Recursively trim non-alphanumeric characters from the end of a string.

gistfile1.php

$string = "Mary had a little lamb :#$";
$length = strlen($string);
$regex  = "/[^a-z0-9]/i";

for ($i = 1; $i <= $length; $i++) {
	if (preg_match($regex, substr($string, $length - 1, 1))) {
		$string = substr($string, 0, $length - 1);
		$length = $length - 1;
	} else {
		break;
	}
}

echo "Looped $i times. Result: " . $string;

// Looped 5 times. Result: Mary had a little lamb

I suppose $length = $length - 1; could be --$length;

Why not just:

$string = preg_replace('/[^a-z0-9]+$/i', '', $string);

The above finds sequences of non alpha-numeric characters at the end of a string, and removes them. No need to loop, when PCRE takes care of it for you.

Yep, I was going to say just do it in one regex like @weierophinney suggests. No need to make it any more complicated.

Awesome, thanks! It seems the magic bit I was missing was +$. Looks like I have a bit of PCRE studying to do. Thanks again!