summaryrefslogtreecommitdiffstats
path: root/Lib/test/test_pep277.py
blob: 98c716b4c7a6ac99352a6303d4b1031fcf08189f (plain)
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
# Test the Unicode versions of normal file functions
# open, os.open, os.stat. os.listdir, os.rename, os.remove, os.mkdir, os.chdir, os.rmdir
import os
import sys
import unittest
import warnings
from unicodedata import normalize
from test import support

filenames = [
    '1_abc',
    '2_ascii',
    '3_Gr\xfc\xdf-Gott',
    '4_\u0393\u03b5\u03b9\u03ac-\u03c3\u03b1\u03c2',
    '5_\u0417\u0434\u0440\u0430\u0432\u0441\u0442\u0432\u0443\u0439\u0442\u0435',
    '6_\u306b\u307d\u3093',
    '7_\u05d4\u05e9\u05e7\u05e6\u05e5\u05e1',
    '8_\u66e8\u66e9\u66eb',
    '9_\u66e8\u05e9\u3093\u0434\u0393\xdf',
    # Specific code points: fn, NFC(fn) and NFKC(fn) all differents
    '10_\u1fee\u1ffd',
    ]

# Mac OS X decomposes Unicode names, using Normal Form D.
# http://developer.apple.com/mac/library/qa/qa2001/qa1173.html
# "However, most volume formats do not follow the exact specification for
# these normal forms.  For example, HFS Plus uses a variant of Normal Form D
# in which U+2000 through U+2FFF, U+F900 through U+FAFF, and U+2F800 through
# U+2FAFF are not decomposed."
if sys.platform != 'darwin':
    filenames.extend([
        # Specific code points: NFC(fn), NFD(fn), NFKC(fn) and NFKD(fn) all differents
        '11_\u0385\u03d3\u03d4',
        '12_\u00a8\u0301\u03d2\u0301\u03d2\u0308', # == NFD('\u0385\u03d3\u03d4')
        '13_\u0020\u0308\u0301\u038e\u03ab',       # == NFKC('\u0385\u03d3\u03d4')
        '14_\u1e9b\u1fc1\u1fcd\u1fce\u1fcf\u1fdd\u1fde\u1fdf\u1fed',

        # Specific code points: fn, NFC(fn) and NFKC(fn) all differents
        '15_\u1fee\u1ffd\ufad1',
        '16_\u2000\u2000\u2000A',
        '17_\u2001\u2001\u2001A',
        '18_\u2003\u2003\u2003A',  # == NFC('\u2001\u2001\u2001A')
        '19_\u0020\u0020\u0020A',  # '\u0020' == ' ' == NFKC('\u2000') ==
                                   #  NFKC('\u2001') == NFKC('\u2003')
    ])


# Is it Unicode-friendly?
if not os.path.supports_unicode_filenames:
    fsencoding = sys.getfilesystemencoding()
    try:
        for name in filenames:
            name.encode(fsencoding)
    except UnicodeEncodeError:
        raise unittest.SkipTest("only NT+ and systems with "
                                "Unicode-friendly filesystem encoding")


class UnicodeFileTests(unittest.TestCase):
    files = set(filenames)
    normal_form = None

    def setUp(self):
        try:
            os.mkdir(support.TESTFN)
        except FileExistsError:
            pass
        self.addCleanup(support.rmtree, support.TESTFN)

        files = set()
        for name in self.files:
            name = os.path.join(support.TESTFN, self.norm(name))
            with open(name, 'wb') as f:
                f.write((name+'\n').encode("utf-8"))
            os.stat(name)
            files.add(name)
        self.files = files

    def norm(self, s):
        if self.normal_form:
            return normalize(self.normal_form, s)
        return s

