Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 1 | """ Test script for the Unicode implementation. |
| 2 | |
| 3 | Written by Bill Tutt. |
Fredrik Lundh | 06d1268 | 2001-01-24 07:59:11 +0000 | [diff] [blame] | 4 | Modified for Python 2.0 by Fredrik Lundh (fredrik@pythonware.com) |
Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 5 | |
| 6 | (c) Copyright CNRI, All Rights Reserved. NO WARRANTY. |
| 7 | |
| 8 | """#" |
Marc-André Lemburg | 3661908 | 2001-01-17 19:11:13 +0000 | [diff] [blame] | 9 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 10 | import unittest |
Ezio Melotti | 931b8aa | 2011-10-21 21:57:36 +0300 | [diff] [blame] | 11 | import unicodedata |
Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 12 | |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 13 | from test import support |
Ezio Melotti | 931b8aa | 2011-10-21 21:57:36 +0300 | [diff] [blame] | 14 | from http.client import HTTPException |
| 15 | from test.test_normalization import check_version |
Fredrik Lundh | ee865c6 | 2001-01-19 11:00:42 +0000 | [diff] [blame] | 16 | |
Serhiy Storchaka | 5cfc79d | 2014-02-07 10:06:39 +0200 | [diff] [blame^] | 17 | try: |
| 18 | from _testcapi import INT_MAX, PY_SSIZE_T_MAX, UINT_MAX |
| 19 | except ImportError: |
| 20 | INT_MAX = PY_SSIZE_T_MAX = UINT_MAX = 2**64 - 1 |
| 21 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 22 | class UnicodeNamesTest(unittest.TestCase): |
Fredrik Lundh | ee865c6 | 2001-01-19 11:00:42 +0000 | [diff] [blame] | 23 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 24 | def checkletter(self, name, code): |
| 25 | # Helper that put all \N escapes inside eval'd raw strings, |
Tim Peters | 669454e | 2003-03-07 17:30:48 +0000 | [diff] [blame] | 26 | # to make sure this script runs even if the compiler |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 27 | # chokes on \N escapes |
Guido van Rossum | bf4806b | 2007-07-21 00:15:34 +0000 | [diff] [blame] | 28 | res = eval(r'"\N{%s}"' % name) |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 29 | self.assertEqual(res, code) |
| 30 | return res |
Fredrik Lundh | ee865c6 | 2001-01-19 11:00:42 +0000 | [diff] [blame] | 31 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 32 | def test_general(self): |
| 33 | # General and case insensitivity test: |
| 34 | chars = [ |
| 35 | "LATIN CAPITAL LETTER T", |
| 36 | "LATIN SMALL LETTER H", |
| 37 | "LATIN SMALL LETTER E", |
| 38 | "SPACE", |
| 39 | "LATIN SMALL LETTER R", |
| 40 | "LATIN CAPITAL LETTER E", |
| 41 | "LATIN SMALL LETTER D", |
| 42 | "SPACE", |
| 43 | "LATIN SMALL LETTER f", |
| 44 | "LATIN CAPITAL LeTtEr o", |
| 45 | "LATIN SMaLl LETTER x", |
| 46 | "SPACE", |
| 47 | "LATIN SMALL LETTER A", |
| 48 | "LATIN SMALL LETTER T", |
| 49 | "LATIN SMALL LETTER E", |
| 50 | "SPACE", |
| 51 | "LATIN SMALL LETTER T", |
| 52 | "LATIN SMALL LETTER H", |
| 53 | "LATIN SMALL LETTER E", |
| 54 | "SpAcE", |
| 55 | "LATIN SMALL LETTER S", |
| 56 | "LATIN SMALL LETTER H", |
| 57 | "LATIN small LETTER e", |
| 58 | "LATIN small LETTER e", |
| 59 | "LATIN SMALL LETTER P", |
| 60 | "FULL STOP" |
| 61 | ] |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 62 | string = "The rEd fOx ate the sheep." |
Martin v. Löwis | 8579efc | 2002-11-23 17:11:42 +0000 | [diff] [blame] | 63 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 64 | self.assertEqual( |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 65 | "".join([self.checkletter(*args) for args in zip(chars, string)]), |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 66 | string |
| 67 | ) |
