[XeTeX] Problem using bidi with hyperref and amsart

Ari Meir Brodsky ari.brodsky at utoronto.ca
Thu Sep 22 15:25:16 CEST 2011


I've been using LaTeX to typeset math for years, but just recently I decided
to upgrade to XeLaTeX in order to include Hebrew.  I've come across the
following problem.  Chapter 2 of the bidi package document (version 1.3.0,
page 15) claims that the bidi package supports the hyperref package as well
as the amsart class.  However, when I use all three together --- amsart,
hyperref, bidi --- the XeTeX engine goes into an infinite loop at the
\makefile command, and exceeds the input stack size capacity.  Here is a
small example:

% !TeX encoding = UTF-8
% !TeX program = xelatex
% test_bidi_hyperref.tex
\title{This is a test file}
\author{Ari Meir Brodsky}

Hi, this is just a test.

Some (hopefully relevant) excerpted lines from the log file are:

This is XeTeX, Version 3.1415926-2.3-0.9997.5 (MiKTeX 2.9) (preloaded
format=xelatex 2011.9.20)  
entering extended mode

LaTeX2e <2011/06/27>
Document Class: amsart 2009/07/02 v2.20.1
Package: hyperref 2011/08/19 v6.82h Hypertext links for LaTeX 
Package hyperref Message: Driver (autodetected): hxetex.
File: hxetex.def 2011/08/19 v6.82h Hyperref driver for XeTeX
Package: bidi 2011/10/01 v1.3.0 Bidirectional typesetting in plain TeX and
File: xetex-bidi.def 2011/07/25 v0.9 bidi adaptations for XeTeX engine
File: amsart-xetex-bidi.def 2011/08/26 v0.6 bidi adaptations for amsart
class for XeTeX engine
File: amsthm-xetex-bidi.def 2010/07/25 v0.1 bidi adaptations for amsthm
package for XeTeX engine
File: hyperref-xetex-bidi.def 2010/07/25 v0.4 bidi adaptations for hyperref
package for XeTeX engine
File: umsa.fd 2009/06/22 v3.00 AMS symbols A
File: umsb.fd 2009/06/22 v3.00 AMS symbols B
Package: nameref 2010/04/30 v2.40 Cross-referencing by name of section

! TeX capacity exceeded, sorry [input stack size=5000]. 
\@ifnextchar #1#2#3->\let \reserved at d =#1
                                         \def \reserved at a {#2}\def
\reserved... l.11 
If you really absolutely need more capacity,
you can ask a wizard to enlarge me.

Here is how much of TeX's memory you used:
 7077 strings out of 429156
 96081 string characters out of 3168943
 193729 words of memory out of 3000000
 10145 multiletter control sequences out of 15000+200000
 7385 words of font info for 29 fonts, out of 3000000 for 9000 
 947 hyphenation exceptions out of 8191 
 5000i,3n,26p,313b,130s stack positions out of
No pages of output.


Any ideas how to fix this?


Ari M. Brodsky
ari.brodsky at utoronto.ca

More information about the XeTeX mailing list