Counting word usage. Could be more efficient ways.

gistfile1.txt

import re

data = open("a bunch of text", 'r')
words = {}

strip_punc = re.compile("[!.,]")
whitespace_collapse = re.compile("\s+")

for line in data.readlines():
    line = strip_punc.sub('', line)
    line = whitespace_collapse.sub(' ', line)
    line = line.lower()
    tokens = line.split(' ')
    for token in tokens:
        if token not in words:
            words[token] = 0
        words[token] += 1

sorted_words = sorted(words.iteritems(), key=lambda x: x[1])
print sorted_words

Nothing like the internet to make you scramble to fix something stupid in your code.