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
202
203
204
205
206
207
208
|
#!/usr/bin/env python3
# encoding: utf-8
import os
import subprocess
from nose import with_setup
from tests.utils import *
from parameterized import parameterized
def create_files():
# Same size, different content.
create_file('a', '1.a')
create_file('b', '1.b')
# Same size, same content.
create_file('a', '2.a')
create_file('a', '2.a_')
# Different content and size
create_file('a' * 3, '3.a')
create_file('a' * 4, '3.a_')
# Size group that will generate ext_cksums for all files
create_file('b' * 2, '4.a')
create_file('b' * 2, '4.b')
create_file('c' * 2, '4.c')
create_file('c' * 2, '4.d')
# duplicate_dirs + with --write_unfinished
create_file('x', 'dir_a/1')
create_file('x', 'dir_b/1')
def check(data, write_cache):
unfinished = [p['path'] for p in data if p['type'] == 'unique_file']
dupe_files = [p['path'] for p in data if p['type'] == 'duplicate_file']
dupe_trees = [p['path'] for p in data if p['type'] == 'duplicate_dir']
path_in = lambda name, paths: os.path.join(TESTDIR_NAME, name) in paths
if write_cache:
assert len(unfinished) == 3
assert path_in('1.b', unfinished)
assert path_in('dir_a/1', unfinished)
assert path_in('dir_b/1', unfinished)
assert len(dupe_trees) == 2
assert path_in('dir_a', dupe_trees)
assert path_in('dir_b', dupe_trees)
assert len(dupe_files) == 7
assert path_in('2.a', dupe_files)
assert path_in('2.a_', dupe_files)
assert path_in('1.a', dupe_files)
assert path_in('4.a', dupe_files)
assert path_in('4.b', dupe_files)
assert path_in('4.c', dupe_files)
assert path_in('4.d', dupe_files)
def test_xattr_basic(usual_setup_usual_teardown):
create_files()
for _ in range(2):
for write_cache in True, False:
if write_cache:
head, *data, footer = run_rmlint('-U -D -S pa --xattr-write')
else:
head, *data, footer = run_rmlint('-D -S pa --xattr-read')
check(data, write_cache)
head, *data, footer = run_rmlint('-D -S pa --xattr-clear')
@pytest.mark.parametrize("extra_opts", ["", "-D"])
def test_xattr_detail(usual_setup_usual_teardown, extra_opts):
if not runs_as_root():
# This tests need a ext4 fs which is created during the test.
# The mount step sadly needs root privileges.
return
with create_special_fs("this-is-not-tmpfs") as ext4_path:
# Keep the checksum fixed, if we change the default we don't want to
# break this test (although I'm sure some tests will break)
base_options = extra_opts + " -T df -S pa -a blake2b "
path_1 = os.path.join(ext4_path, "1")
path_2 = os.path.join(ext4_path, "2")
path_3 = os.path.join(ext4_path, "3")
path_4 = os.path.join(ext4_path, "4")
create_file("abc", path_1)
create_file("abc", path_2)
create_file("def", path_3)
create_file("longer", path_4)
head, *data, footer = run_rmlint(base_options + ' --xattr-write')
assert len(data) == 2
xattr_1 = must_read_xattr(path_1)
xattr_2 = must_read_xattr(path_2)
xattr_3 = must_read_xattr(path_3)
assert xattr_1["user.rmlint.blake2b.cksum"] == \
b'ba80a53f981c4d0d6a2797b69f12f6e94c212f14685ac4b74b12bb6fdbffa2d17d87c5392aab792dc252d5de4533cc9518d38aa8dbf1925ab92386edd4009923\x00'
assert xattr_1 == xattr_2
# no --write-unfinished given.
assert xattr_3 == {}
# Repeating the caching option should have no effect on the output.
for _ in range(10):
head, *data, footer = run_rmlint(base_options + ' --xattr')
# one more due to the unique_file
assert len(data) == 3
xattr_1 = must_read_xattr(path_1)
xattr_2 = must_read_xattr(path_2)
xattr_3 = must_read_xattr(path_3)
assert xattr_1["user.rmlint.blake2b.cksum"] == \
b'ba80a53f981c4d0d6a2797b69f12f6e94c212f14685ac4b74b12bb6fdbffa2d17d87c5392aab792dc252d5de4533cc9518d38aa8dbf1925ab92386edd4009923\x00'
