<div dir="ltr"><font size="2"><font face="verdana,sans-serif"></font></font>On Thu, Oct 21, 2010 at 03:17, Ulrike Fischer <span dir="ltr"><<a href="mailto:news3@nililand.de">news3@nililand.de</a>></span> wrote:<br><div class="gmail_quote">
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Am Thu, 21 Oct 2010 00:20:47 -0400 schrieb Kamal Abdali:<br>
<div>>... </div></blockquote><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div class="im">> Makeindex reads the idx file for sorting the indices. Probably, it finds the<br>
> non-Western numerals undecipherable.<br>
<br>
</div>Yes, makeindex needs to understand numbers to be able to build page<br>
ranges and sort the page numbers. As default it understands the<br>
standard digits, alphabetic and roman page numbers.<br>
<br>
If you would redefine \thepage to e.g. \renewcommand\thepage{Page<br>
\arabic{page}} you would run into problems in an english document<br>
too.<br>
<br>
I would suggest that you try xindy.<div class="h5"> <br>
</div></blockquote></div><br><font size="2"><font>Thanks, Ulrike, for the
xindy recommendation. A recent tendency in Urdu publications (which concern me
most) is to use Western numerals, so Makeindex will work partially. But
in general xindy will be unavoidable.<br>
<br>
Kamal Abdali<br></font></font><br></div>