1 """Regresssion tests for urllib""" 2 3 import urllib 4 import httplib 5 import unittest 6 import os 7 import sys 8 import mimetools 9 import tempfile 10 import StringIO 11 12 from test import test_support 13 from base64 import b64encode 14 15 16 def hexescape(char): 17 """Escape char as RFC 2396 specifies""" 18 hex_repr = hex(ord(char))[2:].upper() 19 if len(hex_repr) == 1: 20 hex_repr = "0%s" % hex_repr 21 return "%" + hex_repr 22 23 24 class FakeHTTPMixin(object): 25 def fakehttp(self, fakedata): 26 class FakeSocket(StringIO.StringIO): 27 28 def sendall(self, data): 29 FakeHTTPConnection.buf = data 30 31 def makefile(self, *args, **kwds): 32 return self 33 34 def read(self, amt=None): 35 if self.closed: 36 return "" 37 return StringIO.StringIO.read(self, amt) 38 39 def readline(self, length=None): 40 if self.closed: 41 return "" 42 return StringIO.StringIO.readline(self, length) 43 44 class FakeHTTPConnection(httplib.HTTPConnection): 45 46 # buffer to store data for verification in urlopen tests. 47 buf = "" 48 49 def connect(self): 50 self.sock = FakeSocket(fakedata) 51 52 assert httplib.HTTP._connection_class == httplib.HTTPConnection 53 54 httplib.HTTP._connection_class = FakeHTTPConnection 55 56 def unfakehttp(self): 57 httplib.HTTP._connection_class = httplib.HTTPConnection 58 59 60 class urlopen_FileTests(unittest.TestCase): 61 """Test urlopen() opening a temporary file. 62 63 Try to test as much functionality as possible so as to cut down on reliance 64 on connecting to the Net for testing. 65 66 """ 67 68 def setUp(self): 69 """Setup of a temp file to use for testing""" 70 self.text = "test_urllib: %s\n" % self.__class__.__name__ 71 FILE = file(test_support.TESTFN, 'wb') 72 try: 73 FILE.write(self.text) 74 finally: 75 FILE.close() 76 self.pathname = test_support.TESTFN 77 self.returned_obj = urllib.urlopen("file:%s" % self.pathname) 78 79 def tearDown(self): 80 """Shut down the open object""" 81 self.returned_obj.close() 82 os.remove(test_support.TESTFN) 83 84 def test_interface(self): 85 # Make sure object returned by urlopen() has the specified methods 86 for attr in ("read", "readline", "readlines", "fileno", 87 "close", "info", "geturl", "getcode", "__iter__"): 88 self.assertTrue(hasattr(self.returned_obj, attr), 89 "object returned by urlopen() lacks %s attribute" % 90 attr) 91 92 def test_read(self): 93 self.assertEqual(self.text, self.returned_obj.read()) 94 95 def test_readline(self): 96 self.assertEqual(self.text, self.returned_obj.readline()) 97 self.assertEqual('', self.returned_obj.readline(), 98 "calling readline() after exhausting the file did not" 99 " return an empty string") 100 101 def test_readlines(self): 102 lines_list = self.returned_obj.readlines() 103 self.assertEqual(len(lines_list), 1, 104 "readlines() returned the wrong number of lines") 105 self.assertEqual(lines_list[0], self.text, 106 "readlines() returned improper text") 107 108 def test_fileno(self): 109 file_num = self.returned_obj.fileno() 110 self.assertIsInstance(file_num, int, "fileno() did not return an int") 111 self.assertEqual(os.read(file_num, len(self.text)), self.text, 112 "Reading on the file descriptor returned by fileno() " 113 "did not return the expected text") 114 115 def test_close(self): 116 # Test close() by calling it hear and then having it be called again 117 # by the tearDown() method for the test 118 self.returned_obj.close() 119 120 def test_info(self): 121 self.assertIsInstance(self.returned_obj.info(), mimetools.Message) 122 123 def test_geturl(self): 124 self.assertEqual(self.returned_obj.geturl(), self.pathname) 125 126 def test_getcode(self): 127 self.assertEqual(self.returned_obj.getcode(), None) 128 129 def test_iter(self): 130 # Test iterator 131 # Don't need to count number of iterations since test would fail the 132 # instant it returned anything beyond the first line from the 133 # comparison 134 for line in self.returned_obj.