esfahbod
Behdad Esfahbod
Sorry, I quoted the wrong place. The relevant piece is in the `kern` table spec: https://developer.apple.com/fonts/TrueType-Reference-Manual/RM06/Chap6kern.html > Because the stateTableOffset in the state table header is (strictly speaking) redundant, some...
Thanks Andrew. I made another change, can you please test again?
I've lost track of all the requirements. @anthrotype can you take this over by any chance?
Humm. `CFF2ToCFF` still fails with `recalcBBoxes=True`. Is rather hard to fix, since the CFF table has cid names and the rest of the font have the original names...
> Humm. `CFF2ToCFF` still fails with `recalcBBoxes=True`. Is rather hard to fix, since the CFF table has cid names and the rest of the font have the original names... Since...
We really need a CFF lawyer. Maybe I can get help from Ken Lunde.
> Since a CFF2 font from CID-keyed sources doesn't have any recollection of the ROS there's no way to know the CID mapping unless that can be passed along as...
> sets the ROS to Adobe-Identity-1 which isn't a thing What do you mean it isn't a thing?
Oh, I didn't realize there's Adobe-Identity-0 and Adobe-Identity-1. That said, HB code doesn't do anything with the ROS. So it's basically assuming an identity map?