Reports a count of words for a given file.

word_counter.py

import re
from collections import defaultdict

SHORT_WORD_LEN = 4 
WORD_REGEX = '\w+'
TEXT_FILE_PATH = 'text.txt'

word_counter = defaultdict(int)

# read in all the words, cull out the ones that aren't long enough
words = re.findall( WORD_REGEX, open( TEXT_FILE_PATH ).read().lower() )
no_short_words = [ x for x in words if len(x) > SHORT_WORD_LEN ]

# tally up the words
for w in no_short_words:
    word_counter[w] += 1

# remove all words that have not been repeated
no_singleton_words = [ (v, k) for k,v in word_counter.items() if v > 1 ] 

# dump out the list of words
for c,w in reversed( sorted( no_singleton_words ) ):
    print "%s: %s" % ( str(c).rjust(3), w )