As some of you may know websites like Facebook or Twitter allow you to fetch user profile images if you know id of the given users. For example, you can fetch the Facebook image for Mark Zuckerberg by going to http://graph.facebook.com/4/picture (4 is Mark's user id), however if you pay attention to the address bar you will notice that the the aforementioned URL actually redirects to http://profile.ak.fbcdn.net/hprofile-ak-snc4/157340_4_3955636_q.jpg.
Same goes for Twitter - I could fetch my user picture by going to http://api.twitter.com/1/users/profile_image/codeaid but it actually redirects to https://si0.twimg.com/profile_images/321201130/twitter_bigger_normal.png.
In one of the projects I've been working on lately, I added a functionality for people to log in using their Facebook, Twitter and other service credentials. I am also storing URLs of their user images in the database for displaying on the comment pages. As you saw earlier it is easy to generate URLs redirecting to the correct destination, however as you can imagine it doesn't make sense to store them as they are as you would have to execute those redirects every time you access the image. Imagine if you have a page with comments and every comment has a picture - it could and would take a long time to resolve and load them all.
That lead me to writing a simple function, which takes the starting URL, follows all the redirects and returns the last effective URL for it. From examples above the last effective URL for http://graph.facebook.com/4/picture would be http://profile.ak.fbcdn.net/hprofile-ak-snc4/157340_4_3955636_q.jpg.