| 1 |
# Copyright (C) 2001 Python Software Foundation |
| 2 |
# csv package unit tests |
| 3 |
|
| 4 |
import copy |
| 5 |
import sys |
| 6 |
import unittest |
| 7 |
from io import StringIO |
| 8 |
from tempfile import TemporaryFile |
| 9 |
import csv |
| 10 |
import gc |
| 11 |
import pickle |
| 12 |
from test import support |
| 13 |
from test.support import cpython_only, import_helper, check_disallow_instantiation |
| 14 |
from test.support.import_helper import ensure_lazy_imports |
| 15 |
from itertools import permutations |
| 16 |
from textwrap import dedent |
| 17 |
from collections import OrderedDict |
| 18 |
|
| 19 |
|
| 20 |
class BadIterable: |
| 21 |
def __iter__(self): |
| 22 |
raise OSError |
| 23 |
|
| 24 |
|
| 25 |
class Test_Csv(unittest.TestCase): |
| 26 |
""" |
| 27 |
Test the underlying C csv parser in ways that are not appropriate |
| 28 |
from the high level interface. Further tests of this nature are done |
| 29 |
in TestDialectRegistry. |
| 30 |
""" |
| 31 |
def _test_arg_valid(self, ctor, arg): |
| 32 |
ctor(arg) |
| 33 |
self.assertRaises(TypeError, ctor) |
| 34 |
self.assertRaises(TypeError, ctor, None) |
| 35 |
self.assertRaises(TypeError, ctor, arg, bad_attr=0) |
| 36 |
self.assertRaises(TypeError, ctor, arg, delimiter='') |
| 37 |
self.assertRaises(TypeError, ctor, arg, escapechar='') |
| 38 |
self.assertRaises(TypeError, ctor, arg, quotechar='') |
| 39 |
self.assertRaises(TypeError, ctor, arg, delimiter='^^') |
| 40 |
self.assertRaises(TypeError, ctor, arg, escapechar='^^') |
| 41 |
self.assertRaises(TypeError, ctor, arg, quotechar='^^') |
| 42 |
self.assertRaises(csv.Error, ctor, arg, 'foo') |
| 43 |
self.assertRaises(TypeError, ctor, arg, delimiter=None) |
| 44 |
self.assertRaises(TypeError, ctor, arg, delimiter=1) |
| 45 |
self.assertRaises(TypeError, ctor, arg, escapechar=1) |
| 46 |
self.assertRaises(TypeError, ctor, arg, quotechar=1) |
| 47 |
self.assertRaises(TypeError, ctor, arg, lineterminator=None) |
| 48 |
self.assertRaises(TypeError, ctor, arg, lineterminator=1) |
| 49 |
self.assertRaises(TypeError, ctor, arg, quoting=None) |
| 50 |
self.assertRaises(TypeError, ctor, arg, |
| 51 |
quoting=csv.QUOTE_ALL, quotechar='') |
| 52 |
self.assertRaises(TypeError, ctor, arg, |
| 53 |
quoting=csv.QUOTE_ALL, quotechar=None) |
| 54 |
self.assertRaises(TypeError, ctor, arg, |
| 55 |
quoting=csv.QUOTE_NONE, quotechar='') |
| 56 |
self.assertRaises(ValueError, ctor, arg, delimiter='\n') |
| 57 |
self.assertRaises(ValueError, ctor, arg, escapechar='\n') |
| 58 |
self.assertRaises(ValueError, ctor, arg, quotechar='\n') |
| 59 |
self.assertRaises(ValueError, ctor, arg, delimiter='\r') |
| 60 |
self.assertRaises(ValueError, ctor, arg, escapechar='\r') |
| 61 |
self.assertRaises(ValueError, ctor, arg, quotechar='\r') |
| 62 |
ctor(arg, delimiter=' ') |
| 63 |
ctor(arg, escapechar=' ') |
| 64 |
ctor(arg, quotechar=' ') |
| 65 |
ctor(arg, delimiter='\t', skipinitialspace=True) |
| 66 |
ctor(arg, escapechar='\t', skipinitialspace=True) |
| 67 |
ctor(arg, quotechar='\t', skipinitialspace=True) |
| 68 |
ctor(arg, delimiter=' ', skipinitialspace=True) |
| 69 |
self.assertRaises(ValueError, ctor, arg, |
| 70 |
escapechar=' ', skipinitialspace=True) |
| 71 |
self.assertRaises(ValueError, ctor, arg, |
| 72 |
quotechar=' ', skipinitialspace=True) |
| 73 |
ctor(arg, delimiter='^') |
| 74 |
ctor(arg, escapechar='^') |
| 75 |
ctor(arg, quotechar='^') |
| 76 |
self.assertRaises(ValueError, ctor, arg, delimiter='^', escapechar='^') |
| 77 |
self.assertRaises(ValueError, ctor, arg, delimiter='^', quotechar='^') |
| 78 |
self.assertRaises(ValueError, ctor, arg, escapechar='^', quotechar='^') |
| 79 |
ctor(arg, delimiter='\x85') |
| 80 |
ctor(arg, escapechar='\x85') |
| 81 |
ctor(arg, quotechar='\x85') |
| 82 |
ctor(arg, lineterminator='\x85') |
| 83 |
self.assertRaises(ValueError, ctor, arg, |
| 84 |
delimiter='\x85', lineterminator='\x85') |
| 85 |
self.assertRaises(ValueError, ctor, arg, |
| 86 |
escapechar='\x85', lineterminator='\x85') |
| 87 |
self.assertRaises(ValueError, ctor, arg, |
| 88 |
quotechar='\x85', lineterminator='\x85') |
| 89 |
|
| 90 |
def test_reader_arg_valid(self): |
| 91 |
self._test_arg_valid(csv.reader, []) |
| 92 |
self.assertRaises(OSError, csv.reader, BadIterable()) |
| 93 |
|
| 94 |
def test_writer_arg_valid(self): |
| 95 |
self._test_arg_valid(csv.writer, StringIO()) |
| 96 |
class BadWriter: |
| 97 |
@property |
| 98 |
def write(self): |
| 99 |
raise OSError |
| 100 |
self.assertRaises(OSError, csv.writer, BadWriter()) |
| 101 |
|
| 102 |
def _test_default_attrs(self, ctor, *args): |
| 103 |
obj = ctor(*args) |
| 104 |
# Check defaults |
| 105 |
self.assertEqual(obj.dialect.delimiter, ',') |
| 106 |
self.assertIs(obj.dialect.doublequote, True) |
| 107 |
self.assertEqual(obj.dialect.escapechar, None) |
| 108 |
self.assertEqual(obj.dialect.lineterminator, "\r\n") |
| 109 |
self.assertEqual(obj.dialect.quotechar, '"') |
| 110 |
self.assertEqual(obj.dialect.quoting, csv.QUOTE_MINIMAL) |
| 111 |
self.assertIs(obj.dialect.skipinitialspace, False) |
| 112 |
self.assertIs(obj.dialect.strict, False) |
| 113 |
# Try deleting or changing attributes (they are read-only) |
| 114 |
self.assertRaises(AttributeError, delattr, obj.dialect, 'delimiter') |
| 115 |
self.assertRaises(AttributeError, setattr, obj.dialect, 'delimiter', ':') |
| 116 |
self.assertRaises(AttributeError, delattr, obj.dialect, 'quoting') |
| 117 |
self.assertRaises(AttributeError, setattr, obj.dialect, |
| 118 |
'quoting', None) |
| 119 |
|
| 120 |
def test_reader_attrs(self): |
| 121 |
self._test_default_attrs(csv.reader, []) |
| 122 |
|
| 123 |
def test_writer_attrs(self): |
| 124 |
self._test_default_attrs(csv.writer, StringIO()) |
| 125 |
|
| 126 |
def _test_kw_attrs(self, ctor, *args): |
| 127 |
# Now try with alternate options |
| 128 |
kwargs = dict(delimiter=':', doublequote=False, escapechar='\\', |
| 129 |
lineterminator='\r', quotechar='*', |
| 130 |
quoting=csv.QUOTE_NONE, skipinitialspace=True, |
| 131 |
strict=True) |
| 132 |
obj = ctor(*args, **kwargs) |
| 133 |
self.assertEqual(obj.dialect.delimiter, ':') |
| 134 |
self.assertIs(obj.dialect.doublequote, False) |
| 135 |
self.assertEqual(obj.dialect.escapechar, '\\') |
| 136 |
self.assertEqual(obj.dialect.lineterminator, "\r") |
| 137 |
self.assertEqual(obj.dialect.quotechar, '*') |
| 138 |
self.assertEqual(obj.dialect.quoting, csv.QUOTE_NONE) |
| 139 |
self.assertIs(obj.dialect.skipinitialspace, True) |
| 140 |
self.assertIs(obj.dialect.strict, True) |
| 141 |
|
| 142 |
def test_reader_kw_attrs(self): |
| 143 |
self._test_kw_attrs(csv.reader, []) |
| 144 |
|
| 145 |
def test_writer_kw_attrs(self): |
| 146 |
self._test_kw_attrs(csv.writer, StringIO()) |
| 147 |
|
| 148 |
def _test_dialect_attrs(self, ctor, *args): |
| 149 |
# Now try with dialect-derived options |
| 150 |
class dialect: |
| 151 |
delimiter='-' |
| 152 |
doublequote=False |
| 153 |
escapechar='^' |
| 154 |
lineterminator='$' |
| 155 |
quotechar='#' |
| 156 |
quoting=csv.QUOTE_ALL |
| 157 |
skipinitialspace=True |
| 158 |
strict=False |
| 159 |
args = args + (dialect,) |
| 160 |
obj = ctor(*args) |
| 161 |
self.assertEqual(obj.dialect.delimiter, '-') |
| 162 |
self.assertIs(obj.dialect.doublequote, False) |
| 163 |
self.assertEqual(obj.dialect.escapechar, '^') |
| 164 |
self.assertEqual(obj.dialect.lineterminator, "$") |
| 165 |
self.assertEqual(obj.dialect.quotechar, '#') |
| 166 |
self.assertEqual(obj.dialect.quoting, csv.QUOTE_ALL) |
| 167 |
self.assertIs(obj.dialect.skipinitialspace, True) |
| 168 |
self.assertIs(obj.dialect.strict, False) |
| 169 |
|
| 170 |
def test_reader_dialect_attrs(self): |
| 171 |
self._test_dialect_attrs(csv.reader, []) |
| 172 |
|
| 173 |
def test_writer_dialect_attrs(self): |
| 174 |
self._test_dialect_attrs(csv.writer, StringIO()) |
| 175 |
|
| 176 |
|
| 177 |
def _write_test(self, fields, expect, **kwargs): |
| 178 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 179 |
writer = csv.writer(fileobj, **kwargs) |
| 180 |
writer.writerow(fields) |
| 181 |
fileobj.seek(0) |
| 182 |
self.assertEqual(fileobj.read(), |
| 183 |
expect + writer.dialect.lineterminator) |
| 184 |
|
| 185 |
def _write_error_test(self, exc, fields, **kwargs): |
| 186 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 187 |
writer = csv.writer(fileobj, **kwargs) |
| 188 |
with self.assertRaises(exc): |
| 189 |
writer.writerow(fields) |
| 190 |
fileobj.seek(0) |
| 191 |
self.assertEqual(fileobj.read(), '') |
| 192 |
|
| 193 |
def test_write_arg_valid(self): |
| 194 |
self._write_error_test(csv.Error, None) |
| 195 |
# Check that exceptions are passed up the chain |
| 196 |
self._write_error_test(OSError, BadIterable()) |
