Welcome, guest | Sign In | My Account | Store | Cart

Lists all duplicate file. This also demonstrates the effective use of generators and the use of functional composition in Python to solve the problem.

Python, 165 lines
  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
#!/usr/bin/env python
#
# find-duplicates
# by Keith Gaughan <http://talideon.com/>
#
# Finds an lists any duplicate files in the given directories.
#
# Copyright (c) Keith Gaughan, 2008.
# All Rights Reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
#  1. Redistributions of source code must retain the above copyright
#     notice, this list of conditions and the following disclaimer.
#
#  2. Redistributions in binary form must reproduce the above copyright
#     notice, this list of conditions and the following disclaimer in the
#     documentation and/or other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY AUTHOR AND CONTRIBUTORS "AS IS" AND ANY
# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL AUTHOR OR CONTRIBUTORS BE LIABLE FOR ANY
# DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
# ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
# THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# This license is subject to the laws and courts of the Republic of Ireland.
#

from __future__ import with_statement
import sys
import os
import hashlib
import getopt
import filecmp


USAGE = "Usage: %s [-h] [-m<crc|md5>] <dir>*"


class crc:
    """
    Wraps up zlib.crc32 to make it suitable for use as a faster but less
    accurate alternative to the hashlib.* classes.
    """
    def __init__(self, initial=None):
        self.crc = 0
        if initial is not None:
            self.update(initial)
    def update(self, block):
        import zlib
        self.crc = zlib.crc32(block, self.crc)
    def hexdigest(self):
        return "%X" % self.crc
    def digest(self):
        # Er...
        return self.crc


def all_files(*tops):
    """Lists all files in the given directories."""
    for top in tops:
        for dirname, _, filenames in os.walk(top):
            for f in filenames:
                path = os.path.join(dirname, f)
                if os.path.isfile(path):
                    yield path


def digest(file, method=hashlib.md5):
    with open(file) as f:
        h = method(f.read()).digest()
    return h


def true_duplicates(files):
    """
    Compare the given files, breaking them down into groups with identical
    content.
    """
    while len(files) > 1:
        next_set = []
        this_set = []
        master = files[0]
        this_set.append(master)
        for other in files[1:]:
            if filecmp.cmp(master, other, False):
                this_set.append(other)
            else:
                next_set.append(other)
        if len(this_set) > 1:
            yield this_set
        files = next_set


def group_by(groups, grouper, min_size=1):
    """Breaks each of the groups into smaller subgroups."""
    for group in groups:
        subgroups = {}
        for item in group:
            g = grouper(item)
            if not subgroups.has_key(g):
                subgroups[g] = []
            subgroups[g].append(item)
        for g in subgroups.itervalues():
            if len(g) >= min_size:
                yield g


def usage(message=None):
    global USAGE
    fh = sys.stdout
    exit_code = 0
    if message:
        fh = sys.stderr
        exit_code = 2
        print >>fh, str(message)
    name = os.path.basename(sys.argv[0])
    print >>fh, USAGE % (name,)
    sys.exit(exit_code)


def main():
    try:
        opts, paths = getopt.getopt(sys.argv[1:], "hm:")
    except getopt.GetoptError, err:
        usage(err)
    method = crc
    for o, a in opts:
        if o == "-m":
            if a == "crc":
                method = crc
            elif a == "md5":
                method = hashlib.md5
            else:
                usage("Unknown grouping method: %s" % (a,))
        elif o == "-h":
            usage()
        else:
            usage("Unknown option: %s%s" % (o, a))

    if len(paths) == 0:
        paths = ["."]

    first = True
    groups = [all_files(*paths)]
    for grouper in [os.path.getsize, lambda file: digest(file, method)]:
        groups = group_by(groups, grouper, 2)
    for group in groups:
        for files in sorted(true_duplicates(group)):
            if first:
                first = False
            else:
                print
            for file in files:
                print file


if __name__ == "__main__":
    main()

This script uses several different methods to progressively whittle the list of files into groups, each one progressively slower but more accurate. If first groups on file size, then on either a CRC or an MD5 hash. When that's done, it then explicitly compares the files. This method is faster and more robust than simply creating an MD5 hash of all the files.

Originally posted at http://talideon.com/weblog/2008/02/find-duplicates.cfm

1 comment

Koorma Pramodh 13 years, 9 months ago  # | flag

Instead of calculating crc, md5 hash, how about comparing only when the fileSizes match?