__iter__(): 135 self.assertEqual(line, self.text) 136 137 def test_relativelocalfile(self): 138 self.assertRaises(ValueError,urllib.urlopen,'./' + self.pathname) 139 140 class ProxyTests(unittest.TestCase): 141 142 def setUp(self): 143 # Records changes to env vars 144 self.env = test_support.EnvironmentVarGuard() 145 # Delete all proxy related env vars 146 for k in os.environ.keys(): 147 if 'proxy' in k.lower(): 148 self.env.unset(k) 149 150 def tearDown(self): 151 # Restore all proxy related env vars 152 self.env.__exit__() 153 del self.env 154 155 def test_getproxies_environment_keep_no_proxies(self): 156 self.env.set('NO_PROXY', 'localhost') 157 proxies = urllib.getproxies_environment() 158 # getproxies_environment use lowered case truncated (no '_proxy') keys 159 self.assertEqual('localhost', proxies['no']) 160 # List of no_proxies with space. 161 self.env.set('NO_PROXY', 'localhost, anotherdomain.com, newdomain.com') 162 self.assertTrue(urllib.proxy_bypass_environment('anotherdomain.com')) 163 164 165 class urlopen_HttpTests(unittest.TestCase, FakeHTTPMixin): 166 """Test urlopen() opening a fake http connection.""" 167 168 def test_read(self): 169 self.fakehttp('Hello!') 170 try: 171 fp = urllib.urlopen("http://python.org/") 172 self.assertEqual(fp.readline(), 'Hello!') 173 self.assertEqual(fp.readline(), '') 174 self.assertEqual(fp.geturl(), 'http://python.org/') 175 self.assertEqual(fp.getcode(), 200) 176 finally: 177 self.unfakehttp() 178 179 def test_url_fragment(self): 180 # Issue #11703: geturl() omits fragments in the original URL. 181 url = 'http://docs.python.org/library/urllib.html#OK' 182 self.fakehttp('Hello!') 183 try: 184 fp = urllib.urlopen(url) 185 self.assertEqual(fp.geturl(), url) 186 finally: 187 self.unfakehttp() 188 189 def test_read_bogus(self): 190 # urlopen() should raise IOError for many error codes. 191 self.fakehttp('''HTTP/1.1 401 Authentication Required 192 Date: Wed, 02 Jan 2008 03:03:54 GMT 193 Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e 194 Connection: close 195 Content-Type: text/html; charset=iso-8859-1 196 ''') 197 try: 198 self.assertRaises(IOError, urllib.urlopen, "http://python.org/") 199 finally: 200 self.unfakehttp() 201 202 def test_invalid_redirect(self): 203 # urlopen() should raise IOError for many error codes. 204 self.fakehttp("""HTTP/1.1 302 Found 205 Date: Wed, 02 Jan 2008 03:03:54 GMT 206 Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e 207 Location: file:README 208 Connection: close 209 Content-Type: text/html; charset=iso-8859-1 210 """) 211 try: 212 self.assertRaises(IOError, urllib.urlopen, "http://python.org/") 213 finally: 214 self.unfakehttp() 215 216 def test_empty_socket(self): 217 # urlopen() raises IOError if the underlying socket does not send any 218 # data. (#1680230) 219 self.fakehttp('') 220 try: 221 self.assertRaises(IOError, urllib.urlopen, 'http://something') 222 finally: 223 self.unfakehttp() 224 225 def test_missing_localfile(self): 226 self.assertRaises(IOError, urllib.urlopen, 227 'file://localhost/a/missing/file.py') 228 fd, tmp_file = tempfile.mkstemp() 229 tmp_fileurl = 'file://localhost/' + tmp_file.replace(os.path.sep, '/') 230 try: 231 self.assertTrue(os.path.exists(tmp_file)) 232 fp = urllib.urlopen(tmp_fileurl) 233 finally: 234 os.close(fd) 235 fp.close() 236 os.unlink(tmp_file) 237 238 self.assertFalse(os.path.exists(tmp_file)) 239 self.assertRaises(IOError, urllib.urlopen, tmp_fileurl) 240 241 def test_ftp_nonexisting(self): 242 self.assertRaises(IOError, urllib.urlopen, 243 'ftp://localhost/not/existing/file.py') 244 245 246 def test_userpass_inurl(self): 247 self.fakehttp('Hello!') 248 try: 249 fakehttp_wrapper = httplib.HTTP._connection_class 250 fp = urllib.urlopen("http://user:pass@python.org/") 251 authorization = ("Authorization: Basic %s\r\n" % 252 b64encode('user:pass')) 253 # The authorization header must be in place 254 self.assertIn(authorization, fakehttp_wrapper.buf) 255 self.assertEqual(fp.readline(), "Hello!") 