assert xattr_1 == xattr_2
# --write-unfinished will also write the unfinished one.
xattr_3 = must_read_xattr(path_3)
assert xattr_3["user.rmlint.blake2b.cksum"] == \
b'36badf2227521b798b78d1bd43c62520a35b9b541547ff223f35f74b1168da2cd3c8d102aaee1a0cc217b601258d80151067cdee3a6352517b8fc7f7106902d3\x00'
# unique file which was not hashed -> does not need to be touched.
xattr_4 = must_read_xattr(path_4)
assert xattr_4 == {}
# Try clearing the attributes:
head, *data, footer = run_rmlint(base_options + '--xattr-clear')
assert len(data) == 2
assert must_read_xattr(path_1) == {}
assert must_read_xattr(path_2) == {}
assert must_read_xattr(path_3) == {}
assert must_read_xattr(path_4) == {}
# regression test for GitHub issue #475
# NB: this test is only effective if RM_TS_DIR is on an xattr-capable filesystem
@with_setup(usual_setup_func, usual_teardown_func)
def test_treemerge_xattr_hardlink():
create_file('xxx', 'a/x')
create_file('yyy', 'a/y')
create_file('xxx', 'b/x')
create_file('yyy', 'b/y')
sh_path = os.path.join(TESTDIR_NAME, 'rmlint.sh')
head, *data, foot = run_rmlint('--xattr-write -o sh:{p} -c sh:hardlink'.format(p=sh_path))
assert len(data) == 4
# run script to hardlink files
subprocess.check_output([sh_path, '-d'])
# This used to fail with 'rm_shred_group_free: assertion failed: (self->num_pending == 0)'
head, *data, foot = run_rmlint('-D --xattr-read')
assert len(data) == 6
# NB: this test is only effective if RM_TS_DIR is on an xattr-capable filesystem
@parameterized([('-q 1',), ('-Q 1',), ('-q 50%',), ('-Q 50%',)])
@with_setup(usual_setup_func, usual_teardown_func)
def test_clamp_xattr_false_negative(clamp):
create_file('xxx', 'a')
create_file('yyy', 'b')
# we used to write xattrs even when clamping is used
head, *data, foot = run_rmlint('--xattr', clamp)
assert all(e['type'] == 'unique_file' for e in data)
create_file('xxx', 'c')
# the first run after creating 'c' is ok...
head, *data, foot = run_rmlint('--xattr', force_no_pendantic=True)
assert len([e for e in data if e['type'] == 'duplicate_file']) == 2 # 'a' matches 'c'
# but we would get a false negative here, as the xattrs didn't match
head, *data, foot = run_rmlint('--xattr', force_no_pendantic=True)
assert len([e for e in data if e['type'] == 'duplicate_file']) == 2 # do they still match?
# NB: this test is only effective if RM_TS_DIR is on an xattr-capable filesystem
@parameterized([('-q 2',), ('-Q 1',), ('-q 70%',), ('-Q 50%',)])
@with_setup(usual_setup_func, usual_teardown_func)
def test_clamp_xattr_false_positive(clamp):
# directories 'a' and 'b' obviously do not match
# extra files are needed to satisfy preprocessing, which compares file size
create_file('xxx', '1')
create_file('xxx', 'a/1')
create_file('x', '2')
create_file('x', 'b/2')
# we used to write xattrs even when clamping is used
head, *data, foot = run_rmlint('--xattr --size 3', clamp)
assert len([e for e in data if e['type'] == 'duplicate_file']) == 2 # '1' matches 'a/1'
# fill in other xattrs
head, *data, foot = run_rmlint('--xattr', force_no_pendantic=True)
assert len([e for e in data if e['type'] == 'duplicate_file']) == 4 # '1' matches 'a/1', '2' matches 'b/2'
# we would get a false positive here, as the xattrs matched
head, *data, foot = run_rmlint('--xattr -T dd', force_no_pendantic=True)
assert not any(e['type'] == 'duplicate_dir' for e in data) # do 'a' and 'b' match?
|
