blob: 7bf383f6bc04c3d5b50d0da1ac116e821bf63519 [file] [log] [blame]
Brett Cannon74bfd702003-04-25 09:39:47 +00001"""Regresssion tests for urllib"""
2
Jeremy Hylton6102e292000-08-31 15:48:10 +00003import urllib
Hye-Shik Chang39aef792004-06-05 13:30:56 +00004import httplib
Brett Cannon74bfd702003-04-25 09:39:47 +00005import unittest
6from test import test_support
7import os
8import mimetools
Georg Brandl5a650a22005-08-26 08:51:34 +00009import tempfile
Hye-Shik Chang39aef792004-06-05 13:30:56 +000010import StringIO
Jeremy Hylton6102e292000-08-31 15:48:10 +000011
Brett Cannon74bfd702003-04-25 09:39:47 +000012def hexescape(char):
13 """Escape char as RFC 2396 specifies"""
14 hex_repr = hex(ord(char))[2:].upper()
15 if len(hex_repr) == 1:
16 hex_repr = "0%s" % hex_repr
17 return "%" + hex_repr
Jeremy Hylton6102e292000-08-31 15:48:10 +000018
Brett Cannon74bfd702003-04-25 09:39:47 +000019class urlopen_FileTests(unittest.TestCase):
20 """Test urlopen() opening a temporary file.
Jeremy Hylton6102e292000-08-31 15:48:10 +000021
Brett Cannon74bfd702003-04-25 09:39:47 +000022 Try to test as much functionality as possible so as to cut down on reliance
Andrew M. Kuchlingf1a2f9e2004-06-29 13:07:53 +000023 on connecting to the Net for testing.
Jeremy Hylton7ae51bf2000-09-14 16:59:07 +000024
Brett Cannon74bfd702003-04-25 09:39:47 +000025 """
Jeremy Hylton7ae51bf2000-09-14 16:59:07 +000026
Brett Cannon74bfd702003-04-25 09:39:47 +000027 def setUp(self):
28 """Setup of a temp file to use for testing"""
29 self.text = "test_urllib: %s\n" % self.__class__.__name__
Guido van Rossum51735b02003-04-25 15:01:05 +000030 FILE = file(test_support.TESTFN, 'wb')
Brett Cannon74bfd702003-04-25 09:39:47 +000031 try:
32 FILE.write(self.text)
33 finally:
34 FILE.close()
35 self.pathname = test_support.TESTFN
36 self.returned_obj = urllib.urlopen("file:%s" % self.pathname)
Jeremy Hylton7ae51bf2000-09-14 16:59:07 +000037
Brett Cannon74bfd702003-04-25 09:39:47 +000038 def tearDown(self):
39 """Shut down the open object"""
40 self.returned_obj.close()
Brett Cannon19691362003-04-29 05:08:06 +000041 os.remove(test_support.TESTFN)
Jeremy Hylton7ae51bf2000-09-14 16:59:07 +000042
Brett Cannon74bfd702003-04-25 09:39:47 +000043 def test_interface(self):
44 # Make sure object returned by urlopen() has the specified methods
45 for attr in ("read", "readline", "readlines", "fileno",
Georg Brandl9b0d46d2008-01-20 11:43:03 +000046 "close", "info", "geturl", "getcode", "__iter__"):
Benjamin Peterson5c8da862009-06-30 22:57:08 +000047 self.assertTrue(hasattr(self.returned_obj, attr),
Brett Cannon74bfd702003-04-25 09:39:47 +000048 "object returned by urlopen() lacks %s attribute" %
49 attr)
Skip Montanaroe78b92a2001-01-20 20:22:30 +000050
Brett Cannon74bfd702003-04-25 09:39:47 +000051 def test_read(self):
52 self.assertEqual(self.text, self.returned_obj.read())
Skip Montanaro080c9972001-01-28 21:12:22 +000053
Brett Cannon74bfd702003-04-25 09:39:47 +000054 def test_readline(self):
55 self.assertEqual(self.text, self.returned_obj.readline())
56 self.assertEqual('', self.returned_obj.readline(),
57 "calling readline() after exhausting the file did not"
58 " return an empty string")
Skip Montanaro080c9972001-01-28 21:12:22 +000059
Brett Cannon74bfd702003-04-25 09:39:47 +000060 def test_readlines(self):
61 lines_list = self.returned_obj.readlines()
62 self.assertEqual(len(lines_list), 1,
63 "readlines() returned the wrong number of lines")
64 self.assertEqual(lines_list[0], self.text,
65 "readlines() returned improper text")
Skip Montanaro080c9972001-01-28 21:12:22 +000066
Brett Cannon74bfd702003-04-25 09:39:47 +000067 def test_fileno(self):
68 file_num = self.returned_obj.fileno()
Ezio Melottib0f5adc2010-01-24 16:58:36 +000069 self.assertIsInstance(file_num, int, "fileno() did not return an int")
