Implemented more prefixes. Figured out a good test that should work for all prefixes. Removed binary databases so we can just store human readable csv files.