    def _apply_failure(self, fn, filename,
                       expected_exception=FileNotFoundError,
                       check_filename=True):
        with self.assertRaises(expected_exception) as c:
            fn(filename)
        exc_filename = c.exception.filename
        if check_filename:
            self.assertEqual(exc_filename, filename, "Function '%s(%a) failed "
                             "with bad filename in the exception: %a" %
                             (fn.__name__, filename, exc_filename))

    def test_failures(self):
        # Pass non-existing Unicode filenames all over the place.
        for name in self.files:
            name = "not_" + name
            self._apply_failure(open, name)
            self._apply_failure(os.stat, name)
            self._apply_failure(os.chdir, name)
            self._apply_failure(os.rmdir, name)
            self._apply_failure(os.remove, name)
            self._apply_failure(os.listdir, name)

    if sys.platform == 'win32':
        # Windows is lunatic. Issue #13366.
        _listdir_failure = NotADirectoryError, FileNotFoundError
    else:
        _listdir_failure = NotADirectoryError

    def test_open(self):
        for name in self.files:
            f = open(name, 'wb')
            f.write((name+'\n').encode("utf-8"))
            f.close()
            os.stat(name)
            self._apply_failure(os.listdir, name, self._listdir_failure)

    # Skip the test on darwin, because darwin does normalize the filename to
    # NFD (a variant of Unicode NFD form). Normalize the filename to NFC, NFKC,
    # NFKD in Python is useless, because darwin will normalize it later and so
    # open(), os.stat(), etc. don't raise any exception.
    @unittest.skipIf(sys.platform == 'darwin', 'irrelevant test on Mac OS X')
    def test_normalize(self):
        files = set(self.files)
        others = set()
        for nf in set(['NFC', 'NFD', 'NFKC', 'NFKD']):
            others |= set(normalize(nf, file) for file in files)
        others -= files
        for name in others:
            self._apply_failure(open, name)
            self._apply_failure(os.stat, name)
            self._apply_failure(os.chdir, name)
            self._apply_failure(os.rmdir, name)
            self._apply_failure(os.remove, name)
            self._apply_failure(os.listdir, name)

    # Skip the test on darwin, because darwin uses a normalization different
    # than Python NFD normalization: filenames are different even if we use
    # Python NFD normalization.
    @unittest.skipIf(sys.platform == 'darwin', 'irrelevant test on Mac OS X')
    def test_listdir(self):
        sf0 = set(self.files)
        with warnings.catch_warnings():
            warnings.simplefilter("ignore", DeprecationWarning)
            f1 = os.listdir(support.TESTFN.encode(sys.getfilesystemencoding()))
        f2 = os.listdir(support.TESTFN)
        sf2 = set(os.path.join(support.TESTFN, f) for f in f2)
        self.assertEqual(sf0, sf2, "%a != %a" % (sf0, sf2))
        self.assertEqual(len(f1), len(f2))

    def test_rename(self):
        for name in self.files:
            os.rename(name, "tmp")
            os.rename("tmp", name)

    def test_directory(self):
        dirname = os.path.join(support.TESTFN, 'Gr\xfc\xdf-\u66e8\u66e9\u66eb')
        filename = '\xdf-\u66e8\u66e9\u66eb'
        with support.temp_cwd(dirname):
            with open(filename, 'wb') as f:
                f.write((filename + '\n').encode("utf-8"))
            os.access(filename,os.R_OK)
            os.remove(filename)


class UnicodeNFCFileTests(UnicodeFileTests):
    normal_form = 'NFC'


class UnicodeNFDFileTests(UnicodeFileTests):
    normal_form = 'NFD'


class UnicodeNFKCFileTests(UnicodeFileTests):
    normal_form = 'NFKC'


class UnicodeNFKDFileTests(UnicodeFileTests):
    normal_form = 'NFKD'


def test_main():
    support.run_unittest(
        UnicodeFileTests,
        UnicodeNFCFileTests,
        UnicodeNFDFileTests,
        UnicodeNFKCFileTests,
        UnicodeNFKDFileTests,
    )


if __name__ == "__main__":
    test_main()