256 self.assertEqual(fp.readline(), "") 257 self.assertEqual(fp.geturl(), 'http://user:pass@python.org/') 258 self.assertEqual(fp.getcode(), 200) 259 finally: 260 self.unfakehttp() 261 262 def test_userpass_with_spaces_inurl(self): 263 self.fakehttp('Hello!') 264 try: 265 url = "http://a b:c d (at] python.org/" 266 fakehttp_wrapper = httplib.HTTP._connection_class 267 authorization = ("Authorization: Basic %s\r\n" % 268 b64encode('a b:c d')) 269 fp = urllib.urlopen(url) 270 # The authorization header must be in place 271 self.assertIn(authorization, fakehttp_wrapper.buf) 272 self.assertEqual(fp.readline(), "Hello!") 273 self.assertEqual(fp.readline(), "") 274 # the spaces are quoted in URL so no match 275 self.assertNotEqual(fp.geturl(), url) 276 self.assertEqual(fp.getcode(), 200) 277 finally: 278 self.unfakehttp() 279 280 281 class urlretrieve_FileTests(unittest.TestCase): 282 """Test urllib.urlretrieve() on local files""" 283 284 def setUp(self): 285 # Create a list of temporary files. Each item in the list is a file 286 # name (absolute path or relative to the current working directory). 287 # All files in this list will be deleted in the tearDown method. Note, 288 # this only helps to makes sure temporary files get deleted, but it 289 # does nothing about trying to close files that may still be open. It 290 # is the responsibility of the developer to properly close files even 291 # when exceptional conditions occur. 292 self.tempFiles = [] 293 294 # Create a temporary file. 295 self.registerFileForCleanUp(test_support.TESTFN) 296 self.text = 'testing urllib.urlretrieve' 297 try: 298 FILE = file(test_support.TESTFN, 'wb') 299 FILE.write(self.text) 300 FILE.close() 301 finally: 302 try: FILE.close() 303 except: pass 304 305 def tearDown(self): 306 # Delete the temporary files. 307 for each in self.tempFiles: 308 try: os.remove(each) 309 except: pass 310 311 def constructLocalFileUrl(self, filePath): 312 return "file://%s" % urllib.pathname2url(os.path.abspath(filePath)) 313 314 def createNewTempFile(self, data=""): 315 """Creates a new temporary file containing the specified data, 316 registers the file for deletion during the test fixture tear down, and 317 returns the absolute path of the file.""" 318 319 newFd, newFilePath = tempfile.mkstemp() 320 try: 321 self.registerFileForCleanUp(newFilePath) 322 newFile = os.fdopen(newFd, "wb") 323 newFile.write(data) 324 newFile.close() 325 finally: 326 try: newFile.close() 327 except: pass 328 return newFilePath 329 330 def registerFileForCleanUp(self, fileName): 331 self.tempFiles.append(fileName) 332 333 def test_basic(self): 334 # Make sure that a local file just gets its own location returned and 335 # a headers value is returned. 336 result = urllib.urlretrieve("file:%s" % test_support.TESTFN) 337 self.assertEqual(result[0], test_support.TESTFN) 338 self.assertIsInstance(result[1], mimetools.Message, 339 "did not get a mimetools.Message instance as " 340 "second returned value") 341 342 def test_copy(self): 343 # Test that setting the filename argument works. 344 second_temp = "%s.2" % test_support.TESTFN 345 self.registerFileForCleanUp(second_temp) 346 result = urllib.urlretrieve(self.constructLocalFileUrl( 347 test_support.TESTFN), second_temp) 348 self.assertEqual(second_temp, result[0]) 349 self.assertTrue(os.path.exists(second_temp), "copy of the file was not " 350 "made") 351 FILE = file(second_temp, 'rb') 352 try: 353 text = FILE.read() 354 FILE.close() 355 finally: 356 try: FILE.close() 357 except: pass 358 self.assertEqual(self.text, text) 359 360 def test_reporthook(self): 361 # Make sure that the reporthook works. 