1# Copyright (C) 2001,2002 Python Software Foundation 2# csv package unit tests 3 4import copy 5import sys 6import unittest 7from io import StringIO 8from tempfile import TemporaryFile 9import csv 10import gc 11import pickle 12from test import support 13from test.support import warnings_helper 14from itertools import permutations 15from textwrap import dedent 16from collections import OrderedDict 17 18 19class BadIterable: 20 def __iter__(self): 21 raise OSError 22 23 24class Test_Csv(unittest.TestCase): 25 """ 26 Test the underlying C csv parser in ways that are not appropriate 27 from the high level interface. Further tests of this nature are done 28 in TestDialectRegistry. 29 """ 30 def _test_arg_valid(self, ctor, arg): 31 self.assertRaises(TypeError, ctor) 32 self.assertRaises(TypeError, ctor, None) 33 self.assertRaises(TypeError, ctor, arg, bad_attr = 0) 34 self.assertRaises(TypeError, ctor, arg, delimiter = 0) 35 self.assertRaises(TypeError, ctor, arg, delimiter = 'XX') 36 self.assertRaises(csv.Error, ctor, arg, 'foo') 37 self.assertRaises(TypeError, ctor, arg, delimiter=None) 38 self.assertRaises(TypeError, ctor, arg, delimiter=1) 39 self.assertRaises(TypeError, ctor, arg, quotechar=1) 40 self.assertRaises(TypeError, ctor, arg, lineterminator=None) 41 self.assertRaises(TypeError, ctor, arg, lineterminator=1) 42 self.assertRaises(TypeError, ctor, arg, quoting=None) 43 self.assertRaises(TypeError, ctor, arg, 44 quoting=csv.QUOTE_ALL, quotechar='') 45 self.assertRaises(TypeError, ctor, arg, 46 quoting=csv.QUOTE_ALL, quotechar=None) 47 self.assertRaises(TypeError, ctor, arg, 48 quoting=csv.QUOTE_NONE, quotechar='') 49 50 def test_reader_arg_valid(self): 51 self._test_arg_valid(csv.reader, []) 52 self.assertRaises(OSError, csv.reader, BadIterable()) 53 54 def test_writer_arg_valid(self): 55 self._test_arg_valid(csv.writer, StringIO()) 56 class BadWriter: 57 @property 58 def write(self): 59 raise OSError 60 self.assertRaises(OSError, csv.writer, BadWriter()) 61 62 def _test_default_attrs(self, ctor, *args): 63 obj = ctor(*args) 64 # Check defaults 65 self.assertEqual(obj.dialect.delimiter, ',') 66 self.assertIs(obj.dialect.doublequote, True) 67 self.assertEqual(obj.dialect.escapechar, None) 68 self.assertEqual(obj.dialect.lineterminator, "\r\n") 69 self.assertEqual(obj.dialect.quotechar, '"') 70 self.assertEqual(obj.dialect.quoting, csv.QUOTE_MINIMAL) 71 self.assertIs(obj.dialect.skipinitialspace, False) 72 self.assertIs(obj.dialect.strict, False) 73 # Try deleting or changing attributes (they are read-only) 74 self.assertRaises(AttributeError, delattr, obj.dialect, 'delimiter') 75 self.assertRaises(AttributeError, setattr, obj.dialect, 'delimiter', ':') 76 self.assertRaises(AttributeError, delattr, obj.dialect, 'quoting') 77 self.assertRaises(AttributeError, setattr, obj.dialect, 78 'quoting', None) 79 80 def test_reader_attrs(self): 81 self._test_default_attrs(csv.reader, []) 82 83 def test_writer_attrs(self): 84 self._test_default_attrs(csv.writer, StringIO()) 85 86 def _test_kw_attrs(self, ctor, *args): 87 # Now try with alternate options 88 kwargs = dict(delimiter=':', doublequote=False, escapechar='\\', 89 lineterminator='\r', quotechar='*', 90 quoting=csv.QUOTE_NONE, skipinitialspace=True, 91 strict=True) 92 obj = ctor(*args, **kwargs) 93 self.assertEqual(obj.dialect.delimiter, ':') 94 self.assertIs(obj.dialect.doublequote, False) 95 self.assertEqual(obj.dialect.escapechar, '\\') 96 self.assertEqual(obj.dialect.lineterminator, "\r") 97 self.assertEqual(obj.dialect.quotechar, '*') 98 self.assertEqual(obj.dialect.quoting, csv.QUOTE_NONE) 99 self.assertIs(obj.dialect.skipinitialspace, True) 100 self.assertIs(obj.dialect.strict, True) 101 102 def test_reader_kw_attrs(self): 103 self._test_kw_attrs(csv.reader, []) 104 105 def test_writer_kw_attrs(self): 106 self._test_kw_attrs(csv.writer, StringIO()) 107 108 def _test_dialect_attrs(self, ctor, *args): 109 # Now try with dialect-derived options 110 class dialect: 111 delimiter='-' 112 doublequote=False 113 escapechar='^' 114 lineterminator='$' 115 quotechar='#' 116 quoting=csv.QUOTE_ALL 117 skipinitialspace=True 118 strict=False 119 args = args + (dialect,) 120 obj = ctor(*args) 121 self.assertEqual(obj.dialect.delimiter, '-') 122 self.assertIs(obj.dialect.doublequote, False) 123 self.assertEqual(obj.dialect.escapechar, '^') 124 self.assertEqual(obj.dialect.lineterminator, "$") 125 self.assertEqual(obj.dialect.quotechar, '#') 126 self.assertEqual(obj.dialect.quoting, csv.QUOTE_ALL) 127 self.assertIs(obj.dialect.skipinitialspace, True) 128 self.assertIs(obj.dialect.strict, False) 129 130 def test_reader_dialect_attrs(self): 131 self._test_dialect_attrs(csv.reader, []) 132 133 def test_writer_dialect_attrs(self): 134 self._test_dialect_attrs(csv.writer, StringIO()) 135 136 137 def _write_test(self, fields, expect, **kwargs): 138 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 139 writer = csv.writer(fileobj, **kwargs) 140 writer.writerow(fields) 141 fileobj.seek(0) 142 self.assertEqual(fileobj.read(), 143 expect + writer.dialect.lineterminator) 144 145 def _write_error_test(self, exc, fields, **kwargs): 146 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 147 writer = csv.writer(fileobj, **kwargs) 148 with self.assertRaises(exc): 149 writer.writerow(fields) 150 fileobj.seek(0) 151 self.assertEqual(fileobj.read(), '') 152 153 def test_write_arg_valid(self): 154 self._write_error_test(csv.Error, None) 155 self._write_test((), '') 156 self._write_test([None], '""') 157 self._write_error_test(csv.Error, [None], quoting = csv.QUOTE_NONE) 158 # Check that exceptions are passed up the chain 159 self._write_error_test(OSError, BadIterable()) 160 class BadList: 161 def __len__(self): 162 return 10 163 def __getitem__(self, i): 164 if i > 2: 165 raise OSError 166 self._write_error_test(OSError, BadList()) 167 class BadItem: 168 def __str__(self): 169 raise OSError 170 self._write_error_test(OSError, [BadItem()]) 171 172 def test_write_bigfield(self): 