A-Z regex for international languages

[optik]

Hi

What are the best methods for using a regex to cover a number of international alphabets?

[henry0]

I might be wrong, actually I'll love to see how it can be done but I do not think there is such a (very useful) regex.
I will do a switch and treat one case at a time.
of course it could be done by allowing a long list of special char belonging to each language, such as n tilde, reverse question mark, inflection etc...
But you could also end by allowing too much
Example: allow for reverse question mark thus regular question mark too
Personally I disallow question mark .. see where I go.

[rob7591]

So you're trying to match all characters including ones such as åéáê etc? Do you want punctuation as well?

You can use \x to match a character's hex code (and search through ranges).

I don't know if Ç is a character in any alphabet, but the range from Ç (\x80) to Ñ (\xA5) appears to have most of the special accented characters and what not.
"/[\x80-\xA5]/" should match a good portion of the characters that you are looking for. (of course you can use a bar to use this in conjunction with a-z0-9 etc.)

[phranque]

i would suggest using the \w metacharacter (matches any "word" character) which is available with Perl Compatible Regular Expression Syntax [php.net].

[optik]

the \w works nicely although htmlentities then messes it up and replaces the characters with the wrong codes for example é gets changed to é

[phranque]

have you set the locale to include the language-specific codes?

[optik]

No am I meant to? How would I do that.

[optik]

Here is the how to set htmlentities to include UTF-8 character set

htmlentities($s, ENT_NOQUOTES,"UTF-8");