| 197 |
class BadList: |
| 198 |
def __len__(self): |
| 199 |
return 10 |
| 200 |
def __getitem__(self, i): |
| 201 |
if i > 2: |
| 202 |
raise OSError |
| 203 |
self._write_error_test(OSError, BadList()) |
| 204 |
class BadItem: |
| 205 |
def __str__(self): |
| 206 |
raise OSError |
| 207 |
self._write_error_test(OSError, [BadItem()]) |
| 208 |
def test_write_bigfield(self): |
| 209 |
# This exercises the buffer realloc functionality |
| 210 |
bigstring = 'X' * 50000 |
| 211 |
self._write_test([bigstring,bigstring], '%s,%s' % \ |
| 212 |
(bigstring, bigstring)) |
| 213 |
|
| 214 |
def test_write_quoting(self): |
| 215 |
self._write_test(['a',1,'p,q'], 'a,1,"p,q"') |
| 216 |
self._write_error_test(csv.Error, ['a',1,'p,q'], |
| 217 |
quoting = csv.QUOTE_NONE) |
| 218 |
self._write_test(['a',1,'p,q'], 'a,1,"p,q"', |
| 219 |
quoting = csv.QUOTE_MINIMAL) |
| 220 |
self._write_test(['a',1,'p,q'], '"a",1,"p,q"', |
| 221 |
quoting = csv.QUOTE_NONNUMERIC) |
| 222 |
self._write_test(['a',1,'p,q'], '"a","1","p,q"', |
| 223 |
quoting = csv.QUOTE_ALL) |
| 224 |
self._write_test(['a\nb',1], '"a\nb","1"', |
| 225 |
quoting = csv.QUOTE_ALL) |
| 226 |
self._write_test(['a','',None,1], '"a","",,1', |
| 227 |
quoting = csv.QUOTE_STRINGS) |
| 228 |
self._write_test(['a','',None,1], '"a","",,"1"', |
| 229 |
quoting = csv.QUOTE_NOTNULL) |
| 230 |
|
| 231 |
def test_write_escape(self): |
| 232 |
self._write_test(['a',1,'p,q'], 'a,1,"p,q"', |
| 233 |
escapechar='\\') |
| 234 |
self._write_error_test(csv.Error, ['a',1,'p,"q"'], |
| 235 |
escapechar=None, doublequote=False) |
| 236 |
self._write_test(['a',1,'p,"q"'], 'a,1,"p,\\"q\\""', |
| 237 |
escapechar='\\', doublequote = False) |
| 238 |
self._write_test(['"'], '""""', |
| 239 |
escapechar='\\', quoting = csv.QUOTE_MINIMAL) |
| 240 |
self._write_test(['"'], '\\"', |
| 241 |
escapechar='\\', quoting = csv.QUOTE_MINIMAL, |
| 242 |
doublequote = False) |
| 243 |
self._write_test(['"'], '\\"', |
| 244 |
escapechar='\\', quoting = csv.QUOTE_NONE) |
| 245 |
self._write_test(['a',1,'p,q'], 'a,1,p\\,q', |
| 246 |
escapechar='\\', quoting = csv.QUOTE_NONE) |
| 247 |
self._write_test(['\\', 'a'], '\\\\,a', |
| 248 |
escapechar='\\', quoting=csv.QUOTE_NONE) |
| 249 |
self._write_test(['\\', 'a'], '\\\\,a', |
| 250 |
escapechar='\\', quoting=csv.QUOTE_MINIMAL) |
| 251 |
self._write_test(['\\', 'a'], '"\\\\","a"', |
| 252 |
escapechar='\\', quoting=csv.QUOTE_ALL) |
| 253 |
self._write_test(['\\ ', 'a'], '\\\\ ,a', |
| 254 |
escapechar='\\', quoting=csv.QUOTE_MINIMAL) |
| 255 |
self._write_test(['\\,', 'a'], '\\\\\\,,a', |
| 256 |
escapechar='\\', quoting=csv.QUOTE_NONE) |
| 257 |
self._write_test([',\\', 'a'], '",\\\\",a', |
| 258 |
escapechar='\\', quoting=csv.QUOTE_MINIMAL) |
| 259 |
self._write_test(['C\\', '6', '7', 'X"'], 'C\\\\,6,7,"X"""', |
| 260 |
escapechar='\\', quoting=csv.QUOTE_MINIMAL) |
| 261 |
|
| 262 |
def test_write_lineterminator(self): |
| 263 |
for lineterminator in '\r\n', '\n', '\r', '!@#', '\0': |
| 264 |
with self.subTest(lineterminator=lineterminator): |
| 265 |
with StringIO() as sio: |
| 266 |
writer = csv.writer(sio, lineterminator=lineterminator) |
| 267 |
writer.writerow(['a', 'b']) |
| 268 |
writer.writerow([1, 2]) |
| 269 |
writer.writerow(['\r', '\n']) |
| 270 |
self.assertEqual(sio.getvalue(), |
| 271 |
f'a,b{lineterminator}' |
| 272 |
f'1,2{lineterminator}' |
| 273 |
f'"\r","\n"{lineterminator}') |
| 274 |
|
| 275 |
def test_write_iterable(self): |
| 276 |
self._write_test(iter(['a', 1, 'p,q']), 'a,1,"p,q"') |
| 277 |
self._write_test(iter(['a', 1, None]), 'a,1,') |
| 278 |
self._write_test(iter([]), '') |
| 279 |
self._write_test(iter([None]), '""') |
| 280 |
self._write_error_test(csv.Error, iter([None]), quoting=csv.QUOTE_NONE) |
| 281 |
self._write_test(iter([None, None]), ',') |
| 282 |
|
| 283 |
def test_writerows(self): |
| 284 |
class BrokenFile: |
| 285 |
def write(self, buf): |
| 286 |
raise OSError |
| 287 |
writer = csv.writer(BrokenFile()) |
| 288 |
self.assertRaises(OSError, writer.writerows, [['a']]) |
| 289 |
|
| 290 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 291 |
writer = csv.writer(fileobj) |
| 292 |
self.assertRaises(TypeError, writer.writerows, None) |
| 293 |
writer.writerows([['a', 'b'], ['c', 'd']]) |
| 294 |
fileobj.seek(0) |
| 295 |
self.assertEqual(fileobj.read(), "a,b\r\nc,d\r\n") |
| 296 |
|
| 297 |
def test_writerows_with_none(self): |
| 298 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 299 |
writer = csv.writer(fileobj) |
| 300 |
writer.writerows([['a', None], [None, 'd']]) |
| 301 |
fileobj.seek(0) |
| 302 |
self.assertEqual(fileobj.read(), "a,\r\n,d\r\n") |
| 303 |
|
| 304 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 305 |
writer = csv.writer(fileobj) |
| 306 |
writer.writerows([[None], ['a']]) |
| 307 |
fileobj.seek(0) |
| 308 |
self.assertEqual(fileobj.read(), '""\r\na\r\n') |
| 309 |
|
| 310 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 311 |
writer = csv.writer(fileobj) |
| 312 |
writer.writerows([['a'], [None]]) |
| 313 |
fileobj.seek(0) |
| 314 |
self.assertEqual(fileobj.read(), 'a\r\n""\r\n') |
| 315 |
|
| 316 |
|
| 317 |
def test_write_empty_fields(self): |
| 318 |
self._write_test((), '') |
| 319 |
self._write_test([''], '""') |
| 320 |
self._write_error_test(csv.Error, [''], quoting=csv.QUOTE_NONE) |
| 321 |
self._write_test([''], '""', quoting=csv.QUOTE_STRINGS) |
| 322 |
self._write_test([''], '""', quoting=csv.QUOTE_NOTNULL) |
| 323 |
self._write_test([None], '""') |
| 324 |
self._write_error_test(csv.Error, [None], quoting=csv.QUOTE_NONE) |
| 325 |
self._write_error_test(csv.Error, [None], quoting=csv.QUOTE_STRINGS) |
| 326 |
self._write_error_test(csv.Error, [None], quoting=csv.QUOTE_NOTNULL) |
| 327 |
self._write_test(['', ''], ',') |
| 328 |
self._write_test([None, None], ',') |
| 329 |
|
| 330 |
def test_write_empty_fields_space_delimiter(self): |
| 331 |
self._write_test([''], '""', delimiter=' ', skipinitialspace=False) |
| 332 |
self._write_test([''], '""', delimiter=' ', skipinitialspace=True) |
| 333 |
self._write_test([None], '""', delimiter=' ', skipinitialspace=False) |
| 334 |
self._write_test([None], '""', delimiter=' ', skipinitialspace=True) |
| 335 |
|
| 336 |
self._write_test(['', ''], ' ', delimiter=' ', skipinitialspace=False) |
| 337 |
self._write_test(['', ''], '"" ""', delimiter=' ', skipinitialspace=True) |
| 338 |
self._write_test([None, None], ' ', delimiter=' ', skipinitialspace=False) |
| 339 |
self._write_test([None, None], '"" ""', delimiter=' ', skipinitialspace=True) |
| 340 |
|
| 341 |
self._write_test(['', ''], ' ', delimiter=' ', skipinitialspace=False, |
| 342 |
quoting=csv.QUOTE_NONE) |
| 343 |
self._write_error_test(csv.Error, ['', ''], |
| 344 |
delimiter=' ', skipinitialspace=True, |
| 345 |
quoting=csv.QUOTE_NONE) |
| 346 |
for quoting in csv.QUOTE_STRINGS, csv.QUOTE_NOTNULL: |
| 347 |
self._write_test(['', ''], '"" ""', delimiter=' ', skipinitialspace=False, |
| 348 |
quoting=quoting) |
| 349 |
self._write_test(['', ''], '"" ""', delimiter=' ', skipinitialspace=True, |
| 350 |
quoting=quoting) |
| 351 |
|
| 352 |
for quoting in csv.QUOTE_NONE, csv.QUOTE_STRINGS, csv.QUOTE_NOTNULL: |
| 353 |
self._write_test([None, None], ' ', delimiter=' ', skipinitialspace=False, |
| 354 |
quoting=quoting) |
| 355 |
self._write_error_test(csv.Error, [None, None], |
| 356 |
delimiter=' ', skipinitialspace=True, |
| 357 |
quoting=quoting) |
| 358 |
|
| 359 |
def test_writerows_errors(self): |
| 360 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 361 |
writer = csv.writer(fileobj) |
| 362 |
self.assertRaises(TypeError, writer.writerows, None) |
| 363 |
self.assertRaises(OSError, writer.writerows, BadIterable()) |
| 364 |
|
| 365 |
def _read_test(self, input, expect, **kwargs): |
| 366 |
reader = csv.reader(input, **kwargs) |
| 367 |
result = list(reader) |
| 368 |
self.assertEqual(result, expect) |
| 369 |
|
| 370 |
def test_read_oddinputs(self): |
| 371 |
self._read_test([], []) |
| 372 |
self._read_test([''], [[]]) |
| 373 |
self.assertRaises(csv.Error, self._read_test, |
| 374 |
['"ab"c'], None, strict = 1) |
| 375 |
self._read_test(['"ab"c'], [['abc']], doublequote = 0) |
| 376 |
|
| 377 |
self.assertRaises(csv.Error, self._read_test, |
| 378 |
[b'abc'], None) |
| 379 |
|
| 380 |
def test_read_eol(self): |
| 381 |
self._read_test(['a,b', 'c,d'], [['a','b'], ['c','d']]) |
| 382 |
self._read_test(['a,b\n', 'c,d\n'], [['a','b'], ['c','d']]) |
| 383 |
self._read_test(['a,b\r\n', 'c,d\r\n'], [['a','b'], ['c','d']]) |
| 384 |
self._read_test(['a,b\r', 'c,d\r'], [['a','b'], ['c','d']]) |
| 385 |
|
| 386 |
errmsg = "with newline=''" |
| 387 |
with self.assertRaisesRegex(csv.Error, errmsg): |
