[XeTeX] xetex+babel, was: missing everything

Yves Codet ycodet at club-internet.fr
Mon Jan 23 10:44:00 CET 2006

Le 19 janv. 06, à 12:44, Bruno Voisin a écrit :

> It may be possible to use or create switches so as to disactivate all 
> but the hyphenation and section naming etc. mechanisms in babel, but I 
> would tend to think this is simply not worth the trouble. In my case I 
> can use babel with XeTeX in documents containing text partly in French 
> and partly in English, but this is because both languages use only the 
> Latin alphabet so that babel does not impose any font or encoding 
> setting, and because the French option provides a switch 
> \StandardLayout to disactivate most of the disastrous initiatives 
> taken by babel in terms of item spacing in lists, choices of bullets 
> also in lists, construction of ellipsis character, etc.

There is also something wrong when we want to add Unicode hyphenation 
patterns through "language.dat". Here is the end of my "xelatex.log". 
The two languages I added (in fact only one, since "uni-grhyph.tex" is 
substituted to "grhyph.tex") appear twice, so two numbers are skipped 
in the list at the very end.

Local configuration file hyphen.cfg used
File: hyphen.cfg 2004/11/20 v3.8d Babel language switching mechanism
\l at american=\language0

\l at USenglish=\language0
\l at english=\language0
\l at french=\language1

frhyph.tex - French hyphenation patterns (V2.12) <2002/12/11>)
\l at patois=\language1
\l at german=\language2

German Traditional Hyphenation Patterns `dehypht' Version 3.2a 
(Formerly known under the name `ghyph31' and `ghyphen'.))
\l at austrian=\language2
\l at greek=\language3

Greek hyphenation patterns `uni-grhyph.tex' (v0.1) <2006/1/9>
\l at latin=\language5

Latin Hyphenation Patterns `lahyph' Version 3.0b <2001/11/21>)
\l at sanskrit=\language6

Sanskrit and Prakrit hyphenation patterns `sanhyph.tex' (v0.1) 
\l at spanish=\language8

Loading hyphenation patterns for spanish.
\l at nohyphenation=\language9

Applying patch file ltpatch.ltx
  ) )
Beginning to dump on file xelatex.fmt
  (format=xelatex 2006.1.23)
69702 strings of total length 59689
44112 memory locations dumped; current usage is 144&42153
3179 multiletter control sequences
3633 words of font info for 14 preloaded fonts
16 hyphenation exceptions
Hyphenation trie of length 25329 has 548 ops out of 35111
   29 for language 8
   10 for language 7
   26 for language 5
   7 for language 4
   207 for language 2
   88 for language 1
   181 for language 0
No pages of output.

So far it seems to be harmless but I suppose it should be fixed. Is 
there another solution? For instance if we properly edit a "hyphen.cfg" 
such as this sample I found:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: hyphen.cfg.orig
Type: application/octet-stream
Size: 973 bytes
Desc: not available
Url : http://tug.org/pipermail/xetex/attachments/20060123/c73af734/hyphen.cfg.obj
-------------- next part --------------

and put it into ~/Library/texmf it will be read instead of Babel's 
"hyphen.cfg", but will Babel accept that and will we be able to use a 
command like "\language=\greek" in the XeTeX source file?

Of course it would be the best if an expert in LaTeX wrote a package 
such as Babel, perhaps not a "xe-" or "uni-babel" but something new, 
without the flaws which Bruno mentioned.

Best wishes,


More information about the XeTeX mailing list