Brett Cannon74bfd702003-04-25 09:39:47 +000070 self.assertEqual(os.read(file_num, len(self.text)), self.text,
71 "Reading on the file descriptor returned by fileno() "
72 "did not return the expected text")
Skip Montanaroe78b92a2001-01-20 20:22:30 +000073
Brett Cannon74bfd702003-04-25 09:39:47 +000074 def test_close(self):
75 # Test close() by calling it hear and then having it be called again
76 # by the tearDown() method for the test
77 self.returned_obj.close()
Skip Montanaro080c9972001-01-28 21:12:22 +000078
Brett Cannon74bfd702003-04-25 09:39:47 +000079 def test_info(self):
Ezio Melottib0f5adc2010-01-24 16:58:36 +000080 self.assertIsInstance(self.returned_obj.info(), mimetools.Message)
Skip Montanaroe78b92a2001-01-20 20:22:30 +000081
Brett Cannon74bfd702003-04-25 09:39:47 +000082 def test_geturl(self):
83 self.assertEqual(self.returned_obj.geturl(), self.pathname)
Skip Montanaro080c9972001-01-28 21:12:22 +000084
Georg Brandl9b0d46d2008-01-20 11:43:03 +000085 def test_getcode(self):
86 self.assertEqual(self.returned_obj.getcode(), None)
87
Brett Cannon74bfd702003-04-25 09:39:47 +000088 def test_iter(self):
89 # Test iterator
90 # Don't need to count number of iterations since test would fail the
91 # instant it returned anything beyond the first line from the
92 # comparison
93 for line in self.returned_obj.__iter__():
94 self.assertEqual(line, self.text)
Skip Montanaro080c9972001-01-28 21:12:22 +000095
Benjamin Peterson2c7470d2008-09-21 21:27:51 +000096class ProxyTests(unittest.TestCase):
97
98 def setUp(self):
Walter Dörwald4b965f62009-04-26 20:51:44 +000099 # Records changes to env vars
100 self.env = test_support.EnvironmentVarGuard()
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000101 # Delete all proxy related env vars
Senthil Kumaran7a2ee0b2010-01-08 19:20:25 +0000102 for k in os.environ.keys():
Walter Dörwald4b965f62009-04-26 20:51:44 +0000103 if 'proxy' in k.lower():
Senthil Kumarandc61ec32009-10-01 01:50:13 +0000104 self.env.unset(k)
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000105
106 def tearDown(self):
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000107 # Restore all proxy related env vars
Walter Dörwald4b965f62009-04-26 20:51:44 +0000108 self.env.__exit__()
109 del self.env
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000110
111 def test_getproxies_environment_keep_no_proxies(self):
Walter Dörwald4b965f62009-04-26 20:51:44 +0000112 self.env.set('NO_PROXY', 'localhost')
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000113 proxies = urllib.getproxies_environment()
114 # getproxies_environment use lowered case truncated (no '_proxy') keys
Ezio Melotti2623a372010-11-21 13:34:58 +0000115 self.assertEqual('localhost', proxies['no'])
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000116
117
Hye-Shik Chang39aef792004-06-05 13:30:56 +0000118class urlopen_HttpTests(unittest.TestCase):
119 """Test urlopen() opening a fake http connection."""
120
121 def fakehttp(self, fakedata):
122 class FakeSocket(StringIO.StringIO):
123 def sendall(self, str): pass
124 def makefile(self, mode, name): return self
125 def read(self, amt=None):
126 if self.closed: return ''
127 return StringIO.StringIO.read(self, amt)
128 def readline(self, length=None):
129 if self.closed: return ''
130 return StringIO.StringIO.readline(self, length)
131 class FakeHTTPConnection(httplib.HTTPConnection):
132 def connect(self):
133 self.sock = FakeSocket(fakedata)
134 assert httplib.HTTP._connection_class == httplib.HTTPConnection
135 httplib.HTTP._connection_class = FakeHTTPConnection
136
137 def unfakehttp(self):
138 httplib.HTTP._connection_class = httplib.HTTPConnection
139
140 def test_read(self):
141 self.fakehttp('Hello!')
