[SOLVED] preg_replace URLS

[karldesign]

I have the following funtion:

function parseUrl($str_haystack)
{
$str_haystack = str_replace('http://', '', $str_haystack);
$str_needle = preg_replace("/(www.)([^\s,]*)/i", "<a href='http://$1$2' target='_blank' class='ld'>$1$2</a>", $str_haystack);
return $str_needle;
}

 

I am parsing URLs from a string, which works fine until I have a URL as such: www.domain.com/subdomain.php?id=234...

 

Any ideas why? Not familiar with preg_replace in honesty...

[effigy]

Is '...' part of the URL? If so, is there any surrounding content that hints at what it represents?

[dsaba]

what is your problem?

[karldesign]

No, the '...' was just me typing.

 

function nl2p($str)
{
$new_str = preg_replace('/<br \\/>\s*<br \\/>/', "</p><p>", nl2br($str));
return '<p>' . $new_str . '</p>';
}

 

This is the function I perform before the parseUrl function (above). So when I have www.domain.com its fine. But if I have www.domain.com/subdomain.php?id=1234 the page it links to is www.domain.com/subdomain.php?id=1234</p><p> . Now I took out the </p><p> from the nl2p function and it worked, but the formatting of the text went to pot...

[karldesign]

Found a solution... not too sure if its the best way or the right way, but its working...

 

I added a ' ' (whitespace) before each <p> and </p>. This seemed to end the regex...

 

Hope this is useful to anyone else too!

 

Feel free to use it.

[karldesign]

Having said that  , I have found that https: urls are causing me massive concerns!!!

[karldesign]

Ok, I have found a way of allowing all URLs, https: and http...

So if you are interested, here it is.

 

function parseUrl($str_haystack)
{
$str_haystack = str_replace('http://', '', $str_haystack);
$str_haystack = str_replace('www.', 'http://www.', $str_haystack);
$str_needle = preg_replace("/(http:\/\/|https:\/\/)([^\s,]*)/i","<a href='$1$2' target='_blank' class='ld'>$2</a>",$str_haystack);
return $str_needle;
}

 

Please notice how the output doesn't show $1, this is because I didn't want to show http: or https: in the output link... I expect you could add ftp: and any others to the | (or) list... I would however suggest testing it fully before using it, as I haven't done any testing.

 

Hope this helps!

[effigy]

function parseUrl($str_haystack)
{
	$str_needle = preg_replace("#(https?://|www\.)([-a-z0-9+.%/?=]+)#i", "<a href='http://$1$2' target='_blank' class='ld'>$1$2</a>", $str_haystack);
	return $str_needle;
}