1""" Test script for the Unicode implementation. 2 3Written by Marc-Andre Lemburg (mal@lemburg.com). 4 5(c) Copyright CNRI, All Rights Reserved. NO WARRANTY. 6 7""" 8import _string 9import codecs 10import itertools 11import operator 12import struct 13import sys 14import unicodedata 15import unittest 16import warnings 17from test import support, string_tests 18 19# Error handling (bad decoder return) 20def search_function(encoding): 21 def decode1(input, errors="strict"): 22 return 42 # not a tuple 23 def encode1(input, errors="strict"): 24 return 42 # not a tuple 25 def encode2(input, errors="strict"): 26 return (42, 42) # no unicode 27 def decode2(input, errors="strict"): 28 return (42, 42) # no unicode 29 if encoding=="test.unicode1": 30 return (encode1, decode1, None, None) 31 elif encoding=="test.unicode2": 32 return (encode2, decode2, None, None) 33 else: 34 return None 35codecs.register(search_function) 36 37def duplicate_string(text): 38 """ 39 Try to get a fresh clone of the specified text: 40 new object with a reference count of 1. 41 42 This is a best-effort: latin1 single letters and the empty 43 string ('') are singletons and cannot be cloned. 44 """ 45 return text.encode().decode() 46 47class StrSubclass(str): 48 pass 49 50class UnicodeTest(string_tests.CommonTest, 51 string_tests.MixinStrUnicodeUserStringTest, 52 string_tests.MixinStrUnicodeTest, 53 unittest.TestCase): 54 55 type2test = str 56 57 def checkequalnofix(self, result, object, methodname, *args): 58 method = getattr(object, methodname) 59 realresult = method(*args) 60 self.assertEqual(realresult, result) 61 self.assertTrue(type(realresult) is type(result)) 62 63 # if the original is returned make sure that 64 # this doesn't happen with subclasses 65 if realresult is object: 66 class usub(str): 67 def __repr__(self): 68 return 'usub(%r)' % str.__repr__(self) 69 object = usub(object) 70 method = getattr(object, methodname) 71 realresult = method(*args) 72 self.assertEqual(realresult, result) 73 self.assertTrue(object is not realresult) 74 75 def test_literals(self): 76 self.assertEqual('\xff', '\u00ff') 77 self.assertEqual('\uffff', '\U0000ffff') 78 self.assertRaises(SyntaxError, eval, '\'\\Ufffffffe\'') 79 self.assertRaises(SyntaxError, eval, '\'\\Uffffffff\'') 80 self.assertRaises(SyntaxError, eval, '\'\\U%08x\'' % 0x110000) 81 # raw strings should not have unicode escapes 82 self.assertNotEqual(r"\u0020", " ") 83 84 def test_ascii(self): 85 if not sys.platform.startswith('java'): 86 # Test basic sanity of repr() 87 self.assertEqual(ascii('abc'), "'abc'") 88 self.assertEqual(ascii('ab\\c'), "'ab\\\\c'") 89 self.assertEqual(ascii('ab\\'), "'ab\\\\'") 90 self.assertEqual(ascii('\\c'), "'\\\\c'") 91 self.assertEqual(ascii('\\'), "'\\\\'") 92 self.assertEqual(ascii('\n'), "'\\n'") 93 self.assertEqual(ascii('\r'), "'\\r'") 94 self.assertEqual(ascii('\t'), "'\\t'") 95 self.assertEqual(ascii('\b'), "'\\x08'") 96 self.assertEqual(ascii("'\""), """'\\'"'""") 97 self.assertEqual(ascii("'\""), """'\\'"'""") 98 self.assertEqual(ascii("'"), '''"'"''') 99 self.assertEqual(ascii('"'), """'"'""") 100 latin1repr = ( 101 "'\\x00\\x01\\x02\\x03\\x04\\x05\\x06\\x07\\x08\\t\\n\\x0b\\x0c\\r" 102 "\\x0e\\x0f\\x10\\x11\\x12\\x13\\x14\\x15\\x16\\x17\\x18\\x19\\x1a" 103 "\\x1b\\x1c\\x1d\\x1e\\x1f !\"#$%&\\'()*+,-./0123456789:;<=>?@ABCDEFGHI" 104 "JKLMNOPQRSTUVWXYZ[\\\\]^_`abcdefghijklmnopqrstuvwxyz{|}~\\x7f" 105 "\\x80\\x81\\x82\\x83\\x84\\x85\\x86\\x87\\x88\\x89\\x8a\\x8b\\x8c\\x8d" 106 "\\x8e\\x8f\\x90\\x91\\x92\\x93\\x94\\x95\\x96\\x97\\x98\\x99\\x9a\\x9b" 107 "\\x9c\\x9d\\x9e\\x9f\\xa0\\xa1\\xa2\\xa3\\xa4\\xa5\\xa6\\xa7\\xa8\\xa9" 108 "\\xaa\\xab\\xac\\xad\\xae\\xaf\\xb0\\xb1\\xb2\\xb3\\xb4\\xb5\\xb6\\xb7" 109 "\\xb8\\xb9\\xba\\xbb\\xbc\\xbd\\xbe\\xbf\\xc0\\xc1\\xc2\\xc3\\xc4\\xc5" 110 "\\xc6\\xc7\\xc8\\xc9\\xca\\xcb\\xcc\\xcd\\xce\\xcf\\xd0\\xd1\\xd2\\xd3" 111 "\\xd4\\xd5\\xd6\\xd7\\xd8\\xd9\\xda\\xdb\\xdc\\xdd\\xde\\xdf\\xe0\\xe1" 112 "\\xe2\\xe3\\xe4\\xe5\\xe6\\xe7\\xe8\\xe9\\xea\\xeb\\xec\\xed\\xee\\xef" 113 "\\xf0\\xf1\\xf2\\xf3\\xf4\\xf5\\xf6\\xf7\\xf8\\xf9\\xfa\\xfb\\xfc\\xfd" 114 "\\xfe\\xff'") 115 testrepr = ascii(''.join(map(chr, range(256)))) 116 self.assertEqual(testrepr, latin1repr) 117 # Test ascii works on wide unicode escapes without overflow. 118 self.assertEqual(ascii("\U00010000" * 39 + "\uffff" * 4096), 119 ascii("\U00010000" * 39 + "\uffff" * 4096)) 120 121 class WrongRepr: 122 def __repr__(self): 123 return b'byte-repr' 124 self.assertRaises(TypeError, ascii, WrongRepr()) 125 126 def test_repr(self): 127 if not sys.platform.startswith('java'): 128 # Test basic sanity of repr() 129 self.assertEqual(repr('abc'), "'abc'") 130 self.assertEqual(repr('ab\\c'), "'ab\\\\c'") 131 self.assertEqual(repr('ab\\'), "'ab\\\\'") 132 self.assertEqual(repr('\\c'), "'\\\\c'") 133 self.assertEqual(repr('\\'), "'\\\\'") 134 self.assertEqual(repr('\n'), "'\\n'") 135 self.assertEqual(repr('\r'), "'\\r'") 136 self.assertEqual(repr('\t'), "'\\t'") 137 self.assertEqual(repr('\b'), "'\\x08'") 138 self.assertEqual(repr("'\""), """'\\'"'""") 139 self.assertEqual(repr("'\""), """'\\'"'""") 140 self.assertEqual(repr("'"), '''"'"''') 141 self.assertEqual(repr('"'), """'"'""") 142 latin1repr = ( 143 "'\\x00\\x01\\x02\\x03\\x04\\x05\\x06\\x07\\x08\\t\\n\\x0b\\x0c\\r" 144 "\\x0e\\x0f\\x10\\x11\\x12\\x13\\x14\\x15\\x16\\x17\\x18\\x19\\x1a" 145 "\\x1b\\x1c\\x1d\\x1e\\x1f !\"#$%&\\'()*+,-./0123456789:;<=>?@ABCDEFGHI" 146 "JKLMNOPQRSTUVWXYZ[\\\\]^_`abcdefghijklmnopqrstuvwxyz{|}~\\x7f" 147 "\\x80\\x81\\x82\\x83\\x84\\x85\\x86\\x87\\x88\\x89\\x8a\\x8b\\x8c\\x8d" 148 "\\x8e\\x8f\\x90\\x91\\x92\\x93\\x94\\x95\\x96\\x97\\x98\\x99\\x9a\\x9b" 149 "\\x9c\\x9d\\x9e\\x9f\\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9" 150 "\xaa\xab\xac\\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7" 151 "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5" 152 "\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3" 153 "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1" 154 "\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef" 155 "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd" 156 "\xfe\xff'") 157 testrepr = repr(''.join(map(chr, range(256)))) 158 self.assertEqual(testrepr, latin1repr) 159 # Test repr works on wide unicode escapes without overflow. 160 self.assertEqual(repr("\U00010000" * 39 + "\uffff" * 4096), 161 repr("\U00010000" * 39 + "\uffff" * 4096)) 162 163 class WrongRepr: 164 def __repr__(self): 165 return b'byte-repr' 166 self.assertRaises(TypeError, repr, WrongRepr()) 167 168 def test_iterators(self): 169 # Make sure unicode objects have an __iter__ method 170 it = "\u1111\u2222\u3333".__iter__() 171 self.assertEqual(next(it), "\u1111") 172 self.assertEqual(next(it), "\u2222") 173 self.assertEqual(next(it), "\u3333") 174 self.assertRaises(StopIteration, next, it) 175 176 def test_count(self): 177 string_tests.CommonTest.test_count(self) 178 # check mixed argument types 179 self.checkequalnofix(3, 'aaa', 'count', 'a') 180 self.checkequalnofix(0, 'aaa', 'count', 'b') 181 self.checkequalnofix(3, 'aaa', 'count', 'a') 182 self.checkequalnofix(0, 'aaa', 'count', 'b') 183 self.checkequalnofix(0, 'aaa', 'count', 'b') 184 self.checkequalnofix(1, 'aaa', 'count', 'a', -1) 185 self.checkequalnofix(3, 'aaa', 'count', 'a', -10) 186 self.checkequalnofix(2, 'aaa', 'count', 'a', 0, -1) 187 self.checkequalnofix(0, 'aaa', 'count', 'a', 0, -10) 188 # test mixed kinds 189 self.checkequal(10, '\u0102' + 'a' * 10, 'count', 'a') 190 self.checkequal(10, '\U00100304' + 'a' * 10, 'count', 'a') 191 self.checkequal(10, '\U00100304' + '\u0102' * 10, 'count', '\u0102') 192 self.checkequal(0, 'a' * 10, 'count', '\u0102') 193 self.checkequal(0, 'a' * 10, 'count', '\U00100304') 194 self.checkequal(0, '\u0102' * 10, 'count', '\U00100304') 195 self.checkequal(10, '\u0102' + 'a_' * 10, 'count', 'a_') 196 self.checkequal(10, '\U00100304' + 'a_' * 10, 'count', 'a_') 197 self.checkequal(10, '\U00100304' + '\u0102_' * 10, 'count', '\u0102_') 198 self.checkequal(0, 'a' * 10, 'count', 'a\u0102') 199 self.checkequal(0, 'a' * 10, 'count', 'a\U00100304') 200 self.checkequal(0, '\u0102' * 10, 'count', '\u0102\U00100304') 201 202 def test_find(self): 203 string_tests.CommonTest.test_find(self) 204 # test implementation details of the memchr fast path 205 self.checkequal(100, 'a' * 100 + '\u0102', 'find', '\u0102') 206 self.checkequal(-1, 'a' * 100 + '\u0102', 'find', '\u0201') 207 self.checkequal(-1, 'a' * 100 + '\u0102', 'find', '\u0120') 208 self.checkequal(-1, 'a' * 100 + '\u0102', 'find', '\u0220') 209 self.checkequal(100, 'a' * 100 + '\U00100304', 'find', '\U00100304') 210 self.checkequal(-1, 'a' * 100 + '\U00100304', 'find', '\U00100204') 211 self.checkequal(-1, 'a' * 100 + '\U00100304', 'find', '\U00102004') 212 # check mixed argument types 213 self.checkequalnofix(0, 'abcdefghiabc', 'find', 'abc') 214 self.checkequalnofix(9, 'abcdefghiabc', 'find', 'abc', 1) 215 self.checkequalnofix(-1, 'abcdefghiabc', 'find', 'def', 4) 216 217 self.assertRaises(TypeError, 'hello'.find) 218 self.assertRaises(TypeError, 'hello'.find, 42) 219 # test mixed kinds 220 self.checkequal(100, '\u0102' * 100 + 'a', 'find', 'a') 221 self.checkequal(100, '\U00100304' * 100 + 'a', 'find', 'a') 222 self.checkequal(100, '\U00100304' * 100 + '\u0102', 'find', '\u0102') 223 self.checkequal(-1, 'a' * 100, 'find', '\u0102') 224 self.checkequal(-1, 'a' * 100, 'find', '\U00100304') 225 self.checkequal(-1, '\u0102' * 100, 'find', '\U00100304') 226 self.checkequal(100, '\u0102' * 100 + 'a_', 'find', 'a_') 227 self.checkequal(100, '\U00100304' * 100 + 'a_', 'find', 'a_') 228 self.checkequal(100, '\U00100304' * 100 + '\u0102_', 'find', '\u0102_') 229 self.checkequal(-1, 'a' * 100, 'find', 'a\u0102') 230 self.checkequal(-1, 'a' * 100, 'find', 'a\U00100304') 231 self.checkequal(-1, '\u0102' * 100, 'find', '\u0102\U00100304') 232 233 def test_rfind(self): 234 string_tests.CommonTest.test_rfind(self) 235 # test implementation details of the memrchr fast path 236 self.checkequal(0, '\u0102' + 'a' * 100 , 'rfind', '\u0102') 237 self.checkequal(-1, '\u0102' + 'a' * 100 , 'rfind', '\u0201') 238 self.checkequal(-1, '\u0102' + 'a' * 100 , 'rfind', '\u0120') 239 self.checkequal(-1, '\u0102' + 'a' * 100 , 'rfind', '\u0220') 240 self.checkequal(0, '\U00100304' + 'a' * 100, 'rfind', '\U00100304') 241 self.checkequal(-1, '\U00100304' + 'a' * 100, 'rfind', '\U00100204') 242 self.checkequal(-1, '\U00100304' + 'a' * 100, 'rfind', '\U00102004') 243 # check mixed argument types 244 self.checkequalnofix(9, 'abcdefghiabc', 'rfind', 'abc') 245 self.checkequalnofix(12, 'abcdefghiabc', 'rfind', '') 246 self.checkequalnofix(12, 'abcdefghiabc', 'rfind', '') 247 # test mixed kinds 248 self.checkequal(0, 'a' + '\u0102' * 100, 'rfind', 'a') 249 self.checkequal(0, 'a' + '\U00100304' * 100, 'rfind', 'a') 250 self.checkequal(0, '\u0102' + '\U00100304' * 100, 'rfind', '\u0102') 251 self.checkequal(-1, 'a' * 100, 'rfind', '\u0102') 252 self.checkequal(-1, 'a' * 100, 'rfind', '\U00100304') 253 self.checkequal(-1, '\u0102' * 100, 'rfind', '\U00100304') 254 self.checkequal(0, '_a' + '\u0102' * 100, 'rfind', '_a') 255 self.checkequal(0, '_a' + '\U00100304' * 100, 'rfind', '_a') 256 self.checkequal(0, '_\u0102' + '\U00100304' * 100, 'rfind', '_\u0102') 257 self.checkequal(-1, 'a' * 100, 'rfind', '\u0102a') 258 self.checkequal(-1, 'a' * 100, 'rfind', '\U00100304a') 259 self.checkequal(-1, '\u0102' * 100, 'rfind', '\U00100304\u0102') 260 261 def test_index(self): 262 string_tests.CommonTest.test_index(self) 263 self.checkequalnofix(0, 'abcdefghiabc', 'index', '') 264 self.checkequalnofix(3, 'abcdefghiabc', 'index', 'def') 265 self.checkequalnofix(0, 'abcdefghiabc', 'index', 'abc') 266 self.checkequalnofix(9, 'abcdefghiabc', 'index', 'abc', 1) 267 self.assertRaises(ValueError, 'abcdefghiabc'.index, 'hib') 268 self.assertRaises(ValueError, 'abcdefghiab'.index, 'abc', 1) 269 self.assertRaises(ValueError, 'abcdefghi'.index, 'ghi', 8) 270 self.assertRaises(ValueError, 'abcdefghi'.index, 'ghi', -1) 271 # test mixed kinds 272 self.checkequal(100, '\u0102' * 100 + 'a', 'index', 'a') 273 self.checkequal(100, '\U00100304' * 100 + 'a', 'index', 'a') 274 self.checkequal(100, '\U00100304' * 100 + '\u0102', 'index', '\u0102') 275 self.assertRaises(ValueError, ('a' * 100).index, '\u0102') 276 self.assertRaises(ValueError, ('a' * 100).index, '\U00100304') 277 self.assertRaises(ValueError, ('\u0102' * 100).index, '\U00100304') 278 self.checkequal(100, '\u0102' * 100 + 'a_', 'index', 'a_') 279 self.checkequal(100, '\U00100304' * 100 + 'a_', 'index', 'a_') 280 self.checkequal(100, '\U00100304' * 100 + '\u0102_', 'index', '\u0102_') 281 self.assertRaises(ValueError, ('a' * 100).index, 'a\u0102') 282 self.assertRaises(ValueError, ('a' * 100).index, 'a\U00100304') 283 self.assertRaises(ValueError, ('\u0102' * 100).index, '\u0102\U00100304') 284 285 def test_rindex(self): 286 string_tests.CommonTest.test_rindex(self) 287 self.checkequalnofix(12, 'abcdefghiabc', 'rindex', '') 288 self.checkequalnofix(3, 'abcdefghiabc', 'rindex', 'def') 289 self.checkequalnofix(9, 'abcdefghiabc', 'rindex', 'abc') 290 self.checkequalnofix(0, 'abcdefghiabc', 'rindex', 'abc', 0, -1) 291 292 self.assertRaises(ValueError, 'abcdefghiabc'.rindex, 'hib') 293 self.assertRaises(ValueError, 'defghiabc'.rindex, 'def', 1) 294 self.assertRaises(ValueError, 'defghiabc'.rindex, 'abc', 0, -1) 295 self.assertRaises(ValueError, 'abcdefghi'.rindex, 'ghi', 0, 8) 296 self.assertRaises(ValueError, 'abcdefghi'.rindex, 'ghi', 0, -1) 297 # test mixed kinds 298 self.checkequal(0, 'a' + '\u0102' * 100, 'rindex', 'a') 299 self.checkequal(0, 'a' + '\U00100304' * 100, 'rindex', 'a') 300 self.checkequal(0, '\u0102' + '\U00100304' * 100, 'rindex', '\u0102') 301 self.assertRaises(ValueError, ('a' * 100).rindex, '\u0102') 302 self.assertRaises(ValueError, ('a' * 100).rindex, '\U00100304') 303 self.assertRaises(ValueError, ('\u0102' * 100).rindex, '\U00100304') 304 self.checkequal(0, '_a' + '\u0102' * 100, 'rindex', '_a') 305 self.checkequal(0, '_a' + '\U00100304' * 100, 'rindex', '_a') 306 self.checkequal(0, '_\u0102' + '\U00100304' * 100, 'rindex', '_\u0102') 307 self.assertRaises(ValueError, ('a' * 100).rindex, '\u0102a') 308 self.assertRaises(ValueError, ('a' * 100).rindex, '\U00100304a') 309 self.assertRaises(ValueError, ('\u0102' * 100).rindex, '\U00100304\u0102') 310 311 def test_maketrans_translate(self): 312 # these work with plain translate() 313 self.checkequalnofix('bbbc', 'abababc', 'translate', 314 {ord('a'): None}) 315 self.checkequalnofix('iiic', 'abababc', 'translate', 316 {ord('a'): None, ord('b'): ord('i')}) 317 self.checkequalnofix('iiix', 'abababc', 'translate', 318 {ord('a'): None, ord('b'): ord('i'), ord('c'): 'x'}) 319 self.checkequalnofix('c', 'abababc', 'translate', 320 {ord('a'): None, ord('b'): ''}) 321 self.checkequalnofix('xyyx', 'xzx', 'translate', 322 {ord('z'): 'yy'}) 323 324 # this needs maketrans() 325 self.checkequalnofix('abababc', 'abababc', 'translate', 326 {'b': '<i>'}) 327 tbl = self.type2test.maketrans({'a': None, 'b': '<i>'}) 328 self.checkequalnofix('<i><i><i>c', 'abababc', 'translate', tbl) 329 # test alternative way of calling maketrans() 330 tbl = self.type2test.maketrans('abc', 'xyz', 'd') 331 self.checkequalnofix('xyzzy', 'abdcdcbdddd', 'translate', tbl) 332 333 # various tests switching from ASCII to latin1 or the opposite; 334 # same length, remove a letter, or replace with a longer string. 