[fontforge-users] Crash on contextual alternates subtable

[marty39]

I've built a font with all caps (LC identical to UC), intended for closely
spaced lines, so the descent is zero and the caps use the whole ascent. I
have glyphs for Latin, Greek and Cyrillic and I built small caps for them. I
want to substitute a small cap when a combining diacritic is added, so that
the resulting character stays within the em box.

The image shows how I built a contextual substitution.
<Loading Image...>
When I hit Next, and then OK, Fontforge quits and disappears (even if I use
only 'smcp' or 'c2sc' instead of both). What am I doing wrong? Or is this a
continuation of an old bug? How can I get the result I'm looking for?

I'm using Fontforge version 23:52 PDT 26-May-2015-ML-D in Mac OS X version
10.11.1, installed from a DMG. Is there another version that would work
better? I have a laptop running Linux Mint, if that would help.



--
View this message in context: http://fontforge.10959.n7.nabble.com/Crash-on-contextual-alternates-subtable-tp14943.html
Sent from the User mailing list archive at Nabble.com.

------------------------------------------------------------------------------

[marty39]

Reply to myself: why even try to do that? No ordinary word processor
available now implements the feature you're trying to add. NeoOffice (but
not OpenOffice or LibreOffice) knows how to use combining diacritics, but
none of them, nor Pages, nor TextEdit can use conditional substitutions.

Of course I have a modest assortment of precomposed accented glyphs. To
accommodate combining diacritics above, I made the lower case glyphs
smaller. To accommodate combining diacritics below, I would have to
conditionally raise the base glyph, which won't happen in the word processor
even if I could make fontforge build the feature. So things like comma below
will only work if the accented character is precomposed.

Any guesses as to when these things will work? That is: (1) fontforge builds
conditional vkern or conditional substitution without crashing, and (2)
ordinary word processors implement these features?



--
View this message in context: http://fontforge.10959.n7.nabble.com/Crash-on-contextual-alternates-subtable-tp14943p14946.html
Sent from the User mailing list archive at Nabble.com.

------------------------------------------------------------------------------