It's a Naïve Bayesian Text Classifier on Redis (aka spam filter.) I wrote this to filter spammy comments from a high traffic forum website and it worked pretty well. It can work for you too :) It's not tied to any particular format like email, it just deals with the raw text.
This is probably the only spam filtering library you'll find for Python that's simple (170 lines of code), works (30 lines of test code), and doesn't suck.
sudo python setup.py install
sudo pip install redisbayes
sudo pip install git+git://github.com/jart/redisbayes.git
import redis, redisbayes rb = redisbayes.RedisBayes(redis=redis.Redis()) rb.train('good', 'sunshine God love sex lobster sloth') rb.train('bad', 'fear death horror government zombie') assert rb.classify('sloths are so cute i love them') == 'good' assert rb.classify('i am a zombie and love the government') == 'bad' print rb.score('i fear God and hate the government') rb.untrain('good', 'sunshine God love sex lobster sloth') rb.untrain('bad', 'fear death horror government zombie')