| 388 |
next(csv.reader(['a,b\rc,d'])) |
| 389 |
with self.assertRaisesRegex(csv.Error, errmsg): |
| 390 |
next(csv.reader(['a,b\nc,d'])) |
| 391 |
with self.assertRaisesRegex(csv.Error, errmsg): |
| 392 |
next(csv.reader(['a,b\r\nc,d'])) |
| 393 |
|
| 394 |
def test_read_eof(self): |
| 395 |
self._read_test(['a,"'], [['a', '']]) |
| 396 |
self._read_test(['"a'], [['a']]) |
| 397 |
self._read_test(['^'], [['\n']], escapechar='^') |
| 398 |
self.assertRaises(csv.Error, self._read_test, ['a,"'], [], strict=True) |
| 399 |
self.assertRaises(csv.Error, self._read_test, ['"a'], [], strict=True) |
| 400 |
self.assertRaises(csv.Error, self._read_test, |
| 401 |
['^'], [], escapechar='^', strict=True) |
| 402 |
|
| 403 |
def test_read_nul(self): |
| 404 |
self._read_test(['\0'], [['\0']]) |
| 405 |
self._read_test(['a,\0b,c'], [['a', '\0b', 'c']]) |
| 406 |
self._read_test(['a,b\0,c'], [['a', 'b\0', 'c']]) |
| 407 |
self._read_test(['a,b\\\0,c'], [['a', 'b\0', 'c']], escapechar='\\') |
| 408 |
self._read_test(['a,"\0b",c'], [['a', '\0b', 'c']]) |
| 409 |
|
| 410 |
def test_read_delimiter(self): |
| 411 |
self._read_test(['a,b,c'], [['a', 'b', 'c']]) |
| 412 |
self._read_test(['a;b;c'], [['a', 'b', 'c']], delimiter=';') |
| 413 |
self._read_test(['a\0b\0c'], [['a', 'b', 'c']], delimiter='\0') |
| 414 |
|
| 415 |
def test_read_escape(self): |
| 416 |
self._read_test(['a,\\b,c'], [['a', 'b', 'c']], escapechar='\\') |
| 417 |
self._read_test(['a,b\\,c'], [['a', 'b,c']], escapechar='\\') |
| 418 |
self._read_test(['a,"b\\,c"'], [['a', 'b,c']], escapechar='\\') |
| 419 |
self._read_test(['a,"b,\\c"'], [['a', 'b,c']], escapechar='\\') |
| 420 |
self._read_test(['a,"b,c\\""'], [['a', 'b,c"']], escapechar='\\') |
| 421 |
self._read_test(['a,"b,c"\\'], [['a', 'b,c\\']], escapechar='\\') |
| 422 |
self._read_test(['a,^b,c'], [['a', 'b', 'c']], escapechar='^') |
| 423 |
self._read_test(['a,\0b,c'], [['a', 'b', 'c']], escapechar='\0') |
| 424 |
self._read_test(['a,\\b,c'], [['a', '\\b', 'c']], escapechar=None) |
| 425 |
self._read_test(['a,\\b,c'], [['a', '\\b', 'c']]) |
| 426 |
|
| 427 |
def test_read_quoting(self): |
| 428 |
self._read_test(['1,",3,",5'], [['1', ',3,', '5']]) |
| 429 |
self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']], |
| 430 |
quotechar=None, escapechar='\\') |
| 431 |
self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']], |
| 432 |
quoting=csv.QUOTE_NONE, escapechar='\\') |
| 433 |
# will this fail where locale uses comma for decimals? |
| 434 |
self._read_test([',3,"5",7.3, 9'], [['', 3, '5', 7.3, 9]], |
| 435 |
quoting=csv.QUOTE_NONNUMERIC) |
| 436 |
self._read_test([',3,"5",7.3, 9'], [[None, '3', '5', '7.3', ' 9']], |
| 437 |
quoting=csv.QUOTE_NOTNULL) |
| 438 |
self._read_test([',3,"5",7.3, 9'], [[None, 3, '5', 7.3, 9]], |
| 439 |
quoting=csv.QUOTE_STRINGS) |
| 440 |
|
| 441 |
self._read_test([',,"",'], [['', '', '', '']]) |
| 442 |
self._read_test([',,"",'], [['', '', '', '']], |
| 443 |
quoting=csv.QUOTE_NONNUMERIC) |
| 444 |
self._read_test([',,"",'], [[None, None, '', None]], |
| 445 |
quoting=csv.QUOTE_NOTNULL) |
| 446 |
self._read_test([',,"",'], [[None, None, '', None]], |
| 447 |
quoting=csv.QUOTE_STRINGS) |
| 448 |
|
| 449 |
self._read_test(['"a\nb", 7'], [['a\nb', ' 7']]) |
| 450 |
self.assertRaises(ValueError, self._read_test, |
| 451 |
['abc,3'], [[]], |
| 452 |
quoting=csv.QUOTE_NONNUMERIC) |
| 453 |
self.assertRaises(ValueError, self._read_test, |
| 454 |
['abc,3'], [[]], |
| 455 |
quoting=csv.QUOTE_STRINGS) |
| 456 |
self._read_test(['1,@,3,@,5'], [['1', ',3,', '5']], quotechar='@') |
| 457 |
self._read_test(['1,\0,3,\0,5'], [['1', ',3,', '5']], quotechar='\0') |
| 458 |
self._read_test(['1\\.5,\\.5,.5'], [[1.5, 0.5, 0.5]], |
| 459 |
quoting=csv.QUOTE_NONNUMERIC, escapechar='\\') |
| 460 |
self._read_test(['1\\.5,\\.5,"\\.5"'], [[1.5, 0.5, ".5"]], |
| 461 |
quoting=csv.QUOTE_STRINGS, escapechar='\\') |
| 462 |
|
| 463 |
def test_read_skipinitialspace(self): |
| 464 |
self._read_test(['no space, space, spaces,\ttab'], |
| 465 |
[['no space', 'space', 'spaces', '\ttab']], |
| 466 |
skipinitialspace=True) |
| 467 |
self._read_test([' , , '], |
| 468 |
[['', '', '']], |
| 469 |
skipinitialspace=True) |
| 470 |
self._read_test([' , , '], |
| 471 |
[[None, None, None]], |
| 472 |
skipinitialspace=True, quoting=csv.QUOTE_NOTNULL) |
| 473 |
self._read_test([' , , '], |
| 474 |
[[None, None, None]], |
| 475 |
skipinitialspace=True, quoting=csv.QUOTE_STRINGS) |
| 476 |
|
| 477 |
def test_read_space_delimiter(self): |
| 478 |
self._read_test(['a b', ' a ', ' ', ''], |
| 479 |
[['a', '', '', 'b'], ['', '', 'a', '', ''], ['', '', ''], []], |
| 480 |
delimiter=' ', skipinitialspace=False) |
| 481 |
self._read_test(['a b', ' a ', ' ', ''], |
| 482 |
[['a', 'b'], ['a', ''], [''], []], |
| 483 |
delimiter=' ', skipinitialspace=True) |
| 484 |
|
| 485 |
def test_read_bigfield(self): |
| 486 |
# This exercises the buffer realloc functionality and field size |
| 487 |
# limits. |
| 488 |
limit = csv.field_size_limit() |
| 489 |
try: |
| 490 |
size = 50000 |
| 491 |
bigstring = 'X' * size |
| 492 |
bigline = '%s,%s' % (bigstring, bigstring) |
| 493 |
self._read_test([bigline], [[bigstring, bigstring]]) |
| 494 |
csv.field_size_limit(size) |
| 495 |
self._read_test([bigline], [[bigstring, bigstring]]) |
| 496 |
self.assertEqual(csv.field_size_limit(), size) |
| 497 |
csv.field_size_limit(size-1) |
| 498 |
self.assertRaises(csv.Error, self._read_test, [bigline], []) |
| 499 |
self.assertRaises(TypeError, csv.field_size_limit, None) |
| 500 |
self.assertRaises(TypeError, csv.field_size_limit, 1, None) |
| 501 |
finally: |
| 502 |
csv.field_size_limit(limit) |
| 503 |
|
| 504 |
def test_read_linenum(self): |
| 505 |
r = csv.reader(['line,1', 'line,2', 'line,3']) |
| 506 |
self.assertEqual(r.line_num, 0) |
| 507 |
next(r) |
| 508 |
self.assertEqual(r.line_num, 1) |
| 509 |
next(r) |
| 510 |
self.assertEqual(r.line_num, 2) |
| 511 |
next(r) |
| 512 |
self.assertEqual(r.line_num, 3) |
| 513 |
self.assertRaises(StopIteration, next, r) |
| 514 |
self.assertEqual(r.line_num, 3) |
| 515 |
|
| 516 |
def test_roundtrip_quoteed_newlines(self): |
| 517 |
rows = [ |
| 518 |
['\na', 'b\nc', 'd\n'], |
| 519 |
['\re', 'f\rg', 'h\r'], |
| 520 |
['\r\ni', 'j\r\nk', 'l\r\n'], |
| 521 |
['\n\rm', 'n\n\ro', 'p\n\r'], |
| 522 |
['\r\rq', 'r\r\rs', 't\r\r'], |
| 523 |
['\n\nu', 'v\n\nw', 'x\n\n'], |
| 524 |
] |
| 525 |
for lineterminator in '\r\n', '\n', '\r': |
| 526 |
with self.subTest(lineterminator=lineterminator): |
| 527 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 528 |
writer = csv.writer(fileobj, lineterminator=lineterminator) |
| 529 |
writer.writerows(rows) |
| 530 |
fileobj.seek(0) |
| 531 |
for i, row in enumerate(csv.reader(fileobj)): |
| 532 |
self.assertEqual(row, rows[i]) |
| 533 |
|
| 534 |
def test_roundtrip_escaped_unquoted_newlines(self): |
| 535 |
rows = [ |
| 536 |
['\na', 'b\nc', 'd\n'], |
| 537 |
['\re', 'f\rg', 'h\r'], |
| 538 |
['\r\ni', 'j\r\nk', 'l\r\n'], |
| 539 |
['\n\rm', 'n\n\ro', 'p\n\r'], |
| 540 |
['\r\rq', 'r\r\rs', 't\r\r'], |
| 541 |
['\n\nu', 'v\n\nw', 'x\n\n'], |
| 542 |
] |
| 543 |
for lineterminator in '\r\n', '\n', '\r': |
| 544 |
with self.subTest(lineterminator=lineterminator): |
| 545 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 546 |
writer = csv.writer(fileobj, lineterminator=lineterminator, |
| 547 |
quoting=csv.QUOTE_NONE, escapechar="\\") |
| 548 |
writer.writerows(rows) |
| 549 |
fileobj.seek(0) |
| 550 |
for i, row in enumerate(csv.reader(fileobj, |
| 551 |
quoting=csv.QUOTE_NONE, |
| 552 |
escapechar="\\")): |
| 553 |
self.assertEqual(row, rows[i]) |
| 554 |
|
| 555 |
|
| 556 |
def test_reader_reentrant_iterator(self): |
| 557 |
# gh-145105: re-entering the reader from the iterator must not crash. |
| 558 |
class ReentrantIter: |
| 559 |
def __init__(self): |
| 560 |
self.reader = None |
| 561 |
self.n = 0 |
| 562 |
def __iter__(self): |
| 563 |
return self |
| 564 |
def __next__(self): |
| 565 |
self.n += 1 |
| 566 |
if self.n == 1: |
| 567 |
try: |
| 568 |
next(self.reader) |
| 569 |
except StopIteration: |
| 570 |
pass |
| 571 |
return "a,b" |
| 572 |
if self.n == 2: |
| 573 |
return "x" |
| 574 |
raise StopIteration |
| 575 |
|
| 576 |
it = ReentrantIter() |
| 577 |
reader = csv.reader(it) |
| 578 |
it.reader = reader |
| 579 |
with self.assertRaises(csv.Error): |
| 580 |
next(reader) |
| 581 |
|
| 582 |
|
| 583 |
class TestDialectRegistry(unittest.TestCase): |
| 584 |
def test_registry_badargs(self): |
| 585 |
self.assertRaises(TypeError, csv.list_dialects, None) |
| 586 |
self.assertRaises(TypeError, csv.get_dialect) |
| 587 |
self.assertRaises(csv.Error, csv.get_dialect, None) |
