Bob Hallissy

> Font Validator should not be raising an error (or warning) if startCode + idDelta is < 0. Rather, it should be computing (startCode + idDelta) mod 65536 to ensure...

Notes: https://docs.google.com/document/d/1kZmZWO_dH8H91Kue0r2-m7ob9X1y31W-KdpHf23XpLw/edit

Does the existing rendering difference occur only with cursive attachment (type 3 lookups) or is it also problematic for single- or pair- adjustment kerning (type 1 and 2 lookups, or...

I can confirm that in mark-to-base lookups, if a base glyph does not have an anchor for a particular mark class, then FontVal generates E4100 with a bogus AnchorFormat value....

Most, if not all, of the tracebacks are from `deepcopy()` being used to make a copy of a `TTFont` object, but I don't get such if I do something simple,...

Could be same as #5007

Noto Sans v2.015 also reports those glyphs; can confirm that with PR #5008 the font passes.

PR #5008 has been merged so you might want to re-test your font and see if this issue can be closed.

Thanks! Confirmed we're again able to build, though with the fallback warning.