335 self.assertEqual("[a]".translate(str.maketrans('a', 'X')), 336 "[X]") 337 self.assertEqual("[a]".translate(str.maketrans({'a': 'X'})), 338 "[X]") 339 self.assertEqual("[a]".translate(str.maketrans({'a': None})), 340 "[]") 341 self.assertEqual("[a]".translate(str.maketrans({'a': 'XXX'})), 342 "[XXX]") 343 self.assertEqual("[a]".translate(str.maketrans({'a': '\xe9'})), 344 "[\xe9]") 345 self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '123'})), 346 "x123") 347 self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '\xe9'})), 348 "x\xe9") 349 350 # test non-ASCII (don't take the fast-path) 351 self.assertEqual("[a]".translate(str.maketrans({'a': '<\xe9>'})), 352 "[<\xe9>]") 353 self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': 'a'})), 354 "[a]") 355 self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': None})), 356 "[]") 357 self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': '123'})), 358 "[123]") 359 self.assertEqual("[a\xe9]".translate(str.maketrans({'a': '<\u20ac>'})), 360 "[<\u20ac>\xe9]") 361 362 # invalid Unicode characters 363 invalid_char = 0x10ffff+1 364 for before in "a\xe9\u20ac\U0010ffff": 365 mapping = str.maketrans({before: invalid_char}) 366 text = "[%s]" % before 367 self.assertRaises(ValueError, text.translate, mapping) 368 369 # errors 370 self.assertRaises(TypeError, self.type2test.maketrans) 371 self.assertRaises(ValueError, self.type2test.maketrans, 'abc', 'defg') 372 self.assertRaises(TypeError, self.type2test.maketrans, 2, 'def') 373 self.assertRaises(TypeError, self.type2test.maketrans, 'abc', 2) 374 self.assertRaises(TypeError, self.type2test.maketrans, 'abc', 'def', 2) 375 self.assertRaises(ValueError, self.type2test.maketrans, {'xy': 2}) 376 self.assertRaises(TypeError, self.type2test.maketrans, {(1,): 2}) 377 378 self.assertRaises(TypeError, 'hello'.translate) 379 self.assertRaises(TypeError, 'abababc'.translate, 'abc', 'xyz') 380 381 def test_split(self): 382 string_tests.CommonTest.test_split(self) 383 384 # test mixed kinds 385 for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'): 386 left *= 9 387 right *= 9 388 for delim in ('c', '\u0102', '\U00010302'): 389 self.checkequal([left + right], 390 left + right, 'split', delim) 391 self.checkequal([left, right], 392 left + delim + right, 'split', delim) 393 self.checkequal([left + right], 394 left + right, 'split', delim * 2) 395 self.checkequal([left, right], 396 left + delim * 2 + right, 'split', delim *2) 397 398 def test_rsplit(self): 399 string_tests.CommonTest.test_rsplit(self) 400 # test mixed kinds 401 for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'): 402 left *= 9 403 right *= 9 404 for delim in ('c', '\u0102', '\U00010302'): 405 self.checkequal([left + right], 406 left + right, 'rsplit', delim) 407 self.checkequal([left, right], 408 left + delim + right, 'rsplit', delim) 409 self.checkequal([left + right], 410 left + right, 'rsplit', delim * 2) 411 self.checkequal([left, right], 412 left + delim * 2 + right, 'rsplit', delim *2) 413 414 def test_partition(self): 415 string_tests.MixinStrUnicodeUserStringTest.test_partition(self) 416 # test mixed kinds 417 self.checkequal(('ABCDEFGH', '', ''), 'ABCDEFGH', 'partition', '\u4200') 418 for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'): 419 left *= 9 420 right *= 9 421 for delim in ('c', '\u0102', '\U00010302'): 422 self.checkequal((left + right, '', ''), 423 left + right, 'partition', delim) 424 self.checkequal((left, delim, right), 425 left + delim + right, 'partition', delim) 426 self.checkequal((left + right, '', ''), 427 left + right, 'partition', delim * 2) 428 self.checkequal((left, delim * 2, right), 429 left + delim * 2 + right, 'partition', delim * 2) 430 431 def test_rpartition(self): 432 string_tests.MixinStrUnicodeUserStringTest.test_rpartition(self) 433 # test mixed kinds 434 self.checkequal(('', '', 'ABCDEFGH'), 'ABCDEFGH', 'rpartition', '\u4200') 435 for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'): 436 left *= 9 437 right *= 9 438 for delim in ('c', '\u0102', '\U00010302'): 439 self.checkequal(('', '', left + right), 440 left + right, 'rpartition', delim) 441 self.checkequal((left, delim, right), 442 left + delim + right, 'rpartition', delim) 443 self.checkequal(('', '', left + right), 444 left + right, 'rpartition', delim * 2) 445 self.checkequal((left, delim * 2, right), 446 left + delim * 2 + right, 'rpartition', delim * 2) 447 448 def test_join(self): 449 string_tests.MixinStrUnicodeUserStringTest.test_join(self) 450 451 class MyWrapper: 452 def __init__(self, sval): self.sval = sval 453 def __str__(self): return self.sval 454 455 # mixed arguments 456 self.checkequalnofix('a b c d', ' ', 'join', ['a', 'b', 'c', 'd']) 457 self.checkequalnofix('abcd', '', 'join', ('a', 'b', 'c', 'd')) 458 self.checkequalnofix('w x y z', ' ', 'join', string_tests.Sequence('wxyz')) 459 self.checkequalnofix('a b c d', ' ', 'join', ['a', 'b', 'c', 'd']) 460 self.checkequalnofix('a b c d', ' ', 'join', ['a', 'b', 'c', 'd']) 461 self.checkequalnofix('abcd', '', 'join', ('a', 'b', 'c', 'd')) 462 self.checkequalnofix('w x y z', ' ', 'join', string_tests.Sequence('wxyz')) 463 self.checkraises(TypeError, ' ', 'join', ['1', '2', MyWrapper('foo')]) 464 self.checkraises(TypeError, ' ', 'join', ['1', '2', '3', bytes()]) 465 self.checkraises(TypeError, ' ', 'join', [1, 2, 3]) 466 self.checkraises(TypeError, ' ', 'join', ['1', '2', 3]) 467 468 @unittest.skipIf(sys.maxsize > 2**32, 469 'needs too much memory on a 64-bit platform') 470 def test_join_overflow(self): 471 size = int(sys.maxsize**0.5) + 1 472 seq = ('A' * size,) * size 473 self.assertRaises(OverflowError, ''.join, seq) 474 475 def test_replace(self): 476 string_tests.CommonTest.test_replace(self) 477 478 # method call forwarded from str implementation because of unicode argument 479 self.checkequalnofix('one@two!three!', 'one!two!three!', 'replace', '!', '@', 1) 480 self.assertRaises(TypeError, 'replace'.replace, "r", 42) 481 # test mixed kinds 482 for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'): 483 left *= 9 484 right *= 9 485 for delim in ('c', '\u0102', '\U00010302'): 486 for repl in ('d', '\u0103', '\U00010303'): 487 self.checkequal(left + right, 488 left + right, 'replace', delim, repl) 489 self.checkequal(left + repl + right, 490 left + delim + right, 491 'replace', delim, repl) 492 self.checkequal(left + right, 493 left + right, 'replace', delim * 2, repl) 494 self.checkequal(left + repl + right, 495 left + delim * 2 + right, 496 'replace', delim * 2, repl) 497 498 @support.cpython_only 499 def test_replace_id(self): 500 pattern = 'abc' 501 text = 'abc def' 502 self.assertIs(text.replace(pattern, pattern), text) 503 504 def test_bytes_comparison(self): 505 with support.check_warnings(): 506 warnings.simplefilter('ignore', BytesWarning) 507 self.assertEqual('abc' == b'abc', False) 508 self.assertEqual('abc' != b'abc', True) 509 self.assertEqual('abc' == bytearray(b'abc'), False) 510 self.assertEqual('abc' != bytearray(b'abc'), True) 511 512 def test_comparison(self): 513 # Comparisons: 514 self.assertEqual('abc', 'abc') 515 self.assertTrue('abcd' > 'abc') 516 self.assertTrue('abc' < 'abcd') 517 518 if 0: 519 # Move these tests to a Unicode collation module test... 520 # Testing UTF-16 code point order comparisons... 521 522 # No surrogates, no fixup required. 523 self.assertTrue('\u0061' < '\u20ac') 524 # Non surrogate below surrogate value, no fixup required 525 self.assertTrue('\u0061' < '\ud800\udc02') 526 527 # Non surrogate above surrogate value, fixup required 528 def test_lecmp(s, s2): 529 self.assertTrue(s < s2) 530 531 def test_fixup(s): 532 s2 = '\ud800\udc01' 533 test_lecmp(s, s2) 534 s2 = '\ud900\udc01' 535 test_lecmp(s, s2) 536 s2 = '\uda00\udc01' 537 test_lecmp(s, s2) 538 s2 = '\udb00\udc01' 539 test_lecmp(s, s2) 540 s2 = '\ud800\udd01' 541 test_lecmp(s, s2) 542 s2 = '\ud900\udd01' 543 test_lecmp(s, s2) 544 s2 = '\uda00\udd01' 545 test_lecmp(s, s2) 546 s2 = '\udb00\udd01' 547 test_lecmp(s, s2) 548 s2 = '\ud800\ude01' 549 test_lecmp(s, s2) 550 s2 = '\ud900\ude01' 551 test_lecmp(s, s2) 552 s2 = '\uda00\ude01' 553 test_lecmp(s, s2) 554 s2 = '\udb00\ude01' 555 test_lecmp(s, s2) 556 s2 = '\ud800\udfff' 557 test_lecmp(s, s2) 558 s2 = '\ud900\udfff' 559 test_lecmp(s, s2) 560 s2 = '\uda00\udfff' 561 test_lecmp(s, s2) 562 s2 = '\udb00\udfff' 563 test_lecmp(s, s2) 564 565 test_fixup('\ue000') 566 test_fixup('\uff61') 567 568 # Surrogates on both sides, no fixup required 569 self.assertTrue('\ud800\udc02' < '\ud84d\udc56') 570 571 def test_islower(self): 572 super().test_islower() 573 self.checkequalnofix(False, '\u1FFc', 'islower') 574 self.assertFalse('\u2167'.islower()) 575 self.assertTrue('\u2177'.islower()) 576 # non-BMP, uppercase 577 self.assertFalse('\U00010401'.islower()) 578 self.assertFalse('\U00010427'.islower()) 579 # non-BMP, lowercase 580 self.assertTrue('\U00010429'.islower()) 581 self.assertTrue('\U0001044E'.islower()) 582 # non-BMP, non-cased 583 self.assertFalse('\U0001F40D'.islower()) 584 self.assertFalse('\U0001F46F'.islower()) 585 586 def test_isupper(self): 587 super().test_isupper() 588 if not sys.platform.startswith('java'): 589 self.checkequalnofix(False, '\u1FFc', 'isupper') 590 self.assertTrue('\u2167'.isupper()) 591 self.assertFalse('\u2177'.isupper()) 592 # non-BMP, uppercase 593 self.assertTrue('\U00010401'.isupper()) 594 self.assertTrue('\U00010427'.isupper()) 595 # non-BMP, lowercase 596 self.assertFalse('\U00010429'.isupper()) 597 self.assertFalse('\U0001044E'.isupper()) 598 # non-BMP, non-cased 599 self.assertFalse('\U0001F40D'.isupper()) 600 self.assertFalse('\U0001F46F'.isupper()) 601 602 def test_istitle(self): 603 super().test_istitle() 604 self.checkequalnofix(True, '\u1FFc', 'istitle') 605 self.checkequalnofix(True, 'Greek \u1FFcitlecases ...', 'istitle') 606 607 # non-BMP, uppercase + lowercase 608 self.assertTrue('\U00010401\U00010429'.istitle()) 609 self.assertTrue('\U00010427\U0001044E'.istitle()) 610 # apparently there are no titlecased (Lt) non-BMP chars in Unicode 6 611 for ch in ['\U00010429', '\U0001044E', '\U0001F40D', '\U0001F46F']: 612 self.assertFalse(ch.istitle(), '{!a} is not title'.format(ch)) 613 614 def test_isspace(self): 615 super().test_isspace() 616 self.checkequalnofix(True, '\u2000', 'isspace') 617 self.checkequalnofix(True, '\u200a', 'isspace') 618 self.checkequalnofix(False, '\u2014', 'isspace') 619 # There are no non-BMP whitespace chars as of Unicode 12. 620 for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E', 621 '\U0001F40D', '\U0001F46F']: 622 self.assertFalse(ch.isspace(), '{!a} is not space.'.format(ch)) 623 624 @support.requires_resource('cpu') 625 def test_isspace_invariant(self): 626 for codepoint in range(sys.maxunicode + 1): 627 char = chr(codepoint) 628 bidirectional = unicodedata.bidirectional(char) 629 category = unicodedata.category(char) 630 self.assertEqual(char.isspace(), 631 (bidirectional in ('WS', 'B', 'S') 632 or category == 'Zs')) 633 634 def test_isalnum(self): 635 super().test_isalnum() 636 for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E', 637 '\U0001D7F6', '\U00011066', '\U000104A0', '\U0001F107']: 638 self.assertTrue(ch.isalnum(), '{!a} is alnum.'.format(ch)) 639 640 def test_isalpha(self): 641 super().test_isalpha() 642 self.checkequalnofix(True, '\u1FFc', 'isalpha') 643 # non-BMP, cased 644 self.assertTrue('\U00010401'.isalpha()) 645 self.assertTrue('\U00010427'.isalpha()) 646 self.assertTrue('\U00010429'.isalpha()) 647 self.assertTrue('\U0001044E'.isalpha()) 648 # non-BMP, non-cased 649 self.assertFalse('\U0001F40D'.isalpha()) 650 self.assertFalse('\U0001F46F'.isalpha()) 651 652 def test_isascii(self): 653 super().test_isascii() 654 self.assertFalse("\u20ac".isascii()) 655 self.assertFalse("\U0010ffff".isascii()) 656 657 def test_isdecimal(self): 658 self.checkequalnofix(False, '', 'isdecimal') 659 self.checkequalnofix(False, 'a', 'isdecimal') 660 self.checkequalnofix(True, '0', 'isdecimal') 661 self.checkequalnofix(False, '\u2460', 'isdecimal') # CIRCLED DIGIT ONE 662 self.checkequalnofix(False, '\xbc', 'isdecimal') # VULGAR FRACTION ONE QUARTER 663 self.checkequalnofix(True, '\u0660', 'isdecimal') # ARABIC-INDIC DIGIT ZERO 664 self.checkequalnofix(True, '0123456789', 'isdecimal') 665 self.checkequalnofix(False, '0123456789a', 'isdecimal') 666 667 self.checkraises(TypeError, 'abc', 'isdecimal', 42) 668 669 for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E', 670 '\U0001F40D', '\U0001F46F', '\U00011065', '\U0001F107']: 671 self.assertFalse(ch.isdecimal(), '{!a} is not decimal.'.format(ch)) 672 for ch in ['\U0001D7F6', '\U00011066', '\U000104A0']: 673 self.assertTrue(ch.isdecimal(), '{!a} is decimal.'.format(ch)) 674 675 def test_isdigit(self): 676 super().test_isdigit() 677 self.checkequalnofix(True, '\u2460', 'isdigit') 678 self.checkequalnofix(False, '\xbc', 'isdigit') 679 self.checkequalnofix(True, '\u0660', 'isdigit') 680 681 for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E', 682 '\U0001F40D', '\U0001F46F', '\U00011065']: 683 self.assertFalse(ch.isdigit(), '{!a} is not a digit.'.format(ch)) 684 for ch in ['\U0001D7F6', '\U00011066', '\U000104A0', '\U0001F107']: 685 self.assertTrue(ch.isdigit(), '{!a} is a digit.'.format(ch)) 686 687 def test_isnumeric(self): 688 self.checkequalnofix(False, '', 'isnumeric') 689 self.checkequalnofix(False, 'a', 'isnumeric') 690 self.checkequalnofix(True, '0', 'isnumeric') 691 self.checkequalnofix(True, '\u2460', 'isnumeric') 692 self.checkequalnofix(True, '\xbc', 'isnumeric') 693 self.checkequalnofix(True, '\u0660', 'isnumeric') 694 self.checkequalnofix(True, '0123456789', 'isnumeric') 695 self.checkequalnofix(False, '0123456789a', 'isnumeric') 696 697 self.assertRaises(TypeError, "abc".isnumeric, 42) 698 699 for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E', 700 '\U0001F40D', '\U0001F46F']: 701 self.assertFalse(ch.isnumeric(), '{!a} is not numeric.'.format(ch)) 702 for ch in ['\U00011065', '\U0001D7F6', '\U00011066', 703 '\U000104A0', '\U0001F107']: 704 self.assertTrue(ch.isnumeric(), '{!a} is numeric.'