Re: Regex Masters: Please inspect this regex- Pulling urls out of page
| From: | Christian Reiniger | Date: | Thu, 08 Mar 2001 09:35:22 +0000 |
| Subject: | Re: Regex Masters: Please inspect this regex- Pulling urls out of page | ||
| References: | 1 | Groups: | php.general |
| Request: | Send a blank email to php-general+get-43013@lists.php.net to get a copy of this message | ||
On Thursday 08 March 2001 09:19, you wrote:
> I'm putting together a regex to pull all of the urls out of a web page.
> Not the href tag, but just the url part of that tag.
>
> Here's what I've come up with:
>
>
> preg_match_all('/<.*href\s*=\s*(\"|\')?(.*?)(\s|\"|\'|>)/i',
> $html,
> $matches);
> foreach($matches[2] as $m) print "<P>$m\n";
>
> All regex masters please tell me if I'm missing something. It's working
> well, but I'm still learning about perl regex and I'd like any input if
> at all possible.
Pretty good. Some minor things:
(1) "<.*href" will also match <a name='foo'><h2> example of
href="hello"</h2>
(2) You're pretty lax on the tag syntax - you don't require quotes to
match, don't require quotes at all etc
I'd rewrite it to
'/<\s*a\s*href\s*=\s*("|\')(.*?)\\1\s*>/i'
> What's a good way to exclude things like javascript: urls and other non
> URI info? I guess what I'm really looking for is all the http urls, no
> ftp, mms etc... or anything like that.
Use the following regexes on your result to get only the types you want:
$HostName = '([a-zA-Z][\w-]*(\.[a-zA-Z][\w-]*)+)';
$HostIP = '(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})';
$Host = '(' . $HostName . '|' . $HostIP . ')';
$HTTPPath = '(([\w\.\/~\?%=\-]|&)+)';
$FTPPath = '(\/[^\/\s]*)*\/?';
$Port = '(:\d+)?';
$HTTPURL = 'http:\/\/' . $Host . $Port . $HTTPPath;
$WWWAddress = 'www\.' . $HostName . $Port . $HTTPPath;
$FTPURL = 'ftp:\/\/' . $Host . $FTPPath;
--
Christian Reiniger
LGDC Webmaster (http://sunsite.dk/lgdc/)
Very funny, Scotty! Now beam up my clothes...