Martin v. Löwis | ef7fe2e | 2002-11-23 18:01:32 +0000 | [diff] [blame] | 68 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 69 | def test_ascii_letters(self): |
Guido van Rossum | 805365e | 2007-05-07 22:24:25 +0000 | [diff] [blame] | 70 | for char in "".join(map(chr, range(ord("a"), ord("z")))): |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 71 | name = "LATIN SMALL LETTER %s" % char.upper() |
| 72 | code = unicodedata.lookup(name) |
| 73 | self.assertEqual(unicodedata.name(code), name) |
Fredrik Lundh | 2acb54a | 2001-01-19 11:13:46 +0000 | [diff] [blame] | 74 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 75 | def test_hangul_syllables(self): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 76 | self.checkletter("HANGUL SYLLABLE GA", "\uac00") |
| 77 | self.checkletter("HANGUL SYLLABLE GGWEOSS", "\uafe8") |
| 78 | self.checkletter("HANGUL SYLLABLE DOLS", "\ub3d0") |
| 79 | self.checkletter("HANGUL SYLLABLE RYAN", "\ub7b8") |
| 80 | self.checkletter("HANGUL SYLLABLE MWIK", "\ubba0") |
| 81 | self.checkletter("HANGUL SYLLABLE BBWAEM", "\ubf88") |
| 82 | self.checkletter("HANGUL SYLLABLE SSEOL", "\uc370") |
| 83 | self.checkletter("HANGUL SYLLABLE YI", "\uc758") |
| 84 | self.checkletter("HANGUL SYLLABLE JJYOSS", "\ucb40") |
| 85 | self.checkletter("HANGUL SYLLABLE KYEOLS", "\ucf28") |
| 86 | self.checkletter("HANGUL SYLLABLE PAN", "\ud310") |
| 87 | self.checkletter("HANGUL SYLLABLE HWEOK", "\ud6f8") |
| 88 | self.checkletter("HANGUL SYLLABLE HIH", "\ud7a3") |
Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 89 | |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 90 | self.assertRaises(ValueError, unicodedata.name, "\ud7a4") |
Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 91 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 92 | def test_cjk_unified_ideographs(self): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 93 | self.checkletter("CJK UNIFIED IDEOGRAPH-3400", "\u3400") |
| 94 | self.checkletter("CJK UNIFIED IDEOGRAPH-4DB5", "\u4db5") |
| 95 | self.checkletter("CJK UNIFIED IDEOGRAPH-4E00", "\u4e00") |
Martin v. Löwis | 5cbc71e | 2010-11-22 09:00:02 +0000 | [diff] [blame] | 96 | self.checkletter("CJK UNIFIED IDEOGRAPH-9FCB", "\u9fCB") |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 97 | self.checkletter("CJK UNIFIED IDEOGRAPH-20000", "\U00020000") |
| 98 | self.checkletter("CJK UNIFIED IDEOGRAPH-2A6D6", "\U0002a6d6") |
Martin v. Löwis | 5cbc71e | 2010-11-22 09:00:02 +0000 | [diff] [blame] | 99 | self.checkletter("CJK UNIFIED IDEOGRAPH-2A700", "\U0002A700") |
| 100 | self.checkletter("CJK UNIFIED IDEOGRAPH-2B734", "\U0002B734") |
| 101 | self.checkletter("CJK UNIFIED IDEOGRAPH-2B740", "\U0002B740") |
| 102 | self.checkletter("CJK UNIFIED IDEOGRAPH-2B81D", "\U0002B81D") |
Fredrik Lundh | 0fdb90c | 2001-01-19 09:45:02 +0000 | [diff] [blame] | 103 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 104 | def test_bmp_characters(self): |
Guido van Rossum | 805365e | 2007-05-07 22:24:25 +0000 | [diff] [blame] | 105 | for code in range(0x10000): |
Guido van Rossum | 84fc66d | 2007-05-03 17:18:26 +0000 | [diff] [blame] | 106 | char = chr(code) |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 107 | name = unicodedata.name(char, None) |
| 108 | if name is not None: |
| 109 | self.assertEqual(unicodedata.lookup(name), char) |
Marc-André Lemburg | 6cdec2e | 2000-06-30 09:45:20 +0000 | [diff] [blame] | 110 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 111 | def test_misc_symbols(self): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 112 | self.checkletter("PILCROW SIGN", "\u00b6") |
