On Fri, Oct 22, 2010 at 08:47:04PM -0400, Kamal Abdali wrote: > HI Houda and Ross, > > Ross is of the opinion that the job of indexing in Arabic can be done with > makeindex. I did a little search and reading, and concluded that the job is > not > one that can be done in an afternoon (or two). At least not without some hints > from Ross (and others). > > For the xindy route, there are files that accompany the xepersian package. > Since the Persian alphabet is a superset of the Arabic one, those files should > suffice without change to create an Arabic index (assuming that they work > properly in Persian). Since the Urdu alphabet is a superset of the Persian > one, the Persian xindy still needs to be generalized somewhat. Sorry to keep > inserting Urdu in the discussion, but that concerns me most these days. > Moreover, a solution for Urdu should also work for Arabic. As I've never used > xindy before, the task again seems much more complex than an afternoon's > hacking :-).
I think collation rules are language dependant, some letters get different sort order in Arabic and Persian, so the files should be checked carefully before use in other languages. Some, if not most depending on the field, Arabic authors will prefer Abjadi order which is totally different from Alphabetical order. Regards, Khaled -- Khaled Hosny Arabic localiser and member of Arabeyes.org team Free font developer -------------------------------------------------- Subscriptions, Archive, and List information, etc.: http://tug.org/mailman/listinfo/xetex