362 def hooktester(count, block_size, total_size, count_holder=[0]): 363 self.assertIsInstance(count, int) 364 self.assertIsInstance(block_size, int) 365 self.assertIsInstance(total_size, int) 366 self.assertEqual(count, count_holder[0]) 367 count_holder[0] = count_holder[0] + 1 368 second_temp = "%s.2" % test_support.TESTFN 369 self.registerFileForCleanUp(second_temp) 370 urllib.urlretrieve(self.constructLocalFileUrl(test_support.TESTFN), 371 second_temp, hooktester) 372 373 def test_reporthook_0_bytes(self): 374 # Test on zero length file. Should call reporthook only 1 time. 375 report = [] 376 def hooktester(count, block_size, total_size, _report=report): 377 _report.append((count, block_size, total_size)) 378 srcFileName = self.createNewTempFile() 379 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName), 380 test_support.TESTFN, hooktester) 381 self.assertEqual(len(report), 1) 382 self.assertEqual(report[0][2], 0) 383 384 def test_reporthook_5_bytes(self): 385 # Test on 5 byte file. Should call reporthook only 2 times (once when 386 # the "network connection" is established and once when the block is 387 # read). Since the block size is 8192 bytes, only one block read is 388 # required to read the entire file. 389 report = [] 390 def hooktester(count, block_size, total_size, _report=report): 391 _report.append((count, block_size, total_size)) 392 srcFileName = self.createNewTempFile("x" * 5) 393 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName), 394 test_support.TESTFN, hooktester) 395 self.assertEqual(len(report), 2) 396 self.assertEqual(report[0][1], 8192) 397 self.assertEqual(report[0][2], 5) 398 399 def test_reporthook_8193_bytes(self): 400 # Test on 8193 byte file. Should call reporthook only 3 times (once 401 # when the "network connection" is established, once for the next 8192 402 # bytes, and once for the last byte). 403 report = [] 404 def hooktester(count, block_size, total_size, _report=report): 405 _report.append((count, block_size, total_size)) 406 srcFileName = self.createNewTempFile("x" * 8193) 407 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName), 408 test_support.TESTFN, hooktester) 409 self.assertEqual(len(report), 3) 410 self.assertEqual(report[0][1], 8192) 411 self.assertEqual(report[0][2], 8193) 412 413 414 class urlretrieve_HttpTests(unittest.TestCase, FakeHTTPMixin): 415 """Test urllib.urlretrieve() using fake http connections""" 416 417 def test_short_content_raises_ContentTooShortError(self): 418 self.fakehttp('''HTTP/1.1 200 OK 419 Date: Wed, 02 Jan 2008 03:03:54 GMT 420 Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e 421 Connection: close 422 Content-Length: 100 423 Content-Type: text/html; charset=iso-8859-1 424 425 FF 426 ''') 427 428 def _reporthook(par1, par2, par3): 429 pass 430 431 try: 432 self.assertRaises(urllib.ContentTooShortError, urllib.urlretrieve, 433 'http://example.com', reporthook=_reporthook) 434 finally: 435 self.unfakehttp() 436 437 def test_short_content_raises_ContentTooShortError_without_reporthook(self): 438 self.fakehttp('''HTTP/1.1 200 OK 439 Date: Wed, 02 Jan 2008 03:03:54 GMT 440 Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e 441 Connection: close 442 Content-Length: 100 443 Content-Type: text/html; charset=iso-8859-1 444 445 FF 446 ''') 447 try: 448 self.assertRaises(urllib.ContentTooShortError, urllib.urlretrieve, 'http://example.com/') 449 finally: 450 self.unfakehttp() 451 452 class QuotingTests(unittest.TestCase): 453 """Tests for urllib.quote() and urllib.quote_plus() 454 455 According to RFC 2396 ("Uniform Resource Identifiers), to escape a 456 character you write it as '%' + <2 character US-ASCII hex value>. The Python 457 code of ``'%' + hex(ord(<character>))[2:]`` escapes a character properly. 458 Case does not matter on the hex letters. 459 460 The various character sets specified are: 461 462 Reserved characters : ";/?:@&=+$," 463 Have special meaning in URIs and must be escaped if not being used for 464 their special meaning 465 Data characters : letters, digits, and "-_.!