Working with the Enron dataset

Every now and then I need to do some work with text, and the Enron data set is one of the most well known corpuses. I ended up writing the parsing code for that so many times that it isn't even funny.