173 # This exercises the buffer realloc functionality 174 bigstring = 'X' * 50000 175 self._write_test([bigstring,bigstring], '%s,%s' % \ 176 (bigstring, bigstring)) 177 178 def test_write_quoting(self): 179 self._write_test(['a',1,'p,q'], 'a,1,"p,q"') 180 self._write_error_test(csv.Error, ['a',1,'p,q'], 181 quoting = csv.QUOTE_NONE) 182 self._write_test(['a',1,'p,q'], 'a,1,"p,q"', 183 quoting = csv.QUOTE_MINIMAL) 184 self._write_test(['a',1,'p,q'], '"a",1,"p,q"', 185 quoting = csv.QUOTE_NONNUMERIC) 186 self._write_test(['a',1,'p,q'], '"a","1","p,q"', 187 quoting = csv.QUOTE_ALL) 188 self._write_test(['a\nb',1], '"a\nb","1"', 189 quoting = csv.QUOTE_ALL) 190 191 def test_write_escape(self): 192 self._write_test(['a',1,'p,q'], 'a,1,"p,q"', 193 escapechar='\\') 194 self._write_error_test(csv.Error, ['a',1,'p,"q"'], 195 escapechar=None, doublequote=False) 196 self._write_test(['a',1,'p,"q"'], 'a,1,"p,\\"q\\""', 197 escapechar='\\', doublequote = False) 198 self._write_test(['"'], '""""', 199 escapechar='\\', quoting = csv.QUOTE_MINIMAL) 200 self._write_test(['"'], '\\"', 201 escapechar='\\', quoting = csv.QUOTE_MINIMAL, 202 doublequote = False) 203 self._write_test(['"'], '\\"', 204 escapechar='\\', quoting = csv.QUOTE_NONE) 205 self._write_test(['a',1,'p,q'], 'a,1,p\\,q', 206 escapechar='\\', quoting = csv.QUOTE_NONE) 207 self._write_test(['\\', 'a'], '\\\\,a', 208 escapechar='\\', quoting=csv.QUOTE_NONE) 209 self._write_test(['\\', 'a'], '\\\\,a', 210 escapechar='\\', quoting=csv.QUOTE_MINIMAL) 211 self._write_test(['\\', 'a'], '"\\\\","a"', 212 escapechar='\\', quoting=csv.QUOTE_ALL) 213 self._write_test(['\\ ', 'a'], '\\\\ ,a', 214 escapechar='\\', quoting=csv.QUOTE_MINIMAL) 215 self._write_test(['\\,', 'a'], '\\\\\\,,a', 216 escapechar='\\', quoting=csv.QUOTE_NONE) 217 self._write_test([',\\', 'a'], '",\\\\",a', 218 escapechar='\\', quoting=csv.QUOTE_MINIMAL) 219 self._write_test(['C\\', '6', '7', 'X"'], 'C\\\\,6,7,"X"""', 220 escapechar='\\', quoting=csv.QUOTE_MINIMAL) 221 222 def test_write_lineterminator(self): 223 for lineterminator in '\r\n', '\n', '\r', '!@#', '\0': 224 with self.subTest(lineterminator=lineterminator): 225 with StringIO() as sio: 226 writer = csv.writer(sio, lineterminator=lineterminator) 227 writer.writerow(['a', 'b']) 228 writer.writerow([1, 2]) 229 self.assertEqual(sio.getvalue(), 230 f'a,b{lineterminator}' 231 f'1,2{lineterminator}') 232 233 def test_write_iterable(self): 234 self._write_test(iter(['a', 1, 'p,q']), 'a,1,"p,q"') 235 self._write_test(iter(['a', 1, None]), 'a,1,') 236 self._write_test(iter([]), '') 237 self._write_test(iter([None]), '""') 238 self._write_error_test(csv.Error, iter([None]), quoting=csv.QUOTE_NONE) 239 self._write_test(iter([None, None]), ',') 240 241 def test_writerows(self): 242 class BrokenFile: 243 def write(self, buf): 244 raise OSError 245 writer = csv.writer(BrokenFile()) 246 self.assertRaises(OSError, writer.writerows, [['a']]) 247 248 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 249 writer = csv.writer(fileobj) 250 self.assertRaises(TypeError, writer.writerows, None) 251 writer.writerows([['a', 'b'], ['c', 'd']]) 252 fileobj.seek(0) 253 self.assertEqual(fileobj.read(), "a,b\r\nc,d\r\n") 254 255 def test_writerows_with_none(self): 256 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 257 writer = csv.writer(fileobj) 258 writer.writerows([['a', None], [None, 'd']]) 259 fileobj.seek(0) 260 self.assertEqual(fileobj.read(), "a,\r\n,d\r\n") 261 262 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 263 writer = csv.writer(fileobj) 264 writer.writerows([[None], ['a']]) 265 fileobj.seek(0) 266 self.assertEqual(fileobj.read(), '""\r\na\r\n') 267 268 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 269 writer = csv.writer(fileobj) 270 writer.writerows([['a'], [None]]) 271 fileobj.seek(0) 272 self.assertEqual(fileobj.read(), 'a\r\n""\r\n') 273 274 def test_writerows_errors(self): 275 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 276 writer = csv.writer(fileobj) 277 self.assertRaises(TypeError, writer.writerows, None) 278 self.assertRaises(OSError, writer.writerows, BadIterable()) 279 280 @support.cpython_only 281 @support.requires_legacy_unicode_capi 282 @warnings_helper.ignore_warnings(category=DeprecationWarning) 283 def test_writerows_legacy_strings(self): 284 import _testcapi 285 c = _testcapi.unicode_legacy_string('a') 286 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 287 writer = csv.writer(fileobj) 288 writer.writerows([[c]]) 289 fileobj.seek(0) 290 self.assertEqual(fileobj.read(), "a\r\n") 291 292 def _read_test(self, input, expect, **kwargs): 293 reader = csv.reader(input, **kwargs) 294 result = list(reader) 295 self.assertEqual(result, expect) 296 297 def test_read_oddinputs(self): 298 self._read_test([], []) 299 self._read_test([''], [[]]) 300 self.assertRaises(csv.Error, self._read_test, 301 ['"ab"c'], None, strict = 1) 302 self._read_test(['"ab"c'], [['abc']], doublequote = 0) 303 304 self.assertRaises(csv.Error, self._read_test, 305 [b'abc'], None) 306 307 def test_read_eol(self): 308 self._read_test(['a,b'], [['a','b']]) 309 self._read_test(['a,b\n'], [['a','b']]) 310 self._read_test(['a,b\r\n'], [['a','b']]) 311 self._read_test(['a,b\r'], [['a','b']]) 312 self.assertRaises(csv.Error, self._read_test, ['a,b\rc,d'], []) 313 self.assertRaises(csv.Error, self._read_test, ['a,b\nc,d'], []) 314 self.assertRaises(csv.Error, self._read_test, ['a,b\r\nc,d'], []) 315 316 def test_read_eof(self): 317 self._read_test(['a,"'], [['a', '']]) 318 self._read_test(['"a'], [['a']]) 319 self._read_test(['^'], [['\n']], escapechar='^') 320 self.assertRaises(csv.Error, self._read_test, ['a,"'], [], strict=True) 321 self.assertRaises(csv.Error, self._read_test, ['"a'], [], strict=True) 322 self.assertRaises(csv.Error, self._read_test, 323 ['^'], [], escapechar='^', strict=True) 324 325 def test_read_nul(self): 326 self._read_test(['\0'], [['\0']]) 327 self._read_test(['a,\0b,c'], [['a', '\0b', 'c']]) 328 