Reply To: Applying Diacritical marks for transliteration of Arabic language

You can change the Unicode in the *header line* for each character. That is where the script gets the value from. But be warned that if you change your existing font, all its current codes are off! You could bite the bullet and check your entire document again, or (perhaps preferable) you could save the adjusted font under a subtly different name. Then you can keep track of what characters are still to be changed – those in the “old” font only.

As for kerning: hum hah. Creating a kerning table out of the blue is quite hard to begin with, but aside from the technical issue, me and Marc have no idea of how we should let a user input all those values. For a typical (complete) font, there can be hundreds or even thousands of kerning pairs … Thinking out of the box, I suppose you could use a free tool such as TTX to add kerning, but I have no idea how that would work in practice.