This is a little script I’ve been playing with. I’m using it on my pet project GoGuide.co.za, a social bookmarking site. The plan is for the site to pull a thumbnail image of a submitted URL’s page. The script scans specified sites for the first image with a width greater than 200px. I suggest 200px as in most cases blogs and news sites will not have any images with widths greater than 200px other than in their content.

The script is simple. It grabs the HTML source of a page. Scans for HTML image tags. Grabs the path of an image and copies it to your server.

Download source - ZIP
Download source - RAR