[XeTeX] arabic

Kamal Abdali k.abdali at acm.org
Thu Oct 21 06:20:47 CEST 2010

Hi Ross,

On Wed, Oct 20, 2010 at 19:12, Ross Moore <ross.moore at mq.edu.au> wrote:

> ...
> To find out what needs updating, put  \listfiles
> in the preamble (e.g., after \makeindex )
> Then compare the list of package dates to those for what I'm using:
> viz.
>  *File List*
>    book.cls    2007/10/19 v1.4h Standard LaTeX document class
>    bk10.clo    2007/10/19 v1.4h Standard LaTeX file (size option)
> ...

I compared, and saw  that some of your files (e.g., *l3* files, xltxtra)
need updating  -:) .

But here's a serious problem. My suggestion to Houda was that it might be
worthwhile to try xepersian or polyglossia, since these will use the Persian
words for "chapter", "contents", etc. But running the polyglossia code, I
notice that an index is generated only if the language declaration specifies
that Western numerals be used. That is,

The same problem occurs with Arabic and Urdu. Here is a cut down variant of
Houda's code that will illustrate the complaint.

\setmainlanguage{farsi}    %[numerals=western]{farsi}
\title{روش تولید نمایه توسط زیلاتک}
\author{مهدی امیدعلی}
\chapter{ امتحان }

این اولین نمایه این متن است که برای امتحان آورده شده است.

\index{نمایہ }

If the western numerals are not specified, this is what Makeindex says:
Scanning input file xetexQ5.idx...done (0 entries accepted, 1 rejected).
Nothing written in xetexQ5.ind.
Makeindex reads the idx file for sorting the indices. Probably, it finds the
non-Western numerals undecipherable.

Kamal Abdali
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tug.org/pipermail/xetex/attachments/20101021/f68cdd78/attachment.html>

More information about the XeTeX mailing list