| 113 | self.checkletter("REPLACEMENT CHARACTER", "\uFFFD") |
| 114 | self.checkletter("HALFWIDTH KATAKANA SEMI-VOICED SOUND MARK", "\uFF9F") |
| 115 | self.checkletter("FULLWIDTH LATIN SMALL LETTER A", "\uFF41") |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 116 | |
Ezio Melotti | 931b8aa | 2011-10-21 21:57:36 +0300 | [diff] [blame] | 117 | def test_aliases(self): |
| 118 | # Check that the aliases defined in the NameAliases.txt file work. |
| 119 | # This should be updated when new aliases are added or the file |
| 120 | # should be downloaded and parsed instead. See #12753. |
| 121 | aliases = [ |
| 122 | ('LATIN CAPITAL LETTER GHA', 0x01A2), |
| 123 | ('LATIN SMALL LETTER GHA', 0x01A3), |
| 124 | ('KANNADA LETTER LLLA', 0x0CDE), |
| 125 | ('LAO LETTER FO FON', 0x0E9D), |
| 126 | ('LAO LETTER FO FAY', 0x0E9F), |
| 127 | ('LAO LETTER RO', 0x0EA3), |
| 128 | ('LAO LETTER LO', 0x0EA5), |
| 129 | ('TIBETAN MARK BKA- SHOG GI MGO RGYAN', 0x0FD0), |
| 130 | ('YI SYLLABLE ITERATION MARK', 0xA015), |
| 131 | ('PRESENTATION FORM FOR VERTICAL RIGHT WHITE LENTICULAR BRACKET', 0xFE18), |
| 132 | ('BYZANTINE MUSICAL SYMBOL FTHORA SKLIRON CHROMA VASIS', 0x1D0C5) |
| 133 | ] |
| 134 | for alias, codepoint in aliases: |
| 135 | self.checkletter(alias, chr(codepoint)) |
| 136 | name = unicodedata.name(chr(codepoint)) |
| 137 | self.assertNotEqual(name, alias) |
| 138 | self.assertEqual(unicodedata.lookup(alias), |
| 139 | unicodedata.lookup(name)) |
| 140 | with self.assertRaises(KeyError): |
| 141 | unicodedata.ucd_3_2_0.lookup(alias) |
| 142 | |
| 143 | def test_aliases_names_in_pua_range(self): |
| 144 | # We are storing aliases in the PUA 15, but their names shouldn't leak |
| 145 | for cp in range(0xf0000, 0xf0100): |
| 146 | with self.assertRaises(ValueError) as cm: |
| 147 | unicodedata.name(chr(cp)) |
| 148 | self.assertEqual(str(cm.exception), 'no such name') |
| 149 | |
| 150 | def test_named_sequences_names_in_pua_range(self): |
| 151 | # We are storing named seq in the PUA 15, but their names shouldn't leak |
| 152 | for cp in range(0xf0100, 0xf0fff): |
| 153 | with self.assertRaises(ValueError) as cm: |
| 154 | unicodedata.name(chr(cp)) |
| 155 | self.assertEqual(str(cm.exception), 'no such name') |
| 156 | |
| 157 | def test_named_sequences_sample(self): |
| 158 | # Check a few named sequences. See #12753. |
| 159 | sequences = [ |
| 160 | ('LATIN SMALL LETTER R WITH TILDE', '\u0072\u0303'), |
| 161 | ('TAMIL SYLLABLE SAI', '\u0BB8\u0BC8'), |
| 162 | ('TAMIL SYLLABLE MOO', '\u0BAE\u0BCB'), |
| 163 | ('TAMIL SYLLABLE NNOO', '\u0BA3\u0BCB'), |
| 164 | ('TAMIL CONSONANT KSS', '\u0B95\u0BCD\u0BB7\u0BCD'), |
| 165 | ] |
| 166 | for seqname, codepoints in sequences: |
| 167 | self.assertEqual(unicodedata.lookup(seqname), codepoints) |
| 168 | with self.assertRaises(SyntaxError): |
| 169 | self.checkletter(seqname, None) |
| 170 | with self.assertRaises(KeyError): |
| 171 | unicodedata.ucd_3_2_0.lookup(seqname) |
| 172 | |
| 173 | def test_named_sequences_full(self): |
| 174 | # Check all the named sequences |
| 175 | url = ("http://www.unicode.org/Public/%s/ucd/NamedSequences.txt" % |
| 176 | unicodedata.unidata_version) |
| 177 | try: |
| 178 | testdata = support.open_urlresource(url, encoding="utf-8", |
| 179 | check=check_version) |
| 180 | except (IOError, HTTPException): |
| 181 | self.skipTest("Could not retrieve " + url) |
| 182 | self.addCleanup(testdata.close) |
| 183 | for line in testdata: |
| 184 | line = line.strip() |
| 185 | if not line or line.startswith('#'): |
| 186 | continue |
| 187 | seqname, codepoints = line.split(';') |
