[alane]

Hi

I am updating a script wrote a script few years ago using the following

PHP Code:

$entry = ereg_replace("[^A-Za-z0-9]", "", $entry) 

to remove everything but alpha numatic characters but I understand that the function will no longer exists in PHP6 and that the use now of this function is 'highly discouraged'

I think the correct replacement is preg_replace. Is this correct and does everything within the brackets remain unchanged?

Further quick question in the first part within brackets ("[^A-Za-z0-9]" what does the ^ actually do? I can't for the life of me remember why it's there.

Thanks

[captcha]

preg_replace (in the form of '/[^A-Za-z0-9]/') should work, yes.

The ^ negates everything in the [] brackets. you are matching everything that is NOT (^) a character between A-Z, a-z and 0-9 and replace it with nothing ('')

[alane]

Thanks for that.

May I ask, just for the 'learning curve' the purpose of the two / in the function please.

[End User]

Originally Posted by alane

May I ask, just for the 'learning curve' the purpose of the two / in the function please.

They're the delimiters that mark the start and stop of the regular expression. You can use any delimiter you want, but slashes and pipes (|) are common. If you use one that's also a meta character you may run into problems when the regex is evaluated.

[captcha]

http://php.net/manual/en/intro.pcre.php is a good place to start reading about delimiters / pattern modifiers / general (PHP) regex handling