~*'()" 466 Unreserved and do not need to be escaped; can be, though, if desired 467 Control characters : 0x00 - 0x1F, 0x7F 468 Have no use in URIs so must be escaped 469 space : 0x20 470 Must be escaped 471 Delimiters : '<>#%"' 472 Must be escaped 473 Unwise : "{}|\^[]`" 474 Must be escaped 475 476 """ 477 478 def test_never_quote(self): 479 # Make sure quote() does not quote letters, digits, and "_,.-" 480 do_not_quote = '' .join(["ABCDEFGHIJKLMNOPQRSTUVWXYZ", 481 "abcdefghijklmnopqrstuvwxyz", 482 "0123456789", 483 "_.-"]) 484 result = urllib.quote(do_not_quote) 485 self.assertEqual(do_not_quote, result, 486 "using quote(): %s != %s" % (do_not_quote, result)) 487 result = urllib.quote_plus(do_not_quote) 488 self.assertEqual(do_not_quote, result, 489 "using quote_plus(): %s != %s" % (do_not_quote, result)) 490 491 def test_default_safe(self): 492 # Test '/' is default value for 'safe' parameter 493 self.assertEqual(urllib.quote.func_defaults[0], '/') 494 495 def test_safe(self): 496 # Test setting 'safe' parameter does what it should do 497 quote_by_default = "<>" 498 result = urllib.quote(quote_by_default, safe=quote_by_default) 499 self.assertEqual(quote_by_default, result, 500 "using quote(): %s != %s" % (quote_by_default, result)) 501 result = urllib.quote_plus(quote_by_default, safe=quote_by_default) 502 self.assertEqual(quote_by_default, result, 503 "using quote_plus(): %s != %s" % 504 (quote_by_default, result)) 505 506 def test_default_quoting(self): 507 # Make sure all characters that should be quoted are by default sans 508 # space (separate test for that). 509 should_quote = [chr(num) for num in range(32)] # For 0x00 - 0x1F 510 should_quote.append('<>#%"{}|\^[]`') 511 should_quote.append(chr(127)) # For 0x7F 512 should_quote = ''.join(should_quote) 513 for char in should_quote: 514 result = urllib.quote(char) 515 self.assertEqual(hexescape(char), result, 516 "using quote(): %s should be escaped to %s, not %s" % 517 (char, hexescape(char), result)) 518 result = urllib.quote_plus(char) 519 self.assertEqual(hexescape(char), result, 520 "using quote_plus(): " 521 "%s should be escapes to %s, not %s" % 522 (char, hexescape(char), result)) 523 del should_quote 524 partial_quote = "ab[]cd" 525 expected = "ab%5B%5Dcd" 526 result = urllib.quote(partial_quote) 527 self.assertEqual(expected, result, 528 "using quote(): %s != %s" % (expected, result)) 529 result = urllib.quote_plus(partial_quote) 530 self.assertEqual(expected, result, 531 "using quote_plus(): %s != %s" % (expected, result)) 532 self.assertRaises(TypeError, urllib.quote, None) 533 534 def test_quoting_space(self): 535 # Make sure quote() and quote_plus() handle spaces as specified in 536 # their unique way 537 result = urllib.quote(' ') 538 self.assertEqual(result, hexescape(' '), 539 "using quote(): %s != %s" % (result, hexescape(' '))) 540 result = urllib.quote_plus(' ') 541 self.assertEqual(result, '+', 542 "using quote_plus(): %s != +" % result) 543 given = "a b cd e f" 544 expect = given.replace(' ', hexescape(' ')) 545 result = urllib.quote(given) 546 self.assertEqual(expect, result, 547 "using quote(): %s != %s" % (expect, result)) 548 expect = given.replace(' ', '+') 549 result = urllib.quote_plus(given) 550 self.assertEqual(expect, result, 551 "using quote_plus(): %s != %s" % (expect, result)) 552 553 def test_quoting_plus(self): 554 self.assertEqual(urllib.quote_plus('alpha+beta gamma'), 555 'alpha%2Bbeta+gamma') 556 self.assertEqual(urllib.quote_plus('alpha+beta gamma', '+'), 557 'alpha+beta+gamma') 558 559 class UnquotingTests(unittest.TestCase): 560 """Tests for unquote() and unquote_plus() 561 562 See the doc string for quoting_Tests for details on quoting and such. 563 564 """ 565 566 def test_unquoting(self): 567 # Make sure unquoting of all ASCII values works 568 escape_list = [] 569 for num in range(128): 570 given = hexescape(chr(num)) 571 expect = chr(num) 572 result = urllib.unquote(given) 