1"""Test cases for the fnmatch module."""
2
3import unittest
4import os
5import string
6import warnings
7
8from fnmatch import fnmatch, fnmatchcase, translate, filter
9
10class FnmatchTestCase(unittest.TestCase):
11
12    def check_match(self, filename, pattern, should_match=True, fn=fnmatch):
13        if should_match:
14            self.assertTrue(fn(filename, pattern),
15                         "expected %r to match pattern %r"
16                         % (filename, pattern))
17        else:
18            self.assertFalse(fn(filename, pattern),
19                         "expected %r not to match pattern %r"
20                         % (filename, pattern))
21
22    def test_fnmatch(self):
23        check = self.check_match
24        check('abc', 'abc')
25        check('abc', '?*?')
26        check('abc', '???*')
27        check('abc', '*???')
28        check('abc', '???')
29        check('abc', '*')
30        check('abc', 'ab[cd]')
31        check('abc', 'ab[!de]')
32        check('abc', 'ab[de]', False)
33        check('a', '??', False)
34        check('a', 'b', False)
35
36        # these test that '\' is handled correctly in character sets;
37        # see SF bug #409651
38        check('\\', r'[\]')
39        check('a', r'[!\]')
40        check('\\', r'[!\]', False)
41
42        # test that filenames with newlines in them are handled correctly.
43        # http://bugs.python.org/issue6665
44        check('foo\nbar', 'foo*')
45        check('foo\nbar\n', 'foo*')
46        check('\nfoo', 'foo*', False)
47        check('\n', '*')
48
49    def test_slow_fnmatch(self):
50        check = self.check_match
51        check('a' * 50, '*a*a*a*a*a*a*a*a*a*a')
52        # The next "takes forever" if the regexp translation is
53        # straightforward.  See bpo-40480.
54        check('a' * 50 + 'b', '*a*a*a*a*a*a*a*a*a*a', False)
55
56    def test_mix_bytes_str(self):
57        self.assertRaises(TypeError, fnmatch, 'test', b'*')
58        self.assertRaises(TypeError, fnmatch, b'test', '*')
59        self.assertRaises(TypeError, fnmatchcase, 'test', b'*')
60        self.assertRaises(TypeError, fnmatchcase, b'test', '*')
61
62    def test_fnmatchcase(self):
63        check = self.check_match
64        check('abc', 'abc', True, fnmatchcase)
65        check('AbC', 'abc', False, fnmatchcase)
66        check('abc', 'AbC', False, fnmatchcase)
67        check('AbC', 'AbC', True, fnmatchcase)
68
69        check('usr/bin', 'usr/bin', True, fnmatchcase)
70        check('usr\\bin', 'usr/bin', False, fnmatchcase)
71        check('usr/bin', 'usr\\bin', False, fnmatchcase)
72        check('usr\\bin', 'usr\\bin', True, fnmatchcase)
73
74    def test_bytes(self):
75        self.check_match(b'test', b'te*')
76        self.check_match(b'test\xff', b'te*\xff')
77        self.check_match(b'foo\nbar', b'foo*')
78
79    def test_case(self):
80        ignorecase = os.path.normcase('ABC') == os.path.normcase('abc')
81        check = self.check_match
82        check('abc', 'abc')
83        check('AbC', 'abc', ignorecase)
84        check('abc', 'AbC', ignorecase)
85        check('AbC', 'AbC')
86
87    def test_sep(self):
88        normsep = os.path.normcase('\\') == os.path.normcase('/')
89        check = self.check_match
90        check('usr/bin', 'usr/bin')
91        check('usr\\bin', 'usr/bin', normsep)
92        check('usr/bin', 'usr\\bin', normsep)
93        check('usr\\bin', 'usr\\bin')
94
95    def test_char_set(self):
96        ignorecase = os.path.normcase('ABC') == os.path.normcase('abc')
97        check = self.check_match
98        tescases = string.ascii_lowercase + string.digits + string.punctuation
99        for c in tescases:
100            check(c, '[az]', c in 'az')
101            check(c, '[!az]', c not in 'az')
102        # Case insensitive.
103        for c in tescases:
104            check(c, '[AZ]', (c in 'az') and ignorecase)
105            check(c, '[!AZ]', (c not in 'az') or not ignorecase)
106        for c in string.ascii_uppercase:
107            check(c, '[az]', (c in 'AZ') and ignorecase)
108            check(c, '[!az]', (c not in 'AZ') or not ignorecase)
109        # Repeated same character.
