URL Regular Expression Pattern

[gaza165]

// Pattern and pregmatches etc \\

			$pattern = "%http://\S+\.(?:jpe?g|png|gif)%i";
			$match = preg_match_all($pattern, $body, $matches,1);
			$gaz = $body;	
			$i = 0;	

				foreach($matches[0] as $url) {	
					$replacement = "%TOKEN".$i."%";	
					$gaz = preg_replace("@$url@", $replacement, $gaz, 1);
					$i++;
				}	

				// End of replace with Tokens!! \\	

			$string = nl2br(htmlentities($gaz));
			$i= 0;

				foreach($matches[0] as $value) {
					$replace = "<img src='".$value."'></img>";
					$search = "%TOKEN".$i."%";
				    $string = str_replace($search, $replace, $string);
					$i++;
				}


			$linkpattern = "/http:\/\/[A-Za-z0-9\-_]+\\.+[A-Za-z0-9\.\/%&=\?\-_]+/i";
			$linkmatch = preg_match_all($linkpattern, $body, $linkmatches,1);
			$i = 0;	

				foreach($linkmatches[0] as $linkurl) {	
					$replacement = "<a href='".$linkurl."'>$linkurl</a>";	
					$new = preg_replace("@$linkurl@", $replacement, $string, 1);
					$i++;
				}	


			echo $new; //output the blog...

 

I am creating a blog that can search for image links and urls for example

 

http://www.domain.com/image1.jpg and http://www.google.com

 

Can someone look at these two patterns and tell me if they are the right ones to use to filter out IMAGE URLS from REGULAR URLS.

 

Thanks

 

Garry

[effigy]

The images could be expanded if needed. For example, do you need to look for bmps? As for the URLs, it seems overcomplicated to me. If you only want domains, use a similar approach: http://\S+\.[a-z]+\b. I would also add the \b to the end of the images pattern so something like "http://www.domain.com/images.gifs/1.gif" will not half-match.