PHP Regex to Make Twitter Links Clickable

this is awesome, i made one modification – your first two replacements forgot to close the link.
.-= mike´s last blog ..the family at home =-.

Not if this matters but I was getting “bad escape” warnings from Zend Studio. You can see the changes I made below.

$string = preg_replace(“#(^|[\n ])([\\w]+?://[\\w]+[^ \"\n\r\t< ]*)#", "\\1\\2“, $string);
$string = preg_replace(“#(^|[\n ])((www|ftp)\\.[^ \"\t\n\r< ]*)#", "\\1\\2“, $string);
$string = preg_replace(“/@(\\w+)/”, “@\\1“, $string);
$string = preg_replace(“/#(\\w+)/”, “#\\1“, $string);

Wait, I caught a bug.
If you parse a message like this “blah..blah..blah http://www.google.com?“,
the question mark will be included in the url.

That's very cool tips. Thanks for your post.

You can't use this on HTML code without risking to destroy your HTML. Example:

Don't break!

(… let's hope this thing doesn't post my link as link but as HTML code )

Thanks buddy, that made my day

You wouldn't use it in a tweet or RSS, you would use it in a PHP script that parses tweets or tweet RSS feeds.

echo twitterify($tweet_content);

You don't implement it on your Twitter – Twitter already does auto-linking for you. You would use this in an application or web page that is pulling directly from your Twitter feed and emebdding in the page.

LOL It's good to know you're not prone to hyperbole or anything. I'm glad it helped!

This totally kicks ass and saved me a good hour+ of work and I'm really good with regexs.

Thanks!!

This is great! I'm just learning preg replace and regex and had already spent some time trying to figure this out with no result in sight. Thank you very much.

One quickie, though: email addresses. Some tweets contain 'you@gmail.com' with the predictable result of only '@gmail' linking to http://www.twitter.com/gmail. Any thoughts on a fix?

Thanks again!

I think that's the exact same set of regular expressions I'm using to parse links on my website home page.

This is awesome, thanks! I started to write my own with a bunch of if's, else's and substr_replace(), and substr() but it quickly became a mess. I really need to learn regex!

Thanks a lot!!! There is problem with some characters in parsed text. For exapmle #Nestlé is not parsed right

I converted this function to javascript

function twitterify(ret) {

ret = ret.replace(/(^|[n ])([w]+?://[w]+[^ "nrt< ]*)/, “$1$2“);
ret = ret.replace(/(^|[n ])((www|ftp).[^ "tnr< ]*)/, “$1$2“);
ret = ret.replace(/@(w+)/, “@$1“);
ret = ret.replace(/#(w+)/, “#$1“);

return ret;

}

The HTML tags seem to have gone missing from the replace parts of those statements! Anyway, the basic idea is to replace \1 with $1. The hash delimiters should be replaced with forward slashes, which means any other forward slashes need escaping with backslashes!

Nice function.But, I need some help.
The URL “http://test.com/test.htm#test”does'nt links correctly. Twitter parsed this as complete url (no hastag). Your function parsed a hastag, wich is not correct.

Can you fix this?

Damn! There was an mistake in my posted code, sorry!
This code should be work:

function twitterify($ret) {
$ret = preg_replace(“#(^|[n ])([w]+?://[w]+[^ "nrt< ]*)#”, “\1\2“, $ret);
$ret = preg_replace(“#(^|[n ])((www|ftp).[^ "tnr< ]*)#”, “\1\2“, $ret);
$ret = preg_replace(“/(^| )@(w+)/”, “\1@\2“, $ret);
$ret = preg_replace(“/(^| )#(w+)/”, “\1#\2“, $ret);
return $ret;
}

This is really cool, but it looks like a 'smart' apostrophe (’)will break it