Regex to get the Facebook Page ID from a given URL

gistfile1.php

<?php
$tests = array(
	'http://www.facebook.com/page_id'						=> 'page_id',
	'https://www.facebook.com/page_id'						=> 'page_id',
	'http://www.facebook.com/#!/page_id'						=> 'page_id',
	'http://www.facebook.com/pages/Parisé-France/Vanity-Url/123456?v=app_555'	=> '123456',
	'http://www.facebook.com/pages/Vanity-Url/45678'				=> '45678',
	'http://www.facebook.com/#!/page_with_1_number'					=> 'page_with_1_number',
	'http://www.facebook.com/bounce_page#!/pages/Vanity-Url/45678'			=> '45678',
	'http://www.facebook.com/bounce_page#!/my_page_id?v=app_166292090072334'	=> 'my_page_id',
	'http://www.facebook.com/pages/some-café-or-èàù-url/123456'			=> '123456',
	'http://www.facebook.com/some-vanity-url/123456'				=> '123456',
	'http://www.facebook.com/some.page.9'						=> 'some.page.9',
	'http://www.facebook.com/a_page_with_id/123456789?ref=hl'			=> '123456789',
	'vanityurl/123456789?ref=hl'							=> '123456789',
	'pages/really-long-vanity-url-page/123456789?ref=hl'				=> '123456789',
	'http://www.facebook.com/profile.php?id=123456789'				=> '123456789'
);
echo '<table>';
foreach ($tests as $url => $result) {
	echo '<tr><td>';
	echo $url;
	echo '</td><td>';
	echo preg_replace(
		'#'
		. '(?:https?://)?'
		. '(?:www.)?'
		. '(?:facebook.com/)?'
		. '(?:(?:\w)*\#!/)?'
		. '(?:pages/)?'
		. '(?:[?\p{L}\-_]*/)?'
		. '(?:[?\w\-_]*/)?'
		. '(?:profile.php\?id=(?=\d.*))?'
		. '([\d\-]*)?'
		. '(?:\?.*)?'
		. '#u',
		'$1',
		$url
	);
	echo '</td></tr>';
}
echo '</table>';
?>

Based on a small sample of real world data, I added a few use cases and improved on the regex:

• handles https pages
• works without the "facebook.com" part
• changed the delimiter from / to # to be more readable (based on a comment this SO question: http://stackoverflow.com/q/11907178/233404)
• added the u modifier and replaced \w by \p{L} to account for possible accents

And I wrapped the whole in a small test to easily add more use cases. Hope this helps.