On Thursday 02 February 2006 07:52, you wrote: > (almost ?) all indic scripts are affected by this, AFAICT. hi, pa, bn > would have to be treated separately, like we did for ja, zh_*, ko... What exactly do you mean by "treat separately"? > This happens because we hardcoded that the default font is freefont. Yes. But you are not taking into account the Freefont _is already stripped_. So there _should_ be no overlap from Freefont to Indic fonts. If there is an overlap, either the stripping of Freefont was not done correctly, or some other font that is "higher" in the fontconfig list also provides characters from Indic scripts. What I'm asking for is someone to research that. Knowing which glyphs have problems would help with that. Only if there is an overlap _between_ fonts used for Indic scripts, will we absolutely need to change the default font when a language is selected. General question: starting with an UTF-8 encoded string/file, what is the easiest way to get the hex UTF-8 codes of the characters in it?
Attachment:
pgp4nw29vYg1C.pgp
Description: PGP signature