| 588 |
self.assertRaises(csv.Error, csv.get_dialect, "nonesuch") |
| 589 |
self.assertRaises(TypeError, csv.unregister_dialect) |
| 590 |
self.assertRaises(csv.Error, csv.unregister_dialect, None) |
| 591 |
self.assertRaises(csv.Error, csv.unregister_dialect, "nonesuch") |
| 592 |
self.assertRaises(TypeError, csv.register_dialect, None) |
| 593 |
self.assertRaises(TypeError, csv.register_dialect, None, None) |
| 594 |
self.assertRaises(TypeError, csv.register_dialect, "nonesuch", 0, 0) |
| 595 |
self.assertRaises(TypeError, csv.register_dialect, "nonesuch", |
| 596 |
badargument=None) |
| 597 |
self.assertRaises(TypeError, csv.register_dialect, "nonesuch", |
| 598 |
quoting=None) |
| 599 |
self.assertRaises(TypeError, csv.register_dialect, []) |
| 600 |
|
| 601 |
def test_registry(self): |
| 602 |
class myexceltsv(csv.excel): |
| 603 |
delimiter = "\t" |
| 604 |
name = "myexceltsv" |
| 605 |
expected_dialects = csv.list_dialects() + [name] |
| 606 |
expected_dialects.sort() |
| 607 |
csv.register_dialect(name, myexceltsv) |
| 608 |
self.addCleanup(csv.unregister_dialect, name) |
| 609 |
self.assertEqual(csv.get_dialect(name).delimiter, '\t') |
| 610 |
got_dialects = sorted(csv.list_dialects()) |
| 611 |
self.assertEqual(expected_dialects, got_dialects) |
| 612 |
|
| 613 |
def test_register_kwargs(self): |
| 614 |
name = 'fedcba' |
| 615 |
csv.register_dialect(name, delimiter=';') |
| 616 |
self.addCleanup(csv.unregister_dialect, name) |
| 617 |
self.assertEqual(csv.get_dialect(name).delimiter, ';') |
| 618 |
self.assertEqual([['X', 'Y', 'Z']], list(csv.reader(['X;Y;Z'], name))) |
| 619 |
|
| 620 |
def test_register_kwargs_override(self): |
| 621 |
class mydialect(csv.Dialect): |
| 622 |
delimiter = "\t" |
| 623 |
quotechar = '"' |
| 624 |
doublequote = True |
| 625 |
skipinitialspace = False |
| 626 |
lineterminator = '\r\n' |
| 627 |
quoting = csv.QUOTE_MINIMAL |
| 628 |
|
| 629 |
name = 'test_dialect' |
| 630 |
csv.register_dialect(name, mydialect, |
| 631 |
delimiter=';', |
| 632 |
quotechar="'", |
| 633 |
doublequote=False, |
| 634 |
skipinitialspace=True, |
| 635 |
lineterminator='\n', |
| 636 |
quoting=csv.QUOTE_ALL) |
| 637 |
self.addCleanup(csv.unregister_dialect, name) |
| 638 |
|
| 639 |
# Ensure that kwargs do override attributes of a dialect class: |
| 640 |
dialect = csv.get_dialect(name) |
| 641 |
self.assertEqual(dialect.delimiter, ';') |
| 642 |
self.assertEqual(dialect.quotechar, "'") |
| 643 |
self.assertEqual(dialect.doublequote, False) |
| 644 |
self.assertEqual(dialect.skipinitialspace, True) |
| 645 |
self.assertEqual(dialect.lineterminator, '\n') |
| 646 |
self.assertEqual(dialect.quoting, csv.QUOTE_ALL) |
| 647 |
|
| 648 |
def test_incomplete_dialect(self): |
| 649 |
class myexceltsv(csv.Dialect): |
| 650 |
delimiter = "\t" |
| 651 |
self.assertRaises(csv.Error, myexceltsv) |
| 652 |
|
| 653 |
def test_space_dialect(self): |
| 654 |
class space(csv.excel): |
| 655 |
delimiter = " " |
| 656 |
quoting = csv.QUOTE_NONE |
| 657 |
escapechar = "\\" |
| 658 |
|
| 659 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 660 |
fileobj.write("abc def\nc1ccccc1 benzene\n") |
| 661 |
fileobj.seek(0) |
| 662 |
reader = csv.reader(fileobj, dialect=space()) |
| 663 |
self.assertEqual(next(reader), ["abc", "", "", "def"]) |
| 664 |
self.assertEqual(next(reader), ["c1ccccc1", "benzene"]) |
| 665 |
|
| 666 |
def compare_dialect_123(self, expected, *writeargs, **kwwriteargs): |
| 667 |
|
| 668 |
with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: |
| 669 |
|
| 670 |
writer = csv.writer(fileobj, *writeargs, **kwwriteargs) |
| 671 |
writer.writerow([1,2,3]) |
| 672 |
fileobj.seek(0) |
| 673 |
self.assertEqual(fileobj.read(), expected) |
| 674 |
|
| 675 |
def test_dialect_apply(self): |
| 676 |
class testA(csv.excel): |
| 677 |
delimiter = "\t" |
| 678 |
class testB(csv.excel): |
| 679 |
delimiter = ":" |
| 680 |
class testC(csv.excel): |
| 681 |
delimiter = "|" |
| 682 |
class testUni(csv.excel): |
| 683 |
delimiter = "\u039B" |
| 684 |
|
| 685 |
class unspecified(): |
| 686 |
# A class to pass as dialect but with no dialect attributes. |
| 687 |
pass |
| 688 |
|
| 689 |
csv.register_dialect('testC', testC) |
| 690 |
try: |
| 691 |
self.compare_dialect_123("1,2,3\r\n") |
| 692 |
self.compare_dialect_123("1,2,3\r\n", dialect=None) |
| 693 |
self.compare_dialect_123("1,2,3\r\n", dialect=unspecified) |
| 694 |
self.compare_dialect_123("1\t2\t3\r\n", testA) |
| 695 |
self.compare_dialect_123("1:2:3\r\n", dialect=testB()) |
| 696 |
self.compare_dialect_123("1|2|3\r\n", dialect='testC') |
| 697 |
self.compare_dialect_123("1;2;3\r\n", dialect=testA, |
| 698 |
delimiter=';') |
| 699 |
self.compare_dialect_123("1\u039B2\u039B3\r\n", |
| 700 |
dialect=testUni) |
| 701 |
|
| 702 |
finally: |
| 703 |
csv.unregister_dialect('testC') |
| 704 |
|
| 705 |
def test_copy(self): |
| 706 |
for name in csv.list_dialects(): |
| 707 |
dialect = csv.get_dialect(name) |
| 708 |
self.assertRaises(TypeError, copy.copy, dialect) |
| 709 |
|
| 710 |
def test_pickle(self): |
| 711 |
for name in csv.list_dialects(): |
| 712 |
dialect = csv.get_dialect(name) |
| 713 |
for proto in range(pickle.HIGHEST_PROTOCOL + 1): |
| 714 |
self.assertRaises(TypeError, pickle.dumps, dialect, proto) |
| 715 |
|
| 716 |
class TestCsvBase(unittest.TestCase): |
| 717 |
def readerAssertEqual(self, input, expected_result): |
| 718 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 719 |
fileobj.write(input) |
| 720 |
fileobj.seek(0) |
| 721 |
reader = csv.reader(fileobj, dialect = self.dialect) |
| 722 |
fields = list(reader) |
| 723 |
self.assertEqual(fields, expected_result) |
| 724 |
|
| 725 |
def writerAssertEqual(self, input, expected_result): |
| 726 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 727 |
writer = csv.writer(fileobj, dialect = self.dialect) |
| 728 |
writer.writerows(input) |
| 729 |
fileobj.seek(0) |
| 730 |
self.assertEqual(fileobj.read(), expected_result) |
| 731 |
|
| 732 |
class TestDialectExcel(TestCsvBase): |
| 733 |
dialect = 'excel' |
| 734 |
|
| 735 |
def test_single(self): |
| 736 |
self.readerAssertEqual('abc', [['abc']]) |
| 737 |
|
| 738 |
def test_simple(self): |
| 739 |
self.readerAssertEqual('1,2,3,4,5', [['1','2','3','4','5']]) |
| 740 |
|
| 741 |
def test_blankline(self): |
| 742 |
self.readerAssertEqual('', []) |
| 743 |
|
| 744 |
def test_empty_fields(self): |
| 745 |
self.readerAssertEqual(',', [['', '']]) |
| 746 |
|
| 747 |
def test_singlequoted(self): |
| 748 |
self.readerAssertEqual('""', [['']]) |
| 749 |
|
| 750 |
def test_singlequoted_left_empty(self): |
| 751 |
self.readerAssertEqual('"",', [['','']]) |
| 752 |
|
| 753 |
def test_singlequoted_right_empty(self): |
| 754 |
self.readerAssertEqual(',""', [['','']]) |
| 755 |
|
| 756 |
def test_single_quoted_quote(self): |
| 757 |
self.readerAssertEqual('""""', [['"']]) |
| 758 |
|
| 759 |
def test_quoted_quotes(self): |
| 760 |
self.readerAssertEqual('""""""', [['""']]) |
| 761 |
|
| 762 |
def test_inline_quote(self): |
| 763 |
self.readerAssertEqual('a""b', [['a""b']]) |
| 764 |
|
| 765 |
def test_inline_quotes(self): |
| 766 |
self.readerAssertEqual('a"b"c', [['a"b"c']]) |
| 767 |
|
| 768 |
def test_quotes_and_more(self): |
| 769 |
# Excel would never write a field containing '"a"b', but when |
| 770 |
# reading one, it will return 'ab'. |
| 771 |
self.readerAssertEqual('"a"b', [['ab']]) |
| 772 |
|
| 773 |
def test_lone_quote(self): |
| 774 |
self.readerAssertEqual('a"b', [['a"b']]) |
| 775 |
|
| 776 |
def test_quote_and_quote(self): |
| 777 |
# Excel would never write a field containing '"a" "b"', but when |
| 778 |
# reading one, it will return 'a "b"'. |
| 779 |
self.readerAssertEqual('"a" "b"', [['a "b"']]) |
| 780 |
|
| 781 |
def test_space_and_quote(self): |
| 782 |
self.readerAssertEqual(' "a"', [[' "a"']]) |
| 783 |
|
| 784 |
def test_quoted(self): |
| 785 |
self.readerAssertEqual('1,2,3,"I think, therefore I am",5,6', |
| 786 |
[['1', '2', '3', |
| 787 |
'I think, therefore I am', |
| 788 |
'5', '6']]) |
| 789 |
|
| 790 |
def test_quoted_quote(self): |
| 791 |
self.readerAssertEqual('1,2,3,"""I see,"" said the blind man","as he picked up his hammer and saw"', |
| 792 |
[['1', '2', '3', |
| 793 |
'"I see," said the blind man', |
| 794 |
'as he picked up his hammer and saw']]) |
| 795 |
|
| 796 |
def test_quoted_nl(self): |
| 797 |
input = '''\ |
| 798 |
1,2,3,"""I see,"" |
| 799 |
said the blind man","as he picked up his |
| 800 |
hammer and saw" |
| 801 |
9,8,7,6''' |
| 802 |
self.readerAssertEqual(input, |
| 803 |
[['1', '2', '3', |
| 804 |
'"I see,"\nsaid the blind man', |
| 805 |
'as he picked up his\nhammer and saw'], |
| 806 |
['9','8','7','6']]) |
| 807 |
|
| 808 |
