Python中使用md5sum检查目录中相同文件代码分享

代码如下:
"""This module contains code from
Think Python by Allen B. Downey

http://thinkpython.com

Copyright 2012 Allen B. Downey
License: GNU GPLv3 http://www.gnu.org/licenses/gpl.html

"""

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):
            names.append(path)
        else:
            names.extend(walk(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 = fp.read()
    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].append(name)