Behdad Esfahbod

Thanks! I like the code to still build with GL2 as well. Any chance you can work on that?

Copying templatization results from https://github.com/googlefonts/colr-gradients-spec/issues/371#issuecomment-1666262521: The savings unfortunately are meager. It saves 2.84% of the COLR table size, which translates to only 0.63% of the total font. This is expected...

For `PaintGlyphSelf` / `PaintGlyphDelta`: ``` $ fonttools colorLib.COLRv1ToCOLRv2 Nupuram-Color.colrv1.ttf 841 glyph paints 1682 layer paints Original COLRv1 table is 27016 bytes Detecting self/delta glyph paints. 2 unique templates for 841...

I think what you are observing is that the viewpoint is always scaled to contain the whole text, no matter what font size you use. Then again, I haven't looked...

FWIW the main problem is this: ``` ./hb-iter.hh:359:1: note: candidate template ignored: substitution failure [with Lhs = hb_filter_iter_t, Rhs = hb_map_iter_factory_t]: no type named 'type' in 'hb_enable_if' operator | (Lhs&&...

I'm also skeptical of `C2PA`. But that aside, from a technical point of view, I like to see at least one implementation before it's put in the spec. Currently it's...

Yeah this one was done intentionally. And is unfortunate. Let me think how to do it the way the others do. It's not as trivial. We need to me right...

Yes I'm aware of this the tragic recent events. Hopefully someone will fork this repo and maintain Saber's legacy. @eliheuer has already expressed some interest.

> > I suppose when valueFormat2 is non-zero, we should record it as "unsafe" to break between the pair even if the values applied were zero? > > The problem...

> Since valueFormat2 of nonzero seems to be rare, I suppose we can mark unsafe-to-break between the second glyph and the glyph after that... https://github.com/harfbuzz/harfbuzz/pull/3874