def test_dubious_quote(self): |
| 809 |
self.readerAssertEqual('12,12,1",', [['12', '12', '1"', '']]) |
| 810 |
|
| 811 |
def test_null(self): |
| 812 |
self.writerAssertEqual([], '') |
| 813 |
|
| 814 |
def test_single_writer(self): |
| 815 |
self.writerAssertEqual([['abc']], 'abc\r\n') |
| 816 |
|
| 817 |
def test_simple_writer(self): |
| 818 |
self.writerAssertEqual([[1, 2, 'abc', 3, 4]], '1,2,abc,3,4\r\n') |
| 819 |
|
| 820 |
def test_quotes(self): |
| 821 |
self.writerAssertEqual([[1, 2, 'a"bc"', 3, 4]], '1,2,"a""bc""",3,4\r\n') |
| 822 |
|
| 823 |
def test_quote_fieldsep(self): |
| 824 |
self.writerAssertEqual([['abc,def']], '"abc,def"\r\n') |
| 825 |
|
| 826 |
def test_newlines(self): |
| 827 |
self.writerAssertEqual([[1, 2, 'a\nbc', 3, 4]], '1,2,"a\nbc",3,4\r\n') |
| 828 |
|
| 829 |
class EscapedExcel(csv.excel): |
| 830 |
quoting = csv.QUOTE_NONE |
| 831 |
escapechar = '\\' |
| 832 |
|
| 833 |
class TestEscapedExcel(TestCsvBase): |
| 834 |
dialect = EscapedExcel() |
| 835 |
|
| 836 |
def test_escape_fieldsep(self): |
| 837 |
self.writerAssertEqual([['abc,def']], 'abc\\,def\r\n') |
| 838 |
|
| 839 |
def test_read_escape_fieldsep(self): |
| 840 |
self.readerAssertEqual('abc\\,def\r\n', [['abc,def']]) |
| 841 |
|
| 842 |
class TestDialectUnix(TestCsvBase): |
| 843 |
dialect = 'unix' |
| 844 |
|
| 845 |
def test_simple_writer(self): |
| 846 |
self.writerAssertEqual([[1, 'abc def', 'abc']], '"1","abc def","abc"\n') |
| 847 |
|
| 848 |
def test_simple_reader(self): |
| 849 |
self.readerAssertEqual('"1","abc def","abc"\n', [['1', 'abc def', 'abc']]) |
| 850 |
|
| 851 |
class QuotedEscapedExcel(csv.excel): |
| 852 |
quoting = csv.QUOTE_NONNUMERIC |
| 853 |
escapechar = '\\' |
| 854 |
|
| 855 |
class TestQuotedEscapedExcel(TestCsvBase): |
| 856 |
dialect = QuotedEscapedExcel() |
| 857 |
|
| 858 |
def test_write_escape_fieldsep(self): |
| 859 |
self.writerAssertEqual([['abc,def']], '"abc,def"\r\n') |
| 860 |
|
| 861 |
def test_read_escape_fieldsep(self): |
| 862 |
self.readerAssertEqual('"abc\\,def"\r\n', [['abc,def']]) |
| 863 |
|
| 864 |
class TestDictFields(unittest.TestCase): |
| 865 |
### "long" means the row is longer than the number of fieldnames |
| 866 |
### "short" means there are fewer elements in the row than fieldnames |
| 867 |
def test_writeheader_return_value(self): |
| 868 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 869 |
writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) |
| 870 |
writeheader_return_value = writer.writeheader() |
| 871 |
self.assertEqual(writeheader_return_value, 10) |
| 872 |
|
| 873 |
def test_write_simple_dict(self): |
| 874 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 875 |
writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) |
| 876 |
writer.writeheader() |
| 877 |
fileobj.seek(0) |
| 878 |
self.assertEqual(fileobj.readline(), "f1,f2,f3\r\n") |
| 879 |
writer.writerow({"f1": 10, "f3": "abc"}) |
| 880 |
fileobj.seek(0) |
| 881 |
fileobj.readline() # header |
| 882 |
self.assertEqual(fileobj.read(), "10,,abc\r\n") |
| 883 |
|
| 884 |
def test_write_multiple_dict_rows(self): |
| 885 |
fileobj = StringIO() |
| 886 |
writer = csv.DictWriter(fileobj, fieldnames=["f1", "f2", "f3"]) |
| 887 |
writer.writeheader() |
| 888 |
self.assertEqual(fileobj.getvalue(), "f1,f2,f3\r\n") |
| 889 |
writer.writerows([{"f1": 1, "f2": "abc", "f3": "f"}, |
| 890 |
{"f1": 2, "f2": 5, "f3": "xyz"}]) |
| 891 |
self.assertEqual(fileobj.getvalue(), |
| 892 |
"f1,f2,f3\r\n1,abc,f\r\n2,5,xyz\r\n") |
| 893 |
|
| 894 |
def test_write_no_fields(self): |
| 895 |
fileobj = StringIO() |
| 896 |
self.assertRaises(TypeError, csv.DictWriter, fileobj) |
| 897 |
|
| 898 |
def test_write_fields_not_in_fieldnames(self): |
| 899 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 900 |
writer = csv.DictWriter(fileobj, fieldnames = ["f1", "f2", "f3"]) |
| 901 |
# Of special note is the non-string key (issue 19449) |
| 902 |
with self.assertRaises(ValueError) as cx: |
| 903 |
writer.writerow({"f4": 10, "f2": "spam", 1: "abc"}) |
| 904 |
exception = str(cx.exception) |
| 905 |
self.assertIn("fieldnames", exception) |
| 906 |
self.assertIn("'f4'", exception) |
| 907 |
self.assertNotIn("'f2'", exception) |
| 908 |
self.assertIn("1", exception) |
| 909 |
|
| 910 |
def test_typo_in_extrasaction_raises_error(self): |
| 911 |
fileobj = StringIO() |
| 912 |
self.assertRaises(ValueError, csv.DictWriter, fileobj, ['f1', 'f2'], |
| 913 |
extrasaction="raised") |
| 914 |
|
| 915 |
def test_write_field_not_in_field_names_raise(self): |
| 916 |
fileobj = StringIO() |
| 917 |
writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="raise") |
| 918 |
dictrow = {'f0': 0, 'f1': 1, 'f2': 2, 'f3': 3} |
| 919 |
self.assertRaises(ValueError, csv.DictWriter.writerow, writer, dictrow) |
| 920 |
|
| 921 |
# see bpo-44512 (differently cased 'raise' should not result in 'ignore') |
| 922 |
writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="RAISE") |
| 923 |
self.assertRaises(ValueError, csv.DictWriter.writerow, writer, dictrow) |
| 924 |
|
| 925 |
def test_write_field_not_in_field_names_ignore(self): |
| 926 |
fileobj = StringIO() |
| 927 |
writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="ignore") |
| 928 |
dictrow = {'f0': 0, 'f1': 1, 'f2': 2, 'f3': 3} |
| 929 |
csv.DictWriter.writerow(writer, dictrow) |
| 930 |
self.assertEqual(fileobj.getvalue(), "1,2\r\n") |
| 931 |
|
| 932 |
# bpo-44512 |
| 933 |
writer = csv.DictWriter(fileobj, ['f1', 'f2'], extrasaction="IGNORE") |
| 934 |
csv.DictWriter.writerow(writer, dictrow) |
| 935 |
|
| 936 |
def test_dict_reader_fieldnames_accepts_iter(self): |
| 937 |
fieldnames = ["a", "b", "c"] |
| 938 |
f = StringIO() |
| 939 |
reader = csv.DictReader(f, iter(fieldnames)) |
| 940 |
self.assertEqual(reader.fieldnames, fieldnames) |
| 941 |
|
| 942 |
def test_dict_reader_fieldnames_accepts_list(self): |
| 943 |
fieldnames = ["a", "b", "c"] |
| 944 |
f = StringIO() |
| 945 |
reader = csv.DictReader(f, fieldnames) |
| 946 |
self.assertEqual(reader.fieldnames, fieldnames) |
| 947 |
|
| 948 |
def test_dict_reader_set_fieldnames(self): |
| 949 |
fieldnames = ["a", "b", "c"] |
| 950 |
f = StringIO() |
| 951 |
reader = csv.DictReader(f) |
| 952 |
self.assertIsNone(reader.fieldnames) |
| 953 |
reader.fieldnames = fieldnames |
| 954 |
self.assertEqual(reader.fieldnames, fieldnames) |
| 955 |
|
| 956 |
def test_dict_writer_fieldnames_rejects_iter(self): |
| 957 |
fieldnames = ["a", "b", "c"] |
| 958 |
f = StringIO() |
| 959 |
writer = csv.DictWriter(f, iter(fieldnames)) |
| 960 |
self.assertEqual(writer.fieldnames, fieldnames) |
| 961 |
|
| 962 |
def test_dict_writer_fieldnames_accepts_list(self): |
| 963 |
fieldnames = ["a", "b", "c"] |
| 964 |
f = StringIO() |
| 965 |
writer = csv.DictWriter(f, fieldnames) |
| 966 |
self.assertEqual(writer.fieldnames, fieldnames) |
| 967 |
|
| 968 |
def test_dict_reader_fieldnames_is_optional(self): |
| 969 |
f = StringIO() |
| 970 |
reader = csv.DictReader(f, fieldnames=None) |
| 971 |
self.assertIsNone(reader.fieldnames) |
| 972 |
|
| 973 |
def test_read_dict_fields(self): |
| 974 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 975 |
fileobj.write("1,2,abc\r\n") |
| 976 |
fileobj.seek(0) |
| 977 |
reader = csv.DictReader(fileobj, |
| 978 |
fieldnames=["f1", "f2", "f3"]) |
| 979 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) |
| 980 |
|
| 981 |
def test_read_dict_no_fieldnames(self): |
| 982 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 983 |
fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") |
| 984 |
fileobj.seek(0) |
| 985 |
reader = csv.DictReader(fileobj) |
| 986 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) |
| 987 |
self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) |
| 988 |
|
| 989 |
# Two test cases to make sure existing ways of implicitly setting |
| 990 |
# fieldnames continue to work. Both arise from discussion in issue3436. |
| 991 |
def test_read_dict_fieldnames_from_file(self): |
| 992 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 993 |
fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") |
| 994 |
fileobj.seek(0) |
| 995 |
reader = csv.DictReader(fileobj, |
| 996 |
fieldnames=next(csv.reader(fileobj))) |
| 997 |
self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) |
| 998 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', "f3": 'abc'}) |
| 999 |
|
| 1000 |
def test_read_dict_fieldnames_chain(self): |
| 1001 |
import itertools |
| 1002 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 1003 |
fileobj.write("f1,f2,f3\r\n1,2,abc\r\n") |
| 1004 |
fileobj.seek(0) |
| 1005 |
reader = csv.DictReader(fileobj) |
| 1006 |
first = next(reader) |
| 1007 |
for row in itertools.chain([first], reader): |
