“””this module contains code from
think python by allen b. downey

copyright 2012 allen b. downey
license: gnu gplv3


import os

def walk(dirname):
“””finds the names of all files in dirname and its subdirectories.

dirname: string name of directory
names = []
for name in os.listdir(dirname):
path = os.path.join(dirname, name)

if os.path.isfile(path):
return names

def compute_checksum(filename):
“””computes the md5 checksum of the contents of a file.

filename: string
cmd = ‘md5sum ‘ + filename
return pipe(cmd)

def check_diff(name1, name2):
“””computes the difference between the contents of two files.

name1, name2: string filenames
cmd = ‘diff %s %s’ % (name1, name2)
return pipe(cmd)

def pipe(cmd):
“””runs a command in a subprocess.

cmd: string unix command

returns (res, stat), the output of the subprocess and the exit status.
fp = os.popen(cmd)
res =
stat = fp.close()
assert stat is none
return res, stat

def compute_checksums(dirname, suffix):
“””computes checksums for all files with the given suffix.

dirname: string name of directory to search
suffix: string suffix to match

returns: map from checksum to list of files with that checksum
names = walk(dirname)

d = {}
for name in names:
if name.endswith(suffix):
res, stat = compute_checksum(name)
checksum, _ = res.split()

if checksum in d:
d[checksum] = [name]

return d

def check_pairs(names):
“””checks whether any in a list of files differs from the others.

names: list of string filenames
for name1 in names:
for name2 in names:
if name1 < name2: res, stat = check_diff(name1, name2) if res: return false return true

def print_duplicates(d):
“””checks for duplicate files.

reports any files with the same checksum and checks whether they
are, in fact, identical.

d: map from checksum to list of files with that checksum
for key, names in d.iteritems():
if len(names) > 1:
print ‘the following files have the same checksum:’
for name in names:
print name

if check_pairs(names):
print ‘and they are identical.’

if __name__ == ‘__main__’:
d = compute_checksums(dirname=’.’, suffix=’.py’)

Posted in 未分类