.format(ch)) 705 706 def test_isidentifier(self): 707 self.assertTrue("a".isidentifier()) 708 self.assertTrue("Z".isidentifier()) 709 self.assertTrue("_".isidentifier()) 710 self.assertTrue("b0".isidentifier()) 711 self.assertTrue("bc".isidentifier()) 712 self.assertTrue("b_".isidentifier()) 713 self.assertTrue("µ".isidentifier()) 714 self.assertTrue("".isidentifier()) 715 716 self.assertFalse(" ".isidentifier()) 717 self.assertFalse("[".isidentifier()) 718 self.assertFalse("©".isidentifier()) 719 self.assertFalse("0".isidentifier()) 720 721 def test_isprintable(self): 722 self.assertTrue("".isprintable()) 723 self.assertTrue(" ".isprintable()) 724 self.assertTrue("abcdefg".isprintable()) 725 self.assertFalse("abcdefg\n".isprintable()) 726 # some defined Unicode character 727 self.assertTrue("\u0374".isprintable()) 728 # undefined character 729 self.assertFalse("\u0378".isprintable()) 730 # single surrogate character 731 self.assertFalse("\ud800".isprintable()) 732 733 self.assertTrue('\U0001F46F'.isprintable()) 734 self.assertFalse('\U000E0020'.isprintable()) 735 736 def test_surrogates(self): 737 for s in ('a\uD800b\uDFFF', 'a\uDFFFb\uD800', 738 'a\uD800b\uDFFFa', 'a\uDFFFb\uD800a'): 739 self.assertTrue(s.islower()) 740 self.assertFalse(s.isupper()) 741 self.assertFalse(s.istitle()) 742 for s in ('A\uD800B\uDFFF', 'A\uDFFFB\uD800', 743 'A\uD800B\uDFFFA', 'A\uDFFFB\uD800A'): 744 self.assertFalse(s.islower()) 745 self.assertTrue(s.isupper()) 746 self.assertTrue(s.istitle()) 747 748 for meth_name in ('islower', 'isupper', 'istitle'): 749 meth = getattr(str, meth_name) 750 for s in ('\uD800', '\uDFFF', '\uD800\uD800', '\uDFFF\uDFFF'): 751 self.assertFalse(meth(s), '%a.%s() is False' % (s, meth_name)) 752 753 for meth_name in ('isalpha', 'isalnum', 'isdigit', 'isspace', 754 'isdecimal', 'isnumeric', 755 'isidentifier', 'isprintable'): 756 meth = getattr(str, meth_name) 757 for s in ('\uD800', '\uDFFF', '\uD800\uD800', '\uDFFF\uDFFF', 758 'a\uD800b\uDFFF', 'a\uDFFFb\uD800', 759 'a\uD800b\uDFFFa', 'a\uDFFFb\uD800a'): 760 self.assertFalse(meth(s), '%a.%s() is False' % (s, meth_name)) 761 762 763 def test_lower(self): 764 string_tests.CommonTest.test_lower(self) 765 self.assertEqual('\U00010427'.lower(), '\U0001044F') 766 self.assertEqual('\U00010427\U00010427'.lower(), 767 '\U0001044F\U0001044F') 768 self.assertEqual('\U00010427\U0001044F'.lower(), 769 '\U0001044F\U0001044F') 770 self.assertEqual('X\U00010427x\U0001044F'.lower(), 771 'x\U0001044Fx\U0001044F') 772 self.assertEqual('fi'.lower(), 'fi') 773 self.assertEqual('\u0130'.lower(), '\u0069\u0307') 774 # Special case for GREEK CAPITAL LETTER SIGMA U+03A3 775 self.assertEqual('\u03a3'.lower(), '\u03c3') 776 self.assertEqual('\u0345\u03a3'.lower(), '\u0345\u03c3') 777 self.assertEqual('A\u0345\u03a3'.lower(), 'a\u0345\u03c2') 778 self.assertEqual('A\u0345\u03a3a'.lower(), 'a\u0345\u03c3a') 779 self.assertEqual('A\u0345\u03a3'.lower(), 'a\u0345\u03c2') 780 self.assertEqual('A\u03a3\u0345'.lower(), 'a\u03c2\u0345') 781 self.assertEqual('\u03a3\u0345 '.lower(), '\u03c3\u0345 ') 782 self.assertEqual('\U0008fffe'.lower(), '\U0008fffe') 783 self.assertEqual('\u2177'.lower(), '\u2177') 784 785 def test_casefold(self): 786 self.assertEqual('hello'.casefold(), 'hello') 787 self.assertEqual('hELlo'.casefold(), 'hello') 788 self.assertEqual('ß'.casefold(), 'ss') 789 self.assertEqual('fi'.casefold(), 'fi') 790 self.assertEqual('\u03a3'.casefold(), '\u03c3') 791 self.assertEqual('A\u0345\u03a3'.casefold(), 'a\u03b9\u03c3') 792 self.assertEqual('\u00b5'.casefold(), '\u03bc') 793 794 def test_upper(self): 795 string_tests.CommonTest.test_upper(self) 796 self.assertEqual('\U0001044F'.upper(), '\U00010427') 797 self.assertEqual('\U0001044F\U0001044F'.upper(), 798 '\U00010427\U00010427') 799 self.assertEqual('\U00010427\U0001044F'.upper(), 800 '\U00010427\U00010427') 801 self.assertEqual('X\U00010427x\U0001044F'.upper(), 802 'X\U00010427X\U00010427') 803 self.assertEqual('fi'.upper(), 'FI') 804 self.assertEqual('\u0130'.upper(), '\u0130') 805 self.assertEqual('\u03a3'.upper(), '\u03a3') 806 self.assertEqual('ß'.upper(), 'SS') 807 self.assertEqual('\u1fd2'.upper(), '\u0399\u0308\u0300') 808 self.assertEqual('\U0008fffe'.upper(), '\U0008fffe') 809 self.assertEqual('\u2177'.upper(), '\u2167') 810 811 def test_capitalize(self): 812 string_tests.CommonTest.test_capitalize(self) 813 self.assertEqual('\U0001044F'.capitalize(), '\U00010427') 814 self.assertEqual('\U0001044F\U0001044F'.capitalize(), 815 '\U00010427\U0001044F') 816 self.assertEqual('\U00010427\U0001044F'.capitalize(), 817 '\U00010427\U0001044F') 818 self.assertEqual('\U0001044F\U00010427'.capitalize(), 819 '\U00010427\U0001044F') 820 self.assertEqual('X\U00010427x\U0001044F'.capitalize(), 821 'X\U0001044Fx\U0001044F') 822 self.assertEqual('h\u0130'.capitalize(), 'H\u0069\u0307') 823 exp = '\u0399\u0308\u0300\u0069\u0307' 824 self.assertEqual('\u1fd2\u0130'.capitalize(), exp) 825 self.assertEqual('finnish'.capitalize(), 'FInnish') 826 self.assertEqual('A\u0345\u03a3'.capitalize(), 'A\u0345\u03c2') 827 828 def test_title(self): 829 super().test_title() 830 self.assertEqual('\U0001044F'.title(), '\U00010427') 831 self.assertEqual('\U0001044F\U0001044F'.title(), 832 '\U00010427\U0001044F') 833 self.assertEqual('\U0001044F\U0001044F \U0001044F\U0001044F'.title(), 834 '\U00010427\U0001044F \U00010427\U0001044F') 835 self.assertEqual('\U00010427\U0001044F \U00010427\U0001044F'.title(), 836 '\U00010427\U0001044F \U00010427\U0001044F') 837 self.assertEqual('\U0001044F\U00010427 \U0001044F\U00010427'.title(), 838 '\U00010427\U0001044F \U00010427\U0001044F') 839 self.assertEqual('X\U00010427x\U0001044F X\U00010427x\U0001044F'.title(), 840 'X\U0001044Fx\U0001044F X\U0001044Fx\U0001044F') 841 self.assertEqual('fiNNISH'.title(), 'Finnish') 842 self.assertEqual('A\u03a3 \u1fa1xy'.title(), 'A\u03c2 \u1fa9xy') 843 self.assertEqual('A\u03a3A'.title(), 'A\u03c3a') 844 845 def test_swapcase(self): 846 string_tests.CommonTest.test_swapcase(self) 847 self.assertEqual('\U0001044F'.swapcase(), '\U00010427') 848 self.assertEqual('\U00010427'.swapcase(), '\U0001044F') 849 self.assertEqual('\U0001044F\U0001044F'.swapcase(), 850 '\U00010427\U00010427') 851 self.assertEqual('\U00010427\U0001044F'.swapcase(), 852 '\U0001044F\U00010427') 853 self.assertEqual('\U0001044F\U00010427'.swapcase(), 854 '\U00010427\U0001044F') 855 self.assertEqual('X\U00010427x\U0001044F'.swapcase(), 856 'x\U0001044FX\U00010427') 857 self.assertEqual('fi'.swapcase(), 'FI') 858 self.assertEqual('\u0130'.swapcase(), '\u0069\u0307') 859 # Special case for GREEK CAPITAL LETTER SIGMA U+03A3 860 self.assertEqual('\u03a3'.swapcase(), '\u03c3') 861 self.assertEqual('\u0345\u03a3'.swapcase(), '\u0399\u03c3') 862 self.assertEqual('A\u0345\u03a3'.swapcase(), 'a\u0399\u03c2') 863 self.assertEqual('A\u0345\u03a3a'.swapcase(), 'a\u0399\u03c3A') 864 self.assertEqual('A\u0345\u03a3'.swapcase(), 'a\u0399\u03c2') 865 self.assertEqual('A\u03a3\u0345'.swapcase(), 'a\u03c2\u0399') 866 self.assertEqual('\u03a3\u0345 '.swapcase(), '\u03c3\u0399 ') 867 self.assertEqual('\u03a3'.swapcase(), '\u03c3') 868 self.assertEqual('ß'.swapcase(), 'SS') 869 self.assertEqual('\u1fd2'.swapcase(), '\u0399\u0308\u0300') 870 871 def test_center(self): 872 string_tests.CommonTest.test_center(self) 873 self.assertEqual('x'.center(2, '\U0010FFFF'), 874 'x\U0010FFFF') 875 self.assertEqual('x'.center(3, '\U0010FFFF'), 876 '\U0010FFFFx\U0010FFFF') 877 self.assertEqual('x'.center(4, '\U0010FFFF'), 878 '\U0010FFFFx\U0010FFFF\U0010FFFF') 879 880 @unittest.skipUnless(sys.maxsize == 2**31 - 1, "requires 32-bit system") 881 @support.cpython_only 882 def test_case_operation_overflow(self): 883 # Issue #22643 884 size = 2**32//12 + 1 885 try: 886 s = "ü" * size 887 except MemoryError: 888 self.skipTest('no enough memory (%.0f MiB required)' % (size / 2**20)) 889 try: 890 self.assertRaises(OverflowError, s.upper) 891 finally: 892 del s 893 894 def test_contains(self): 895 # Testing Unicode contains method 896 self.assertIn('a', 'abdb') 897 self.assertIn('a', 'bdab') 898 self.assertIn('a', 'bdaba') 899 self.assertIn('a', 'bdba') 900 self.assertNotIn('a', 'bdb') 901 self.assertIn('a', 'bdba') 902 self.assertIn('a', ('a',1,None)) 903 self.assertIn('a', (1,None,'a')) 904 self.assertIn('a', ('a',1,None)) 905 self.assertIn('a', (1,None,'a')) 906 self.assertNotIn('a', ('x',1,'y')) 907 self.assertNotIn('a', ('x',1,None)) 908 self.assertNotIn('abcd', 'abcxxxx') 909 self.assertIn('ab', 'abcd') 910 self.assertIn('ab', 'abc') 911 self.assertIn('ab', (1,None,'ab')) 912 self.assertIn('', 'abc') 913 self.assertIn('', '') 914 self.assertIn('', 'abc') 915 self.assertNotIn('\0', 'abc') 916 self.assertIn('\0', '\0abc') 917 self.assertIn('\0', 'abc\0') 918 self.assertIn('a', '\0abc') 919 self.assertIn('asdf', 'asdf') 920 self.assertNotIn('asdf', 'asd') 921 self.assertNotIn('asdf', '') 922 923 self.assertRaises(TypeError, "abc".__contains__) 924 # test mixed kinds 925 for fill in ('a', '\u0100', '\U00010300'): 926 fill *= 9 927 for delim in ('c', '\u0102', '\U00010302'): 928 self.assertNotIn(delim, fill) 929 self.assertIn(delim, fill + delim) 930 self.assertNotIn(delim * 2, fill) 931 self.assertIn(delim * 2, fill + delim * 2) 932 933 def test_issue18183(self): 934 '\U00010000\U00100000'.lower() 935 '\U00010000\U00100000'.casefold() 936 '\U00010000\U00100000'.upper() 937 '\U00010000\U00100000'.capitalize() 938 '\U00010000\U00100000'.title() 939 '\U00010000\U00100000'.swapcase() 940 '\U00100000'.center(3, '\U00010000') 941 '\U00100000'.ljust(3, '\U00010000') 942 '\U00100000'.rjust(3, '\U00010000') 943 944 def test_format(self): 945 self.assertEqual(''.format(), '') 946 self.assertEqual('a'.format(), 'a') 947 self.assertEqual('ab'.format(), 'ab') 948 self.assertEqual('a{{'.format(), 'a{') 949 self.assertEqual('a}}'.format(), 'a}') 950 self.assertEqual('{{b'.format(), '{b') 951 self.assertEqual('}}b'.format(), '}b') 952 self.assertEqual('a{{b'.format(), 'a{b') 953 954 # examples from the PEP: 955 import datetime 956 self.assertEqual("My name is {0}".format('Fred'), "My name is Fred") 957 self.assertEqual("My name is {0[name]}".format(dict(name='Fred')), 958 "My name is Fred") 959 self.assertEqual("My name is {0} :-{{}}".format('Fred'), 960 "My name is Fred :-{}") 961 962 d = datetime.date(2007, 8, 18) 963 self.assertEqual("The year is {0.year}".format(d), 964 "The year is 2007") 965 966 # classes we'll use for testing 967 class C: 968 def __init__(self, x=100): 969 self._x = x 970 def __format__(self, spec): 971 return spec 972 973 class D: 974 def __init__(self, x): 975 self.x = x 976 def __format__(self, spec): 977 return str(self.x) 978 979 # class with __str__, but no __format__ 980 class E: 981 def __init__(self, x): 982 self.x = x 983 def __str__(self): 984 return 'E(' + self.x + ')' 985 986 # class with __repr__, but no __format__ or __str__ 987 class F: 988 def __init__(self, x): 989 self.x = x 990 def __repr__(self): 991 return 'F(' + self.x + ')' 992 993 # class with __format__ that forwards to string, for some format_spec's 994 class G: 995 def __init__(self, x): 996 self.x = x 997 def __str__(self): 998 return "string is " + self.x 999 def __format__(self, format_spec): 1000 if format_spec == 'd': 1001 return 'G(' + self.x + ')' 1002 return object.__format__(self, format_spec) 1003 1004 class I(datetime.date): 1005 def __format__(self, format_spec): 1006 return self.strftime(format_spec) 1007 1008 class J(int): 1009 def __format__(self, format_spec): 1010 return int.__format__(self * 2, format_spec) 1011 1012 class M: 1013 def __init__(self, x): 1014 self.x = x 1015 def __repr__(self): 1016 return 'M(' + self.x + ')' 1017 __str__ = None 1018 1019 class N: 1020 def __init__(self, x): 1021 self.x = x 1022 def __repr__(self): 1023 return 'N(' + self.x + ')' 1024 __format__ = None 1025 1026 self.assertEqual(''.format(), '') 1027 self.assertEqual('abc'.format(), 'abc') 1028 self.assertEqual('{0}'.format('abc'), 'abc') 1029 self.assertEqual('{0:}'.format('abc'), 'abc') 1030# self.assertEqual('{ 0 }'.format('abc'), 'abc') 1031 self.assertEqual('X{0}'.format('abc'), 'Xabc') 1032 self.assertEqual('{0}X'.format('abc'), 'abcX') 1033 self.assertEqual('X{0}Y'.format('abc'), 'XabcY') 1034 self.assertEqual('{1}'.format(1, 'abc'), 'abc') 1035 self.assertEqual('X{1}'.format(1, 'abc'), 'Xabc') 1036 self.assertEqual('{1}X'.format(1, 'abc'), 'abcX') 1037 self.assertEqual('X{1}Y'.format(1, 'abc'), 'XabcY') 1038 self.assertEqual('{0}'.format(-15), '-15') 1039 self.assertEqual('{0}{1}'.format(-15, 'abc'), '-15abc') 1040 self.assertEqual('{0}X{1}'.format(-15, 'abc'), '-15Xabc') 1041 self.assertEqual('{{'.format(), '{') 1042 self.assertEqual('}}'.format(), '}') 1043 self.assertEqual('{{}}'.format(), '{}') 1044 self.assertEqual('{{x}}'.format(), '{x}') 1045 self.assertEqual('{{{0}}}'.format(123), '{123}') 1046 self.assertEqual('{{{{0}}}}'.format(), '{{0}}') 1047 self.assertEqual('}}{{'.format(), '}{') 1048 self.assertEqual('}}x{{'.format(), '}x{') 1049 1050 # weird field names 1051 self.assertEqual("{0[foo-bar]}".format({'foo-bar':'baz'}), 'baz') 1052 self.assertEqual("{0[foo bar]}".format({'foo bar':'baz'}), 'baz') 1053 self.assertEqual("{0[ ]}".format({' ':3}), '3') 1054 1055 self.assertEqual('{foo._x}'.format(foo=C(20)), '20') 1056 self.assertEqual('{1}{0}'.format(D(10), D(20)), '2010') 1057 self.assertEqual('{0._x.x}'.format(C(D('abc'))), 'abc') 1058 self.assertEqual('{0[0]}'.format(['abc', 'def']), 'abc') 1059 self.assertEqual('{0[1]}'.format(['abc', 'def']), 'def') 1060 self.assertEqual('{0[1][0]}'.format(['abc', ['def']]), 'def') 1061 self.assertEqual('{0[1][0].x}'.format(['abc', [D('def')]]), 'def') 1062 1063 # strings 1064 self.assertEqual('{0:.3s}'.format('abc'), 'abc') 1065 self.assertEqual('{0:.3s}'.format('ab'), 'ab') 1066 self.assertEqual('{0:.3s}'.format('abcdef'), 'abc') 1067 self.assertEqual('{0:.0s}'.format('abcdef'), '') 1068 self.assertEqual('{0:3.3s}'.format('abc'), 'abc') 1069 self.assertEqual('{0:2.3s}'.format('abc'), 'abc') 1070 self.assertEqual('{0:2.2s}'.format('abc'), 'ab') 1071 self.assertEqual('{0:3.2s}'.format('abc'), 'ab ') 1072 self.assertEqual('{0:x<0s}'.format('result'), 'result') 1073 self.assertEqual('{0:x<5s}'.format('result'), 'result') 1074 self.assertEqual('{0:x<6s}'.format('result'), 'result') 1075 self.assertEqual('{0:x<7s}'.format('result'), 'resultx') 1076 self.assertEqual('{0:x<8s}'.format('result'), 'resultxx') 1077 self.assertEqual('{0: <7s}'.format('result'), 'result ') 1078 self.assertEqual('{0:<7s}'.format('result'), 'result ') 1079 self.assertEqual('{0:>7s}'.format('result'), ' result') 1080 self.assertEqual('{0:>8s}'.format('result'), ' result') 1081 self.assertEqual('{0:^8s}'.format('result'), ' result ') 1082 self.assertEqual('{0:^9s}'.format('result'), ' result ') 1083 self.assertEqual('{0:^10s}'.format('result'), ' result ') 1084 self.assertEqual('{0:10000}'.format('a'), 'a' + ' ' * 9999) 1085 self.assertEqual('{0:10000}'.format(''), ' ' * 10000) 1086 self.assertEqual('{0:10000000}'.format(''), ' ' * 10000000) 1087 1088 # issue 12546: use \x00 as a fill character 1089 self.assertEqual('{0:\x00<6s}'.format('foo'), 'foo\x00\x00\x00') 1090 self.assertEqual('{0:\x01<6s}'.format('foo'), 'foo\x01\x01\x01') 1091 self.assertEqual('{0:\x00^6s}'.format('foo'), '\x00foo\x00\x00') 1092 self.assertEqual('{0:^6s}'.format('foo'), ' foo ') 1093 1094 self.assertEqual('{0:\x00<6}'.format(3), '3\x00\x00\x00\x00\x00') 1095 self.assertEqual('{0:\x01<6}'.format(3), '3\x01\x01\x01\x01\x01') 1096 self.assertEqual('{0:\x00^6}'.format(3), '\x00\x003\x00\x00\x00') 1097 self.assertEqual('{0:<6}'.format(3), '3 ') 1098 1099 self.assertEqual('{0:\x00<6}'.format(3.14), '3.14\x00\x00') 1100 self.assertEqual('{0:\x01<6}'.format(3.14), '3.14\x01\x01') 1101 self.assertEqual('{0:\x00^6}'.format(3.14), '\x003.14\x00') 1102 self.assertEqual('{0:^6}'.format(3.14), ' 3.14 ') 1103 1104 self.assertEqual('{0:\x00<12}'.format(3+2.0j), '(3+2j)\x00\x00\x00\x00\x00\x00') 1105 self.assertEqual('{0:\x01<12}'.format(3+2.0j), '(3+2j)\x01\x01\x01\x01\x01\x01') 1106 self.assertEqual('{0:\x00^12}'.format(3+2.0j), '\x00\x00\x00(3+2j)\x00\x00\x00') 1107 self.assertEqual('{0:^12}'.format(3+2.0j), ' (3+2j) ') 1108 1109 # format specifiers for user defined type 1110 self.assertEqual('{0:abc}'.format(C()), 'abc') 1111 1112 # !r, !s and !a coercions 1113 self.assertEqual('{0!s}'.format('Hello'), 'Hello') 1114 self.assertEqual('{0!s:}'.format('Hello'), 'Hello') 1115 self.assertEqual('{0!s:15}'.format('Hello'), 'Hello ') 1116 self.assertEqual('{0!s:15s}'.format('Hello'), 'Hello ') 1117 self.assertEqual('{0!r}'.format('Hello'), "'Hello'") 1118 self.assertEqual('{0!r:}'.format('Hello'), "'Hello'") 1119 self.assertEqual('{0!r}'.format(F('Hello')), 'F(Hello)') 1120 self.assertEqual('{0!r}'.format('\u0378'), "'\\u0378'") # nonprintable 1121 self.assertEqual('{0!r}'.format('\u0374'), "'\u0374'") # printable 1122 self.assertEqual('{0!r}'.format(F('\u0374')), 'F(\u0374)') 1123 self.assertEqual('{0!a}'.format('Hello'), "'Hello'") 1124 self.assertEqual('{0!a}'.format('\u0378'), "'\\u0378'") # nonprintable 1125 self.assertEqual('{0!a}'.format('\u0374'), "'\\u0374'") # printable 1126 self.assertEqual('{0!a:}'.format('Hello'), "'Hello'") 1127 self.assertEqual('{0!a}'.format(F('Hello')), 'F(Hello)') 1128 self.assertEqual('{0!a}'.format(F('\u0374')), 'F(\\u0374)') 1129 1130 # test fallback to object.__format__ 1131 self.assertEqual('{0}'.format({}), '{}') 1132 self.assertEqual('{0}'.format([]), '[]') 1133 self.assertEqual('{0}'.format([1]), '[1]') 1134 1135 self.assertEqual('{0:d}'.format(G('data')), 'G(data)') 1136 self.assertEqual('{0!s}'.format(G('data')), 'string is data') 1137 1138 self.assertRaises(TypeError, '{0:^10}'.format, E('data')) 1139 self.assertRaises(TypeError, '{0:^10s}'.format, E('data')) 1140 self.assertRaises(TypeError, '{0:>15s}'.format, G('data')) 1141 1142 self.assertEqual("{0:date: %Y-%m-%d}".format(I(year=2007, 1143 month=8, 1144 day=27)), 1145 "date: 2007-08-27") 1146 1147 # test deriving from a builtin type and overriding __format__ 1148 self.assertEqual("{0}".format(J(10)), "20") 1149 1150 1151 # string format specifiers 1152 self.assertEqual('{0:}'.format('a'), 'a') 1153 1154 # computed format specifiers 1155 self.assertEqual("{0:.{1}}".format('hello world', 5), 'hello') 1156 self.assertEqual("{0:.{1}s}".format('hello world', 5), 'hello') 1157 self.assertEqual("{0:.{precision}s}".format('hello world', precision=5), 'hello') 1158 self.assertEqual("{0:{width}.{precision}s}".format('hello world', width=10, precision=5), 'hello ') 1159 self.assertEqual("{0:{width}.{precision}s}".format('hello world', width='10', precision='5'), 'hello ') 1160 1161 # test various errors 1162 self.assertRaises(ValueError, '{'.format) 1163 self.assertRaises(ValueError, '}'.format) 1164 self.assertRaises(ValueError, 'a{'.format) 1165 self.assertRaises(ValueError, 'a}'.format) 1166 self.assertRaises(ValueError, '{a'.format) 1167 self.assertRaises(ValueError, '}a'.format) 1168 self.assertRaises(IndexError, '{0}'.format) 1169 self.assertRaises(IndexError, '{1}'.format, 'abc') 1170 self.assertRaises(KeyError, '{x}'.format) 1171 self.assertRaises(ValueError, "}{".format) 1172 self.assertRaises(ValueError, "abc{0:{}".format) 1173 self.assertRaises(ValueError, "{0".format) 1174 self.assertRaises(IndexError, "{0.}".format) 1175 self.assertRaises(ValueError, "{0.}".format, 0) 1176 self.assertRaises(ValueError, "{0[}".format) 1177 self.assertRaises(ValueError, "{0[}".format, []) 1178 self.assertRaises(KeyError, "{0]}".format) 1179 self.assertRaises(ValueError, "{0.[]}".format, 0) 1180 self.assertRaises(ValueError, "{0..foo}".format, 0) 1181 self.assertRaises(ValueError, "{0[0}".format, 0) 1182 self.assertRaises(ValueError, "{0[0:foo}".format, 0) 1183 self.assertRaises(KeyError, "{c]}".format) 1184 self.assertRaises(ValueError, "{{ {{{0}}".format, 0) 1185 self.assertRaises(ValueError, "{0}}".format, 0) 1186 self.assertRaises(KeyError, "{foo}".format, bar=3) 1187 self.assertRaises(ValueError, "{0!x}".format, 3) 1188 self.assertRaises(ValueError, "{0!}".format, 0) 1189 self.assertRaises(ValueError, "{0!rs}".format, 0) 1190 self.assertRaises(ValueError, "{!}".format) 1191 self.assertRaises(IndexError, "{:}".format) 1192 self.assertRaises(IndexError, "{:s}".format) 1193 self.assertRaises(IndexError, "{}".format) 1194 big = "23098475029384702983476098230754973209482573" 1195 self.assertRaises(ValueError, ("{" + big + "}").format) 1196 self.assertRaises(ValueError, ("{[" + big + "]}").format, [0]) 1197 1198 # issue 6089 1199 self.assertRaises(ValueError, "{0[0]x}".format, [None]) 1200 self.assertRaises(ValueError, "{0[0](10)}".format, [None]) 1201 1202 # can't have a replacement on the field name portion 1203 self.assertRaises(TypeError, '{0[{1}]}'.format, 'abcdefg', 4) 1204 1205 # exceed maximum recursion depth 1206 self.assertRaises(ValueError, "{0:{1:{2}}}".format, 'abc', 's', '') 1207 self.assertRaises(ValueError, "{0:{1:{2:{3:{4:{5:{6}}}}}}}".format, 1208 0, 1, 2, 3, 4, 5, 6, 7) 1209 1210 # string format spec errors 1211 self.assertRaises(ValueError, "{0:-s}".format, '') 1212 self.assertRaises(ValueError, format, "", "-") 1213 self.assertRaises(ValueError, "{0:=s}".format, '') 1214 1215 # Alternate formatting is not supported 1216 self.assertRaises(ValueError, format, '', '#') 1217 self.assertRaises(ValueError, format, '', '#20') 1218 1219 # Non-ASCII 1220 self.assertEqual("{0:s}{1:s}".format("ABC", "\u0410\u0411\u0412"), 1221 'ABC\u0410\u0411\u0412') 1222 self.assertEqual("{0:.3s}".format("ABC\u0410\u0411\u0412"), 1223 'ABC') 1224 self.assertEqual("{0:.0s}".format("ABC\u0410\u0411\u0412"), 1225 '') 1226 1227 self.assertEqual("{[{}]}".format({"{}": 5}), "5") 1228 self.assertEqual("{[{}]}".format({"{}" : "a"}), "a") 1229 self.assertEqual("{[{]}".format({"{" : "a"}), "a") 1230 self.assertEqual("{[}]}".format({"}" : "a"}), "a") 1231 self.assertEqual("{[[]}".format({"[" : "a"}), "a") 1232 self.assertEqual("{[!]}".format({"!" : "a"}), "a") 1233 self.assertRaises(ValueError, "{a{}b}".format, 42) 1234 self.assertRaises(ValueError, "{a{b}".format, 42) 1235 self.assertRaises(ValueError, "{[}".format, 42) 1236 1237 self.assertEqual("0x{:0{:d}X}".format(0x0,16), "0x0000000000000000") 1238 1239 # Blocking fallback 1240 m = M('data') 1241 self.assertEqual("{!r}".format(m), 'M(data)') 1242 self.assertRaises(TypeError, "{!s}".format, m) 1243 self.assertRaises(TypeError, "{}".format, m) 1244 n = N('data') 1245 self.assertEqual("{!r}".format(n), 'N(data)') 1246 self.assertEqual("{!s}".format(n), 'N(data)') 1247 self.assertRaises(TypeError, "{}".format, n) 1248 1249 def test_format_map(self): 1250 self.assertEqual(''.format_map({}), '') 1251 self.assertEqual('a'.format_map({}), 'a') 1252 self.assertEqual('ab'.format_map({}), 'ab') 1253 self.assertEqual('a{{'.format_map({}), 'a{') 1254 self.assertEqual('a}}'.format_map({}), 'a}') 1255 self.assertEqual('{{b'.format_map({}), '{b') 1256 self.assertEqual('}}b'.format_map({}), '}b') 1257 self.assertEqual('a{{b'.format_map({}), 'a{b') 1258 1259 # using mappings 1260 class Mapping(dict): 1261 def __missing__(self, key): 1262 return key 1263 self.assertEqual('{hello}'.format_map(Mapping()), 'hello') 1264 self.assertEqual('{a} {world}'.format_map(Mapping(a='hello')), 'hello world') 1265 1266 class InternalMapping: 1267 def __init__(self): 1268 self.mapping = {'a': 'hello'} 1269 def __getitem__(self, key): 1270 return self.mapping[key] 1271 self.assertEqual('{a}'.format_map(InternalMapping()), 'hello') 1272 1273 1274 class C: 1275 def __init__(self, x=100): 1276 self._x = x 1277 def __format__(self, spec): 1278 return spec 1279 self.assertEqual('{foo._x}'.format_map({'foo': C(20)}), '20') 1280 1281 # test various errors 1282 self.assertRaises(TypeError, ''.format_map) 1283 self.assertRaises(TypeError, 'a'.format_map) 1284 1285 self.assertRaises(ValueError, '{'.format_map, {}) 1286 self.assertRaises(ValueError, '}'.format_map, {}) 1287 self.assertRaises(ValueError, 'a{'.format_map, {}) 1288 self.assertRaises(ValueError, 'a}'.format_map, {}) 1289 self.assertRaises(ValueError, '{a'.format_map, {}) 1290 self.assertRaises(ValueError, '}a'.format_map, {}) 1291 1292 # issue #12579: can't supply positional params to format_map 1293 self.assertRaises(ValueError, '{}'.format_map, {'a' : 2}) 1294 self.assertRaises(ValueError, '{}'.format_map, 'a') 1295 self.assertRaises(ValueError, '{a} {}'.format_map, {"a" : 2, "b" : 1}) 1296 1297 class BadMapping: 1298 def __getitem__(self, key): 1299 return 1/0 1300 self.assertRaises(KeyError, '{a}'.format_map, {}) 1301 self.assertRaises(TypeError, '{a}'.format_map, []) 1302 self.assertRaises(ZeroDivisionError, '{a}'.format_map, BadMapping()) 1303 1304 def test_format_huge_precision(self): 1305 format_string = ".{}f".format(sys.maxsize + 1) 1306 with self.assertRaises(ValueError): 1307 result = format(2.34, format_string) 1308 1309 def test_format_huge_width(self): 1310 format_string = "{}f".format(sys.maxsize + 1) 1311 with self.assertRaises(ValueError): 1312 result = format(2.34, format_string) 1313 1314 def test_format_huge_item_number(self): 1315 format_string = "{{{}:.6f}}".format(sys.maxsize + 1) 1316 with self.assertRaises(ValueError): 1317 result = format_string.format(2.34) 1318 1319 def test_format_auto_numbering(self): 1320 class C: 1321 def __init__(self, x=100): 1322 self._x = x 1323 def __format__(self, spec): 1324 return spec 1325 1326 self.assertEqual('{}'.format(10), '10') 1327 self.assertEqual('{:5}'.format('s'), 's ') 1328 self.assertEqual('{!r}'.format('s'), "'s'") 1329 self.assertEqual('{._x}'.format(C(10)), '10') 1330 self.assertEqual('{[1]}'.format([1, 2]), '2') 1331 self.assertEqual('{[a]}'.format({'a':4, 'b':2}), '4') 1332 self.assertEqual('a{}b{}c'.format(0, 1), 'a0b1c') 1333 1334 self.assertEqual('a{:{}}b'.format('x', '^10'), 'a x b') 1335 self.assertEqual('a{:{}x}b'.format(20, '#'), 'a0x14b') 1336 1337 # can't mix and match numbering and auto-numbering 1338 self.assertRaises(ValueError, '{}{1}'.format, 1, 2) 1339 self.assertRaises(ValueError, '{1}{}'.format, 1, 2) 1340 self.assertRaises(ValueError, '{:{1}}'.format, 1, 2) 1341 self.assertRaises(ValueError, '{0:{}}'.format, 1, 2) 1342 1343 # can mix and match auto-numbering and named 1344 self.assertEqual('{f}{}'.format(4, f='test'), 'test4') 1345 self.assertEqual('{}{f}'.format(4, f='test'), '4test') 1346 self.assertEqual('{:{f}}{g}{}'.format(1, 3, g='g', f=2), ' 1g3') 1347 self.assertEqual('{f:{}}{}{g}'.format(2, 4, f=1, g='g'), ' 14g') 1348 1349 def test_formatting(self): 1350 string_tests.MixinStrUnicodeUserStringTest.test_formatting(self) 1351 # Testing Unicode formatting strings... 1352 self.assertEqual("%s, %s" % ("abc", "abc"), 'abc, abc') 1353 self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", 1, 2, 3), 'abc, abc, 1, 2.000000, 3.00') 1354 self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", 1, -2, 3), 'abc, abc, 1, -2.000000, 3.00') 1355 self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", -1, -2, 3.5), 'abc, abc, -1, -2.000000, 3.50') 1356 self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", -1, -2, 3.57), 'abc, abc, -1, -2.000000, 3.57') 1357 self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", -1, -2, 1003.57), 'abc, abc, -1, -2.000000, 1003.57') 1358 if not sys.platform.startswith('java'): 1359 self.assertEqual("%r, %r" % (b"abc", "abc"), "b'abc', 'abc'") 1360 self.assertEqual("%r" % ("\u1234",), "'\u1234'") 1361 self.assertEqual("%a" % ("\u1234",), "'\\u1234'") 1362 self.assertEqual("%(x)s, %(y)s" % {'x':"abc", 'y':"def"}, 'abc, def') 1363 self.assertEqual("%(x)s, %(\xfc)s" % {'x':"abc", '\xfc':"def"}, 'abc, def') 1364 1365 self.assertEqual('%c' % 0x1234, '\u1234') 1366 self.assertEqual('%c' % 0x21483, '\U00021483') 1367 self.assertRaises(OverflowError, "%c".__mod__, (0x110000,)) 1368 self.assertEqual('%c' % '\U00021483', '\U00021483') 1369 self.assertRaises(TypeError, "%c".__mod__, "aa") 1370 self.assertRaises(ValueError, "%.1\u1032f".__mod__, (1.0/3)) 1371 self.assertRaises(TypeError, "%i".__mod__, "aa") 1372 1373 # formatting jobs delegated from the string implementation: 1374 self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...') 1375 self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...') 1376 self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...') 1377 self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...') 1378 self.assertEqual('...%(foo)s...' % {'foo':"abc",'def':123}, '...abc...') 1379 self.assertEqual('...%(foo)s...' % {'foo':"abc",'def':123}, '...abc...') 1380 self.assertEqual('...%s...%s...%s...%s...' % (1,2,3,"abc"), '...1...2...3...abc...') 1381 self.assertEqual('...%%...%%s...%s...%s...%s...%s...' % (1,2,3,"abc"), '...%...%s...1...2...3...abc...') 1382 self.assertEqual('...%s...' % "abc", '...abc...') 1383 self.assertEqual('%*s' % (5,'abc',), ' abc') 1384 self.assertEqual('%*s' % (-5,'abc',), 'abc ') 1385 self.assertEqual('%*.*s' % (5,2,'abc',), ' ab') 1386 self.assertEqual('%*.*s' % (5,3,'abc',), ' abc') 1387 self.assertEqual('%i %*.*s' % (10, 5,3,'abc',), '10 abc') 1388 self.assertEqual('%i%s %*.