573 self.assertEqual(expect, result, 574 "using unquote(): %s != %s" % (expect, result)) 575 result = urllib.unquote_plus(given) 576 self.assertEqual(expect, result, 577 "using unquote_plus(): %s != %s" % 578 (expect, result)) 579 escape_list.append(given) 580 escape_string = ''.join(escape_list) 581 del escape_list 582 result = urllib.unquote(escape_string) 583 self.assertEqual(result.count('%'), 1, 584 "using quote(): not all characters escaped; %s" % 585 result) 586 result = urllib.unquote(escape_string) 587 self.assertEqual(result.count('%'), 1, 588 "using unquote(): not all characters escaped: " 589 "%s" % result) 590 591 def test_unquoting_badpercent(self): 592 # Test unquoting on bad percent-escapes 593 given = '%xab' 594 expect = given 595 result = urllib.unquote(given) 596 self.assertEqual(expect, result, "using unquote(): %r != %r" 597 % (expect, result)) 598 given = '%x' 599 expect = given 600 result = urllib.unquote(given) 601 self.assertEqual(expect, result, "using unquote(): %r != %r" 602 % (expect, result)) 603 given = '%' 604 expect = given 605 result = urllib.unquote(given) 606 self.assertEqual(expect, result, "using unquote(): %r != %r" 607 % (expect, result)) 608 609 def test_unquoting_mixed_case(self): 610 # Test unquoting on mixed-case hex digits in the percent-escapes 611 given = '%Ab%eA' 612 expect = '\xab\xea' 613 result = urllib.unquote(given) 614 self.assertEqual(expect, result, "using unquote(): %r != %r" 615 % (expect, result)) 616 617 def test_unquoting_parts(self): 618 # Make sure unquoting works when have non-quoted characters 619 # interspersed 620 given = 'ab%sd' % hexescape('c') 621 expect = "abcd" 622 result = urllib.unquote(given) 623 self.assertEqual(expect, result, 624 "using quote(): %s != %s" % (expect, result)) 625 result = urllib.unquote_plus(given) 626 self.assertEqual(expect, result, 627 "using unquote_plus(): %s != %s" % (expect, result)) 628 629 def test_unquoting_plus(self): 630 # Test difference between unquote() and unquote_plus() 631 given = "are+there+spaces..." 632 expect = given 633 result = urllib.unquote(given) 634 self.assertEqual(expect, result, 635 "using unquote(): %s != %s" % (expect, result)) 636 expect = given.replace('+', ' ') 637 result = urllib.unquote_plus(given) 638 self.assertEqual(expect, result, 639 "using unquote_plus(): %s != %s" % (expect, result)) 640 641 def test_unquote_with_unicode(self): 642 r = urllib.unquote(u'br%C3%BCckner_sapporo_20050930.doc') 643 self.assertEqual(r, u'br\xc3\xbcckner_sapporo_20050930.doc') 644 645 class urlencode_Tests(unittest.TestCase): 646 """Tests for urlencode()""" 647 648 def help_inputtype(self, given, test_type): 649 """Helper method for testing different input types. 650 651 'given' must lead to only the pairs: 652 * 1st, 1 653 * 2nd, 2 654 * 3rd, 3 655 656 Test cannot assume anything about order. Docs make no guarantee and 657 have possible dictionary input. 658 659 """ 660 expect_somewhere = ["1st=1", "2nd=2", "3rd=3"] 661 result = urllib.urlencode(given) 662 for expected in expect_somewhere: 663 self.assertIn(expected, result, 664 "testing %s: %s not found in %s" % 665 (test_type, expected, result)) 666 self.assertEqual(result.count('&'), 2, 667 "testing %s: expected 2 '&'s; got %s" % 668 (test_type, result.count('&'))) 669 amp_location = result.index('&') 670 on_amp_left = result[amp_location - 1] 671 on_amp_right = result[amp_location + 1] 672 self.assertTrue(on_amp_left.isdigit() and on_amp_right.isdigit(), 673 "testing %s: '&' not located in proper place in %s" % 674 (test_type, result)) 675 self.assertEqual(len(result), (5 * 3) + 2, #5 chars per thing and amps 676 "testing %s: " 677 "unexpected number of characters: %s != %s" % 678 (test_type, len(result), (5 * 3) + 2)) 679 680 def test_using_mapping(self): 681 # Test passing in a mapping object as an argument. 