Hi all!
One user of Debian wanted to build a luatex based latex format (he
wanted to test some examples from the luatex.org documentation slides,
using the old \lua{...} code, anyway).
Unfortunaltely this will not work in any way because there are so many
hyphenation patterns using non-ascii and non-utf8 content. All those of
course break.
Is there any suggested way how to cope with this?
Best wishes
Norbert
-------------------------------------------------------------------------------
Dr. Norbert Preining
Norbert Preining wrote:
Hi all!
One user of Debian wanted to build a luatex based latex format (he wanted to test some examples from the luatex.org documentation slides, using the old \lua{...} code, anyway).
Unfortunaltely this will not work in any way because there are so many hyphenation patterns using non-ascii and non-utf8 content. All those of course break.
Is there any suggested way how to cope with this?
It should be possible to use the (UTF-8) pattern files from xelatex, but I don't know how those are packaged. Best wishes, Taco
On 10/31/07, Taco Hoekwater wrote:
Norbert Preining wrote:
Hi all!
One user of Debian wanted to build a luatex based latex format (he wanted to test some examples from the luatex.org documentation slides, using the old \lua{...} code, anyway).
Unfortunaltely this will not work in any way because there are so many hyphenation patterns using non-ascii and non-utf8 content. All those of course break.
Is there any suggested way how to cope with this?
It should be possible to use the (UTF-8) pattern files from xelatex, but I don't know how those are packaged.
XeTeX uses a kind-of-quick-and-dirty-hack, but it should work for LuaTeX. Take a look into xu-XXhyph.tex (where XX denotes the language), where you can find the following lines: \expandafter\ifx\csname XeTeXrevision\endcsname\relax \else It now recognises XeTeX, but you can modify it to recognise LuaTeX as well. On the other hand, stary by removing (commenting) languages from language.dat, so that there will be far less files to modify for the start. Mojca
Is there any suggested way how to cope with this?
Depending on what he wants to do with LaTeX-luatex, he might: 1. (The smartest) Use a solution like XeTeX's xu-hyphen package to wrap the pattern-loading routine into something that looks more like UTF-8 (I don't know how this is done exactly so I cannot be more precise). 2. (Quick and dirty) De-activate the troublesome hyphen files by either using a custom language.dat, or creating empty files to override the standard ones. In TeX Live 2007 The files in questions should be -rw-r--r-- 1 areutena promo8 0 sep 26 18:42 bghyphsi.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:43 gahyph.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:43 huhyphn.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:44 ruhyphen.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:44 srhyphc.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:44 trhyph.tex -rw-r--r-- 1 areutena promo8 0 sep 26 18:44 ukrhypmp.tex 3. (Even faster) If he's only interested in testing LuaTeX features, use plain LuaTeX. Arthur
Norbert Preining
Hi all!
One user of Debian wanted to build a luatex based latex format (he wanted to test some examples from the luatex.org documentation slides, using the old \lua{...} code, anyway).
Unfortunaltely this will not work in any way because there are so many hyphenation patterns using non-ascii and non-utf8 content. All those of course break.
Is there any suggested way how to cope with this?
I've posted example code for this to the list once (basically converting LuaTeX input into byte streams which is what LaTeX expects), but can't at the moment find the (afterwards improved) sources on my computers. Maybe a list search will turn up the first versions. -- David Kastrup, Kriemhildstr. 15, 44793 Bochum
participants (5)
-
Arthur Reutenauer -
David Kastrup -
Mojca Miklavec -
Norbert Preining -
Taco Hoekwater