| 1008 |
self.assertEqual(reader.fieldnames, ["f1", "f2", "f3"]) |
| 1009 |
self.assertEqual(row, {"f1": '1', "f2": '2', "f3": 'abc'}) |
| 1010 |
|
| 1011 |
def test_read_long(self): |
| 1012 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 1013 |
fileobj.write("1,2,abc,4,5,6\r\n") |
| 1014 |
fileobj.seek(0) |
| 1015 |
reader = csv.DictReader(fileobj, |
| 1016 |
fieldnames=["f1", "f2"]) |
| 1017 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', |
| 1018 |
None: ["abc", "4", "5", "6"]}) |
| 1019 |
|
| 1020 |
def test_read_long_with_rest(self): |
| 1021 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 1022 |
fileobj.write("1,2,abc,4,5,6\r\n") |
| 1023 |
fileobj.seek(0) |
| 1024 |
reader = csv.DictReader(fileobj, |
| 1025 |
fieldnames=["f1", "f2"], restkey="_rest") |
| 1026 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', |
| 1027 |
"_rest": ["abc", "4", "5", "6"]}) |
| 1028 |
|
| 1029 |
def test_read_long_with_rest_no_fieldnames(self): |
| 1030 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 1031 |
fileobj.write("f1,f2\r\n1,2,abc,4,5,6\r\n") |
| 1032 |
fileobj.seek(0) |
| 1033 |
reader = csv.DictReader(fileobj, restkey="_rest") |
| 1034 |
self.assertEqual(reader.fieldnames, ["f1", "f2"]) |
| 1035 |
self.assertEqual(next(reader), {"f1": '1', "f2": '2', |
| 1036 |
"_rest": ["abc", "4", "5", "6"]}) |
| 1037 |
|
| 1038 |
def test_read_short(self): |
| 1039 |
with TemporaryFile("w+", encoding="utf-8") as fileobj: |
| 1040 |
fileobj.write("1,2,abc,4,5,6\r\n1,2,abc\r\n") |
| 1041 |
fileobj.seek(0) |
| 1042 |
reader = csv.DictReader(fileobj, |
| 1043 |
fieldnames="1 2 3 4 5 6".split(), |
| 1044 |
restval="DEFAULT") |
| 1045 |
self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', |
| 1046 |
"4": '4', "5": '5', "6": '6'}) |
| 1047 |
self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', |
| 1048 |
"4": 'DEFAULT', "5": 'DEFAULT', |
| 1049 |
"6": 'DEFAULT'}) |
| 1050 |
|
| 1051 |
def test_read_multi(self): |
| 1052 |
sample = [ |
| 1053 |
'2147483648,43.0e12,17,abc,def\r\n', |
| 1054 |
'147483648,43.0e2,17,abc,def\r\n', |
| 1055 |
'47483648,43.0,170,abc,def\r\n' |
| 1056 |
] |
| 1057 |
|
| 1058 |
reader = csv.DictReader(sample, |
| 1059 |
fieldnames="i1 float i2 s1 s2".split()) |
| 1060 |
self.assertEqual(next(reader), {"i1": '2147483648', |
| 1061 |
"float": '43.0e12', |
| 1062 |
"i2": '17', |
| 1063 |
"s1": 'abc', |
| 1064 |
"s2": 'def'}) |
| 1065 |
|
| 1066 |
def test_read_with_blanks(self): |
| 1067 |
reader = csv.DictReader(["1,2,abc,4,5,6\r\n","\r\n", |
| 1068 |
"1,2,abc,4,5,6\r\n"], |
| 1069 |
fieldnames="1 2 3 4 5 6".split()) |
| 1070 |
self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', |
| 1071 |
"4": '4', "5": '5', "6": '6'}) |
| 1072 |
self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', |
| 1073 |
"4": '4', "5": '5', "6": '6'}) |
| 1074 |
|
| 1075 |
def test_read_semi_sep(self): |
| 1076 |
reader = csv.DictReader(["1;2;abc;4;5;6\r\n"], |
| 1077 |
fieldnames="1 2 3 4 5 6".split(), |
| 1078 |
delimiter=';') |
| 1079 |
self.assertEqual(next(reader), {"1": '1', "2": '2', "3": 'abc', |
| 1080 |
"4": '4', "5": '5', "6": '6'}) |
| 1081 |
|
| 1082 |
class TestArrayWrites(unittest.TestCase): |
| 1083 |
def test_int_write(self): |
| 1084 |
import array |
| 1085 |
contents = [(20-i) for i in range(20)] |
| 1086 |
a = array.array('i', contents) |
| 1087 |
|
| 1088 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 1089 |
writer = csv.writer(fileobj, dialect="excel") |
| 1090 |
writer.writerow(a) |
| 1091 |
expected = ",".join([str(i) for i in a])+"\r\n" |
| 1092 |
fileobj.seek(0) |
| 1093 |
self.assertEqual(fileobj.read(), expected) |
| 1094 |
|
| 1095 |
def test_double_write(self): |
| 1096 |
import array |
| 1097 |
contents = [(20-i)*0.1 for i in range(20)] |
| 1098 |
a = array.array('d', contents) |
| 1099 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 1100 |
writer = csv.writer(fileobj, dialect="excel") |
| 1101 |
writer.writerow(a) |
| 1102 |
expected = ",".join([str(i) for i in a])+"\r\n" |
| 1103 |
fileobj.seek(0) |
| 1104 |
self.assertEqual(fileobj.read(), expected) |
| 1105 |
|
| 1106 |
def test_float_write(self): |
| 1107 |
import array |
| 1108 |
contents = [(20-i)*0.1 for i in range(20)] |
| 1109 |
a = array.array('f', contents) |
| 1110 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 1111 |
writer = csv.writer(fileobj, dialect="excel") |
| 1112 |
writer.writerow(a) |
| 1113 |
expected = ",".join([str(i) for i in a])+"\r\n" |
| 1114 |
fileobj.seek(0) |
| 1115 |
self.assertEqual(fileobj.read(), expected) |
| 1116 |
|
| 1117 |
def test_char_write(self): |
| 1118 |
import array, string |
| 1119 |
a = array.array('w', string.ascii_letters) |
| 1120 |
|
| 1121 |
with TemporaryFile("w+", encoding="utf-8", newline='') as fileobj: |
| 1122 |
writer = csv.writer(fileobj, dialect="excel") |
| 1123 |
writer.writerow(a) |
| 1124 |
expected = ",".join(a)+"\r\n" |
| 1125 |
fileobj.seek(0) |
| 1126 |
self.assertEqual(fileobj.read(), expected) |
| 1127 |
|
| 1128 |
class TestDialectValidity(unittest.TestCase): |
| 1129 |
def test_quoting(self): |
| 1130 |
class mydialect(csv.Dialect): |
| 1131 |
delimiter = ";" |
| 1132 |
escapechar = '\\' |
| 1133 |
doublequote = False |
| 1134 |
skipinitialspace = True |
| 1135 |
lineterminator = '\r\n' |
| 1136 |
quoting = csv.QUOTE_NONE |
| 1137 |
d = mydialect() |
| 1138 |
self.assertEqual(d.quoting, csv.QUOTE_NONE) |
| 1139 |
|
| 1140 |
mydialect.quoting = None |
| 1141 |
self.assertRaises(csv.Error, mydialect) |
| 1142 |
|
| 1143 |
mydialect.quoting = 42 |
| 1144 |
with self.assertRaises(csv.Error) as cm: |
| 1145 |
mydialect() |
| 1146 |
self.assertEqual(str(cm.exception), |
| 1147 |
'bad "quoting" value') |
| 1148 |
|
| 1149 |
mydialect.doublequote = True |
| 1150 |
mydialect.quoting = csv.QUOTE_ALL |
| 1151 |
mydialect.quotechar = '"' |
| 1152 |
d = mydialect() |
| 1153 |
self.assertEqual(d.quoting, csv.QUOTE_ALL) |
| 1154 |
self.assertEqual(d.quotechar, '"') |
| 1155 |
self.assertTrue(d.doublequote) |
| 1156 |
|
| 1157 |
mydialect.quotechar = "" |
| 1158 |
with self.assertRaises(csv.Error) as cm: |
| 1159 |
mydialect() |
| 1160 |
self.assertEqual(str(cm.exception), |
| 1161 |
'"quotechar" must be a unicode character or None, ' |
| 1162 |
'not a string of length 0') |
| 1163 |
|
| 1164 |
mydialect.quotechar = "''" |
| 1165 |
with self.assertRaises(csv.Error) as cm: |
| 1166 |
mydialect() |
| 1167 |
self.assertEqual(str(cm.exception), |
| 1168 |
'"quotechar" must be a unicode character or None, ' |
| 1169 |
'not a string of length 2') |
| 1170 |
|
| 1171 |
mydialect.quotechar = 4 |
| 1172 |
with self.assertRaises(csv.Error) as cm: |
| 1173 |
mydialect() |
| 1174 |
self.assertEqual(str(cm.exception), |
| 1175 |
'"quotechar" must be a unicode character or None, ' |
| 1176 |
'not int') |
| 1177 |
|
| 1178 |
def test_delimiter(self): |
| 1179 |
class mydialect(csv.Dialect): |
| 1180 |
delimiter = ";" |
| 1181 |
escapechar = '\\' |
| 1182 |
doublequote = False |
| 1183 |
skipinitialspace = True |
| 1184 |
lineterminator = '\r\n' |
| 1185 |
quoting = csv.QUOTE_NONE |
| 1186 |
d = mydialect() |
| 1187 |
self.assertEqual(d.delimiter, ";") |
| 1188 |
|
| 1189 |
mydialect.delimiter = ":::" |
| 1190 |
with self.assertRaises(csv.Error) as cm: |
| 1191 |
mydialect() |
| 1192 |
self.assertEqual(str(cm.exception), |
| 1193 |
'"delimiter" must be a unicode character, ' |
| 1194 |
'not a string of length 3') |
| 1195 |
|
| 1196 |
mydialect.delimiter = "" |
| 1197 |
with self.assertRaises(csv.Error) as cm: |
| 1198 |
mydialect() |
| 1199 |
self.assertEqual(str(cm.exception), |
| 1200 |
'"delimiter" must be a unicode character, not a string of length 0') |
| 1201 |
|
| 1202 |
mydialect.delimiter = b"," |
| 1203 |
with self.assertRaises(csv.Error) as cm: |
| 1204 |
mydialect() |
| 1205 |
self.assertEqual(str(cm.exception), |
| 1206 |
'"delimiter" must be a unicode character, not bytes') |
| 1207 |
|
| 1208 |
mydialect.delimiter = 4 |
| 1209 |
with self.assertRaises(csv.Error) as cm: |
| 1210 |
mydialect() |
| 1211 |
self.assertEqual(str(cm.exception), |
| 1212 |
'"delimiter" must be a unicode character, not int') |
| 1213 |
|
| 1214 |
mydialect.delimiter = None |
| 1215 |
with self.assertRaises(csv.Error) as cm: |
| 1216 |
mydialect() |
| 1217 |
self.assertEqual(str(cm.exception), |
| 1218 |
'"delimiter" must be a unicode character, not NoneType') |
| 1219 |
|
| 1220 |
def test_escapechar(self): |
| 1221 |
class mydialect(csv.Dialect): |
| 1222 |
delimiter = ";" |
| 1223 |
escapechar = '\\' |
| 1224 |
doublequote = False |
| 1225 |
skipinitialspace = True |
| 1226 |
lineterminator = '\r\n' |
| 1227 |
quoting = csv.QUOTE_NONE |
| 1228 |
d = mydialect() |
| 1229 |