self._read_test(['a,b\0,c'], [['a', 'b\0', 'c']]) 329 self._read_test(['a,b\\\0,c'], [['a', 'b\0', 'c']], escapechar='\\') 330 self._read_test(['a,"\0b",c'], [['a', '\0b', 'c']]) 331 332 def test_read_delimiter(self): 333 self._read_test(['a,b,c'], [['a', 'b', 'c']]) 334 self._read_test(['a;b;c'], [['a', 'b', 'c']], delimiter=';') 335 self._read_test(['a\0b\0c'], [['a', 'b', 'c']], delimiter='\0') 336 337 def test_read_escape(self): 338 self._read_test(['a,\\b,c'], [['a', 'b', 'c']], escapechar='\\') 339 self._read_test(['a,b\\,c'], [['a', 'b,c']], escapechar='\\') 340 self._read_test(['a,"b\\,c"'], [['a', 'b,c']], escapechar='\\') 341 self._read_test(['a,"b,\\c"'], [['a', 'b,c']], escapechar='\\') 342 self._read_test(['a,"b,c\\""'], [['a', 'b,c"']], escapechar='\\') 343 self._read_test(['a,"b,c"\\'], [['a', 'b,c\\']], escapechar='\\') 344 self._read_test(['a,^b,c'], [['a', 'b', 'c']], escapechar='^') 345 self._read_test(['a,\0b,c'], [['a', 'b', 'c']], escapechar='\0') 346 self._read_test(['a,\\b,c'], [['a', '\\b', 'c']], escapechar=None) 347 self._read_test(['a,\\b,c'], [['a', '\\b', 'c']]) 348 349 def test_read_quoting(self): 350 self._read_test(['1,",3,",5'], [['1', ',3,', '5']]) 351 self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']], 352 quotechar=None, escapechar='\\') 353 self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']], 354 quoting=csv.QUOTE_NONE, escapechar='\\') 355 # will this fail where locale uses comma for decimals? 356 self._read_test([',3,"5",7.3, 9'], [['', 3, '5', 7.3, 9]], 357 quoting=csv.QUOTE_NONNUMERIC) 358 self._read_test(['"a\nb", 7'], [['a\nb', ' 7']]) 359 self.assertRaises(ValueError, self._read_test, 360 ['abc,3'], [[]], 361 quoting=csv.QUOTE_NONNUMERIC) 362 self._read_test(['1,@,3,@,5'], [['1', ',3,', '5']], quotechar='@') 363 self._read_test(['1,\0,3,\0,5'], [['1', ',3,', '5']], quotechar='\0') 364 365 def test_read_bigfield(self): 366 # This exercises the buffer realloc functionality and field size 367 # limits. 368 limit = csv.field_size_limit() 369 try: 370 size = 50000 371 bigstring = 'X' * size 372 bigline = '%s,%s' % (bigstring, bigstring) 373 self._read_test([bigline], [[bigstring, bigstring]]) 374 csv.field_size_limit(size) 375 self._read_test([bigline], [[bigstring, bigstring]]) 376 self.assertEqual(csv.field_size_limit(), size) 377 csv.field_size_limit(size-1) 378 self.assertRaises(csv.Error, self._read_test, [bigline], []) 379 self.assertRaises(TypeError, csv.field_size_limit, None) 380 self.assertRaises(TypeError, csv.field_size_limit, 1, None) 381 finally: 382 csv.field_size_limit(limit) 383 384 def test_read_linenum(self): 385 r = csv.reader(['line,1', 'line,2', 'line,3']) 386 self.assertEqual(r.line_num, 0) 387 next(r) 388 self.assertEqual(r.line_num, 1) 389 next(r) 390 self.assertEqual(r.line_num, 2) 391 next(r) 392 self.assertEqual(r.line_num, 3) 393 self.assertRaises(StopIteration, next, r) 394 self.assertEqual(r.line_num, 3) 395 396 def test_roundtrip_quoteed_newlines(self): 397 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 398 writer = csv.writer(fileobj) 399 rows = [['a\nb','b'],['c','x\r\nd']] 400 writer.writerows(rows) 401 fileobj.seek(0) 402 for i, row in enumerate(csv.reader(fileobj)): 403 self.assertEqual(row, rows[i]) 404 405 def test_roundtrip_escaped_unquoted_newlines(self): 406 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 407 writer = csv.writer(fileobj,quoting=csv.QUOTE_NONE,escapechar="\\") 408 rows = [['a\nb','b'],['c','x\r\nd']] 409 writer.writerows(rows) 410 fileobj.seek(0) 411 for i, row in enumerate(csv.reader(fileobj,quoting=csv.QUOTE_NONE,escapechar="\\")): 412 self.assertEqual(row,rows[i]) 413 414class TestDialectRegistry(unittest.TestCase): 415 def test_registry_badargs(self): 416 self.assertRaises(TypeError, csv.list_dialects, None) 417 self.assertRaises(TypeError, csv.get_dialect) 418 self.assertRaises(csv.Error, csv.get_dialect, None) 419 self.assertRaises(csv.Error, csv.get_dialect, "nonesuch") 420 self.assertRaises(TypeError, csv.unregister_dialect) 421 self.assertRaises(csv.Error, csv.unregister_dialect, None) 422 self.assertRaises(csv.Error, csv.unregister_dialect, "nonesuch") 423 self.assertRaises(TypeError, csv.register_dialect, None) 424 self.assertRaises(TypeError, csv.register_dialect, None, None) 425 self.assertRaises(TypeError, csv.register_dialect, "nonesuch", 0, 0) 426 self.assertRaises(TypeError, csv.register_dialect, "nonesuch", 427 badargument=None) 428 self.assertRaises(TypeError, csv.register_dialect, "nonesuch", 429 quoting=None) 430 self.assertRaises(TypeError, csv.register_dialect, []) 431 432 def test_registry(self): 433 class myexceltsv(csv.excel): 434 delimiter = "\t" 435 name = "myexceltsv" 436 expected_dialects = csv.list_dialects() + [name] 437 expected_dialects.sort() 438 csv.register_dialect(name, myexceltsv) 439 self.addCleanup(csv.unregister_dialect, name) 440 self.assertEqual(csv.get_dialect(name).delimiter, '\t') 441 got_dialects = sorted(csv.list_dialects()) 442 self.assertEqual(expected_dialects, got_dialects) 443 444 def test_register_kwargs(self): 445 name = 'fedcba' 446 csv.register_dialect(name, delimiter=';') 447 self.addCleanup(csv.unregister_dialect, name) 448 self.assertEqual(csv.get_dialect(name).delimiter, ';') 449 self.assertEqual([['X', 'Y', 'Z']], list(csv.reader(['X;Y;Z'], name))) 450 451 def test_incomplete_dialect(self): 452 class myexceltsv(csv.Dialect): 453 delimiter = "\t" 454 self.assertRaises(csv.Error, myexceltsv) 455 456 def test_space_dialect(self): 457 class space(csv.excel): 458 delimiter = " " 459 quoting = csv.QUOTE_NONE 460 escapechar = "\\" 461 462 with TemporaryFile("w+", encoding="utf-8") as fileobj: 463 fileobj.write("abc def\nc1ccccc1 benzene\n") 464 fileobj.seek(0) 465 reader = csv.reader(fileobj, dialect=space()) 466 self.assertEqual(next(reader), ["abc", "def"]) 467 self.assertEqual(next(reader), ["c1ccccc1", "benzene"]) 468 469 def compare_dialect_123(self, expected, *writeargs, **kwwriteargs): 470 471 with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: 472 473 writer = csv.writer(fileobj, *writeargs, **kwwriteargs) 474 writer.writerow([1,2,3]) 475 fileobj.seek(0) 476 self.assertEqual(fileobj.read(), expected) 477 478 def test_dialect_apply(self): 479 class testA(csv.excel): 480 delimiter = "\t" 481 class testB(csv.excel): 482 delimiter = ":" 483 class testC(csv.excel): 484 delimiter = "|" 485 class testUni(csv.excel): 486 delimiter = "\u039B" 487 488 class unspecified(): 489 # A class to pass as dialect but with no dialect attributes. 