*s' % (10, 3, 5, 3, 'abc',), '103 abc') 1389 self.assertEqual('%c' % 'a', 'a') 1390 class Wrapper: 1391 def __str__(self): 1392 return '\u1234' 1393 self.assertEqual('%s' % Wrapper(), '\u1234') 1394 1395 # issue 3382 1396 NAN = float('nan') 1397 INF = float('inf') 1398 self.assertEqual('%f' % NAN, 'nan') 1399 self.assertEqual('%F' % NAN, 'NAN') 1400 self.assertEqual('%f' % INF, 'inf') 1401 self.assertEqual('%F' % INF, 'INF') 1402 1403 # PEP 393 1404 self.assertEqual('%.1s' % "a\xe9\u20ac", 'a') 1405 self.assertEqual('%.2s' % "a\xe9\u20ac", 'a\xe9') 1406 1407 #issue 19995 1408 class PseudoInt: 1409 def __init__(self, value): 1410 self.value = int(value) 1411 def __int__(self): 1412 return self.value 1413 def __index__(self): 1414 return self.value 1415 class PseudoFloat: 1416 def __init__(self, value): 1417 self.value = float(value) 1418 def __int__(self): 1419 return int(self.value) 1420 pi = PseudoFloat(3.1415) 1421 letter_m = PseudoInt(109) 1422 self.assertEqual('%x' % 42, '2a') 1423 self.assertEqual('%X' % 15, 'F') 1424 self.assertEqual('%o' % 9, '11') 1425 self.assertEqual('%c' % 109, 'm') 1426 self.assertEqual('%x' % letter_m, '6d') 1427 self.assertEqual('%X' % letter_m, '6D') 1428 self.assertEqual('%o' % letter_m, '155') 1429 self.assertEqual('%c' % letter_m, 'm') 1430 self.assertRaisesRegex(TypeError, '%x format: an integer is required, not float', operator.mod, '%x', 3.14), 1431 self.assertRaisesRegex(TypeError, '%X format: an integer is required, not float', operator.mod, '%X', 2.11), 1432 self.assertRaisesRegex(TypeError, '%o format: an integer is required, not float', operator.mod, '%o', 1.79), 1433 self.assertRaisesRegex(TypeError, '%x format: an integer is required, not PseudoFloat', operator.mod, '%x', pi), 1434 self.assertRaises(TypeError, operator.mod, '%c', pi), 1435 1436 def test_formatting_with_enum(self): 1437 # issue18780 1438 import enum 1439 class Float(float, enum.Enum): 1440 PI = 3.1415926 1441 class Int(enum.IntEnum): 1442 IDES = 15 1443 class Str(str, enum.Enum): 1444 ABC = 'abc' 1445 # Testing Unicode formatting strings... 1446 self.assertEqual("%s, %s" % (Str.ABC, Str.ABC), 1447 'Str.ABC, Str.ABC') 1448 self.assertEqual("%s, %s, %d, %i, %u, %f, %5.2f" % 1449 (Str.ABC, Str.ABC, 1450 Int.IDES, Int.IDES, Int.IDES, 1451 Float.PI, Float.PI), 1452 'Str.ABC, Str.ABC, 15, 15, 15, 3.141593, 3.14') 1453 1454 # formatting jobs delegated from the string implementation: 1455 self.assertEqual('...%(foo)s...' % {'foo':Str.ABC}, 1456 '...Str.ABC...') 1457 self.assertEqual('...%(foo)s...' % {'foo':Int.IDES}, 1458 '...Int.IDES...') 1459 self.assertEqual('...%(foo)i...' % {'foo':Int.IDES}, 1460 '...15...') 1461 self.assertEqual('...%(foo)d...' % {'foo':Int.IDES}, 1462 '...15...') 1463 self.assertEqual('...%(foo)u...' % {'foo':Int.IDES, 'def':Float.PI}, 1464 '...15...') 1465 self.assertEqual('...%(foo)f...' % {'foo':Float.PI,'def':123}, 1466 '...3.141593...') 1467 1468 def test_formatting_huge_precision(self): 1469 format_string = "%.{}f".format(sys.maxsize + 1) 1470 with self.assertRaises(ValueError): 1471 result = format_string % 2.34 1472 1473 def test_issue28598_strsubclass_rhs(self): 1474 # A subclass of str with an __rmod__ method should be able to hook 1475 # into the % operator 1476 class SubclassedStr(str): 1477 def __rmod__(self, other): 1478 return 'Success, self.__rmod__({!r}) was called'.format(other) 1479 self.assertEqual('lhs %% %r' % SubclassedStr('rhs'), 1480 "Success, self.__rmod__('lhs %% %r') was called") 1481 1482 @support.cpython_only 1483 def test_formatting_huge_precision_c_limits(self): 1484 from _testcapi import INT_MAX 1485 format_string = "%.{}f".format(INT_MAX + 1) 1486 with self.assertRaises(ValueError): 1487 result = format_string % 2.34 1488 1489 def test_formatting_huge_width(self): 1490 format_string = "%{}f".format(sys.maxsize + 1) 1491 with self.assertRaises(ValueError): 1492 result = format_string % 2.34 1493 1494 def test_startswith_endswith_errors(self): 1495 for meth in ('foo'.startswith, 'foo'.endswith): 1496 with self.assertRaises(TypeError) as cm: 1497 meth(['f']) 1498 exc = str(cm.exception) 1499 self.assertIn('str', exc) 1500 self.assertIn('tuple', exc) 1501 1502 @support.run_with_locale('LC_ALL', 'de_DE', 'fr_FR') 1503 def test_format_float(self): 1504 # should not format with a comma, but always with C locale 1505 self.assertEqual('1.0', '%.1f' % 1.0) 1506 1507 def test_constructor(self): 1508 # unicode(obj) tests (this maps to PyObject_Unicode() at C level) 1509 1510 self.assertEqual( 1511 str('unicode remains unicode'), 1512 'unicode remains unicode' 1513 ) 1514 1515 for text in ('ascii', '\xe9', '\u20ac', '\U0010FFFF'): 1516 subclass = StrSubclass(text) 1517 self.assertEqual(str(subclass), text) 1518 self.assertEqual(len(subclass), len(text)) 1519 if text == 'ascii': 1520 self.assertEqual(subclass.encode('ascii'), b'ascii') 1521 self.assertEqual(subclass.encode('utf-8'), b'ascii') 1522 1523 self.assertEqual( 1524 str('strings are converted to unicode'), 1525 'strings are converted to unicode' 1526 ) 1527 1528 class StringCompat: 1529 def __init__(self, x): 1530 self.x = x 1531 def __str__(self): 1532 return self.x 1533 1534 self.assertEqual( 1535 str(StringCompat('__str__ compatible objects are recognized')), 1536 '__str__ compatible objects are recognized' 1537 ) 1538 1539 # unicode(obj) is compatible to str(): 1540 1541 o = StringCompat('unicode(obj) is compatible to str()') 1542 self.assertEqual(str(o), 'unicode(obj) is compatible to str()') 1543 self.assertEqual(str(o), 'unicode(obj) is compatible to str()') 1544 1545 for obj in (123, 123.45, 123): 1546 self.assertEqual(str(obj), str(str(obj))) 1547 1548 # unicode(obj, encoding, error) tests (this maps to 1549 # PyUnicode_FromEncodedObject() at C level) 1550 1551 if not sys.platform.startswith('java'): 1552 self.assertRaises( 1553 TypeError, 1554 str, 1555 'decoding unicode is not supported', 1556 'utf-8', 1557 'strict' 1558 ) 1559 1560 self.assertEqual( 1561 str(b'strings are decoded to unicode', 'utf-8', 'strict'), 1562 'strings are decoded to unicode' 1563 ) 1564 1565 if not sys.platform.startswith('java'): 1566 self.assertEqual( 1567 str( 1568 memoryview(b'character buffers are decoded to unicode'), 1569 'utf-8', 1570 'strict' 1571 ), 1572 'character buffers are decoded to unicode' 1573 ) 1574 1575 self.assertRaises(TypeError, str, 42, 42, 42) 1576 1577 def test_constructor_keyword_args(self): 1578 """Pass various keyword argument combinations to the constructor.""" 1579 # The object argument can be passed as a keyword. 1580 self.assertEqual(str(object='foo'), 'foo') 1581 self.assertEqual(str(object=b'foo', encoding='utf-8'), 'foo') 1582 # The errors argument without encoding triggers "decode" mode. 1583 self.assertEqual(str(b'foo', errors='strict'), 'foo') # not "b'foo'" 1584 self.assertEqual(str(object=b'foo', errors='strict'), 'foo') 1585 1586 def test_constructor_defaults(self): 1587 """Check the constructor argument defaults.""" 1588 # The object argument defaults to '' or b''. 1589 self.assertEqual(str(), '') 1590 self.assertEqual(str(errors='strict'), '') 1591 utf8_cent = '¢'.encode('utf-8') 1592 # The encoding argument defaults to utf-8. 1593 self.assertEqual(str(utf8_cent, errors='strict'), '¢') 1594 # The errors argument defaults to strict. 1595 self.assertRaises(UnicodeDecodeError, str, utf8_cent, encoding='ascii') 1596 1597 def test_codecs_utf7(self): 1598 utfTests = [ 1599 ('A\u2262\u0391.', b'A+ImIDkQ.'), # RFC2152 example 1600 ('Hi Mom -\u263a-!', b'Hi Mom -+Jjo--!'), # RFC2152 example 1601 ('\u65E5\u672C\u8A9E', b'+ZeVnLIqe-'), # RFC2152 example 1602 ('Item 3 is \u00a31.', b'Item 3 is +AKM-1.'), # RFC2152 example 1603 ('+', b'+-'), 1604 ('+-', b'+--'), 1605 ('+?', b'+-?'), 1606 (r'\?', b'+AFw?'), 1607 ('+?', b'+-?'), 1608 (r'\\?', b'+AFwAXA?'), 1609 (r'\\\?', b'+AFwAXABc?'), 1610 (r'++--', b'+-+---'), 1611 ('\U000abcde', b'+2m/c3g-'), # surrogate pairs 1612 ('/', b'/'), 1613 ] 1614 1615 for (x, y) in utfTests: 1616 self.assertEqual(x.encode('utf-7'), y) 1617 1618 # Unpaired surrogates are passed through 1619 self.assertEqual('\uD801'.encode('utf-7'), b'+2AE-') 1620 self.assertEqual('\uD801x'.encode('utf-7'), b'+2AE-x') 1621 self.assertEqual('\uDC01'.encode('utf-7'), b'+3AE-') 1622 self.assertEqual('\uDC01x'.encode('utf-7'), b'+3AE-x') 1623 self.assertEqual(b'+2AE-'.decode('utf-7'), '\uD801') 1624 self.assertEqual(b'+2AE-x'.decode('utf-7'), '\uD801x') 1625 self.assertEqual(b'+3AE-'.decode('utf-7'), '\uDC01') 1626 self.assertEqual(b'+3AE-x'.decode('utf-7'), '\uDC01x') 1627 1628 self.assertEqual('\uD801\U000abcde'.encode('utf-7'), b'+2AHab9ze-') 1629 self.assertEqual(b'+2AHab9ze-'.decode('utf-7'), '\uD801\U000abcde') 1630 1631 # Issue #2242: crash on some Windows/MSVC versions 1632 self.assertEqual(b'+\xc1'.decode('utf-7', 'ignore'), '') 1633 1634 # Direct encoded characters 1635 set_d = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'(),-./:?" 1636 # Optional direct characters 1637 set_o = '!"#$%&*;<=>@[]^_`{|}' 1638 for c in set_d: 1639 self.assertEqual(c.encode('utf7'), c.encode('ascii')) 1640 self.assertEqual(c.encode('ascii').decode('utf7'), c) 1641 for c in set_o: 1642 self.assertEqual(c.encode('ascii').decode('utf7'), c) 1643 1644 def test_codecs_utf8(self): 1645 self.assertEqual(''.encode('utf-8'), b'') 1646 self.assertEqual('\u20ac'.encode('utf-8'), b'\xe2\x82\xac') 1647 self.assertEqual('\U00010002'.encode('utf-8'), b'\xf0\x90\x80\x82') 1648 self.assertEqual('\U00023456'.encode('utf-8'), b'\xf0\xa3\x91\x96') 1649 self.assertEqual('\ud800'.encode('utf-8', 'surrogatepass'), b'\xed\xa0\x80') 1650 self.assertEqual('\udc00'.encode('utf-8', 'surrogatepass'), b'\xed\xb0\x80') 1651 self.assertEqual(('\U00010002'*10).encode('utf-8'), 1652 b'\xf0\x90\x80\x82'*10) 1653 self.assertEqual( 1654 '\u6b63\u78ba\u306b\u8a00\u3046\u3068\u7ffb\u8a33\u306f' 1655 '\u3055\u308c\u3066\u3044\u307e\u305b\u3093\u3002\u4e00' 1656 '\u90e8\u306f\u30c9\u30a4\u30c4\u8a9e\u3067\u3059\u304c' 1657 '\u3001\u3042\u3068\u306f\u3067\u305f\u3089\u3081\u3067' 1658 '\u3059\u3002\u5b9f\u969b\u306b\u306f\u300cWenn ist das' 1659 ' Nunstuck git und'.encode('utf-8'), 1660 b'\xe6\xad\xa3\xe7\xa2\xba\xe3\x81\xab\xe8\xa8\x80\xe3\x81' 1661 b'\x86\xe3\x81\xa8\xe7\xbf\xbb\xe8\xa8\xb3\xe3\x81\xaf\xe3' 1662 b'\x81\x95\xe3\x82\x8c\xe3\x81\xa6\xe3\x81\x84\xe3\x81\xbe' 1663 b'\xe3\x81\x9b\xe3\x82\x93\xe3\x80\x82\xe4\xb8\x80\xe9\x83' 1664 b'\xa8\xe3\x81\xaf\xe3\x83\x89\xe3\x82\xa4\xe3\x83\x84\xe8' 1665 b'\xaa\x9e\xe3\x81\xa7\xe3\x81\x99\xe3\x81\x8c\xe3\x80\x81' 1666 b'\xe3\x81\x82\xe3\x81\xa8\xe3\x81\xaf\xe3\x81\xa7\xe3\x81' 1667 b'\x9f\xe3\x82\x89\xe3\x82\x81\xe3\x81\xa7\xe3\x81\x99\xe3' 1668 b'\x80\x82\xe5\xae\x9f\xe9\x9a\x9b\xe3\x81\xab\xe3\x81\xaf' 1669 b'\xe3\x80\x8cWenn ist das Nunstuck git und' 1670 ) 1671 1672 # UTF-8 specific decoding tests 1673 self.assertEqual(str(b'\xf0\xa3\x91\x96', 'utf-8'), '\U00023456' ) 1674 self.assertEqual(str(b'\xf0\x90\x80\x82', 'utf-8'), '\U00010002' ) 1675 self.assertEqual(str(b'\xe2\x82\xac', 'utf-8'), '\u20ac' ) 1676 1677 # Other possible utf-8 test cases: 1678 # * strict decoding testing for all of the 1679 # UTF8_ERROR cases in PyUnicode_DecodeUTF8 1680 1681 def test_utf8_decode_valid_sequences(self): 1682 sequences = [ 1683 # single byte 1684 (b'\x00', '\x00'), (b'a', 'a'), (b'\x7f', '\x7f'), 1685 # 2 bytes 1686 (b'\xc2\x80', '\x80'), (b'\xdf\xbf', '\u07ff'), 1687 # 3 bytes 1688 (b'\xe0\xa0\x80', '\u0800'), (b'\xed\x9f\xbf', '\ud7ff'), 1689 (b'\xee\x80\x80', '\uE000'), (b'\xef\xbf\xbf', '\uffff'), 1690 # 4 bytes 1691 (b'\xF0\x90\x80\x80', '\U00010000'), 1692 (b'\xf4\x8f\xbf\xbf', '\U0010FFFF') 1693 ] 1694 for seq, res in sequences: 1695 self.assertEqual(seq.decode('utf-8'), res) 1696 1697 1698 def test_utf8_decode_invalid_sequences(self): 1699 # continuation bytes in a sequence of 2, 3, or 4 bytes 1700 continuation_bytes = [bytes([x]) for x in range(0x80, 0xC0)] 1701 # start bytes of a 2-byte sequence equivalent to code points < 0x7F 1702 invalid_2B_seq_start_bytes = [bytes([x]) for x in range(0xC0, 0xC2)] 1703 # start bytes of a 4-byte sequence equivalent to code points > 0x10FFFF 1704 invalid_4B_seq_start_bytes = [bytes([x]) for x in range(0xF5, 0xF8)] 1705 invalid_start_bytes = ( 1706 continuation_bytes + invalid_2B_seq_start_bytes + 1707 invalid_4B_seq_start_bytes + [bytes([x]) for x in range(0xF7, 0x100)] 1708 ) 1709 1710 for byte in invalid_start_bytes: 1711 self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8') 1712 1713 for sb in invalid_2B_seq_start_bytes: 1714 for cb in continuation_bytes: 1715 self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8') 1716 1717 for sb in invalid_4B_seq_start_bytes: 1718 for cb1 in continuation_bytes[:3]: 1719 for cb3 in continuation_bytes[:3]: 1720 self.assertRaises(UnicodeDecodeError, 1721 (sb+cb1+b'\x80'+cb3).decode, 'utf-8') 1722 1723 for cb in [bytes([x]) for x in range(0x80, 0xA0)]: 1724 self.assertRaises(UnicodeDecodeError, 1725 (b'\xE0'+cb+b'\x80').decode, 'utf-8') 1726 self.assertRaises(UnicodeDecodeError, 1727 (b'\xE0'+cb+b'\xBF').decode, 'utf-8') 1728 # surrogates 1729 for cb in [bytes([x]) for x in range(0xA0, 0xC0)]: 1730 self.assertRaises(UnicodeDecodeError, 1731 (b'\xED'+cb+b'\x80').decode, 'utf-8') 1732 self.assertRaises(UnicodeDecodeError, 1733 (b'\xED'+cb+b'\xBF').decode, 'utf-8') 1734 for cb in [bytes([x]) for x in range(0x80, 0x90)]: 1735 self.assertRaises(UnicodeDecodeError, 1736 (b'\xF0'+cb+b'\x80\x80').decode, 'utf-8') 1737 self.assertRaises(UnicodeDecodeError, 1738 (b'\xF0'+cb+b'\xBF\xBF').decode, 'utf-8') 1739 for cb in [bytes([x]) for x in range(0x90, 0xC0)]: 1740 self.assertRaises(UnicodeDecodeError, 1741 (b'\xF4'+cb+b'\x80\x80').decode, 'utf-8') 1742 self.assertRaises(UnicodeDecodeError, 1743 (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8') 1744 1745 def test_issue8271(self): 1746 # Issue #8271: during the decoding of an invalid UTF-8 byte sequence, 1747 # only the start byte and the continuation byte(s) are now considered 1748 # invalid, instead of the number of bytes specified by the start byte. 1749 # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95, 1750 # table 3-8, Row 2) for more information about the algorithm used. 1751 FFFD = '\ufffd' 1752 sequences = [ 1753 # invalid start bytes 1754 (b'\x80', FFFD), # continuation byte 1755 (b'\x80\x80', FFFD*2), # 2 continuation bytes 1756 (b'\xc0', FFFD), 1757 (b'\xc0\xc0', FFFD*2), 1758 (b'\xc1', FFFD), 1759 (b'\xc1\xc0', FFFD*2), 1760 (b'\xc0\xc1', FFFD*2), 1761 # with start byte of a 2-byte sequence 1762 (b'\xc2', FFFD), # only the start byte 1763 (b'\xc2\xc2', FFFD*2), # 2 start bytes 1764 (b'\xc2\xc2\xc2', FFFD*3), # 3 start bytes 1765 (b'\xc2\x41', FFFD+'A'), # invalid continuation byte 1766 # with start byte of a 3-byte sequence 1767 (b'\xe1', FFFD), # only the start byte 1768 (b'\xe1\xe1', FFFD*2), # 2 start bytes 1769 (b'\xe1\xe1\xe1', FFFD*3), # 3 start bytes 1770 (b'\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes 1771 (b'\xe1\x80', FFFD), # only 1 continuation byte 1772 (b'\xe1\x41', FFFD+'A'), # invalid continuation byte 1773 (b'\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb 1774 (b'\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes 1775 (b'\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte 1776 (b'\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid 1777 (b'\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid 1778 # with start byte of a 4-byte sequence 1779 (b'\xf1', FFFD), # only the start byte 1780 (b'\xf1\xf1', FFFD*2), # 2 start bytes 1781 (b'\xf1\xf1\xf1', FFFD*3), # 3 start bytes 1782 (b'\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes 1783 (b'\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes 1784 (b'\xf1\x80', FFFD), # only 1 continuation bytes 1785 (b'\xf1\x80\x80', FFFD), # only 2 continuation bytes 1786 (b'\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid 1787 (b'\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid 1788 (b'\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid 1789 (b'\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid 1790 (b'\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid 1791 (b'\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid 1792 (b'\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid 1793 (b'\xf1\x41\xf1\x80', FFFD+'A'+FFFD), 1794 (b'\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2), 1795 (b'\xf1\xf1\x80\x41', FFFD*2+'A'), 1796 (b'\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2), 1797 # with invalid start byte of a 4-byte sequence (rfc2279) 1798 (b'\xf5', FFFD), # only the start byte 1799 (b'\xf5\xf5', FFFD*2), # 2 start bytes 1800 (b'\xf5\x80', FFFD*2), # only 1 continuation byte 1801 (b'\xf5\x80\x80', FFFD*3), # only 2 continuation byte 1802 (b'\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes 1803 (b'\xf5\x80\x41', FFFD*2+'A'), # 1 valid cb and 1 invalid 1804 (b'\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD), 1805 (b'\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'), 1806 # with invalid start byte of a 5-byte sequence (rfc2279) 1807 (b'\xf8', FFFD), # only the start byte 1808 (b'\xf8\xf8', FFFD*2), # 2 start bytes 1809 (b'\xf8\x80', FFFD*2), # only one continuation byte 1810 (b'\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid 1811 (b'\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes 1812 # with invalid start byte of a 6-byte sequence (rfc2279) 1813 (b'\xfc', FFFD), # only the start byte 1814 (b'\xfc\xfc', FFFD*2), # 2 start bytes 1815 (b'\xfc\x80\x80', FFFD*3), # only 2 continuation bytes 1816 (b'\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes 1817 # invalid start byte 1818 (b'\xfe', FFFD), 1819 (b'\xfe\x80\x80', FFFD*3), 1820 # other sequences 1821 (b'\xf1\x80\x41\x42\x43', '\ufffd\x41\x42\x43'), 1822 (b'\xf1\x80\xff\x42\x43', '\ufffd\ufffd\x42\x43'), 1823 (b'\xf1\x80\xc2\x81\x43', '\ufffd\x81\x43'), 1824 (b'\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64', 1825 '\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'), 1826 ] 1827 for n, (seq, res) in enumerate(sequences): 1828 self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict') 1829 self.assertEqual(seq.decode('utf-8', 'replace'), res) 1830 self.assertEqual((seq+b'b').decode('utf-8', 'replace'), res+'b') 1831 self.assertEqual(seq.decode('utf-8', 'ignore'), 1832 res.replace('\uFFFD', '')) 1833 1834 def assertCorrectUTF8Decoding(self, seq, res, err): 1835 """ 1836 Check that an invalid UTF-8 sequence raises a UnicodeDecodeError when 1837 'strict' is used, returns res when 'replace' is used, and that doesn't 1838 return anything when 'ignore' is used. 1839 """ 1840 with self.assertRaises(UnicodeDecodeError) as cm: 1841 seq.decode('utf-8') 1842 exc = cm.exception 1843 1844 self.assertIn(err, str(exc)) 1845 self.assertEqual(seq.decode('utf-8', 'replace'), res) 1846 self.assertEqual((b'aaaa' + seq + b'bbbb').decode('utf-8', 'replace'), 1847 'aaaa' + res + 'bbbb') 1848 res = res.replace('\ufffd', '') 1849 self.assertEqual(seq.decode('utf-8', 'ignore'), res) 1850 self.assertEqual((b'aaaa' + seq + b'bbbb').decode('utf-8', 'ignore'), 1851 'aaaa' + res + 'bbbb') 1852 1853 def test_invalid_start_byte(self): 1854 """ 1855 Test that an 'invalid start byte' error is raised when the first byte 1856 is not in the ASCII range or is not a valid start byte of a 2-, 3-, or 1857 4-bytes sequence. The invalid start byte is replaced with a single 1858 U+FFFD when errors='replace'. 1859 E.g. <80> is a continuation byte and can appear only after a start byte. 1860 """ 1861 FFFD = '\ufffd' 1862 for byte in b'\x80\xA0\x9F\xBF\xC0\xC1\xF5\xFF': 1863 self.assertCorrectUTF8Decoding(bytes([byte]), '\ufffd', 1864 'invalid start byte') 1865 1866 def test_unexpected_end_of_data(self): 1867 """ 1868 Test that an 'unexpected end of data' error is raised when the string 1869 ends after a start byte of a 2-, 3-, or 4-bytes sequence without having 1870 enough continuation bytes. The incomplete sequence is replaced with a 1871 single U+FFFD when errors='replace'. 1872 E.g. in the sequence <F3 80 80>, F3 is the start byte of a 4-bytes 1873 sequence, but it's followed by only 2 valid continuation bytes and the 1874 last continuation bytes is missing. 1875 Note: the continuation bytes must be all valid, if one of them is 1876 invalid another error will be raised. 1877 """ 1878 sequences = [ 1879 'C2', 'DF', 1880 'E0 A0', 'E0 BF', 'E1 80', 'E1 BF', 'EC 80', 'EC BF', 1881 'ED 80', 'ED 9F', 'EE 80', 'EE BF', 'EF 80', 'EF BF', 1882 'F0 90', 'F0 BF', 'F0 90 80', 'F0 90 BF', 'F0 BF 80', 'F0 BF BF', 1883 'F1 80', 'F1 BF', 'F1 80 80', 'F1 80 BF', 'F1 BF 80', 'F1 BF BF', 1884 'F3 80', 'F3 BF', 'F3 80 80', 'F3 80 BF', 'F3 BF 80', 'F3 BF BF', 1885 'F4 80', 'F4 8F', 'F4 80 80', 'F4 80 BF', 'F4 8F 80', 'F4 8F BF' 1886 ] 1887 FFFD = '\ufffd' 1888 for seq in sequences: 1889 self.assertCorrectUTF8Decoding(bytes.fromhex(seq), '\ufffd', 1890 'unexpected end of data') 1891 1892 def test_invalid_cb_for_2bytes_seq(self): 1893 """ 1894 Test that an 'invalid continuation byte' error is raised when the 1895 continuation byte of a 2-bytes sequence is invalid. The start byte 1896 is replaced by a single U+FFFD and the second byte is handled 1897 separately when errors='replace'. 1898 E.g. in the sequence <C2 41>, C2 is the start byte of a 2-bytes 1899 sequence, but 41 is not a valid continuation byte because it's the 1900 ASCII letter 'A'. 1901 """ 1902 FFFD = '\ufffd' 1903 FFFDx2 = FFFD * 2 1904 sequences = [ 1905 ('C2 00', FFFD+'\x00'), ('C2 7F', FFFD+'\x7f'), 1906 ('C2 C0', FFFDx2), ('C2 FF', FFFDx2), 1907 ('DF 00', FFFD+'\x00'), ('DF 7F', FFFD+'\x7f'), 1908 ('DF C0', FFFDx2), ('DF FF', FFFDx2), 1909 ] 1910 for seq, res in sequences: 1911 self.assertCorrectUTF8Decoding(bytes.fromhex(seq), res, 1912 'invalid continuation byte') 1913 1914 def test_invalid_cb_for_3bytes_seq(self): 1915 """ 1916 Test that an 'invalid continuation byte' error is raised when the 1917 continuation byte(s) of a 3-bytes sequence are invalid. When 1918 errors='replace', if the first continuation byte is valid, the first 1919 two bytes (start byte + 1st cb) are replaced by a single U+FFFD and the 1920 third byte is handled separately, otherwise only the start byte is 1921 replaced with a U+FFFD and the other continuation bytes are handled 1922 separately. 1923 E.g. in the sequence <E1 80 41>, E1 is the start byte of a 3-bytes 1924 sequence, 80 is a valid continuation byte, but 41 is not a valid cb 1925 because it's the ASCII letter 'A'. 1926 Note: when the start byte is E0 or ED, the valid ranges for the first 1927 continuation byte are limited to A0..BF and 80..9F respectively. 1928 Python 2 used to consider all the bytes in range 80..BF valid when the 1929 start byte was ED. This is fixed in Python 3. 1930 """ 1931 FFFD = '\ufffd' 1932 FFFDx2 = FFFD * 2 1933 sequences = [ 1934 ('E0 00', FFFD+'\x00'), ('E0 7F', FFFD+'\x7f'), ('E0 80', FFFDx2), 1935 ('E0 9F', FFFDx2), ('E0 C0', FFFDx2), ('E0 FF', FFFDx2), 1936 ('E0 A0 00', FFFD+'\x00'), ('E0 A0 7F', FFFD+'\x7f'), 1937 ('E0 A0 C0', FFFDx2), ('E0 A0 FF', FFFDx2), 1938 ('E0 BF 00', FFFD+'\x00'), ('E0 BF 7F', FFFD+'\x7f'), 1939 ('E0 BF C0', FFFDx2), ('E0 BF FF', FFFDx2), ('E1 00', FFFD+'\x00'), 1940 ('E1 7F', FFFD+'\x7f'), ('E1 C0', FFFDx2), ('E1 FF', FFFDx2), 1941 ('E1 80 00', FFFD+'\x00'), ('E1 80 7F', FFFD+'\x7f'), 1942 ('E1 80 C0', FFFDx2), ('E1 80 FF', FFFDx2), 1943 ('E1 BF 00', FFFD+'\x00'), ('E1 BF 7F', FFFD+'\x7f'), 1944 ('E1 BF C0', FFFDx2), ('E1 BF FF', FFFDx2), ('EC 00', FFFD+'\x00'), 1945 ('EC 7F', FFFD+'\x7f'), ('EC C0', FFFDx2), ('EC FF', FFFDx2), 1946 ('EC 80 00', FFFD+'\x00'), ('EC 80 7F', FFFD+'\x7f'), 1947 ('EC 80 C0', FFFDx2), ('EC 80 FF', FFFDx2), 1948 ('EC BF 00', FFFD+'\x00'), ('EC BF 7F', FFFD+'\x7f'), 1949 ('EC BF C0', FFFDx2), ('EC BF FF', FFFDx2), ('ED 00', FFFD+'\x00'), 1950 ('ED 7F', FFFD+'\x7f'), 1951 ('ED A0', FFFDx2), ('ED BF', FFFDx2), # see note ^ 1952 ('ED C0', FFFDx2), ('ED FF', FFFDx2), ('ED 80 00', FFFD+'\x00'), 1953 ('ED 80 7F', FFFD+'\x7f'), ('ED 80 C0', FFFDx2), 1954 ('ED 80 FF', FFFDx2), ('ED 9F 00', FFFD+'\x00'), 1955 ('ED 9F 7F', FFFD+'\x7f'), ('ED 9F C0', FFFDx2), 1956 ('ED 9F FF', FFFDx2), ('EE 00', FFFD+'\x00'), 1957 ('EE 7F', FFFD+'\x7f'), ('EE C0', FFFDx2), ('EE FF', FFFDx2), 1958 ('EE 80 00', FFFD+'\x00'), ('EE 80 7F', FFFD+'\x7f'), 1959 ('EE 80 C0', FFFDx2), ('EE 80 FF', FFFDx2), 1960 ('EE BF 00', FFFD+'\x00'), ('EE BF 7F', FFFD+'\x7f'), 1961 ('EE BF C0', FFFDx2), ('EE BF FF', FFFDx2), ('EF 00', FFFD+'\x00'), 1962 ('EF 7F', FFFD+'\x7f'), ('EF C0', FFFDx2), ('EF FF', FFFDx2), 1963 ('EF 80 00', FFFD+'\x00'), ('EF 80 7F', FFFD+'\x7f'), 1964 ('EF 80 C0', FFFDx2), ('EF 80 FF', FFFDx2), 1965 ('EF BF 00', FFFD+'\x00'), ('EF BF 7F', FFFD+'\x7f'), 1966 ('EF BF C0', FFFDx2), ('EF BF FF', FFFDx2), 1967 ] 1968 for seq, res in sequences: 1969 self.assertCorrectUTF8Decoding(bytes.fromhex(seq), res, 1970 'invalid continuation byte') 1971 1972 def test_invalid_cb_for_4bytes_seq(self): 1973 """ 1974 Test that an 'invalid continuation byte' error is raised when the 1975 continuation byte(s) of a 4-bytes sequence are invalid. When 1976 errors='replace',the start byte and all the following valid 1977 continuation bytes are replaced with a single U+FFFD, and all the bytes 1978 starting from the first invalid continuation bytes (included) are 1979 handled separately. 1980 E.g. in the sequence <E1 80 41>, E1 is the start byte of a 3-bytes 1981 sequence, 80 is a valid continuation byte, but 41 is not a valid cb 1982 because it's the ASCII letter 'A'. 1983 Note: when the start byte is E0 or ED, the valid ranges for the first 1984 continuation byte are limited to A0..BF and 80..9F respectively. 1985 However, when the start byte is ED, Python 2 considers all the bytes 1986 in range 80..BF valid. This is fixed in Python 3. 1987 """ 1988 FFFD = '\ufffd' 1989 FFFDx2 = FFFD * 2 1990 sequences = [ 1991 ('F0 00', FFFD+'\x00'), ('F0 7F', FFFD+'\x7f'), ('F0 80', FFFDx2), 1992 ('F0 8F', FFFDx2), ('F0 C0', FFFDx2), ('F0 FF', FFFDx2), 1993 ('F0 90 00', FFFD+'\x00'), ('F0 90 7F', FFFD+'\x7f'), 1994 ('F0 90 C0', FFFDx2), ('F0 90 FF', FFFDx2), 1995 ('F0 BF 00', FFFD+'\x00'), ('F0 BF 7F', FFFD+'\x7f'), 1996 ('F0 BF C0', FFFDx2), ('F0 BF FF', FFFDx2), 1997 ('F0 90 80 00', FFFD+'\x00'), ('F0 90 80 7F', FFFD+'\x7f'), 1998 ('F0 90 80 C0', FFFDx2), ('F0 90 80 FF', FFFDx2), 1999 ('F0 90 BF 00', FFFD+'\x00'), ('F0 90 BF 7F', FFFD+'\x7f'), 2000 ('F0 90 BF C0', FFFDx2), ('F0 90 BF FF', FFFDx2), 2001 ('F0 BF 80 00', FFFD+'\x00'), ('F0 BF 80 7F', FFFD+'\x7f'), 2002 ('F0 BF 80 C0', FFFDx2), ('F0 BF 80 FF', FFFDx2), 2003 ('F0 BF BF 00', FFFD+'\x00'), ('F0 BF BF 7F', FFFD+'\x7f'), 2004 ('F0 BF BF C0', FFFDx2), ('F0 BF BF FF', FFFDx2), 2005 ('F1 00', FFFD+'\x00'), ('F1 7F', FFFD+'\x7f'), ('F1 C0', FFFDx2), 2006 ('F1 FF', FFFDx2), ('F1 80 00', FFFD+'\x00'), 2007 ('F1 80 7F', FFFD+'\x7f'), ('F1 80 C0', FFFDx2), 2008 ('F1 80 FF', FFFDx2), ('F1 BF 00', FFFD+'\x00'), 2009 ('F1 BF 7F', FFFD+'\x7f'), ('F1 BF C0', FFFDx2), 2010 ('F1 BF FF', FFFDx2), ('F1 80 80 00', FFFD+'\x00'), 2011 ('F1 80 80 7F', FFFD+'\x7f'), ('F1 80 80 C0', FFFDx2), 2012 ('F1 80 80 FF', FFFDx2), ('F1 80 BF 00', FFFD+'\x00'), 2013 ('F1 80 BF 7F', FFFD+'\x7f'), ('F1 80 BF C0', FFFDx2), 2014 ('F1 80 BF FF', FFFDx2), ('F1 BF 80 00', FFFD+'\x00'), 2015 ('F1 BF 80 7F', FFFD+'\x7f'), ('F1 BF 80 C0', FFFDx2), 2016 ('F1 BF 80 FF', FFFDx2), ('F1 BF BF 00', FFFD+'\x00'), 2017 ('F1 BF BF 7F', FFFD+'\x7f'), ('F1 BF BF C0', FFFDx2), 2018 ('F1 BF BF FF', FFFDx2), ('F3 00', FFFD+'\x00'), 2019 ('F3 7F', FFFD+'\x7f'), ('F3 C0', FFFDx2), ('F3 FF', FFFDx2), 2020 ('F3 80 00', FFFD+'\x00'), ('F3 80 7F', FFFD+'\x7f'), 2021 ('F3 80 C0', FFFDx2), ('F3 80 FF', FFFDx2), 2022 ('F3 BF 00', FFFD+'\x00'), ('F3 BF 7F', FFFD+'\x7f'), 2023 ('F3 BF C0', FFFDx2), ('F3 BF FF', FFFDx2), 2024 ('F3 80 80 00', FFFD+'\x00'), ('F3 80 80 7F', FFFD+'\x7f'), 2025 ('F3 80 80 C0', FFFDx2), ('F3 80 80 FF', FFFDx2), 2026 ('F3 80 BF 00', FFFD+'\x00'), ('F3 80 BF 7F', FFFD+'\x7f'), 2027 ('F3 80 BF C0', FFFDx2), ('F3 80 BF FF', FFFDx2), 2028 ('F3 BF 80 00', FFFD+'\x00'), ('F3 BF 80 7F', FFFD+'\x7f'), 2029 ('F3 BF 80 C0', FFFDx2), ('F3 BF 80 FF', FFFDx2), 2030 ('F3 BF BF 00', FFFD+'\x00'), ('F3 BF BF 7F', FFFD+'\x7f'), 2031 ('F3 BF BF C0', FFFDx2), ('F3 BF BF FF', FFFDx2), 2032 ('F4 00', FFFD+'\x00'), ('F4 7F', FFFD+'\x7f'), ('F4 90', FFFDx2), 2033 ('F4 BF', FFFDx2), ('F4 C0', FFFDx2), ('F4 FF', FFFDx2), 2034 ('F4 80 00', FFFD+'\x00'), ('F4 80 7F', FFFD+'\x7f'), 2035 ('F4 80 C0', FFFDx2), ('F4 80 FF', FFFDx2), 2036 ('F4 8F 00', FFFD+'\x00'), ('F4 8F 7F', FFFD+'\x7f'), 2037 ('F4 8F C0', FFFDx2), ('F4 8F FF', FFFDx2), 2038 ('F4 80 80 00', FFFD+'\x00'), ('F4 80 80 7F', FFFD+'\x7f'), 2039 ('F4 80 80 C0', FFFDx2), ('F4 80 80 FF', FFFDx2), 2040 ('F4 80 BF 00', FFFD+'\x00'), ('F4 80 BF 7F', FFFD+'\x7f'), 2041 ('F4 80 BF C0', FFFDx2), ('F4 80 BF FF', FFFDx2), 2042 ('F4 8F 80 00', FFFD+'\x00'), ('F4 8F 80 7F', FFFD+'\x7f'), 2043 ('F4 8F 80 C0', FFFDx2), ('F4 8F 80 FF', FFFDx2), 2044 ('F4 8F BF 00', FFFD+'\x00'), ('F4 8F BF 7F', FFFD+'\x7f'), 2045 ('F4 8F BF C0', FFFDx2), ('F4 8F BF FF', FFFDx2) 2046 ] 2047 for seq, res in sequences: 2048 self.assertCorrectUTF8Decoding(bytes.fromhex(seq), res, 2049 'invalid continuation byte') 2050 2051 def test_codecs_idna(self): 2052 # Test whether trailing dot is preserved 2053 self.assertEqual("www.python.org.".encode("idna"), b"www.python.org.") 2054 2055 def test_codecs_errors(self): 2056 # Error handling (encoding) 2057 self.assertRaises(UnicodeError, 'Andr\202 x'.encode, 'ascii') 2058 self.assertRaises(UnicodeError, 'Andr\202 x'.encode, 'ascii','strict') 2059 self.assertEqual('Andr\202 x'.encode('ascii','ignore'), b"Andr x") 2060 self.assertEqual('Andr\202 x'.encode('ascii','replace'), b"Andr? x") 2061 self.assertEqual('Andr\202 x'.encode('ascii', 'replace'), 2062 'Andr\202 x'.encode('ascii', errors='replace')) 2063 self.assertEqual('Andr\202 x'.encode('ascii', 