Localisation poser : how to update my regex?

[Cups]

Here we go...

I have a few stalwart regexes I have relied similar to:

$str = "122 bc"; // letters or numbers space only
var_dump( preg_match("#^[a-z0-9 ]{3,20}$#i", $str) ) ;
//returns 1 as all is good 

So now I am handling funny french characters like this, and its job stopped.

$str = "12é bc"; // letters or numbers space only
var_dump( preg_match("#^[a-z0-9 ]{3,20}$#i", $str) ) ;
//returns 0 - all is not good 

The encoding is utf-8 and the languages will be western european, so how do I do the equivalent of that regex i.e. how to filter in chars like é ?

I read this info that kyber kindly pointed me to, but the little pea in my head has stopped rolling round now I cannot clearly see how I check for a range of permitted chars in multi-byte character sets.

Starter for 10 for anyone?

[Salathe]

Look at using the \p{xx} escape sequences (manual). E.g. in your case, /^[\pL\d ]{3,20}$/D

Bear in mind that the above will allow any "letter character", including non–western-european characters!

[Cups]

$french = '123 Rué Oçà ?';
var_dump( preg_match('/^[\pL\d ]{3,20}$/Du', $french ) );

Thanks salathe, that works as expected now, I am so happy I asked.

So, add the u switch for unicode, and instead of checking for a-z (which I only used out of laziness because I can never remember the Base Character Classes, and sometimes the escape char disappears when I post on here) I have to use the fairly recent utf-8 specific \p escape sequences.

Thanks again.