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
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
|
#!/usr/bin/python
#
# Find duplicate files and do something with them.
# Copyright 2010 Lars Wirzenius
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
import hashlib
import optparse
import os
import sys
import time
class ProgressReporter(object):
def __init__(self, do_report):
self.written = ''
self.when = 0
self.do_report = do_report
def write(self, msg):
if self.do_report and time.time() - self.when >= 1:
sys.stdout.flush()
sys.stderr.write('\b \b' * len(self.written))
msg = msg[:79] # FIXME: use real screen width
sys.stderr.write(msg)
sys.stderr.flush()
self.written = msg
self.when = time.time()
def finished(self):
self.when = 0
self.write('')
class DuplicateFileFinder(object):
def __init__(self, progress):
self.by_size = dict()
self.progress = progress
def collect(self, root):
for dirname, subdirs, filenames in os.walk(root):
self.progress.write(dirname)
subdirs.sort()
pathnames = [os.path.join(dirname, f) for f in filenames]
for pathname in pathnames:
stat = os.lstat(pathname)
t = (stat.st_dev, stat.st_ino, pathname)
if stat.st_size in self.by_size:
self.by_size[stat.st_size].append(t)
else:
self.by_size[stat.st_size] = [t]
self.progress.finished()
# FIXME: This computed the checksum for each file, the full file.
# It might be faster to read all files (of the same size) in parallel,
# in suitably small chunks, and compute checksums for all of them in
# parallel. When checksums diverge, split the files into smaller
# groups. If a group becomes just a single file, that file is not
# a duplicate and can be dropped completely.
def duplicates(self):
skip = [size for size in self.by_size if len(self.by_size[size]) == 1]
for size in skip:
del self.by_size[size]
total_bytes = sum(len(tuples) * size
for size, tuples in self.by_size.iteritems())
done_bytes = 0
start_time = time.time()
result = []
for size, tuples in self.by_size.iteritems():
if len(set((dev, ino) for dev, ino, pathname in tuples)) == 1:
# All duplicates are hardlinks to the same inode. Skip.
continue
result += self.find_duplicates([p for d, i, p in tuples])
self.progress.finished()
return result
def find_duplicates(self, pathnames):
'''Find sets of duplicate files.
Return list of groups of identical files, where each group is a list
of pathnames to files that are identical.
'''
result = []
identical_groups = [[(x, file(x)) for x in pathnames]]
while identical_groups:
new = []
for group in identical_groups:
done, not_done = self.read_next_chunks(group)
if len(done) > 1:
result.append(done)
while not_done:
key = not_done[0][2]
temp2 = [t for t in not_done if t[2] == key]
not_done = [t for t in not_done if t[2] != key]
if len(temp2) > 1:
new.append([(pathname, f)
for pathname, f, data in temp2])
identical_groups = new
return result
def read_next_chunks(self, group):
'''Read next chunk of data in each file.
group is a list of tuples (pathname, open file). A chunk of data
is read from each file.
Return value is two lists: one with filenames that reached the
end of the file, and one with tuples (pathname, open file,
chunk).
'''
done = []
not_done = []
chunk_size = 64 * 1024
for pathname, f in group:
data = f.read(chunk_size)
if not data:
f.close()
done.append(pathname)
else:
not_done.append((pathname, f, data))
return done, not_done
def duplicates_progress(self, done, total, started):
duration = time.time() - started
if duration < 1:
speed = 0
else:
speed = done / duration
self.progress.write('%s/%s (%.1f%%) done (%s/s)' %
(self.human_size(done), self.human_size(total),
100.0 * float(done) / float(total),
self.human_size(speed)))
def human_size(self, size):
tab = [(1024**3, 'GiB'),
(1024**2, 'MiB'),
(1024**1, 'KiB')]
for limit, unit in tab:
if size >= limit:
return '%.1f %s' % (float(size) / float(limit), unit)
return '0 B'
def make_hardlinks(duplicates):
canonical = duplicates.pop()
for pathname in duplicates:
os.remove(pathname)
os.link(canonical, pathname)
def report(duplicates):
sys.stdout.write('\n'.join(duplicates))
sys.stdout.write('\n\n')
def main():
parser = optparse.OptionParser()
parser.add_option('--make-hardlinks', action='store_true',
help='hardlink duplicate files to each other')
parser.add_option('--progress', action='store_true',
help='report progress')
opts, args = parser.parse_args()
progress = ProgressReporter(opts.progress)
dupfinder = DuplicateFileFinder(progress)
for dirname in sorted(args):
dupfinder.collect(dirname)
for duplicates in dupfinder.duplicates():
if opts.make_hardlinks:
make_hardlinks(duplicates)
else:
report(duplicates)
if __name__ == '__main__':
main()
|