490 pass 491 492 csv.register_dialect('testC', testC) 493 try: 494 self.compare_dialect_123("1,2,3\r\n") 495 self.compare_dialect_123("1,2,3\r\n", dialect=None) 496 self.compare_dialect_123("1,2,3\r\n", dialect=unspecified) 497 self.compare_dialect_123("1\t2\t3\r\n", testA) 498 self.compare_dialect_123("1:2:3\r\n", dialect=testB()) 499 self.compare_dialect_123("1|2|3\r\n", dialect='testC') 500 self.compare_dialect_123("1;2;3\r\n", dialect=testA, 501 delimiter=';') 502 self.compare_dialect_123("1\u039B2\u039B3\r\n", 503 dialect=testUni) 504 505 finally: 506 csv.unregister_dialect('testC') 507 508 def test_bad_dialect(self): 509 # Unknown parameter 510 self.assertRaises(TypeError, csv.reader, [], bad_attr = 0) 511 # Bad values 512 self.assertRaises(TypeError, csv.reader, [], delimiter = None) 513 self.assertRaises(TypeError, csv.reader, [], quoting = -1) 514 self.assertRaises(TypeError, csv.reader, [], quoting = 100) 515 516 def test_copy(self): 517 for name in csv.list_dialects(): 518 dialect = csv.get_dialect(name) 519 self.assertRaises(TypeError, copy.copy, dialect) 520 521 def test_pickle(self): 522 for name in csv.list_dialects(): 523 dialect = csv.get_dialect(name) 524 for proto in range(pickle.HIGHEST_PROTOCOL + 1): 525 self.assertRaises(TypeError, pickle.dumps, dialect, proto) 526 527class TestCsvBase(unittest.TestCase): 528 def readerAssertEqual(self, input, expected_result): 529 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 530 fileobj.write(input) 531 fileobj.seek(0) 532 reader = csv.reader(fileobj, dialect = self.dialect) 533 fields = list(reader) 534 self.assertEqual(fields, expected_result) 535 536 def writerAssertEqual(self, input, expected_result): 537 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 538 writer = csv.writer(fileobj, dialect = self.dialect) 539 writer.writerows(input) 540 fileobj.seek(0) 541 self.assertEqual(fileobj.read(), expected_result) 542 543class TestDialectExcel(TestCsvBase): 544 dialect = 'excel' 545 546 def test_single(self): 547 self.readerAssertEqual('abc', [['abc']]) 548 549 def test_simple(self): 550 self.readerAssertEqual('1,2,3,4,5', [['1','2','3','4','5']]) 551 552 def test_blankline(self): 553 self.readerAssertEqual('', []) 554 555 def test_empty_fields(self): 556 self.readerAssertEqual(',', [['', '']]) 557 558 def test_singlequoted(self): 559 self.readerAssertEqual('""', [['']]) 560 561 def test_singlequoted_left_empty(self): 562 self.readerAssertEqual('"",', [['','']]) 563 564 def test_singlequoted_right_empty(self): 565 self.readerAssertEqual(',""', [['','']]) 566 567 def test_single_quoted_quote(self): 568 self.readerAssertEqual('""""', [['"']]) 569 570 def test_quoted_quotes(self): 571 self.readerAssertEqual('""""""', [['""']]) 572 573 def test_inline_quote(self): 574 self.readerAssertEqual('a""b', [['a""b']]) 575 576 def test_inline_quotes(self): 577 self.readerAssertEqual('a"b"c', [['a"b"c']]) 578 579 def test_quotes_and_more(self): 580 # Excel would never write a field containing '"a"b', but when 581 # reading one, it will return 'ab'. 582 self.readerAssertEqual('"a"b', [['ab']]) 583 584 def test_lone_quote(self): 585 self.readerAssertEqual('a"b', [['a"b']]) 586 587 def test_quote_and_quote(self): 588 # Excel would never write a field containing '"a" "b"', but when 589 # reading one, it will return 'a "b"'. 590 self.readerAssertEqual('"a" "b"', [['a "b"']]) 591 592 def test_space_and_quote(self): 593 self.readerAssertEqual(' "a"', [[' "a"']]) 594 595 def test_quoted(self): 596 self.readerAssertEqual('1,2,3,"I think, therefore I am",5,6', 597 [['1', '2', '3', 598 'I think, therefore I am', 599 '5', '6']]) 600 601 def test_quoted_quote(self): 602 self.readerAssertEqual('1,2,3,"""I see,"" said the blind man","as he picked up his hammer and saw"', 603 [['1', '2', '3', 604 '"I see," said the blind man', 605 'as he picked up his hammer and saw']]) 606 607 def test_quoted_nl(self): 608 input = '''\ 6091,2,3,"""I see,"" 610said the blind man","as he picked up his 611hammer and saw" 6129,8,7,6''' 613 self.readerAssertEqual(input, 614 [['1', '2', '3', 615 '"I see,"\nsaid the blind man', 616 'as he picked up his\nhammer and saw'], 617 ['9','8','7','6']]) 618 619 def test_dubious_quote(self): 620 self.readerAssertEqual('12,12,1",', [['12', '12', '1"', '']]) 621 622 def test_null(self): 623 self.writerAssertEqual([], '') 624 625 def test_single_writer(self): 626 self.writerAssertEqual([['abc']], 'abc\r\n') 627 628 def test_simple_writer(self): 629 self.writerAssertEqual([[1, 2, 'abc', 3, 4]], '1,2,abc,3,4\r\n') 630 631 def test_quotes(self): 632 self.writerAssertEqual([[1, 2, 'a"bc"', 3, 4]], '1,2,"a""bc""",3,4\r\n') 633 634 def test_quote_fieldsep(self): 635 self.writerAssertEqual([['abc,def']], '"abc,def"\r\n') 636 637 def test_newlines(self): 638 self.writerAssertEqual([[1, 2, 'a\nbc', 3, 4]], '1,2,"a\nbc",3,4\r\n') 639 640class EscapedExcel(csv.excel): 641 quoting = csv.QUOTE_NONE 642 escapechar = '\\' 643 644class TestEscapedExcel(TestCsvBase): 645 dialect = EscapedExcel() 646 647 def test_escape_fieldsep(self): 648 self.writerAssertEqual([['abc,def']], 'abc\\,def\r\n') 649 650 def test_read_escape_fieldsep(self): 651 self.readerAssertEqual('abc\\,def\r\n', [['abc,def']]) 652 653class TestDialectUnix(TestCsvBase): 654 dialect = 'unix' 655 656 def test_simple_writer(self): 657 self.writerAssertEqual([[1, 'abc def', 'abc']], '"1","abc def","abc"\n') 658 659 def test_simple_reader(self): 660 self.readerAssertEqual('"1","abc def","abc"\n', [['1', 