682 self.help_inputtype({"1st":'1', "2nd":'2', "3rd":'3'}, 683 "using dict as input type") 684 685 def test_using_sequence(self): 686 # Test passing in a sequence of two-item sequences as an argument. 687 self.help_inputtype([('1st', '1'), ('2nd', '2'), ('3rd', '3')], 688 "using sequence of two-item tuples as input") 689 690 def test_quoting(self): 691 # Make sure keys and values are quoted using quote_plus() 692 given = {"&":"="} 693 expect = "%s=%s" % (hexescape('&'), hexescape('=')) 694 result = urllib.urlencode(given) 695 self.assertEqual(expect, result) 696 given = {"key name":"A bunch of pluses"} 697 expect = "key+name=A+bunch+of+pluses" 698 result = urllib.urlencode(given) 699 self.assertEqual(expect, result) 700 701 def test_doseq(self): 702 # Test that passing True for 'doseq' parameter works correctly 703 given = {'sequence':['1', '2', '3']} 704 expect = "sequence=%s" % urllib.quote_plus(str(['1', '2', '3'])) 705 result = urllib.urlencode(given) 706 self.assertEqual(expect, result) 707 result = urllib.urlencode(given, True) 708 for value in given["sequence"]: 709 expect = "sequence=%s" % value 710 self.assertIn(expect, result) 711 self.assertEqual(result.count('&'), 2, 712 "Expected 2 '&'s, got %s" % result.count('&')) 713 714 class Pathname_Tests(unittest.TestCase): 715 """Test pathname2url() and url2pathname()""" 716 717 def test_basic(self): 718 # Make sure simple tests pass 719 expected_path = os.path.join("parts", "of", "a", "path") 720 expected_url = "parts/of/a/path" 721 result = urllib.pathname2url(expected_path) 722 self.assertEqual(expected_url, result, 723 "pathname2url() failed; %s != %s" % 724 (result, expected_url)) 725 result = urllib.url2pathname(expected_url) 726 self.assertEqual(expected_path, result, 727 "url2pathame() failed; %s != %s" % 728 (result, expected_path)) 729 730 def test_quoting(self): 731 # Test automatic quoting and unquoting works for pathnam2url() and 732 # url2pathname() respectively 733 given = os.path.join("needs", "quot=ing", "here") 734 expect = "needs/%s/here" % urllib.quote("quot=ing") 735 result = urllib.pathname2url(given) 736 self.assertEqual(expect, result, 737 "pathname2url() failed; %s != %s" % 738 (expect, result)) 739 expect = given 740 result = urllib.url2pathname(result) 741 self.assertEqual(expect, result, 742 "url2pathname() failed; %s != %s" % 743 (expect, result)) 744 given = os.path.join("make sure", "using_quote") 745 expect = "%s/using_quote" % urllib.quote("make sure") 746 result = urllib.pathname2url(given) 747 self.assertEqual(expect, result, 748 "pathname2url() failed; %s != %s" % 749 (expect, result)) 750 given = "make+sure/using_unquote" 751 expect = os.path.join("make+sure", "using_unquote") 752 result = urllib.url2pathname(given) 753 self.assertEqual(expect, result, 754 "url2pathname() failed; %s != %s" % 755 (expect, result)) 756 757 @unittest.skipUnless(sys.platform == 'win32', 758 'test specific to the nturl2path library') 759 def test_ntpath(self): 760 given = ('/C:/', '///C:/', '/C|//') 761 expect = 'C:\\' 762 for url in given: 763 result = urllib.url2pathname(url) 764 self.assertEqual(expect, result, 765 'nturl2path.url2pathname() failed; %s != %s' % 766 (expect, result)) 767 given = '///C|/path' 768 expect = 'C:\\path' 769 result = urllib.url2pathname(given) 770 self.assertEqual(expect, result, 771 'nturl2path.url2pathname() failed; %s != %s' % 772 (expect, result)) 773 774 class Utility_Tests(unittest.TestCase): 775 """Testcase to test the various utility functions in the urllib.""" 776 777 def test_splitpasswd(self): 778 """Some of the password examples are not sensible, but it is added to 779 confirming to RFC2617 and addressing issue4675. 