110        for c in tescases:
111            check(c, '[aa]', c == 'a')
112        # Special cases.
113        for c in tescases:
114            check(c, '[^az]', c in '^az')
115            check(c, '[[az]', c in '[az')
116            check(c, r'[!]]', c != ']')
117        check('[', '[')
118        check('[]', '[]')
119        check('[!', '[!')
120        check('[!]', '[!]')
121
122    def test_range(self):
123        ignorecase = os.path.normcase('ABC') == os.path.normcase('abc')
124        normsep = os.path.normcase('\\') == os.path.normcase('/')
125        check = self.check_match
126        tescases = string.ascii_lowercase + string.digits + string.punctuation
127        for c in tescases:
128            check(c, '[b-d]', c in 'bcd')
129            check(c, '[!b-d]', c not in 'bcd')
130            check(c, '[b-dx-z]', c in 'bcdxyz')
131            check(c, '[!b-dx-z]', c not in 'bcdxyz')
132        # Case insensitive.
133        for c in tescases:
134            check(c, '[B-D]', (c in 'bcd') and ignorecase)
135            check(c, '[!B-D]', (c not in 'bcd') or not ignorecase)
136        for c in string.ascii_uppercase:
137            check(c, '[b-d]', (c in 'BCD') and ignorecase)
138            check(c, '[!b-d]', (c not in 'BCD') or not ignorecase)
139        # Upper bound == lower bound.
140        for c in tescases:
141            check(c, '[b-b]', c == 'b')
142        # Special cases.
143        for c in tescases:
144            check(c, '[!-#]', c not in '-#')
145            check(c, '[!--.]', c not in '-.')
146            check(c, '[^-`]', c in '^_`')
147            if not (normsep and c == '/'):
148                check(c, '[[-^]', c in r'[\]^')
149                check(c, r'[\-^]', c in r'\]^')
150            check(c, '[b-]', c in '-b')
151            check(c, '[!b-]', c not in '-b')
152            check(c, '[-b]', c in '-b')
153            check(c, '[!-b]', c not in '-b')
154            check(c, '[-]', c in '-')
155            check(c, '[!-]', c not in '-')
156        # Upper bound is less that lower bound: error in RE.
157        for c in tescases:
158            check(c, '[d-b]', False)
159            check(c, '[!d-b]', True)
160            check(c, '[d-bx-z]', c in 'xyz')
161            check(c, '[!d-bx-z]', c not in 'xyz')
162            check(c, '[d-b^-`]', c in '^_`')
163            if not (normsep and c == '/'):
164                check(c, '[d-b[-^]', c in r'[\]^')
165
166    def test_sep_in_char_set(self):
167        normsep = os.path.normcase('\\') == os.path.normcase('/')
168        check = self.check_match
169        check('/', r'[/]')
170        check('\\', r'[\]')
171        check('/', r'[\]', normsep)
172        check('\\', r'[/]', normsep)
173        check('[/]', r'[/]', False)
174        check(r'[\\]', r'[/]', False)
175        check('\\', r'[\t]')
176        check('/', r'[\t]', normsep)
177        check('t', r'[\t]')
178        check('\t', r'[\t]', False)
179
180    def test_sep_in_range(self):
181        normsep = os.path.normcase('\\') == os.path.normcase('/')
182        check = self.check_match
183        check('a/b', 'a[.-0]b', not normsep)
184        check('a\\b', 'a[.-0]b', False)
185        check('a\\b', 'a[Z-^]b', not normsep)
186        check('a/b', 'a[Z-^]b', False)
187
188        check('a/b', 'a[/-0]b', not normsep)
189        check(r'a\b', 'a[/-0]b', False)
190        check('a[/-0]b', 'a[/-0]b', False)
191        check(r'a[\-0]b', 'a[/-0]b', False)
192
193        check('a/b', 'a[.-/]b')
194        check(r'a\b', 'a[.-/]b', normsep)
195        check('a[.-/]b', 'a[.-/]b', False)
196        check(r'a[.-\]b', 'a[.-/]b', False)
197
198        check(r'a\b', r'a[\-^]b')
199        check('a/b', r'a[\-^]b', normsep)
200        check(r'a[\-^]b', r'a[\-^]b', False)
