[HarfBuzz] bug in hb_ot_layout_lookup_collect_glyphs?

[Werner LEMBERG]

> Not really...  It's perfectly legitimate to use non-existing glyph
> indices as interim number during GSUB...

Just curious: Have you actually tested this behaviour with
existing engines?

> When in doubt, do your own check.

OK.

> In this case, it's probably a font bug.  But I don't think we want
> to do anything about that.

Of course it is a font bug.  Note that round-trip conversion with the
original ttx doesn't work with the previously sent Roboto-Thin font
(ttx aborts with an error) – which is good IMHO, but I have no idea
whether this is the correct action.

However, your ttx version simply accepts `glyph65535' in the GSUB
substitution.  Do you *really* think this is OK?  At least ttx should
emit a warning that this glyph doesn't exist and is never input to
another substitution.


   Werner