'ignore'), 2064 'Andr\202 x'.encode(encoding='ascii', errors='ignore')) 2065 2066 # Error handling (decoding) 2067 self.assertRaises(UnicodeError, str, b'Andr\202 x', 'ascii') 2068 self.assertRaises(UnicodeError, str, b'Andr\202 x', 'ascii', 'strict') 2069 self.assertEqual(str(b'Andr\202 x', 'ascii', 'ignore'), "Andr x") 2070 self.assertEqual(str(b'Andr\202 x', 'ascii', 'replace'), 'Andr\uFFFD x') 2071 self.assertEqual(str(b'\202 x', 'ascii', 'replace'), '\uFFFD x') 2072 2073 # Error handling (unknown character names) 2074 self.assertEqual(b"\\N{foo}xx".decode("unicode-escape", "ignore"), "xx") 2075 2076 # Error handling (truncated escape sequence) 2077 self.assertRaises(UnicodeError, b"\\".decode, "unicode-escape") 2078 2079 self.assertRaises(TypeError, b"hello".decode, "test.unicode1") 2080 self.assertRaises(TypeError, str, b"hello", "test.unicode2") 2081 self.assertRaises(TypeError, "hello".encode, "test.unicode1") 2082 self.assertRaises(TypeError, "hello".encode, "test.unicode2") 2083 2084 # Error handling (wrong arguments) 2085 self.assertRaises(TypeError, "hello".encode, 42, 42, 42) 2086 2087 # Error handling (lone surrogate in 2088 # _PyUnicode_TransformDecimalAndSpaceToASCII()) 2089 self.assertRaises(ValueError, int, "\ud800") 2090 self.assertRaises(ValueError, int, "\udf00") 2091 self.assertRaises(ValueError, float, "\ud800") 2092 self.assertRaises(ValueError, float, "\udf00") 2093 self.assertRaises(ValueError, complex, "\ud800") 2094 self.assertRaises(ValueError, complex, "\udf00") 2095 2096 def test_codecs(self): 2097 # Encoding 2098 self.assertEqual('hello'.encode('ascii'), b'hello') 2099 self.assertEqual('hello'.encode('utf-7'), b'hello') 2100 self.assertEqual('hello'.encode('utf-8'), b'hello') 2101 self.assertEqual('hello'.encode('utf-8'), b'hello') 2102 self.assertEqual('hello'.encode('utf-16-le'), b'h\000e\000l\000l\000o\000') 2103 self.assertEqual('hello'.encode('utf-16-be'), b'\000h\000e\000l\000l\000o') 2104 self.assertEqual('hello'.encode('latin-1'), b'hello') 2105 2106 # Default encoding is utf-8 2107 self.assertEqual('\u2603'.encode(), b'\xe2\x98\x83') 2108 2109 # Roundtrip safety for BMP (just the first 1024 chars) 2110 for c in range(1024): 2111 u = chr(c) 2112 for encoding in ('utf-7', 'utf-8', 'utf-16', 'utf-16-le', 2113 'utf-16-be', 'raw_unicode_escape', 2114 'unicode_escape', 'unicode_internal'): 2115 with warnings.catch_warnings(): 2116 # unicode-internal has been deprecated 2117 warnings.simplefilter("ignore", DeprecationWarning) 2118 2119 self.assertEqual(str(u.encode(encoding),encoding), u) 2120 2121 # Roundtrip safety for BMP (just the first 256 chars) 2122 for c in range(256): 2123 u = chr(c) 2124 for encoding in ('latin-1',): 2125 self.assertEqual(str(u.encode(encoding),encoding), u) 2126 2127 # Roundtrip safety for BMP (just the first 128 chars) 2128 for c in range(128): 2129 u = chr(c) 2130 for encoding in ('ascii',): 2131 self.assertEqual(str(u.encode(encoding),encoding), u) 2132 2133 # Roundtrip safety for non-BMP (just a few chars) 2134 with warnings.catch_warnings(): 2135 # unicode-internal has been deprecated 2136 warnings.simplefilter("ignore", DeprecationWarning) 2137 2138 u = '\U00010001\U00020002\U00030003\U00040004\U00050005' 2139 for encoding in ('utf-8', 'utf-16', 'utf-16-le', 'utf-16-be', 2140 'raw_unicode_escape', 2141 'unicode_escape', 'unicode_internal'): 2142 self.assertEqual(str(u.encode(encoding),encoding), u) 2143 2144 # UTF-8 must be roundtrip safe for all code points 2145 # (except surrogates, which are forbidden). 2146 u = ''.join(map(chr, list(range(0, 0xd800)) + 2147 list(range(0xe000, 0x110000)))) 2148 for encoding in ('utf-8',): 2149 self.assertEqual(str(u.encode(encoding),encoding), u) 2150 2151 def test_codecs_charmap(self): 2152 # 0-127 2153 s = bytes(range(128)) 2154 for encoding in ( 2155 'cp037', 'cp1026', 'cp273', 2156 'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850', 2157 'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862', 2158 'cp863', 'cp865', 'cp866', 'cp1125', 2159 'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15', 2160 'iso8859_2', 'iso8859_3', 'iso8859_4', 'iso8859_5', 'iso8859_6', 2161 'iso8859_7', 'iso8859_9', 2162 'koi8_r', 'koi8_t', 'koi8_u', 'kz1048', 'latin_1', 2163 'mac_cyrillic', 'mac_latin2', 2164 2165 'cp1250', 'cp1251', 'cp1252', 'cp1253', 'cp1254', 'cp1255', 2166 'cp1256', 'cp1257', 'cp1258', 2167 'cp856', 'cp857', 'cp864', 'cp869', 'cp874', 2168 2169 'mac_greek', 'mac_iceland','mac_roman', 'mac_turkish', 2170 'cp1006', 'iso8859_8', 2171 2172 ### These have undefined mappings: 2173 #'cp424', 2174 2175 ### These fail the round-trip: 2176 #'cp875' 2177 2178 ): 2179 self.assertEqual(str(s, encoding).encode(encoding), s) 2180 2181 # 128-255 2182 s = bytes(range(128, 256)) 2183 for encoding in ( 2184 'cp037', 'cp1026', 'cp273', 2185 'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850', 2186 'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862', 2187 'cp863', 'cp865', 'cp866', 'cp1125', 2188 'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15', 2189 'iso8859_2', 'iso8859_4', 'iso8859_5', 2190 'iso8859_9', 'koi8_r', 'koi8_u', 'latin_1', 2191 'mac_cyrillic', 'mac_latin2', 2192 2193 ### These have undefined mappings: 2194 #'cp1250', 'cp1251', 'cp1252', 'cp1253', 'cp1254', 'cp1255', 2195 #'cp1256', 'cp1257', 'cp1258', 2196 #'cp424', 'cp856', 'cp857', 'cp864', 'cp869', 'cp874', 2197 #'iso8859_3', 'iso8859_6', 'iso8859_7', 'koi8_t', 'kz1048', 2198 #'mac_greek', 'mac_iceland','mac_roman', 'mac_turkish', 2199 2200 ### These fail the round-trip: 2201 #'cp1006', 'cp875', 'iso8859_8', 2202 2203 ): 2204 self.assertEqual(str(s, encoding).encode(encoding), s) 2205 2206 def test_concatenation(self): 2207 self.assertEqual(("abc" "def"), "abcdef") 2208 self.assertEqual(("abc" "def"), "abcdef") 2209 self.assertEqual(("abc" "def"), "abcdef") 2210 self.assertEqual(("abc" "def" "ghi"), "abcdefghi") 2211 self.assertEqual(("abc" "def" "ghi"), "abcdefghi") 2212 2213 def test_printing(self): 2214 class BitBucket: 2215 def write(self, text): 2216 pass 2217 2218 out = BitBucket() 2219 print('abc', file=out) 2220 print('abc', 'def', file=out) 2221 print('abc', 'def', file=out) 2222 print('abc', 'def', file=out) 2223 print('abc\n', file=out) 2224 print('abc\n', end=' ', file=out) 2225 print('abc\n', end=' ', file=out) 2226 print('def\n', file=out) 2227 print('def\n', file=out) 2228 2229 def test_ucs4(self): 2230 x = '\U00100000' 2231 y = x.encode("raw-unicode-escape").decode("raw-unicode-escape") 2232 self.assertEqual(x, y) 2233 2234 y = br'\U00100000' 2235 x = y.decode("raw-unicode-escape").encode("raw-unicode-escape") 2236 self.assertEqual(x, y) 2237 y = br'\U00010000' 2238 x = y.decode("raw-unicode-escape").encode("raw-unicode-escape") 2239 self.assertEqual(x, y) 2240 2241 try: 2242 br'\U11111111'.decode("raw-unicode-escape") 2243 except UnicodeDecodeError as e: 2244 self.assertEqual(e.start, 0) 2245 self.assertEqual(e.end, 10) 2246 else: 2247 self.fail("Should have raised UnicodeDecodeError") 2248 2249 def test_conversion(self): 2250 # Make sure __str__() works properly 2251 class ObjectToStr: 2252 def __str__(self): 2253 return "foo" 2254 2255 class StrSubclassToStr(str): 2256 def __str__(self): 2257 return "foo" 2258 2259 class StrSubclassToStrSubclass(str): 2260 def __new__(cls, content=""): 2261 return str.__new__(cls, 2*content) 2262 def __str__(self): 2263 return self 2264 2265 self.assertEqual(str(ObjectToStr()), "foo") 2266 self.assertEqual(str(StrSubclassToStr("bar")), "foo") 2267 s = str(StrSubclassToStrSubclass("foo")) 2268 self.assertEqual(s, "foofoo") 2269 self.assertIs(type(s), StrSubclassToStrSubclass) 2270 s = StrSubclass(StrSubclassToStrSubclass("foo")) 2271 self.assertEqual(s, "foofoo") 2272 self.assertIs(type(s), StrSubclass) 2273 2274 def test_unicode_repr(self): 2275 class s1: 2276 def __repr__(self): 2277 return '\\n' 2278 2279 class s2: 2280 def __repr__(self): 2281 return '\\n' 2282 2283 self.assertEqual(repr(s1()), '\\n') 2284 self.assertEqual(repr(s2()), '\\n') 2285 2286 def test_printable_repr(self): 2287 self.assertEqual(repr('\U00010000'), "'%c'" % (0x10000,)) # printable 2288 self.assertEqual(repr('\U00014000'), "'\\U00014000'") # nonprintable 2289 2290 # This test only affects 32-bit platforms because expandtabs can only take 2291 # an int as the max value, not a 64-bit C long. If expandtabs is changed 2292 # to take a 64-bit long, this test should apply to all platforms. 2293 @unittest.skipIf(sys.maxsize > (1 << 32) or struct.calcsize('P') != 4, 2294 'only applies to 32-bit platforms') 2295 def test_expandtabs_overflows_gracefully(self): 2296 self.assertRaises(OverflowError, 't\tt\t'.expandtabs, sys.maxsize) 2297 2298 @support.cpython_only 2299 def test_expandtabs_optimization(self): 2300 s = 'abc' 2301 self.assertIs(s.expandtabs(), s) 2302 2303 def test_raiseMemError(self): 2304 if struct.calcsize('P') == 8: 2305 # 64 bits pointers 2306 ascii_struct_size = 48 2307 compact_struct_size = 72 2308 else: 2309 # 32 bits pointers 2310 ascii_struct_size = 24 2311 compact_struct_size = 36 2312 2313 for char in ('a', '\xe9', '\u20ac', '\U0010ffff'): 2314 code = ord(char) 2315 if code < 0x100: 2316 char_size = 1 # sizeof(Py_UCS1) 2317 struct_size = ascii_struct_size 2318 elif code < 0x10000: 2319 char_size = 2 # sizeof(Py_UCS2) 2320 struct_size = compact_struct_size 2321 else: 2322 char_size = 4 # sizeof(Py_UCS4) 2323 struct_size = compact_struct_size 2324 # Note: sys.maxsize is half of the actual max allocation because of 2325 # the signedness of Py_ssize_t. Strings of maxlen-1 should in principle 2326 # be allocatable, given enough memory. 2327 maxlen = ((sys.maxsize - struct_size) // char_size) 2328 alloc = lambda: char * maxlen 2329 self.assertRaises(MemoryError, alloc) 2330 self.assertRaises(MemoryError, alloc) 2331 2332 def test_format_subclass(self): 2333 class S(str): 2334 def __str__(self): 2335 return '__str__ overridden' 2336 s = S('xxx') 2337 self.assertEqual("%s" % s, '__str__ overridden') 2338 self.assertEqual("{}".format(s), '__str__ overridden') 2339 2340 def test_subclass_add(self): 2341 class S(str): 2342 def __add__(self, o): 2343 return "3" 2344 self.assertEqual(S("4") + S("5"), "3") 2345 class S(str): 2346 def __iadd__(self, o): 2347 return "3" 2348 s = S("1") 2349 s += "4" 2350 self.assertEqual(s, "3") 2351 2352 def test_getnewargs(self): 2353 text = 'abc' 2354 args = text.__getnewargs__() 2355 self.assertIsNot(args[0], text) 2356 self.assertEqual(args[0], text) 2357 self.assertEqual(len(args), 1) 2358 2359 def test_resize(self): 2360 for length in range(1, 100, 7): 2361 # generate a fresh string (refcount=1) 2362 text = 'a' * length + 'b' 2363 2364 with support.check_warnings(('unicode_internal codec has been ' 2365 'deprecated', DeprecationWarning)): 2366 # fill wstr internal field 2367 abc = text.encode('unicode_internal') 2368 self.assertEqual(abc.decode('unicode_internal'), text) 2369 2370 # resize text: wstr field must be cleared and then recomputed 2371 text += 'c' 2372 abcdef = text.encode('unicode_internal') 2373 self.assertNotEqual(abc, abcdef) 2374 self.assertEqual(abcdef.decode('unicode_internal'), text) 2375 2376 def test_compare(self): 2377 # Issue #17615 2378 N = 10 2379 ascii = 'a' * N 2380 ascii2 = 'z' * N 2381 latin = '\x80' * N 2382 latin2 = '\xff' * N 2383 bmp = '\u0100' * N 2384 bmp2 = '\uffff' * N 2385 astral = '\U00100000' * N 2386 astral2 = '\U0010ffff' * N 2387 strings = ( 2388 ascii, ascii2, 2389 latin, latin2, 2390 bmp, bmp2, 2391 astral, astral2) 2392 for text1, text2 in itertools.combinations(strings, 2): 2393 equal = (text1 is text2) 2394 self.assertEqual(text1 == text2, equal) 2395 self.assertEqual(text1 != text2, not equal) 2396 2397 if equal: 2398 self.assertTrue(text1 <= text2) 2399 self.assertTrue(text1 >= text2) 2400 2401 # text1 is text2: duplicate strings to skip the "str1 == str2" 2402 # optimization in unicode_compare_eq() and really compare 2403 # character per character 2404 copy1 = duplicate_string(text1) 2405 copy2 = duplicate_string(text2) 2406 self.assertIsNot(copy1, copy2) 2407 2408 self.assertTrue(copy1 == copy2) 2409 self.assertFalse(copy1 != copy2) 2410 2411 self.assertTrue(copy1 <= copy2) 2412 self.assertTrue(copy2 >= copy2) 2413 2414 self.assertTrue(ascii < ascii2) 2415 self.assertTrue(ascii < latin) 2416 self.assertTrue(ascii < bmp) 2417 self.assertTrue(ascii < astral) 2418 self.assertFalse(ascii >= ascii2) 2419 self.assertFalse(ascii >= latin) 2420 self.assertFalse(ascii >= bmp) 2421 self.assertFalse(ascii >= astral) 2422 2423 self.assertFalse(latin < ascii) 2424 self.assertTrue(latin < latin2) 2425 self.assertTrue(latin < bmp) 2426 self.assertTrue(latin < astral) 2427 self.assertTrue(latin >= ascii) 2428 self.assertFalse(latin >= latin2) 2429 self.assertFalse(latin >= bmp) 2430 self.assertFalse(latin >= astral) 2431 2432 self.assertFalse(bmp < ascii) 2433 self.assertFalse(bmp < latin) 2434 self.assertTrue(bmp < bmp2) 2435 self.assertTrue(bmp < astral) 2436 self.assertTrue(bmp >= ascii) 2437 self.assertTrue(bmp >= latin) 2438 self.assertFalse(bmp >= bmp2) 2439 self.assertFalse(bmp >= astral) 2440 2441 self.assertFalse(astral < ascii) 2442 self.assertFalse(astral < latin) 2443 self.assertFalse(astral < bmp2) 2444 self.assertTrue(astral < astral2) 2445 self.assertTrue(astral >= ascii) 2446 self.assertTrue(astral >= latin) 2447 self.assertTrue(astral >= bmp2) 2448 self.assertFalse(astral >= astral2) 2449 2450 def test_free_after_iterating(self): 2451 support.check_free_after_iterating(self, iter, str) 2452 support.check_free_after_iterating(self, reversed, str) 2453 2454 2455class CAPITest(unittest.TestCase): 2456 2457 # Test PyUnicode_FromFormat() 2458 def test_from_format(self): 2459 support.import_module('ctypes') 2460 from ctypes import ( 2461 pythonapi, py_object, sizeof, 2462 c_int, c_long, c_longlong, c_ssize_t, 2463 c_uint, c_ulong, c_ulonglong, c_size_t, c_void_p) 2464 name = "PyUnicode_FromFormat" 2465 _PyUnicode_FromFormat = getattr(pythonapi, name) 2466 _PyUnicode_FromFormat.restype = py_object 2467 2468 def PyUnicode_FromFormat(format, *args): 2469 cargs = tuple( 2470 py_object(arg) if isinstance(arg, str) else arg 2471 for arg in args) 2472 return _PyUnicode_FromFormat(format, *cargs) 2473 2474 def check_format(expected, format, *args): 2475 text = PyUnicode_FromFormat(format, *args) 2476 self.assertEqual(expected, text) 2477 2478 # ascii format, non-ascii argument 2479 check_format('ascii\x7f=unicode\xe9', 2480 b'ascii\x7f=%U', 'unicode\xe9') 2481 2482 # non-ascii format, ascii argument: ensure that PyUnicode_FromFormatV() 2483 # raises an error 2484 self.assertRaisesRegex(ValueError, 2485 r'^PyUnicode_FromFormatV\(\) expects an ASCII-encoded format ' 2486 'string, got a non-ASCII byte: 0xe9$', 2487 