Wow, I've been doing some PDF parsing at work and this is going to come in SO handy.
It gives you a JSON representation of the PDF data structure. What's nice is that doesn't hide the underlying format but it takes care of a lot of the low level edge cases for you.