Re: [xindy] Xindy and UTF8 in rule files

[Joachim S. <js...@ac...>]

>>>>> "SS" == Simon Spiegel writes:
SS> On 26.12.2007, at 10:17, Thomas Henlich wrote:
>> Simon Spiegel schrieb:

>>> In my understanding this is because the rules file is UTF8. I'm not
>>> really sure whether xindy doesn't handle UTF8 rules at all
>> 
>> did you take a look at the xindy-make-rules package? It probably does
>> exactly the sort of thing you want to do, or very similar.

SS> Thanks, I actually had a look at make-rules and tried with a UTF8- 
SS> german rule, but this didn't seem to work. From my understanding  
SS> (which may be wrong) this rule helps if your .ind-File contains UTF8  
SS> characters

Yes, that's the case.

SS> but in my case the rules file contained UTF8 and xindy choked on
SS> that.

Principally, xindy should handle rule files with UTF-8 chars in
strings. I tried it some time ago, even though I didn't try your
example now.

Where xindy chokes is when the UTF-8 file starts with a Byte Order
Mark (BOM), see http://en.wikipedia.org/wiki/Byte-order_mark. I.e.,
when the file starts with the three bytes (0xef, 0xbb, 0xbf). Some
editors add such a BOM to any UTF-8 file they create. This is a known
deficiency where I don't know yet how I could handle it. (I can't
ignore these bytes simply in the reader, they might appear as
characters in strings.)

Cheers, and best wishes for the new year,

	Joachim

--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Joachim Schrod			Email: js...@ac...
xindy maintainer		http://www.xindy.org/
Roedermark, Germany