Find duplicated file with python

linux用. md5sum コマンド でmd5sum のファイルリストを作成し

同じ md5sum を持つファイルをぶわーっと書きだす。 

import sys
import re
import itertools


#utility functions
def error():
    print "usage: python finddup.py rootpath extension"
def splitline(line):
   return re.match("([a-z0-9]+)[ ]+(.*)", line).groups()


#command line argument check
if len(sys.argv) == 3:
    os.system(r'find %s -name "*.%s" -exec md5sum {} \; > dup.md5'
              % (sys.argv[1], sys.argv[2]))
else:
    print "reading existing dup.md5..."
    error()

# open file
try:
    files = map(splitline, open("dup.md5"))
except:
    error()
    sys.exit(1)

# find duplicates and print them
for f1, f2 in itertools.product(files, files):
    md5sum1, path1 = f1
    md5sum2, path2 = f2
    if path1 == path2: continue
    if md5sum1 == md5sum2:
        print "duplicated:"
        print "    ", path1
        print "    ", path2