[XeTeX] SPACE in TECkit context mapping

Jonathan Kew jonathan at jfkew.plus.com
Sat Aug 16 00:09:01 CEST 2008


On 15 Aug 2008, at 10:20 PM, Daniel Stender wrote:

> Hi guys,
>
> I am trying something like:
>
> [foo] / _ U+0020 [bar] > [foo] U+094D / _ U+0020 [bar]
>
> which should result something like (plain english):
>
> "for every character in group [foo] before a SPACE and a character  
> of group [bar] add character U+094D (Devanagari Virama, combining  
> sign) to that"

Yes, that looks reasonable in general as TECkit mapping....

>
> Doesn't work. What's wrong?

But it won't work, because in XeTeX, the mappings are applied to  
individual words; TeX doesn't handle inter-word spaces as character U 
+0020 at all, but turns them into glue, which is a different thing  
altogether.

This means that there is no way to apply such mappings across inter- 
word spaces; the "character string" that will be processed is broken  
at this point.

Depending quite what you're trying to achieve, there might be other  
ways to do it, perhaps at the TeX macro level.

JK



More information about the XeTeX mailing list