self.assertEqual(d.escapechar, "\\") |
| 1230 |
|
| 1231 |
mydialect.escapechar = "" |
| 1232 |
with self.assertRaises(csv.Error) as cm: |
| 1233 |
mydialect() |
| 1234 |
self.assertEqual(str(cm.exception), |
| 1235 |
'"escapechar" must be a unicode character or None, ' |
| 1236 |
'not a string of length 0') |
| 1237 |
|
| 1238 |
mydialect.escapechar = "**" |
| 1239 |
with self.assertRaises(csv.Error) as cm: |
| 1240 |
mydialect() |
| 1241 |
self.assertEqual(str(cm.exception), |
| 1242 |
'"escapechar" must be a unicode character or None, ' |
| 1243 |
'not a string of length 2') |
| 1244 |
|
| 1245 |
mydialect.escapechar = b"*" |
| 1246 |
with self.assertRaises(csv.Error) as cm: |
| 1247 |
mydialect() |
| 1248 |
self.assertEqual(str(cm.exception), |
| 1249 |
'"escapechar" must be a unicode character or None, ' |
| 1250 |
'not bytes') |
| 1251 |
|
| 1252 |
mydialect.escapechar = 4 |
| 1253 |
with self.assertRaises(csv.Error) as cm: |
| 1254 |
mydialect() |
| 1255 |
self.assertEqual(str(cm.exception), |
| 1256 |
'"escapechar" must be a unicode character or None, ' |
| 1257 |
'not int') |
| 1258 |
|
| 1259 |
def test_lineterminator(self): |
| 1260 |
class mydialect(csv.Dialect): |
| 1261 |
delimiter = ";" |
| 1262 |
escapechar = '\\' |
| 1263 |
doublequote = False |
| 1264 |
skipinitialspace = True |
| 1265 |
lineterminator = '\r\n' |
| 1266 |
quoting = csv.QUOTE_NONE |
| 1267 |
d = mydialect() |
| 1268 |
self.assertEqual(d.lineterminator, '\r\n') |
| 1269 |
|
| 1270 |
mydialect.lineterminator = ":::" |
| 1271 |
d = mydialect() |
| 1272 |
self.assertEqual(d.lineterminator, ":::") |
| 1273 |
|
| 1274 |
mydialect.lineterminator = 4 |
| 1275 |
with self.assertRaises(csv.Error) as cm: |
| 1276 |
mydialect() |
| 1277 |
self.assertEqual(str(cm.exception), |
| 1278 |
'"lineterminator" must be a string, not int') |
| 1279 |
|
| 1280 |
mydialect.lineterminator = None |
| 1281 |
with self.assertRaises(csv.Error) as cm: |
| 1282 |
mydialect() |
| 1283 |
self.assertEqual(str(cm.exception), |
| 1284 |
'"lineterminator" must be a string, not NoneType') |
| 1285 |
|
| 1286 |
def test_invalid_chars(self): |
| 1287 |
def create_invalid(field_name, value, **kwargs): |
| 1288 |
class mydialect(csv.Dialect): |
| 1289 |
delimiter = ',' |
| 1290 |
quoting = csv.QUOTE_ALL |
| 1291 |
quotechar = '"' |
| 1292 |
lineterminator = '\r\n' |
| 1293 |
setattr(mydialect, field_name, value) |
| 1294 |
for field_name, value in kwargs.items(): |
| 1295 |
setattr(mydialect, field_name, value) |
| 1296 |
d = mydialect() |
| 1297 |
|
| 1298 |
for field_name in ("delimiter", "escapechar", "quotechar"): |
| 1299 |
with self.subTest(field_name=field_name): |
| 1300 |
self.assertRaises(csv.Error, create_invalid, field_name, "") |
| 1301 |
self.assertRaises(csv.Error, create_invalid, field_name, "abc") |
| 1302 |
self.assertRaises(csv.Error, create_invalid, field_name, b'x') |
| 1303 |
self.assertRaises(csv.Error, create_invalid, field_name, 5) |
| 1304 |
self.assertRaises(ValueError, create_invalid, field_name, "\n") |
| 1305 |
self.assertRaises(ValueError, create_invalid, field_name, "\r") |
| 1306 |
if field_name != "delimiter": |
| 1307 |
self.assertRaises(ValueError, create_invalid, field_name, " ", |
| 1308 |
skipinitialspace=True) |
| 1309 |
|
| 1310 |
|
| 1311 |
class TestSniffer(unittest.TestCase): |
| 1312 |
sample1 = """\ |
| 1313 |
Harry's, Arlington Heights, IL, 2/1/03, Kimi Hayes |
| 1314 |
Shark City, Glendale Heights, IL, 12/28/02, Prezence |
| 1315 |
Tommy's Place, Blue Island, IL, 12/28/02, Blue Sunday/White Crow |
| 1316 |
Stonecutters Seafood and Chop House, Lemont, IL, 12/19/02, Week Back |
| 1317 |
""" |
| 1318 |
sample2 = """\ |
| 1319 |
'Harry''s':'Arlington Heights':'IL':'2/1/03':'Kimi Hayes' |
| 1320 |
'Shark City':'Glendale Heights':'IL':'12/28/02':'Prezence' |
| 1321 |
'Tommy''s Place':'Blue Island':'IL':'12/28/02':'Blue Sunday/White Crow' |
| 1322 |
'Stonecutters ''Seafood'' and Chop House':'Lemont':'IL':'12/19/02':'Week Back' |
| 1323 |
""" |
| 1324 |
header1 = '''\ |
| 1325 |
"venue","city","state","date","performers" |
| 1326 |
''' |
| 1327 |
sample3 = '''\ |
| 1328 |
05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 |
| 1329 |
05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 |
| 1330 |
05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03 |
| 1331 |
''' |
| 1332 |
|
| 1333 |
sample4 = '''\ |
| 1334 |
2147483648;43.0e12;17;abc;def |
| 1335 |
147483648;43.0e2;17;abc;def |
| 1336 |
47483648;43.0;170;abc;def |
| 1337 |
''' |
| 1338 |
|
| 1339 |
sample5 = "aaa\tbbb\r\nAAA\t\r\nBBB\t\r\n" |
| 1340 |
sample6 = "a|b|c\r\nd|e|f\r\n" |
| 1341 |
sample7 = "'a'|'b'|'c'\r\n'd'|e|f\r\n" |
| 1342 |
|
| 1343 |
# Issue 18155: Use a delimiter that is a special char to regex: |
| 1344 |
|
| 1345 |
header2 = '''\ |
| 1346 |
"venue"+"city"+"state"+"date"+"performers" |
| 1347 |
''' |
| 1348 |
sample8 = """\ |
| 1349 |
Harry's+ Arlington Heights+ IL+ 2/1/03+ Kimi Hayes |
| 1350 |
Shark City+ Glendale Heights+ IL+ 12/28/02+ Prezence |
| 1351 |
Tommy's Place+ Blue Island+ IL+ 12/28/02+ Blue Sunday/White Crow |
| 1352 |
Stonecutters Seafood and Chop House+ Lemont+ IL+ 12/19/02+ Week Back |
| 1353 |
""" |
| 1354 |
sample9 = """\ |
| 1355 |
'Harry''s'+ Arlington Heights'+ 'IL'+ '2/1/03'+ 'Kimi Hayes' |
| 1356 |
'Shark City'+ Glendale Heights'+' IL'+ '12/28/02'+ 'Prezence' |
| 1357 |
'Tommy''s Place'+ Blue Island'+ 'IL'+ '12/28/02'+ 'Blue Sunday/White Crow' |
| 1358 |
'Stonecutters ''Seafood'' and Chop House'+ 'Lemont'+ 'IL'+ '12/19/02'+ 'Week Back' |
| 1359 |
""" |
| 1360 |
|
| 1361 |
sample10 = dedent(""" |
| 1362 |
abc,def |
| 1363 |
ghijkl,mno |
| 1364 |
ghi,jkl |
| 1365 |
""") |
| 1366 |
|
| 1367 |
sample11 = dedent(""" |
| 1368 |
abc,def |
| 1369 |
ghijkl,mnop |
| 1370 |
ghi,jkl |
| 1371 |
""") |
| 1372 |
|
| 1373 |
sample12 = dedent(""""time","forces" |
| 1374 |
1,1.5 |
| 1375 |
0.5,5+0j |
| 1376 |
0,0 |
| 1377 |
1+1j,6 |
| 1378 |
""") |
| 1379 |
|
| 1380 |
sample13 = dedent(""""time","forces" |
| 1381 |
0,0 |
| 1382 |
1,2 |
| 1383 |
a,b |
| 1384 |
""") |
| 1385 |
|
| 1386 |
sample14 = """\ |
| 1387 |
abc\0def |
| 1388 |
ghijkl\0mno |
| 1389 |
ghi\0jkl |
| 1390 |
""" |
| 1391 |
|
| 1392 |
sample15 = "\n\n\n" |
| 1393 |
sample16 = "abc\ndef\nghi" |
| 1394 |
|
| 1395 |
sample17 = ["letter,offset"] |
| 1396 |
sample17.extend(f"{chr(ord('a') + i)},{i}" for i in range(20)) |
| 1397 |
sample17.append("v,twenty_one") # 'u' was skipped |
| 1398 |
sample17 = '\n'.join(sample17) |
| 1399 |
|
| 1400 |
sample18 = ["letter,offset"] |
| 1401 |
sample18.extend(f"{chr(ord('a') + i)},{i}" for i in range(21)) |
| 1402 |
sample18.append("v,twenty_one") # 'u' was not skipped |
| 1403 |
sample18 = '\n'.join(sample18) |
| 1404 |
|
| 1405 |
def test_issue43625(self): |
| 1406 |
sniffer = csv.Sniffer() |
| 1407 |
self.assertTrue(sniffer.has_header(self.sample12)) |
| 1408 |
self.assertFalse(sniffer.has_header(self.sample13)) |
| 1409 |
|
| 1410 |
def test_has_header_strings(self): |
| 1411 |
"More to document existing (unexpected?) behavior than anything else." |
| 1412 |
sniffer = csv.Sniffer() |
| 1413 |
self.assertFalse(sniffer.has_header(self.sample10)) |
| 1414 |
self.assertFalse(sniffer.has_header(self.sample11)) |
| 1415 |
|
| 1416 |
def test_has_header(self): |
| 1417 |
sniffer = csv.Sniffer() |
| 1418 |
self.assertIs(sniffer.has_header(self.sample1), False) |
| 1419 |
self.assertIs(sniffer.has_header(self.header1 + self.sample1), True) |
| 1420 |
|
| 1421 |
def test_has_header_regex_special_delimiter(self): |
| 1422 |
sniffer = csv.Sniffer() |
| 1423 |
self.assertIs(sniffer.has_header(self.sample8), False) |
| 1424 |
self.assertIs(sniffer.has_header(self.header2 + self.sample8), True) |
| 1425 |
|
| 1426 |
def test_has_header_checks_20_rows(self): |
| 1427 |
sniffer = csv.Sniffer() |
| 1428 |
self.assertFalse(sniffer.has_header(self.sample17)) |
| 1429 |
self.assertTrue(sniffer.has_header(self.sample18)) |
| 1430 |
|
| 1431 |
def test_guess_quote_and_delimiter(self): |
| 1432 |
sniffer = csv.Sniffer() |
| 1433 |
for header in (";'123;4';", "'123;4';", ";'123;4'", "'123;4'"): |
| 1434 |
with self.subTest(header): |
| 1435 |
dialect = sniffer.sniff(header, ",;") |
| 1436 |
self.assertEqual(dialect.delimiter, ';') |
| 1437 |
self.assertEqual(dialect.quotechar, "'") |
| 1438 |
self.assertIs(dialect.doublequote, False) |
| 1439 |
self.assertIs(dialect.skipinitialspace, False) |
| 1440 |
|
| 1441 |
def test_sniff(self): |
| 1442 |
sniffer = csv.Sniffer() |
| 1443 |
dialect = sniffer.sniff(self.sample1) |
| 1444 |
self.assertEqual(dialect.delimiter, ",") |
| 1445 |
self.assertEqual(dialect.quotechar, '"') |
| 1446 |
self.assertIs(dialect.skipinitialspace, True) |
| 1447 |
|
| 1448 |
