id,summary,reporter,owner,description,type,status,priority,milestone,component,version,severity,resolution,keywords,cc
17267,twentyeleven_url_grabber() needs work,nacin,,"{{{
/**
 * Grab the first URL from a Link post
 */
function twentyeleven_url_grabber() {
	global $post, $posts;

	$first_url = '';

	ob_start();
	ob_end_clean();

	$output = preg_match_all('/<a.+href=[\'""]([^\'""]+)[\'""].*>/i', $post->post_content, $matches);

	$first_url = $matches [1] [0];

	if ( empty( $first_url ) )
		return false;

	return $first_url;
}
}}}

The regex needs some work:
 - `a.+href` would match `area href`. While that isn't much of a concern, it would also skip to the second link as it's greedy.
 - Likewise we should probably use a non-greedy `(.+?)` rather than `([\'""]+)` for what we're capturing.
 - We probably want the `s` pattern modifier there, in case there's an `\n` somewhere such as after `<a ` and before `href`.
 - No need for the `.*>` at the end, just grab the href and bail.

Something like this might work: `'/<a\s[^>]+href=[\'""](.*?)[\'""]/is`.

Also:

No need for the `$posts` global to be called upon.

Rather than preg_match_all, we should just use preg_match, since we want the first one. If `! preg_match`, then we should return false. No need to assign the result to `$output`.

We should arguably use get_the_content() since this is being called in within a loop, rather than the raw, unfiltered data.

Anyone know what the output buffering was designed to do?",defect (bug),closed,normal,3.2,Bundled Theme,3.2,normal,fixed,has-patch close,