'abc def', 'abc']]) 661 662class QuotedEscapedExcel(csv.excel): 663 quoting = csv.QUOTE_NONNUMERIC 664 escapechar = '\\' 665 666class TestQuotedEscapedExcel(TestCsvBase): 667 dialect = QuotedEscapedExcel() 668 669 def test_write_escape_fieldsep(self): 670 self.writerAssertEqual([['abc,def']], '"abc,def"\r\n') 671 672 def test_read_escape_fieldsep(self): 673 self.readerAssertEqual('"abc\\,def"\r\n', [['abc,def']]) 674 675class TestDictFields(unittest.TestCase): 676 ### "long" means the row is longer than the number of fieldnames 677 ### "short" means there are fewer elements in the row than fieldnames 678 def test_writeheader_return_value(self): 679 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 680 writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) 681 writeheader_return_value = writer.writeheader() 682 self.assertEqual(writeheader_return_value, 10) 683 684 def test_write_simple_dict(self): 685 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 686 writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) 687 writer.writeheader() 688 fileobj.seek(0) 689 self.assertEqual(fileobj.readline(), "f1,f2,f3\r\n") 690 writer.writerow({"f1": 10, "f3": "abc"}) 691 fileobj.seek(0) 692 fileobj.readline() # header 693 self.assertEqual(fileobj.read(), "10,,abc\r\n") 694 695 def test_write_multiple_dict_rows(self): 696 fileobj = StringIO() 697 writer = csv.DictWriter(fileobj, fieldnames=["f1", "f2", "f3"]) 698 writer.writeheader() 699 self.assertEqual(fileobj.getvalue(), "f1,f2,f3\r\n") 700 writer.writerows([{"f1": 1, "f2": "abc", "f3": "f"}, 701 {"f1": 2, "f2": 5, "f3": "xyz"}]) 702 self.assertEqual(fileobj.getvalue(), 703 "f1,f2,f3\r\n1,abc,f\r\n2,5,xyz\r\n") 704 705 def test_write_no_fields(self): 706 fileobj = StringIO() 707 self.assertRaises(TypeError, csv.DictWriter, fileobj) 708 709 def test_write_fields_not_in_fieldnames(self): 710 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 711 writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) 712 # Of special note is the non-string key (issue 19449) 713 with self.assertRaises(ValueError) as cx: 714 writer.writerow({"f4": 10, "f2": "spam", 1: "abc"}) 715 exception = str(cx.exception) 716 self.assertIn("fieldnames", exception) 717 self.assertIn("'f4'", exception) 718 self.assertNotIn("'f2'", exception) 719 self.assertIn("1", exception) 720 721 def test_typo_in_extrasaction_raises_error(self): 722 fileobj = StringIO() 723 self.assertRaises(ValueError, csv.DictWriter, fileobj, ['f1', 'f2'], 724 extrasaction="raised") 725 726 def test_write_field_not_in_field_names_raise(self): 727 fileobj = StringIO() 728 writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="raise") 729 dictrow = {'f0': 0, 'f1': 1, 'f2': 2, 'f3': 3} 730 self.assertRaises(ValueError, csv.DictWriter.writerow, writer, dictrow) 731 732 def test_write_field_not_in_field_names_ignore(self): 733 fileobj = StringIO() 734 writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="ignore") 735 dictrow = {'f0': 0, 'f1': 1, 'f2': 2, 'f3': 3} 736 csv.DictWriter.writerow(writer, dictrow) 737 self.assertEqual(fileobj.getvalue(), "1,2\r\n") 738 739 def test_read_dict_fields(self): 740 with TemporaryFile("w+", encoding="utf-8") as fileobj: 741 fileobj.write("1,2,abc\r\n") 742 fileobj.seek(0) 743 reader = csv.DictReader(fileobj, 744 fieldnames=["f1", "f2", "f3"]) 745 self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) 746 747 def test_read_dict_no_fieldnames(self): 748 with TemporaryFile("w+", encoding="utf-8") as fileobj: 749 fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") 750 fileobj.seek(0) 751 reader = csv.DictReader(fileobj) 752 self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) 753 self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) 754 755 # Two test cases to make sure existing ways of implicitly setting 756 # fieldnames continue to work. Both arise from discussion in issue3436. 757 def test_read_dict_fieldnames_from_file(self): 758 with TemporaryFile("w+", encoding="utf-8") as fileobj: 759 fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") 760 fileobj.seek(0) 761 reader = csv.DictReader(fileobj, 762 fieldnames=next(csv.reader(fileobj))) 763 self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) 764 self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) 765 766 def test_read_dict_fieldnames_chain(self): 767 import itertools 768 with TemporaryFile("w+", encoding="utf-8") as fileobj: 769 fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") 770 fileobj.seek(0) 771 reader = csv.DictReader(fileobj) 772 first = next(reader) 773 for row in itertools.chain([first], reader): 774 self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) 775 self.assertEqual(row, {"f1": '1', "f2": '2', "f3": 'abc'}) 776 777 def test_read_long(self): 778 with TemporaryFile("w+", encoding="utf-8") as fileobj: 779 fileobj.write("1,2,abc,4,5,6\r\n") 780 fileobj.seek(0) 781 reader = csv.DictReader(fileobj, 782 fieldnames=["f1", "f2"]) 783 self.assertEqual(next(reader), {"f1": '1', "f2": '2', 784 None: ["abc", "4", "5", "6"]}) 785 786 def test_read_long_with_rest(self): 787 with TemporaryFile("w+", encoding="utf-8") as fileobj: 788 fileobj.write("1,2,abc,4,5,6\r\n") 789 fileobj.seek(0) 790 reader = csv.DictReader(fileobj, 791 fieldnames=["f1", "f2"], restkey="_rest") 792 self.assertEqual(next(reader), {"f1": '1', "f2": '2', 793 "_rest": ["abc", "4", "5", "6"]}) 794 795 def test_read_long_with_rest_no_fieldnames(self): 796 with TemporaryFile("w+", encoding="utf-8") as fileobj: 797 fileobj.write("f1,f2\r\n1,2,abc,4,5,6\r\n") 798 fileobj.seek(0) 799 reader = csv.DictReader(fileobj, restkey="_rest") 800 self.assertEqual(reader.fieldnames, ["f1", "f2"]) 801 self.assertEqual(next(reader), {"f1": '1', "f2": '2', 802 "_rest": ["abc", "4", "5", "6"]}) 803 804 def test_read_short(self): 805 with TemporaryFile("w+", encoding="utf-8") as fileobj: 806 fileobj.write("1,2,abc,4,5,6\r\n1,2,abc\r\n") 807 fileobj.seek(0) 808 reader = csv.DictReader(fileobj, 809 fieldnames="1 2 3 4 