780 """ 781 self.assertEqual(('user', 'ab'),urllib.splitpasswd('user:ab')) 782 self.assertEqual(('user', 'a\nb'),urllib.splitpasswd('user:a\nb')) 783 self.assertEqual(('user', 'a\tb'),urllib.splitpasswd('user:a\tb')) 784 self.assertEqual(('user', 'a\rb'),urllib.splitpasswd('user:a\rb')) 785 self.assertEqual(('user', 'a\fb'),urllib.splitpasswd('user:a\fb')) 786 self.assertEqual(('user', 'a\vb'),urllib.splitpasswd('user:a\vb')) 787 self.assertEqual(('user', 'a:b'),urllib.splitpasswd('user:a:b')) 788 self.assertEqual(('user', 'a b'),urllib.splitpasswd('user:a b')) 789 self.assertEqual(('user 2', 'ab'),urllib.splitpasswd('user 2:ab')) 790 self.assertEqual(('user+1', 'a+b'),urllib.splitpasswd('user+1:a+b')) 791 792 793 class URLopener_Tests(unittest.TestCase): 794 """Testcase to test the open method of URLopener class.""" 795 796 def test_quoted_open(self): 797 class DummyURLopener(urllib.URLopener): 798 def open_spam(self, url): 799 return url 800 801 self.assertEqual(DummyURLopener().open( 802 'spam://example/ /'),'//example/%20/') 803 804 # test the safe characters are not quoted by urlopen 805 self.assertEqual(DummyURLopener().open( 806 "spam://c:|windows%/:=&?~#+!$,;'@()*[]|/path/"), 807 "//c:|windows%/:=&?~#+!$,;'@()*[]|/path/") 808 809 810 # Just commented them out. 811 # Can't really tell why keep failing in windows and sparc. 812 # Everywhere else they work ok, but on those machines, sometimes 813 # fail in one of the tests, sometimes in other. I have a linux, and 814 # the tests go ok. 815 # If anybody has one of the problematic enviroments, please help! 816 # . Facundo 817 # 818 # def server(evt): 819 # import socket, time 820 # serv = socket.socket(socket.AF_INET, socket.SOCK_STREAM) 821 # serv.settimeout(3) 822 # serv.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1) 823 # serv.bind(("", 9093)) 824 # serv.listen(5) 825 # try: 826 # conn, addr = serv.accept() 827 # conn.send("1 Hola mundo\n") 828 # cantdata = 0 829 # while cantdata < 13: 830 # data = conn.recv(13-cantdata) 831 # cantdata += len(data) 832 # time.sleep(.3) 833 # conn.send("2 No more lines\n") 834 # conn.close() 835 # except socket.timeout: 836 # pass 837 # finally: 838 # serv.close() 839 # evt.set() 840 # 841 # class FTPWrapperTests(unittest.TestCase): 842 # 843 # def setUp(self): 844 # import ftplib, time, threading 845 # ftplib.FTP.port = 9093 846 # self.evt = threading.Event() 847 # threading.Thread(target=server, args=(self.evt,)).start() 848 # time.sleep(.1) 849 # 850 # def tearDown(self): 851 # self.evt.wait() 852 # 853 # def testBasic(self): 854 # # connects 855 # ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, []) 856 # ftp.close() 857 # 858 # def testTimeoutNone(self): 859 # # global default timeout is ignored 860 # import socket 861 # self.assertTrue(socket.getdefaulttimeout() is None) 862 # socket.setdefaulttimeout(30) 863 # try: 864 # ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, []) 865 # finally: 866 # socket.setdefaulttimeout(None) 867 # self.assertEqual(ftp.ftp.sock.gettimeout(), 30) 868 # ftp.close() 869 # 870 # def testTimeoutDefault(self): 871 # # global default timeout is used 872 # import socket 873 # self.assertTrue(socket.getdefaulttimeout() is None) 874 # socket.setdefaulttimeout(30) 875 # try: 876 # ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, []) 877 # finally: 878 # socket.setdefaulttimeout(None) 879 # self.assertEqual(ftp.ftp.sock.gettimeout(), 30) 880 # ftp.close() 881 # 882 # def testTimeoutValue(self): 883 # ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [], 884 # timeout=30) 885 # self.assertEqual(ftp.ftp.sock.gettimeout(), 30) 886 # ftp.close() 887 888 889 890 def test_main(): 891 import warnings 892 with warnings.catch_warnings(): 893 warnings.filterwarnings('ignore', ".*urllib\.urlopen.*Python 3.0", 894 DeprecationWarning) 895 test_support.run_unittest( 896 urlopen_FileTests, 897 urlopen_HttpTests, 898 urlretrieve_FileTests, 899 urlretrieve_HttpTests, 900 ProxyTests, 901 QuotingTests, 902 UnquotingTests, 903 urlencode_Tests, 904 Pathname_Tests, 905 Utility_Tests, 906 URLopener_Tests, 907 #FTPWrapperTests, 908 ) 909 910 911 912 if __name__ == '__main__': 913 test_main() 914