PyUnicode_FromFormat, b'unicode\xe9=%s', 'ascii') 2488 2489 # test "%c" 2490 check_format('\uabcd', 2491 b'%c', c_int(0xabcd)) 2492 check_format('\U0010ffff', 2493 b'%c', c_int(0x10ffff)) 2494 with self.assertRaises(OverflowError): 2495 PyUnicode_FromFormat(b'%c', c_int(0x110000)) 2496 # Issue #18183 2497 check_format('\U00010000\U00100000', 2498 b'%c%c', c_int(0x10000), c_int(0x100000)) 2499 2500 # test "%" 2501 check_format('%', 2502 b'%') 2503 check_format('%', 2504 b'%%') 2505 check_format('%s', 2506 b'%%s') 2507 check_format('[%]', 2508 b'[%%]') 2509 check_format('%abc', 2510 b'%%%s', b'abc') 2511 2512 # truncated string 2513 check_format('abc', 2514 b'%.3s', b'abcdef') 2515 check_format('abc[\ufffd', 2516 b'%.5s', 'abc[\u20ac]'.encode('utf8')) 2517 check_format("'\\u20acABC'", 2518 b'%A', '\u20acABC') 2519 check_format("'\\u20", 2520 b'%.5A', '\u20acABCDEF') 2521 check_format("'\u20acABC'", 2522 b'%R', '\u20acABC') 2523 check_format("'\u20acA", 2524 b'%.3R', '\u20acABCDEF') 2525 check_format('\u20acAB', 2526 b'%.3S', '\u20acABCDEF') 2527 check_format('\u20acAB', 2528 b'%.3U', '\u20acABCDEF') 2529 check_format('\u20acAB', 2530 b'%.3V', '\u20acABCDEF', None) 2531 check_format('abc[\ufffd', 2532 b'%.5V', None, 'abc[\u20ac]'.encode('utf8')) 2533 2534 # following tests comes from #7330 2535 # test width modifier and precision modifier with %S 2536 check_format("repr= abc", 2537 b'repr=%5S', 'abc') 2538 check_format("repr=ab", 2539 b'repr=%.2S', 'abc') 2540 check_format("repr= ab", 2541 b'repr=%5.2S', 'abc') 2542 2543 # test width modifier and precision modifier with %R 2544 check_format("repr= 'abc'", 2545 b'repr=%8R', 'abc') 2546 check_format("repr='ab", 2547 b'repr=%.3R', 'abc') 2548 check_format("repr= 'ab", 2549 b'repr=%5.3R', 'abc') 2550 2551 # test width modifier and precision modifier with %A 2552 check_format("repr= 'abc'", 2553 b'repr=%8A', 'abc') 2554 check_format("repr='ab", 2555 b'repr=%.3A', 'abc') 2556 check_format("repr= 'ab", 2557 b'repr=%5.3A', 'abc') 2558 2559 # test width modifier and precision modifier with %s 2560 check_format("repr= abc", 2561 b'repr=%5s', b'abc') 2562 check_format("repr=ab", 2563 b'repr=%.2s', b'abc') 2564 check_format("repr= ab", 2565 b'repr=%5.2s', b'abc') 2566 2567 # test width modifier and precision modifier with %U 2568 check_format("repr= abc", 2569 b'repr=%5U', 'abc') 2570 check_format("repr=ab", 2571 b'repr=%.2U', 'abc') 2572 check_format("repr= ab", 2573 b'repr=%5.2U', 'abc') 2574 2575 # test width modifier and precision modifier with %V 2576 check_format("repr= abc", 2577 b'repr=%5V', 'abc', b'123') 2578 check_format("repr=ab", 2579 b'repr=%.2V', 'abc', b'123') 2580 check_format("repr= ab", 2581 b'repr=%5.2V', 'abc', b'123') 2582 check_format("repr= 123", 2583 b'repr=%5V', None, b'123') 2584 check_format("repr=12", 2585 b'repr=%.2V', None, b'123') 2586 check_format("repr= 12", 2587 b'repr=%5.2V', None, b'123') 2588 2589 # test integer formats (%i, %d, %u) 2590 check_format('010', 2591 b'%03i', c_int(10)) 2592 check_format('0010', 2593 b'%0.4i', c_int(10)) 2594 check_format('-123', 2595 b'%i', c_int(-123)) 2596 check_format('-123', 2597 b'%li', c_long(-123)) 2598 check_format('-123', 2599 b'%lli', c_longlong(-123)) 2600 check_format('-123', 2601 b'%zi', c_ssize_t(-123)) 2602 2603 check_format('-123', 2604 b'%d', c_int(-123)) 2605 check_format('-123', 2606 b'%ld', c_long(-123)) 2607 check_format('-123', 2608 b'%lld', c_longlong(-123)) 2609 check_format('-123', 2610 b'%zd', c_ssize_t(-123)) 2611 2612 check_format('123', 2613 b'%u', c_uint(123)) 2614 check_format('123', 2615 b'%lu', c_ulong(123)) 2616 check_format('123', 2617 b'%llu', c_ulonglong(123)) 2618 check_format('123', 2619 b'%zu', c_size_t(123)) 2620 2621 # test long output 2622 min_longlong = -(2 ** (8 * sizeof(c_longlong) - 1)) 2623 max_longlong = -min_longlong - 1 2624 check_format(str(min_longlong), 2625 b'%lld', c_longlong(min_longlong)) 2626 check_format(str(max_longlong), 2627 b'%lld', c_longlong(max_longlong)) 2628 max_ulonglong = 2 ** (8 * sizeof(c_ulonglong)) - 1 2629 check_format(str(max_ulonglong), 2630 b'%llu', c_ulonglong(max_ulonglong)) 2631 PyUnicode_FromFormat(b'%p', c_void_p(-1)) 2632 2633 # test padding (width and/or precision) 2634 check_format('123'.rjust(10, '0'), 2635 b'%010i', c_int(123)) 2636 check_format('123'.rjust(100), 2637 b'%100i', c_int(123)) 2638 check_format('123'.rjust(100, '0'), 2639 b'%.100i', c_int(123)) 2640 check_format('123'.rjust(80, '0').rjust(100), 2641 b'%100.80i', c_int(123)) 2642 2643 check_format('123'.rjust(10, '0'), 2644 b'%010u', c_uint(123)) 2645 check_format('123'.rjust(100), 2646 b'%100u', c_uint(123)) 2647 check_format('123'.rjust(100, '0'), 2648 b'%.100u', c_uint(123)) 2649 check_format('123'.rjust(80, '0').rjust(100), 2650 b'%100.80u', c_uint(123)) 2651 2652 check_format('123'.rjust(10, '0'), 2653 b'%010x', c_int(0x123)) 2654 check_format('123'.rjust(100), 2655 b'%100x', c_int(0x123)) 2656 check_format('123'.rjust(100, '0'), 2657 b'%.100x', c_int(0x123)) 2658 check_format('123'.rjust(80, '0').rjust(100), 2659 b'%100.80x', c_int(0x123)) 2660 2661 # test %A 2662 check_format(r"%A:'abc\xe9\uabcd\U0010ffff'", 2663 b'%%A:%A', 'abc\xe9\uabcd\U0010ffff') 2664 2665 # test %V 2666 check_format('repr=abc', 2667 b'repr=%V', 'abc', b'xyz') 2668 2669 # Test string decode from parameter of %s using utf-8. 2670 # b'\xe4\xba\xba\xe6\xb0\x91' is utf-8 encoded byte sequence of 2671 # '\u4eba\u6c11' 2672 check_format('repr=\u4eba\u6c11', 2673 b'repr=%V', None, b'\xe4\xba\xba\xe6\xb0\x91') 2674 2675 #Test replace error handler. 2676 check_format('repr=abc\ufffd', 2677 b'repr=%V', None, b'abc\xff') 2678 2679 # not supported: copy the raw format string. these tests are just here 2680 # to check for crashes and should not be considered as specifications 2681 check_format('%s', 2682 b'%1%s', b'abc') 2683 check_format('%1abc', 2684 b'%1abc') 2685 check_format('%+i', 2686 b'%+i', c_int(10)) 2687 check_format('%.%s', 2688 b'%.%s', b'abc') 2689 2690 # Issue #33817: empty strings 2691 check_format('', 2692 b'') 2693 check_format('', 2694 b'%s', b'') 2695 2696 # Test PyUnicode_AsWideChar() 2697 @support.cpython_only 2698 def test_aswidechar(self): 2699 from _testcapi import unicode_aswidechar 2700 support.import_module('ctypes') 2701 from ctypes import c_wchar, sizeof 2702 2703 wchar, size = unicode_aswidechar('abcdef', 2) 2704 self.assertEqual(size, 2) 2705 self.assertEqual(wchar, 'ab') 2706 2707 wchar, size = unicode_aswidechar('abc', 3) 2708 self.assertEqual(size, 3) 2709 self.assertEqual(wchar, 'abc') 2710 2711 wchar, size = unicode_aswidechar('abc', 4) 2712 self.assertEqual(size, 3) 2713 self.assertEqual(wchar, 'abc\0') 2714 2715 wchar, size = unicode_aswidechar('abc', 10) 2716 self.assertEqual(size, 3) 2717 self.assertEqual(wchar, 'abc\0') 2718 2719 wchar, size = unicode_aswidechar('abc\0def', 20) 2720 self.assertEqual(size, 7) 2721 self.assertEqual(wchar, 'abc\0def\0') 2722 2723 nonbmp = chr(0x10ffff) 2724 if sizeof(c_wchar) == 2: 2725 buflen = 3 2726 nchar = 2 2727 else: # sizeof(c_wchar) == 4 2728 buflen = 2 2729 nchar = 1 2730 wchar, size = unicode_aswidechar(nonbmp, buflen) 2731 self.assertEqual(size, nchar) 2732 self.assertEqual(wchar, nonbmp + '\0') 2733 2734 # Test PyUnicode_AsWideCharString() 2735 @support.cpython_only 2736 def test_aswidecharstring(self): 2737 from _testcapi import unicode_aswidecharstring 2738 support.import_module('ctypes') 2739 from ctypes import c_wchar, sizeof 2740 2741 wchar, size = unicode_aswidecharstring('abc') 2742 self.assertEqual(size, 3) 2743 self.assertEqual(wchar, 'abc\0') 2744 2745 wchar, size = unicode_aswidecharstring('abc\0def') 2746 self.assertEqual(size, 7) 2747 self.assertEqual(wchar, 'abc\0def\0') 2748 2749 nonbmp = chr(0x10ffff) 2750 if sizeof(c_wchar) == 2: 2751 nchar = 2 2752 else: # sizeof(c_wchar) == 4 2753 nchar = 1 2754 wchar, size = unicode_aswidecharstring(nonbmp) 2755 self.assertEqual(size, nchar) 2756 self.assertEqual(wchar, nonbmp + '\0') 2757 2758 # Test PyUnicode_AsUCS4() 2759 @support.cpython_only 2760 def test_asucs4(self): 2761 from _testcapi import unicode_asucs4 2762 for s in ['abc', '\xa1\xa2', '\u4f60\u597d', 'a\U0001f600', 2763 'a\ud800b\udfffc', '\ud834\udd1e']: 2764 l = len(s) 2765 self.assertEqual(unicode_asucs4(s, l, 1), s+'\0') 2766 self.assertEqual(unicode_asucs4(s, l, 0), s+'\uffff') 2767 self.assertEqual(unicode_asucs4(s, l+1, 1), s+'\0\uffff') 2768 self.assertEqual(unicode_asucs4(s, l+1, 0), s+'\0\uffff') 2769 self.assertRaises(SystemError, unicode_asucs4, s, l-1, 1) 2770 self.assertRaises(SystemError, unicode_asucs4, s, l-2, 0) 2771 s = '\0'.join([s, s]) 2772 self.assertEqual(unicode_asucs4(s, len(s), 1), s+'\0') 2773 self.assertEqual(unicode_asucs4(s, len(s), 0), s+'\uffff') 2774 2775 # Test PyUnicode_FindChar() 2776 @support.cpython_only 2777 def test_findchar(self): 2778 from _testcapi import unicode_findchar 2779 2780 for str in "\xa1", "\u8000\u8080", "\ud800\udc02", "\U0001f100\U0001f1f1": 2781 for i, ch in enumerate(str): 2782 self.assertEqual(unicode_findchar(str, ord(ch), 0, len(str), 1), i) 2783 self.assertEqual(unicode_findchar(str, ord(ch), 0, len(str), -1), i) 2784 2785 str = "!>_<!" 2786 self.assertEqual(unicode_findchar(str, 0x110000, 0, len(str), 1), -1) 2787 self.assertEqual(unicode_findchar(str, 0x110000, 0, len(str), -1), -1) 2788 # start < end 2789 self.assertEqual(unicode_findchar(str, ord('!'), 1, len(str)+1, 1), 4) 2790 self.assertEqual(unicode_findchar(str, ord('!'), 1, len(str)+1, -1), 4) 2791 # start >= end 2792 self.assertEqual(unicode_findchar(str, ord('!'), 0, 0, 1), -1) 2793 self.assertEqual(unicode_findchar(str, ord('!'), len(str), 0, 1), -1) 2794 # negative 2795 self.assertEqual(unicode_findchar(str, ord('!'), -len(str), -1, 1), 0) 2796 self.assertEqual(unicode_findchar(str, ord('!'), -len(str), -1, -1), 0) 2797 2798 # Test PyUnicode_CopyCharacters() 2799 @support.cpython_only 2800 def test_copycharacters(self): 2801 from _testcapi import unicode_copycharacters 2802 2803 strings = [ 2804 'abcde', '\xa1\xa2\xa3\xa4\xa5', 2805 '\u4f60\u597d\u4e16\u754c\uff01', 2806 '\U0001f600\U0001f601\U0001f602\U0001f603\U0001f604' 2807 ] 2808 2809 for idx, from_ in enumerate(strings): 2810 # wide -> narrow: exceed maxchar limitation 2811 for to in strings[:idx]: 2812 self.assertRaises( 2813 SystemError, 2814 unicode_copycharacters, to, 0, from_, 0, 5 2815 ) 2816 # same kind 2817 for from_start in range(5): 2818 self.assertEqual( 2819 unicode_copycharacters(from_, 0, from_, from_start, 5), 2820 (from_[from_start:from_start+5].ljust(5, '\0'), 2821 5-from_start) 2822 ) 2823 for to_start in range(5): 2824 self.assertEqual( 2825 unicode_copycharacters(from_, to_start, from_, to_start, 5), 2826 (from_[to_start:to_start+5].rjust(5, '\0'), 2827 5-to_start) 2828 ) 2829 # narrow -> wide 2830 # Tests omitted since this creates invalid strings. 2831 2832 s = strings[0] 2833 self.assertRaises(IndexError, unicode_copycharacters, s, 6, s, 0, 5) 2834 self.assertRaises(IndexError, unicode_copycharacters, s, -1, s, 0, 5) 2835 self.assertRaises(IndexError, unicode_copycharacters, s, 0, s, 6, 5) 2836 self.assertRaises(IndexError, unicode_copycharacters, s, 0, s, -1, 5) 2837 self.assertRaises(SystemError, unicode_copycharacters, s, 1, s, 0, 5) 2838 self.assertRaises(SystemError, unicode_copycharacters, s, 0, s, 0, -1) 2839 self.assertRaises(SystemError, unicode_copycharacters, s, 0, b'', 0, 0) 2840 2841 @support.cpython_only 2842 def test_encode_decimal(self): 2843 from _testcapi import unicode_encodedecimal 2844 self.assertEqual(unicode_encodedecimal('123'), 2845 b'123') 2846 self.assertEqual(unicode_encodedecimal('\u0663.\u0661\u0664'), 2847 b'3.14') 2848 self.assertEqual(unicode_encodedecimal("\N{EM SPACE}3.14\N{EN SPACE}"), 2849 b' 3.14 ') 2850 self.assertRaises(UnicodeEncodeError, 2851 unicode_encodedecimal, "123\u20ac", "strict") 2852 self.assertRaisesRegex( 2853 ValueError, 2854 "^'decimal' codec can't encode character", 2855 unicode_encodedecimal, "123\u20ac", "replace") 2856 2857 @support.cpython_only 2858 def test_transform_decimal(self): 2859 from _testcapi import unicode_transformdecimaltoascii as transform_decimal 2860 self.assertEqual(transform_decimal('123'), 2861 '123') 2862 self.assertEqual(transform_decimal('\u0663.\u0661\u0664'), 2863 '3.14') 2864 self.assertEqual(transform_decimal("\N{EM SPACE}3.14\N{EN SPACE}"), 2865 "\N{EM SPACE}3.14\N{EN SPACE}") 2866 self.assertEqual(transform_decimal('123\u20ac'), 2867 '123\u20ac') 2868 2869 @support.cpython_only 2870 def test_pep393_utf8_caching_bug(self): 2871 # Issue #25709: Problem with string concatenation and utf-8 cache 2872 from _testcapi import getargs_s_hash 2873 for k in 0x24, 0xa4, 0x20ac, 0x1f40d: 2874 s = '' 2875 for i in range(5): 2876 # Due to CPython specific optimization the 's' string can be 2877 # resized in-place. 2878 s += chr(k) 2879 # Parsing with the "s#" format code calls indirectly 2880 # PyUnicode_AsUTF8AndSize() which creates the UTF-8 2881 # encoded string cached in the Unicode object. 2882 self.assertEqual(getargs_s_hash(s), chr(k).encode() * (i + 1)) 2883 # Check that the second call returns the same result 2884 self.assertEqual(getargs_s_hash(s), chr(k).encode() * (i + 1)) 2885 2886class StringModuleTest(unittest.TestCase): 2887 def test_formatter_parser(self): 2888 def parse(format): 2889 return list(_string.formatter_parser(format)) 2890 2891 formatter = parse("prefix {2!s}xxx{0:^+10.3f}{obj.attr!s} {z[0]!s:10}") 2892 self.assertEqual(formatter, [ 2893 ('prefix ', '2', '', 's'), 2894 ('xxx', '0', '^+10.3f', None), 2895 ('', 'obj.attr', '', 's'), 2896 (' ', 'z[0]', '10', 's'), 2897 ]) 2898 2899 formatter = parse("prefix {} suffix") 2900 self.assertEqual(formatter, [ 2901 ('prefix ', '', '', None), 2902 (' suffix', None, None, None), 2903 ]) 2904 2905 formatter = parse("str") 2906 self.assertEqual(formatter, [ 2907 ('str', None, None, None), 2908 ]) 2909 2910 formatter = parse("") 2911 self.assertEqual(formatter, []) 2912 2913 formatter = parse("{0}") 2914 self.assertEqual(formatter, [ 2915 ('', '0', '', None), 2916 ]) 2917 2918 self.assertRaises(TypeError, _string.formatter_parser, 1) 2919 2920 def test_formatter_field_name_split(self): 2921 def split(name): 2922 items = list(_string.formatter_field_name_split(name)) 2923 items[1] = list(items[1]) 2924 return items 2925 self.assertEqual(split("obj"), ["obj", []]) 2926 self.assertEqual(split("obj.arg"), ["obj", [(True, 'arg')]]) 2927 self.assertEqual(split("obj[key]"), ["obj", [(False, 'key')]]) 2928 self.assertEqual(split("obj.arg[key1][key2]"), [ 2929 "obj", 2930 [(True, 'arg'), 2931 (False, 'key1'), 2932 (False, 'key2'), 2933 ]]) 2934 self.assertRaises(TypeError, _string.formatter_field_name_split, 1) 2935 2936 2937if __name__ == "__main__": 2938 unittest.main() 2939