5 6".split(), 810 restval="DEFAULT") 811 self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', 812 "4": '4', "5": '5', "6": '6'}) 813 self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', 814 "4": 'DEFAULT', "5": 'DEFAULT', 815 "6": 'DEFAULT'}) 816 817 def test_read_multi(self): 818 sample = [ 819 '2147483648,43.0e12,17,abc,def\r\n', 820 '147483648,43.0e2,17,abc,def\r\n', 821 '47483648,43.0,170,abc,def\r\n' 822 ] 823 824 reader = csv.DictReader(sample, 825 fieldnames="i1 float i2 s1 s2".split()) 826 self.assertEqual(next(reader), {"i1": '2147483648', 827 "float": '43.0e12', 828 "i2": '17', 829 "s1": 'abc', 830 "s2": 'def'}) 831 832 def test_read_with_blanks(self): 833 reader = csv.DictReader(["1,2,abc,4,5,6\r\n","\r\n", 834 "1,2,abc,4,5,6\r\n"], 835 fieldnames="1 2 3 4 5 6".split()) 836 self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', 837 "4": '4', "5": '5', "6": '6'}) 838 self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', 839 "4": '4', "5": '5', "6": '6'}) 840 841 def test_read_semi_sep(self): 842 reader = csv.DictReader(["1;2;abc;4;5;6\r\n"], 843 fieldnames="1 2 3 4 5 6".split(), 844 delimiter=';') 845 self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', 846 "4": '4', "5": '5', "6": '6'}) 847 848class TestArrayWrites(unittest.TestCase): 849 def test_int_write(self): 850 import array 851 contents = [(20-i) for i in range(20)] 852 a = array.array('i', contents) 853 854 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 855 writer = csv.writer(fileobj, dialect="excel") 856 writer.writerow(a) 857 expected = ",".join([str(i) for i in a])+"\r\n" 858 fileobj.seek(0) 859 self.assertEqual(fileobj.read(), expected) 860 861 def test_double_write(self): 862 import array 863 contents = [(20-i)*0.1 for i in range(20)] 864 a = array.array('d', contents) 865 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 866 writer = csv.writer(fileobj, dialect="excel") 867 writer.writerow(a) 868 expected = ",".join([str(i) for i in a])+"\r\n" 869 fileobj.seek(0) 870 self.assertEqual(fileobj.read(), expected) 871 872 def test_float_write(self): 873 import array 874 contents = [(20-i)*0.1 for i in range(20)] 875 a = array.array('f', contents) 876 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 877 writer = csv.writer(fileobj, dialect="excel") 878 writer.writerow(a) 879 expected = ",".join([str(i) for i in a])+"\r\n" 880 fileobj.seek(0) 881 self.assertEqual(fileobj.read(), expected) 882 883 def test_char_write(self): 884 import array, string 885 a = array.array('u', string.ascii_letters) 886 887 with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: 888 writer = csv.writer(fileobj, dialect="excel") 889 writer.writerow(a) 890 expected = ",".join(a)+"\r\n" 891 fileobj.seek(0) 892 self.assertEqual(fileobj.read(), expected) 893 894class TestDialectValidity(unittest.TestCase): 895 def test_quoting(self): 896 class mydialect(csv.Dialect): 897 delimiter = ";" 898 escapechar = '\\' 899 doublequote = False 900 skipinitialspace = True 901 lineterminator = '\r\n' 902 quoting = csv.QUOTE_NONE 903 d = mydialect() 904 self.assertEqual(d.quoting, csv.QUOTE_NONE) 905 906 mydialect.quoting = None 907 self.assertRaises(csv.Error, mydialect) 908 909 mydialect.doublequote = True 910 mydialect.quoting = csv.QUOTE_ALL 911 mydialect.quotechar = '"' 912 d = mydialect() 913 self.assertEqual(d.quoting, csv.QUOTE_ALL) 914 self.assertEqual(d.quotechar, '"') 915 self.assertTrue(d.doublequote) 916 917 mydialect.quotechar = "" 918 with self.assertRaises(csv.Error) as cm: 919 mydialect() 920 self.assertEqual(str(cm.exception), 921 '"quotechar" must be a 1-character string') 922 923 mydialect.quotechar = "''" 924 with self.assertRaises(csv.Error) as cm: 925 mydialect() 926 self.assertEqual(str(cm.exception), 927 '"quotechar" must be a 1-character string') 928 929 mydialect.quotechar = 4 930 with self.assertRaises(csv.Error) as cm: 931 mydialect() 932 self.assertEqual(str(cm.exception), 933 '"quotechar" must be string or None, not int') 934 935 def test_delimiter(self): 936 class mydialect(csv.Dialect): 937 delimiter = ";" 938 escapechar = '\\' 939 doublequote = False 940 skipinitialspace = True 941 lineterminator = '\r\n' 942 quoting = csv.QUOTE_NONE 943 d = mydialect() 944 self.assertEqual(d.delimiter, ";") 945 946 mydialect.delimiter = ":::" 947 with self.assertRaises(csv.Error) as cm: 948 mydialect() 949 self.assertEqual(str(cm.exception), 950 '"delimiter" must be a 1-character string') 951 952 mydialect.delimiter = "" 953 with self.assertRaises(csv.Error) as cm: 954 mydialect() 955 self.assertEqual(str(cm.exception), 956 '"delimiter" must be a 1-character string') 957 958 mydialect.delimiter = b"," 959 with self.assertRaises(csv.Error) as cm: 960 mydialect() 961 self.assertEqual(str(cm.exception), 962 '"delimiter" must be string, not bytes') 963 964 mydialect.delimiter = 4 965 with self.assertRaises(csv.Error) as cm: 966 mydialect() 967 self.assertEqual(str(cm.exception), 968 '"delimiter" must be string, not int') 969 970 mydialect.delimiter = None 971 with self.assertRaises(csv.Error) as cm: 972 mydialect() 973 self.assertEqual(str(cm.exception), 974 '"delimiter" must be string, not NoneType') 975 976 def test_escapechar(self): 977 class mydialect(csv.Dialect): 978 delimiter = ";" 979 escapechar = '\\' 980 doublequote = False 981 skipinitialspace = True 982 lineterminator = '\r\n' 983 quoting = csv.QUOTE_NONE 984 d = mydialect() 985 self.assertEqual(d.escapechar, "\\") 986 987 mydialect.escapechar = "" 988 with self.assertRaisesRegex(csv.Error, '"escapechar" must be a 1-character string'): 989 mydialect() 990 991 mydialect.escapechar = "**" 992 with self.assertRaisesRegex(csv.Error, '"escapechar" must be a 1-character string'): 993 mydialect() 994 995 mydialect.escapechar = b"*" 996 with self.assertRaisesRegex(csv.Error, '"escapechar" must be string or None, not bytes'): 997 mydialect() 998 999 mydialect.escapechar = 4 1000 with self.assertRaisesRegex(csv.Error, '"escapechar" must be string or None, not