The upshot is that IPA symbols only go with lowercase letters, and most of the consonants and some of the vowels are already 'normal' characters (like b, p, t, a, e, ...). I've added 19 new glyphs, most of which resemble other glyphs in the font:
ɑ ɒ ɛ ɜ ǝ ɪ ʌ ɔ ʊ ɡ ʒ ʤ ʃ ʧ ʔ ː 〈 〉
Since the font already exists, there are already a lot of classes and kerning pairs set up. Is it best to just add the new glyphs to existing classes? Which classes should I add them to to make sure they are paired with all the other necessary glyphs? Some of them are easy to match, for example:
- ɑ is copied from Greek lower case alpha, so it can go in classes a1 and c
- ɒ is the alpha upside down, so it can go in class b1
- ǝ already exists at another code point, so I can add it to classes a and b1
- ɔ is similar to o, so it can go in classes b1 and c
- ʤ starts with a d, so it can go in class d
- ʧ starts with a t, so it can go in class t
- ː is just a colon with triangles, so it can go in the class with : and ;
So then I wonder if I should just make new classes for each of these glyphs and create a new subtable specifically for IPA, but that seems inefficient and redundant. On the one hand, I'm looking for whatever is quickest and easiest, but on the other hand, I'm also trying to think ahead to when I may add other phonetic glyphs to the font, so maybe it's worth the effort now?
Thanks in advance for any advice or tips.