| 188 | codepoints = ''.join(chr(int(cp, 16)) for cp in codepoints.split()) |
| 189 | self.assertEqual(unicodedata.lookup(seqname), codepoints) |
| 190 | with self.assertRaises(SyntaxError): |
| 191 | self.checkletter(seqname, None) |
| 192 | with self.assertRaises(KeyError): |
| 193 | unicodedata.ucd_3_2_0.lookup(seqname) |
| 194 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 195 | def test_errors(self): |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 196 | self.assertRaises(TypeError, unicodedata.name) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 197 | self.assertRaises(TypeError, unicodedata.name, 'xx') |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 198 | self.assertRaises(TypeError, unicodedata.lookup) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 199 | self.assertRaises(KeyError, unicodedata.lookup, 'unknown') |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 200 | |
Guido van Rossum | bf4806b | 2007-07-21 00:15:34 +0000 | [diff] [blame] | 201 | def test_strict_error_handling(self): |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 202 | # bogus character name |
| 203 | self.assertRaises( |
| 204 | UnicodeError, |
Guido van Rossum | bf4806b | 2007-07-21 00:15:34 +0000 | [diff] [blame] | 205 | str, b"\\N{blah}", 'unicode-escape', 'strict' |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 206 | ) |
| 207 | # long bogus character name |
| 208 | self.assertRaises( |
| 209 | UnicodeError, |
Guido van Rossum | 9c62772 | 2007-08-27 18:31:48 +0000 | [diff] [blame] | 210 | str, bytes("\\N{%s}" % ("x" * 100000), "ascii"), 'unicode-escape', 'strict' |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 211 | ) |
| 212 | # missing closing brace |
| 213 | self.assertRaises( |
| 214 | UnicodeError, |
Guido van Rossum | bf4806b | 2007-07-21 00:15:34 +0000 | [diff] [blame] | 215 | str, b"\\N{SPACE", 'unicode-escape', 'strict' |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 216 | ) |
| 217 | # missing opening brace |
| 218 | self.assertRaises( |
| 219 | UnicodeError, |
Guido van Rossum | bf4806b | 2007-07-21 00:15:34 +0000 | [diff] [blame] | 220 | str, b"\\NSPACE", 'unicode-escape', 'strict' |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 221 | ) |
| 222 | |
Serhiy Storchaka | 5cfc79d | 2014-02-07 10:06:39 +0200 | [diff] [blame^] | 223 | @support.cpython_only |
| 224 | @unittest.skipUnless(INT_MAX < PY_SSIZE_T_MAX, "needs UINT_MAX < SIZE_MAX") |
| 225 | @support.bigmemtest(size=UINT_MAX + 1, memuse=2 + 1, dry_run=False) |
Serhiy Storchaka | e45dac4 | 2013-01-21 20:23:58 +0200 | [diff] [blame] | 226 | def test_issue16335(self, size): |
Serhiy Storchaka | 4f5f0e5 | 2013-01-21 11:38:00 +0200 | [diff] [blame] | 227 | # very very long bogus character name |
Serhiy Storchaka | 5cfc79d | 2014-02-07 10:06:39 +0200 | [diff] [blame^] | 228 | x = b'\\N{SPACE' + b'x' * (UINT_MAX + 1) + b'}' |
| 229 | self.assertEqual(len(x), len(b'\\N{SPACE}') + (UINT_MAX + 1)) |
Serhiy Storchaka | e45dac4 | 2013-01-21 20:23:58 +0200 | [diff] [blame] | 230 | self.assertRaisesRegex(UnicodeError, |
| 231 | 'unknown Unicode character name', |
| 232 | x.decode, 'unicode-escape' |
| 233 | ) |
Serhiy Storchaka | 4f5f0e5 | 2013-01-21 11:38:00 +0200 | [diff] [blame] | 234 | |
| 235 | |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 236 | def test_main(): |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 237 | support.run_unittest(UnicodeNamesTest) |
Walter Dörwald | 37c4728 | 2003-02-26 14:49:41 +0000 | [diff] [blame] | 238 | |
| 239 | if __name__ == "__main__": |
| 240 | test_main() |