142 try:
143 fp = urllib.urlopen("http://python.org/")
144 self.assertEqual(fp.readline(), 'Hello!')
145 self.assertEqual(fp.readline(), '')
Georg Brandl9b0d46d2008-01-20 11:43:03 +0000146 self.assertEqual(fp.geturl(), 'http://python.org/')
147 self.assertEqual(fp.getcode(), 200)
Hye-Shik Chang39aef792004-06-05 13:30:56 +0000148 finally:
149 self.unfakehttp()
150
Senthil Kumaran49c44082011-04-13 07:31:45 +0800151 def test_url_fragment(self):
152 # Issue #11703: geturl() omits fragments in the original URL.
153 url = 'http://docs.python.org/library/urllib.html#OK'
154 self.fakehttp('Hello!')
155 try:
156 fp = urllib.urlopen(url)
157 self.assertEqual(fp.geturl(), url)
158 finally:
159 self.unfakehttp()
160
Kurt B. Kaiser0f7c25d2008-01-02 04:11:28 +0000161 def test_read_bogus(self):
Kurt B. Kaiser0a112322008-01-02 05:23:38 +0000162 # urlopen() should raise IOError for many error codes.
Kurt B. Kaiser0f7c25d2008-01-02 04:11:28 +0000163 self.fakehttp('''HTTP/1.1 401 Authentication Required
164Date: Wed, 02 Jan 2008 03:03:54 GMT
165Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e
166Connection: close
167Content-Type: text/html; charset=iso-8859-1
168''')
169 try:
170 self.assertRaises(IOError, urllib.urlopen, "http://python.org/")
171 finally:
172 self.unfakehttp()
173
guido@google.comf1509302011-03-28 13:47:01 -0700174 def test_invalid_redirect(self):
175 # urlopen() should raise IOError for many error codes.
176 self.fakehttp("""HTTP/1.1 302 Found
177Date: Wed, 02 Jan 2008 03:03:54 GMT
178Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e
179Location: file:README
180Connection: close
181Content-Type: text/html; charset=iso-8859-1
182""")
183 try:
184 self.assertRaises(IOError, urllib.urlopen, "http://python.org/")
185 finally:
186 self.unfakehttp()
187
Georg Brandlf66b6032007-03-14 08:27:52 +0000188 def test_empty_socket(self):
Kurt B. Kaiser0a112322008-01-02 05:23:38 +0000189 # urlopen() raises IOError if the underlying socket does not send any
190 # data. (#1680230)
Georg Brandlf66b6032007-03-14 08:27:52 +0000191 self.fakehttp('')
192 try:
193 self.assertRaises(IOError, urllib.urlopen, 'http://something')
194 finally:
195 self.unfakehttp()
196
Brett Cannon19691362003-04-29 05:08:06 +0000197class urlretrieve_FileTests(unittest.TestCase):
Brett Cannon74bfd702003-04-25 09:39:47 +0000198 """Test urllib.urlretrieve() on local files"""
Skip Montanaro080c9972001-01-28 21:12:22 +0000199
Brett Cannon19691362003-04-29 05:08:06 +0000200 def setUp(self):
Georg Brandl5a650a22005-08-26 08:51:34 +0000201 # Create a list of temporary files. Each item in the list is a file
202 # name (absolute path or relative to the current working directory).
203 # All files in this list will be deleted in the tearDown method. Note,
204 # this only helps to makes sure temporary files get deleted, but it
205 # does nothing about trying to close files that may still be open. It
206 # is the responsibility of the developer to properly close files even
207 # when exceptional conditions occur.
208 self.tempFiles = []
209
Brett Cannon19691362003-04-29 05:08:06 +0000210 # Create a temporary file.
Georg Brandl5a650a22005-08-26 08:51:34 +0000211 self.registerFileForCleanUp(test_support.TESTFN)
Brett Cannon19691362003-04-29 05:08:06 +0000212 self.text = 'testing urllib.urlretrieve'
Georg Brandl5a650a22005-08-26 08:51:34 +0000213 try:
214 FILE = file(test_support.TESTFN, 'wb')
215 FILE.write(self.text)
216 FILE.close()
217 finally:
218 try: FILE.close()
219 except: pass
Brett Cannon19691362003-04-29 05:08:06 +0000220
221 def tearDown(self):
Georg Brandl5a650a22005-08-26 08:51:34 +0000222 # Delete the temporary files.