201        check('a[/-^]b', r'a[\-^]b', False)
202
203        check(r'a\b', r'a[Z-\]b', not normsep)
204        check('a/b', r'a[Z-\]b', False)
205        check(r'a[Z-\]b', r'a[Z-\]b', False)
206        check('a[Z-/]b', r'a[Z-\]b', False)
207
208    def test_warnings(self):
209        with warnings.catch_warnings():
210            warnings.simplefilter('error', Warning)
211            check = self.check_match
212            check('[', '[[]')
213            check('&', '[a&&b]')
214            check('|', '[a||b]')
215            check('~', '[a~~b]')
216            check(',', '[a-z+--A-Z]')
217            check('.', '[a-z--/A-Z]')
218
219
220class TranslateTestCase(unittest.TestCase):
221
222    def test_translate(self):
223        import re
224        self.assertEqual(translate('*'), r'(?s:.*)\Z')
225        self.assertEqual(translate('?'), r'(?s:.)\Z')
226        self.assertEqual(translate('a?b*'), r'(?s:a.b.*)\Z')
227        self.assertEqual(translate('[abc]'), r'(?s:[abc])\Z')
228        self.assertEqual(translate('[]]'), r'(?s:[]])\Z')
229        self.assertEqual(translate('[!x]'), r'(?s:[^x])\Z')
230        self.assertEqual(translate('[^x]'), r'(?s:[\^x])\Z')
231        self.assertEqual(translate('[x'), r'(?s:\[x)\Z')
232        # from the docs
233        self.assertEqual(translate('*.txt'), r'(?s:.*\.txt)\Z')
234        # squash consecutive stars
235        self.assertEqual(translate('*********'), r'(?s:.*)\Z')
236        self.assertEqual(translate('A*********'), r'(?s:A.*)\Z')
237        self.assertEqual(translate('*********A'), r'(?s:.*A)\Z')
238        self.assertEqual(translate('A*********?[?]?'), r'(?s:A.*.[?].)\Z')
239        # fancy translation to prevent exponential-time match failure
240        t = translate('**a*a****a')
241        self.assertEqual(t, r'(?s:(?>.*?a)(?>.*?a).*a)\Z')
242        # and try pasting multiple translate results - it's an undocumented
243        # feature that this works
244        r1 = translate('**a**a**a*')
245        r2 = translate('**b**b**b*')
246        r3 = translate('*c*c*c*')
247        fatre = "|".join([r1, r2, r3])
248        self.assertTrue(re.match(fatre, 'abaccad'))
249        self.assertTrue(re.match(fatre, 'abxbcab'))
250        self.assertTrue(re.match(fatre, 'cbabcaxc'))
251        self.assertFalse(re.match(fatre, 'dabccbad'))
252
253class FilterTestCase(unittest.TestCase):
254
255    def test_filter(self):
256        self.assertEqual(filter(['Python', 'Ruby', 'Perl', 'Tcl'], 'P*'),
257                         ['Python', 'Perl'])
258        self.assertEqual(filter([b'Python', b'Ruby', b'Perl', b'Tcl'], b'P*'),
259                         [b'Python', b'Perl'])
260
261    def test_mix_bytes_str(self):
262        self.assertRaises(TypeError, filter, ['test'], b'*')
263        self.assertRaises(TypeError, filter, [b'test'], '*')
264
265    def test_case(self):
266        ignorecase = os.path.normcase('P') == os.path.normcase('p')
267        self.assertEqual(filter(['Test.py', 'Test.rb', 'Test.PL'], '*.p*'),
268                         ['Test.py', 'Test.PL'] if ignorecase else ['Test.py'])
269        self.assertEqual(filter(['Test.py', 'Test.rb', 'Test.PL'], '*.P*'),
270                         ['Test.py', 'Test.PL'] if ignorecase else ['Test.PL'])
271
272    def test_sep(self):
273        normsep = os.path.normcase('\\') == os.path.normcase('/')
274        self.assertEqual(filter(['usr/bin', 'usr', 'usr\\lib'], 'usr/*'),
275                         ['usr/bin', 'usr\\lib'] if normsep else ['usr/bin'])
276        self.assertEqual(filter(['usr/bin', 'usr', 'usr\\lib'], 'usr\\*'),
277                         ['usr/bin', 'usr\\lib'] if normsep else ['usr\\lib'])
278
279
280if __name__ == "__main__":
281    unittest.main()
282