[htdig-dev] Internationalisation, UTF-8 and entities.

[Michael C. <ch...@mc...>]

I am going to re-read the locales FAQs, at least one more time ... but thus 
far I have had no success on Mandrake 9.2 set-up to run on UTF-8.

UTF-8 characters come out on my machine with htsearch as an entit-ised 
version of Latin-1, e.g. Genève (G-e-n-e_grave-n-e) comes out as 
Genaève.
Thought UTF-8 was the standard language for the web (or at least XHTML) now?

The above may be my mis-settings.

But Палата (that is some Cyrillic 
inserted in an English-language webpage, that is placed in as entities and 
has not been transcribed to UTF-8 (or any other encoding for that matter), 
shows up nicely in the original webpage, but in the htsearch output is 
rendered as above (that is the source code is now Па& 
etc.).

The few non-numeric entities I have checked seem to survive digging/searching 
(& ®).

Is there a way to stop the killing of numeric entities?



Michael Chapman.