dialect = sniffer.sniff(self.sample2) |
| 1449 |
self.assertEqual(dialect.delimiter, ":") |
| 1450 |
self.assertEqual(dialect.quotechar, "'") |
| 1451 |
self.assertIs(dialect.skipinitialspace, False) |
| 1452 |
|
| 1453 |
def test_delimiters(self): |
| 1454 |
sniffer = csv.Sniffer() |
| 1455 |
dialect = sniffer.sniff(self.sample3) |
| 1456 |
# given that all three lines in sample3 are equal, |
| 1457 |
# I think that any character could have been 'guessed' as the |
| 1458 |
# delimiter, depending on dictionary order |
| 1459 |
self.assertIn(dialect.delimiter, self.sample3) |
| 1460 |
dialect = sniffer.sniff(self.sample3, delimiters="?,") |
| 1461 |
self.assertEqual(dialect.delimiter, "?") |
| 1462 |
dialect = sniffer.sniff(self.sample3, delimiters="/,") |
| 1463 |
self.assertEqual(dialect.delimiter, "/") |
| 1464 |
dialect = sniffer.sniff(self.sample4) |
| 1465 |
self.assertEqual(dialect.delimiter, ";") |
| 1466 |
dialect = sniffer.sniff(self.sample5) |
| 1467 |
self.assertEqual(dialect.delimiter, "\t") |
| 1468 |
dialect = sniffer.sniff(self.sample6) |
| 1469 |
self.assertEqual(dialect.delimiter, "|") |
| 1470 |
dialect = sniffer.sniff(self.sample7) |
| 1471 |
self.assertEqual(dialect.delimiter, "|") |
| 1472 |
self.assertEqual(dialect.quotechar, "'") |
| 1473 |
dialect = sniffer.sniff(self.sample8) |
| 1474 |
self.assertEqual(dialect.delimiter, '+') |
| 1475 |
dialect = sniffer.sniff(self.sample9) |
| 1476 |
self.assertEqual(dialect.delimiter, '+') |
| 1477 |
self.assertEqual(dialect.quotechar, "'") |
| 1478 |
dialect = sniffer.sniff(self.sample14) |
| 1479 |
self.assertEqual(dialect.delimiter, '\0') |
| 1480 |
self.assertRaisesRegex(csv.Error, "Could not determine delimiter", |
| 1481 |
sniffer.sniff, self.sample15) |
| 1482 |
self.assertRaisesRegex(csv.Error, "Could not determine delimiter", |
| 1483 |
sniffer.sniff, self.sample16) |
| 1484 |
|
| 1485 |
def test_doublequote(self): |
| 1486 |
sniffer = csv.Sniffer() |
| 1487 |
dialect = sniffer.sniff(self.header1) |
| 1488 |
self.assertFalse(dialect.doublequote) |
| 1489 |
dialect = sniffer.sniff(self.header2) |
| 1490 |
self.assertFalse(dialect.doublequote) |
| 1491 |
dialect = sniffer.sniff(self.sample2) |
| 1492 |
self.assertTrue(dialect.doublequote) |
| 1493 |
dialect = sniffer.sniff(self.sample8) |
| 1494 |
self.assertFalse(dialect.doublequote) |
| 1495 |
dialect = sniffer.sniff(self.sample9) |
| 1496 |
self.assertTrue(dialect.doublequote) |
| 1497 |
|
| 1498 |
class NUL: |
| 1499 |
def write(s, *args): |
| 1500 |
pass |
| 1501 |
writelines = write |
| 1502 |
|
| 1503 |
@unittest.skipUnless(hasattr(sys, "gettotalrefcount"), |
| 1504 |
'requires sys.gettotalrefcount()') |
| 1505 |
class TestLeaks(unittest.TestCase): |
| 1506 |
def test_create_read(self): |
| 1507 |
delta = 0 |
| 1508 |
lastrc = sys.gettotalrefcount() |
| 1509 |
for i in range(20): |
| 1510 |
gc.collect() |
| 1511 |
self.assertEqual(gc.garbage, []) |
| 1512 |
rc = sys.gettotalrefcount() |
| 1513 |
csv.reader(["a,b,c\r\n"]) |
| 1514 |
csv.reader(["a,b,c\r\n"]) |
| 1515 |
csv.reader(["a,b,c\r\n"]) |
| 1516 |
delta = rc-lastrc |
| 1517 |
lastrc = rc |
| 1518 |
# if csv.reader() leaks, last delta should be 3 or more |
| 1519 |
self.assertLess(delta, 3) |
| 1520 |
|
| 1521 |
def test_create_write(self): |
| 1522 |
delta = 0 |
| 1523 |
lastrc = sys.gettotalrefcount() |
| 1524 |
s = NUL() |
| 1525 |
for i in range(20): |
| 1526 |
gc.collect() |
| 1527 |
self.assertEqual(gc.garbage, []) |
| 1528 |
rc = sys.gettotalrefcount() |
| 1529 |
csv.writer(s) |
| 1530 |
csv.writer(s) |
| 1531 |
csv.writer(s) |
| 1532 |
delta = rc-lastrc |
| 1533 |
lastrc = rc |
| 1534 |
# if csv.writer() leaks, last delta should be 3 or more |
| 1535 |
self.assertLess(delta, 3) |
| 1536 |
|
| 1537 |
def test_read(self): |
| 1538 |
delta = 0 |
| 1539 |
rows = ["a,b,c\r\n"]*5 |
| 1540 |
lastrc = sys.gettotalrefcount() |
| 1541 |
for i in range(20): |
| 1542 |
gc.collect() |
| 1543 |
self.assertEqual(gc.garbage, []) |
| 1544 |
rc = sys.gettotalrefcount() |
| 1545 |
rdr = csv.reader(rows) |
| 1546 |
for row in rdr: |
| 1547 |
pass |
| 1548 |
delta = rc-lastrc |
| 1549 |
lastrc = rc |
| 1550 |
# if reader leaks during read, delta should be 5 or more |
| 1551 |
self.assertLess(delta, 5) |
| 1552 |
|
| 1553 |
def test_write(self): |
| 1554 |
delta = 0 |
| 1555 |
rows = [[1,2,3]]*5 |
| 1556 |
s = NUL() |
| 1557 |
lastrc = sys.gettotalrefcount() |
| 1558 |
for i in range(20): |
| 1559 |
gc.collect() |
| 1560 |
self.assertEqual(gc.garbage, []) |
| 1561 |
rc = sys.gettotalrefcount() |
| 1562 |
writer = csv.writer(s) |
| 1563 |
for row in rows: |
| 1564 |
writer.writerow(row) |
| 1565 |
delta = rc-lastrc |
| 1566 |
lastrc = rc |
| 1567 |
# if writer leaks during write, last delta should be 5 or more |
| 1568 |
self.assertLess(delta, 5) |
| 1569 |
|
| 1570 |
class TestUnicode(unittest.TestCase): |
| 1571 |
|
| 1572 |
names = ["Martin von Löwis", |
| 1573 |
"Marc André Lemburg", |
| 1574 |
"Guido van Rossum", |
| 1575 |
"François Pinard"] |
| 1576 |
|
| 1577 |
def test_unicode_read(self): |
| 1578 |
with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: |
| 1579 |
fileobj.write(",".join(self.names) + "\r\n") |
| 1580 |
fileobj.seek(0) |
| 1581 |
reader = csv.reader(fileobj) |
| 1582 |
self.assertEqual(list(reader), [self.names]) |
| 1583 |
|
| 1584 |
|
| 1585 |
def test_unicode_write(self): |
| 1586 |
with TemporaryFile("w+", newline='', encoding="utf-8") as fileobj: |
| 1587 |
writer = csv.writer(fileobj) |
| 1588 |
writer.writerow(self.names) |
| 1589 |
expected = ",".join(self.names)+"\r\n" |
| 1590 |
fileobj.seek(0) |
| 1591 |
self.assertEqual(fileobj.read(), expected) |
| 1592 |
|
| 1593 |
class KeyOrderingTest(unittest.TestCase): |
| 1594 |
|
| 1595 |
def test_ordering_for_the_dict_reader_and_writer(self): |
| 1596 |
resultset = set() |
| 1597 |
for keys in permutations("abcde"): |
| 1598 |
with TemporaryFile('w+', newline='', encoding="utf-8") as fileobject: |
| 1599 |
dw = csv.DictWriter(fileobject, keys) |
| 1600 |
dw.writeheader() |
| 1601 |
fileobject.seek(0) |
| 1602 |
dr = csv.DictReader(fileobject) |
| 1603 |
kt = tuple(dr.fieldnames) |
| 1604 |
self.assertEqual(keys, kt) |
| 1605 |
resultset.add(kt) |
| 1606 |
# Final sanity check: were all permutations unique? |
| 1607 |
self.assertEqual(len(resultset), 120, "Key ordering: some key permutations not collected (expected 120)") |
| 1608 |
|
| 1609 |
def test_ordered_dict_reader(self): |
| 1610 |
data = dedent('''\ |
| 1611 |
FirstName,LastName |
| 1612 |
Eric,Idle |
| 1613 |
Graham,Chapman,Over1,Over2 |
| 1614 |
|
| 1615 |
Under1 |
| 1616 |
John,Cleese |
| 1617 |
''').splitlines() |
| 1618 |
|
| 1619 |
self.assertEqual(list(csv.DictReader(data)), |
| 1620 |
[OrderedDict([('FirstName', 'Eric'), ('LastName', 'Idle')]), |
| 1621 |
OrderedDict([('FirstName', 'Graham'), ('LastName', 'Chapman'), |
| 1622 |
(None, ['Over1', 'Over2'])]), |
| 1623 |
OrderedDict([('FirstName', 'Under1'), ('LastName', None)]), |
| 1624 |
OrderedDict([('FirstName', 'John'), ('LastName', 'Cleese')]), |
| 1625 |
]) |
| 1626 |
|
| 1627 |
self.assertEqual(list(csv.DictReader(data, restkey='OtherInfo')), |
| 1628 |
[OrderedDict([('FirstName', 'Eric'), ('LastName', 'Idle')]), |
| 1629 |
OrderedDict([('FirstName', 'Graham'), ('LastName', 'Chapman'), |
| 1630 |
('OtherInfo', ['Over1', 'Over2'])]), |
| 1631 |
OrderedDict([('FirstName', 'Under1'), ('LastName', None)]), |
| 1632 |
OrderedDict([('FirstName', 'John'), ('LastName', 'Cleese')]), |
| 1633 |
]) |
| 1634 |
|
| 1635 |
del data[0] # Remove the header row |
| 1636 |
self.assertEqual(list(csv.DictReader(data, fieldnames=['fname', 'lname'])), |
| 1637 |
[OrderedDict([('fname', 'Eric'), ('lname', 'Idle')]), |
| 1638 |
OrderedDict([('fname', 'Graham'), ('lname', 'Chapman'), |
| 1639 |
(None, ['Over1', 'Over2'])]), |
| 1640 |
OrderedDict([('fname', 'Under1'), ('lname', None)]), |
| 1641 |
OrderedDict([('fname', 'John'), ('lname', 'Cleese')]), |
| 1642 |
]) |
| 1643 |
|
| 1644 |
|
| 1645 |
class MiscTestCase(unittest.TestCase): |
| 1646 |
def test__all__(self): |
| 1647 |
support.check__all__(self, csv, ('csv', '_csv')) |
| 1648 |
|
| 1649 |
@cpython_only |
| 1650 |
def test_lazy_import(self): |
| 1651 |
ensure_lazy_imports("csv", {"re"}) |
| 1652 |
|
| 1653 |
def test_subclassable(self): |
| 1654 |
# issue 44089 |
| 1655 |
class Foo(csv.Error): ... |
| 1656 |
|
| 1657 |
@support.cpython_only |
| 1658 |
def test_disallow_instantiation(self): |
| 1659 |
_csv = import_helper.import_module("_csv") |
| 1660 |
for tp in _csv.Reader, _csv.Writer: |
| 1661 |
with self.subTest(tp=tp): |
| 1662 |
check_disallow_instantiation(self, tp) |
| 1663 |
|
| 1664 |
if __name__ == '__main__': |
| 1665 |
unittest.main() |