supports L10N by setting per-group locale and character-set. Any language locale supported by your server can be used to create link words.
This group uses the English locale and ISO-8859-1 character-set, and so here only English characters can be used to create links by joining words_with_underscores. But the default group on this site uses the Hebrew locale and ISO-8859-8 character-set, and so allows also Hebrew characters in words_with_underscores.
When you install chiq_chaq, set the $interface_language variable in your .cf file to determine which characters will be allowed in words_with_underscores. (Currently it can be can be either 'English' or 'Hebrew')
If the link words do not work, you may need to change the $locale variable in the file $root_cc_dir/English/phrases.cf or $root_cc_dir/Hebrew/phrases.cf to match the locals available on your system. (on Unix systems, you can list the available locales with locale -a )
See also: Japanese_Localization
I wonder... would there be any way to allow multiple charsets to be supported on the same server, so users can write articles in any language they choose (and have them interlinked within the same wiki?)
Maybe that would be too confusing!
You can use utf8 charset (writh in the in configuration file $charset = 'utf8'
) , and then the contents
of the pages can be multi-lingual. However link_words
will still only be in the language specified by the $locale.
I plan adding free links syntax, and then it will be possible to have any character in link_words
and page titles.
Dan_Mills (Wednesday, March 12, 2003, 03:18):
Where can I find a list of the valid charsets / locales?
Dan_Mills (Sunday, March 16, 2003, 07:34):
I am confused as to how this works. I'm trying to get a spanish locale setup. I can get things working if I set $locale to 'Spanish' and $charset to 'es_VE', which I think is equivalent to 'iso-8859-1' (latin1). Then I just make sure that the html templates and the .cf file are in latin1 and stuff works, with some caveats.
The caveats are:
- My terminal/shell seem to prefer some other encoding, I presume it's utf-8. This makes it very hard to look through / edit the .chiq files with, say, emacs.
- Getting the 'rename' cgi command to work requires escaping all the non-ascii characters in the url, which is less than fun.
The combination of those two caveats is kind of annoying. So I tried to use utf-8 across the board, but that really
doesn't work. I've tried setting $charset to 'en_VE.utf8', 'utf8', and 'utf-8'. None work very well. If I save the html templates in utf-8 I can get those to display ok, but the strings in the .cf file just don't work right unless I use latin1.
I hope I'm not just repeating known issues, this is fairly new to me! I'd love to help, but I'm not sure where to start. Is there any good documentation on this stuff that I should be reading?
Yes, those charsets can be real difficult sometimes! I'm afraid I don't know enough to help. Maybe you can try to using an editor that supports utf8 for editing the cf?
Is a Portuguese (Brazilian) version of chiqchaq available?