int'): 1001 mydialect() 1002 1003 def test_lineterminator(self): 1004 class mydialect(csv.Dialect): 1005 delimiter = ";" 1006 escapechar = '\\' 1007 doublequote = False 1008 skipinitialspace = True 1009 lineterminator = '\r\n' 1010 quoting = csv.QUOTE_NONE 1011 d = mydialect() 1012 self.assertEqual(d.lineterminator, '\r\n') 1013 1014 mydialect.lineterminator = ":::" 1015 d = mydialect() 1016 self.assertEqual(d.lineterminator, ":::") 1017 1018 mydialect.lineterminator = 4 1019 with self.assertRaises(csv.Error) as cm: 1020 mydialect() 1021 self.assertEqual(str(cm.exception), 1022 '"lineterminator" must be a string') 1023 1024 def test_invalid_chars(self): 1025 def create_invalid(field_name, value): 1026 class mydialect(csv.Dialect): 1027 pass 1028 setattr(mydialect, field_name, value) 1029 d = mydialect() 1030 1031 for field_name in ("delimiter", "escapechar", "quotechar"): 1032 with self.subTest(field_name=field_name): 1033 self.assertRaises(csv.Error, create_invalid, field_name, "") 1034 self.assertRaises(csv.Error, create_invalid, field_name, "abc") 1035 self.assertRaises(csv.Error, create_invalid, field_name, b'x') 1036 self.assertRaises(csv.Error, create_invalid, field_name, 5) 1037 1038 1039class TestSniffer(unittest.TestCase): 1040 sample1 = """\ 1041Harry's, Arlington Heights, IL, 2/1/03, Kimi Hayes 1042Shark City, Glendale Heights, IL, 12/28/02, Prezence 1043Tommy's Place, Blue Island, IL, 12/28/02, Blue Sunday/White Crow 1044Stonecutters Seafood and Chop House, Lemont, IL, 12/19/02, Week Back 1045""" 1046 sample2 = """\ 1047'Harry''s':'Arlington Heights':'IL':'2/1/03':'Kimi Hayes' 1048'Shark City':'Glendale Heights':'IL':'12/28/02':'Prezence' 1049'Tommy''s Place':'Blue Island':'IL':'12/28/02':'Blue Sunday/White Crow' 1050'Stonecutters ''Seafood'' and Chop House':'Lemont':'IL':'12/19/02':'Week Back' 1051""" 1052 header1 = '''\ 1053"venue","city","state","date","performers" 1054''' 1055 sample3 = '''\ 105605/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 105705/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 105805/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 1059''' 1060 1061 sample4 = '''\ 10622147483648;43.0e12;17;abc;def 1063147483648;43.0e2;17;abc;def 106447483648;43.0;170;abc;def 1065''' 1066 1067 sample5 = "aaa\tbbb\r\nAAA\t\r\nBBB\t\r\n" 1068 sample6 = "a|b|c\r\nd|e|f\r\n" 1069 sample7 = "'a'|'b'|'c'\r\n'd'|e|f\r\n" 1070 1071# Issue 18155: Use a delimiter that is a special char to regex: 1072 1073 header2 = '''\ 1074"venue"+"city"+"state"+"date"+"performers" 1075''' 1076 sample8 = """\ 1077Harry's+ Arlington Heights+ IL+ 2/1/03+ Kimi Hayes 1078Shark City+ Glendale Heights+ IL+ 12/28/02+ Prezence 1079Tommy's Place+ Blue Island+ IL+ 12/28/02+ Blue Sunday/White Crow 1080Stonecutters Seafood and Chop House+ Lemont+ IL+ 12/19/02+ Week Back 1081""" 1082 sample9 = """\ 1083'Harry''s'+ Arlington Heights'+ 'IL'+ '2/1/03'+ 'Kimi Hayes' 1084'Shark City'+ Glendale Heights'+' IL'+ '12/28/02'+ 'Prezence' 1085'Tommy''s Place'+ Blue Island'+ 'IL'+ '12/28/02'+ 'Blue Sunday/White Crow' 1086'Stonecutters ''Seafood'' and Chop House'+ 'Lemont'+ 'IL'+ '12/19/02'+ 'Week Back' 1087""" 1088 1089 sample10 = dedent(""" 1090 abc,def 1091 ghijkl,mno 1092 ghi,jkl 1093 """) 1094 1095 sample11 = dedent(""" 1096 abc,def 1097 ghijkl,mnop 1098 ghi,jkl 1099 """) 1100 1101 sample12 = dedent(""""time","forces" 1102 1,1.5 1103 0.5,5+0j 1104 0,0 1105 1+1j,6 1106 """) 1107 1108 sample13 = dedent(""""time","forces" 1109 0,0 1110 1,2 1111 a,b 1112 """) 1113 1114 sample14 = """\ 1115abc\0def 1116ghijkl\0mno 1117ghi\0jkl 1118""" 1119 1120 def test_issue43625(self): 1121 sniffer = csv.Sniffer() 1122 self.assertTrue(sniffer.has_header(self.sample12)) 1123 self.assertFalse(sniffer.has_header(self.sample13)) 1124 1125 def test_has_header_strings(self): 1126 "More to document existing (unexpected?) behavior than anything else." 1127 sniffer = csv.Sniffer() 1128 self.assertFalse(sniffer.has_header(self.sample10)) 1129 self.assertFalse(sniffer.has_header(self.sample11)) 1130 1131 def test_has_header(self): 1132 sniffer = csv.Sniffer() 1133 self.assertIs(sniffer.has_header(self.sample1), False) 1134 self.assertIs(sniffer.has_header(self.header1 + self.sample1), True) 1135 1136 def test_has_header_regex_special_delimiter(self): 1137 sniffer = csv.Sniffer() 1138 self.assertIs(sniffer.has_header(self.sample8), False) 1139 self.assertIs(sniffer.has_header(self.header2 + self.sample8), True) 1140 1141 def test_guess_quote_and_delimiter(self): 1142 sniffer = csv.Sniffer() 1143 for header in (";'123;4';", "'123;4';", ";'123;4'", "'123;4'"): 1144 with self.subTest(header): 1145 dialect = sniffer.sniff(header, ",;") 1146 self.assertEqual(dialect.delimiter, ';') 1147 self.assertEqual(dialect.quotechar, "'") 1148 self.assertIs(dialect.doublequote, False) 1149 self.assertIs(dialect.skipinitialspace, False) 1150 1151 def test_sniff(self): 1152 sniffer = csv.Sniffer() 1153 dialect = sniffer.sniff(self.sample1) 1154 self.assertEqual(dialect.delimiter, ",") 1155 self.assertEqual(dialect.quotechar, '"') 1156 self.assertIs(dialect.skipinitialspace, True) 1157 1158 dialect = sniffer.sniff(self.sample2) 1159 self.assertEqual(dialect.delimiter, ":") 1160 self.assertEqual(dialect.quotechar, "'") 1161 self.assertIs(dialect.skipinitialspace, False) 1162 1163 def test_delimiters(self): 1164 sniffer = csv.Sniffer() 1165 dialect = sniffer.sniff(self.sample3) 1166 # given that all three lines in sample3 are equal, 1167 # I think that any character could have been 'guessed' as the 1168 # delimiter, depending on dictionary order 1169 self.assertIn(dialect.delimiter, self.sample3) 1170 dialect = sniffer.sniff(self.sample3, delimiters="?,") 1171 self.assertEqual(dialect.delimiter, "?") 