Actually you no longer need to assign any PUA to glyphs in OpenType. PUA are just intermediate tools that allow evaluating the rendering of glyphs in test pages, to help defining the ligatures and kerning rules.
A finalized font should no longer contain those PUA (except glyphs which are still not directly representable in Unicode strings, or needed for compatiblity with some necessary specific OpenType features that are not recognized by some older renderers), because OpenType does not need any code point to subtitute one or more glyphs or to position them. All you need is to map the actual standard characters. So you should add the final step: dropping those PUA assignments, once you have programmed the lookups and seen them working as intended (and being used in the correct expected order, without unexpected interactions with more prioritary features !). 2012/6/13 Naena Guru <[email protected]>: > My approach was as follows (applies to Indic and perhaps Arabic and Hebrew > too): > Start observing Anglicizing -> > Map sounds to QWERTY extended key layouts adding non-English letters -> > Result: strict, rule based alphabet extending from ASCII to Latin-1 -> > Font: Program all shapes, base as well as conjoint, to the PUA -> > Program look-up tables according to the orthography rules in the grammar of > the language.