1172 dialect = sniffer.sniff(self.sample3, delimiters="/,") 1173 self.assertEqual(dialect.delimiter, "/") 1174 dialect = sniffer.sniff(self.sample4) 1175 self.assertEqual(dialect.delimiter, ";") 1176 dialect = sniffer.sniff(self.sample5) 1177 self.assertEqual(dialect.delimiter, "\t") 1178 dialect = sniffer.sniff(self.sample6) 1179 self.assertEqual(dialect.delimiter, "|") 1180 dialect = sniffer.sniff(self.sample7) 1181 self.assertEqual(dialect.delimiter, "|") 1182 self.assertEqual(dialect.quotechar, "'") 1183 dialect = sniffer.sniff(self.sample8) 1184 self.assertEqual(dialect.delimiter, '+') 1185 dialect = sniffer.sniff(self.sample9) 1186 self.assertEqual(dialect.delimiter, '+') 1187 self.assertEqual(dialect.quotechar, "'") 1188 dialect = sniffer.sniff(self.sample14) 1189 self.assertEqual(dialect.delimiter, '\0') 1190 1191 def test_doublequote(self): 1192 sniffer = csv.Sniffer() 1193 dialect = sniffer.sniff(self.header1) 1194 self.assertFalse(dialect.doublequote) 1195 dialect = sniffer.sniff(self.header2) 1196 self.assertFalse(dialect.doublequote) 1197 dialect = sniffer.sniff(self.sample2) 1198 self.assertTrue(dialect.doublequote) 1199 dialect = sniffer.sniff(self.sample8) 1200 self.assertFalse(dialect.doublequote) 1201 dialect = sniffer.sniff(self.sample9) 1202 self.assertTrue(dialect.doublequote) 1203 1204class NUL: 1205 def write(s, *args): 1206 pass 1207 writelines = write 1208 1209@unittest.skipUnless(hasattr(sys, "gettotalrefcount"), 1210 'requires sys.gettotalrefcount()') 1211class TestLeaks(unittest.TestCase): 1212 def test_create_read(self): 1213 delta = 0 1214 lastrc = sys.gettotalrefcount() 1215 for i in range(20): 1216 gc.collect() 1217 self.assertEqual(gc.garbage, []) 1218 rc = sys.gettotalrefcount() 1219 csv.reader(["a,b,c\r\n"]) 1220 csv.reader(["a,b,c\r\n"]) 1221 csv.reader(["a,b,c\r\n"]) 1222 delta = rc-lastrc 1223 lastrc = rc 1224 # if csv.reader() leaks, last delta should be 3 or more 1225 self.assertLess(delta, 3) 1226 1227 def test_create_write(self): 1228 delta = 0 1229 lastrc = sys.gettotalrefcount() 1230 s = NUL() 1231 for i in range(20): 1232 gc.collect() 1233 self.assertEqual(gc.garbage, []) 1234 rc = sys.gettotalrefcount() 1235 csv.writer(s) 1236 csv.writer(s) 1237 csv.writer(s) 1238 delta = rc-lastrc 1239 lastrc = rc 1240 # if csv.writer() leaks, last delta should be 3 or more 1241 self.assertLess(delta, 3) 1242 1243 def test_read(self): 1244 delta = 0 1245 rows = ["a,b,c\r\n"]*5 1246 lastrc = sys.gettotalrefcount() 1247 for i in range(20): 1248 gc.collect() 1249 self.assertEqual(gc.garbage, []) 1250 rc = sys.gettotalrefcount() 1251 rdr = csv.reader(rows) 1252 for row in rdr: 1253 pass 1254 delta = rc-lastrc 1255 lastrc = rc 1256 # if reader leaks during read, delta should be 5 or more 1257 self.assertLess(delta, 5) 1258 1259 def test_write(self): 1260 delta = 0 1261 rows = [[1,2,3]]*5 1262 s = NUL() 1263 lastrc = sys.gettotalrefcount() 1264 for i in range(20): 1265 gc.collect() 1266 self.assertEqual(gc.garbage, []) 1267 rc = sys.gettotalrefcount() 1268 writer = csv.writer(s) 1269 for row in rows: 1270 writer.writerow(row) 1271 delta = rc-lastrc 1272 lastrc = rc 1273 # if writer leaks during write, last delta should be 5 or more 1274 self.assertLess(delta, 5) 1275 1276class TestUnicode(unittest.TestCase): 1277 1278 names = ["Martin von Löwis", 1279 "Marc André Lemburg", 1280 "Guido van Rossum", 1281 "François Pinard"] 1282 1283 def test_unicode_read(self): 1284 with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: 1285 fileobj.write(",".join(self.names) + "\r\n") 1286 fileobj.seek(0) 1287 reader = csv.reader(fileobj) 1288 self.assertEqual(list(reader), [self.names]) 1289 1290 1291 def test_unicode_write(self): 1292 with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: 1293 writer = csv.writer(fileobj) 1294 writer.writerow(self.names) 1295 expected = ",".join(self.names)+"\r\n" 1296 fileobj.seek(0) 1297 self.assertEqual(fileobj.read(), expected) 1298 1299class KeyOrderingTest(unittest.TestCase): 1300 1301 def test_ordering_for_the_dict_reader_and_writer(self): 1302 resultset = set() 1303 for keys in permutations("abcde"): 1304 with TemporaryFile('w+', newline='', encoding="utf-8") as fileobject: 1305 dw = csv.DictWriter(fileobject, keys) 1306 dw.writeheader() 1307 fileobject.seek(0) 1308 dr = csv.DictReader(fileobject) 1309 kt = tuple(dr.fieldnames) 1310 self.assertEqual(keys, kt) 1311 resultset.add(kt) 1312 # Final sanity check: were all permutations unique? 1313 self.assertEqual(len(resultset), 120, "Key ordering: some key permutations not collected (expected 120)") 1314 1315 def test_ordered_dict_reader(self): 1316 data = dedent('''\ 1317 FirstName,LastName 1318 Eric,Idle 1319 Graham,Chapman,Over1,Over2 1320 1321 Under1 1322 John,Cleese 1323 ''').splitlines() 1324 1325 self.assertEqual(list(csv.DictReader(data)), 1326 [OrderedDict([('FirstName', 'Eric'), ('LastName', 'Idle')]), 1327 OrderedDict([('FirstName', 'Graham'), ('LastName', 'Chapman'), 1328 (None, ['Over1', 'Over2'])]), 1329 OrderedDict([('FirstName', 'Under1'), ('LastName', None)]), 1330 OrderedDict([('FirstName', 'John'), ('LastName', 'Cleese')]), 1331 ]) 1332 1333 self.assertEqual(list(csv.DictReader(data, restkey='OtherInfo')), 1334 [OrderedDict([('FirstName', 'Eric'), ('LastName', 'Idle')]), 1335 OrderedDict([('FirstName', 'Graham'), ('LastName', 'Chapman'), 1336 ('OtherInfo', ['Over1', 'Over2'])]), 1337 OrderedDict([('FirstName', 'Under1'), ('LastName', None)]), 1338 OrderedDict([('FirstName', 'John'), ('LastName', 'Cleese')]), 1339 ]) 1340 1341 del data[0] # Remove the header row 1342 self.assertEqual(list(csv.DictReader(data, fieldnames=['fname', 'lname'])), 1343 [OrderedDict([('fname', 'Eric'), ('lname', 'Idle')]), 1344 OrderedDict([('fname', 'Graham'), ('lname', 'Chapman'), 1345 (None, ['Over1', 'Over2'])]), 1346 OrderedDict([('fname', 'Under1'), ('lname', None)]), 1347 OrderedDict([('fname', 'John'), ('lname', 'Cleese')]), 1348 ]) 1349 1350 1351class MiscTestCase(unittest.TestCase): 1352 def test__all__(self): 1353 extra = {'__doc__', '__version__'} 1354 support.check__all__(self, csv, ('csv', '_csv'), extra=extra) 1355 1356 def test_subclassable(self): 1357 # issue 44089 1358 class Foo(csv.Error): ... 1359 1360if __name__ == '__main__': 1361 unittest.main() 1362