223 for each in self.tempFiles:
224 try: os.remove(each)
225 except: pass
226
227 def constructLocalFileUrl(self, filePath):
228 return "file://%s" % urllib.pathname2url(os.path.abspath(filePath))
229
230 def createNewTempFile(self, data=""):
231 """Creates a new temporary file containing the specified data,
232 registers the file for deletion during the test fixture tear down, and
233 returns the absolute path of the file."""
234
235 newFd, newFilePath = tempfile.mkstemp()
236 try:
237 self.registerFileForCleanUp(newFilePath)
238 newFile = os.fdopen(newFd, "wb")
239 newFile.write(data)
240 newFile.close()
241 finally:
242 try: newFile.close()
243 except: pass
244 return newFilePath
245
246 def registerFileForCleanUp(self, fileName):
247 self.tempFiles.append(fileName)
Brett Cannon19691362003-04-29 05:08:06 +0000248
249 def test_basic(self):
250 # Make sure that a local file just gets its own location returned and
251 # a headers value is returned.
252 result = urllib.urlretrieve("file:%s" % test_support.TESTFN)
253 self.assertEqual(result[0], test_support.TESTFN)
Ezio Melottib0f5adc2010-01-24 16:58:36 +0000254 self.assertIsInstance(result[1], mimetools.Message,
255 "did not get a mimetools.Message instance as "
256 "second returned value")
Brett Cannon19691362003-04-29 05:08:06 +0000257
258 def test_copy(self):
259 # Test that setting the filename argument works.
260 second_temp = "%s.2" % test_support.TESTFN
Georg Brandl5a650a22005-08-26 08:51:34 +0000261 self.registerFileForCleanUp(second_temp)
262 result = urllib.urlretrieve(self.constructLocalFileUrl(
263 test_support.TESTFN), second_temp)
Brett Cannon19691362003-04-29 05:08:06 +0000264 self.assertEqual(second_temp, result[0])
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000265 self.assertTrue(os.path.exists(second_temp), "copy of the file was not "
Brett Cannon19691362003-04-29 05:08:06 +0000266 "made")
267 FILE = file(second_temp, 'rb')
268 try:
269 text = FILE.read()
Brett Cannon19691362003-04-29 05:08:06 +0000270 FILE.close()
Georg Brandl5a650a22005-08-26 08:51:34 +0000271 finally:
272 try: FILE.close()
273 except: pass
Brett Cannon19691362003-04-29 05:08:06 +0000274 self.assertEqual(self.text, text)
275
276 def test_reporthook(self):
277 # Make sure that the reporthook works.
278 def hooktester(count, block_size, total_size, count_holder=[0]):
Ezio Melottib0f5adc2010-01-24 16:58:36 +0000279 self.assertIsInstance(count, int)
280 self.assertIsInstance(block_size, int)
281 self.assertIsInstance(total_size, int)
Brett Cannon19691362003-04-29 05:08:06 +0000282 self.assertEqual(count, count_holder[0])
283 count_holder[0] = count_holder[0] + 1
284 second_temp = "%s.2" % test_support.TESTFN
Georg Brandl5a650a22005-08-26 08:51:34 +0000285 self.registerFileForCleanUp(second_temp)
286 urllib.urlretrieve(self.constructLocalFileUrl(test_support.TESTFN),
287 second_temp, hooktester)
288
289 def test_reporthook_0_bytes(self):
290 # Test on zero length file. Should call reporthook only 1 time.
291 report = []
292 def hooktester(count, block_size, total_size, _report=report):
293 _report.append((count, block_size, total_size))
294 srcFileName = self.createNewTempFile()
295 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
296 test_support.TESTFN, hooktester)
297 self.assertEqual(len(report), 1)
298 self.assertEqual(report[0][2], 0)
299
300 def test_reporthook_5_bytes(self):
301 # Test on 5 byte file. Should call reporthook only 2 times (once when
302 # the "network connection" is established and once when the block is
303 # read). Since the block size is 8192 bytes, only one block read is
304 # required to read the entire file.
305 report = []
306 def hooktester(count, block_size, total_size, _report=report):
307 _report.append((count, block_size, total_size))
308 srcFileName = self.createNewTempFile("x" * 5)
309 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
310 test_support.TESTFN, hooktester)
311 self.assertEqual(len(report), 2)
312 self.assertEqual(report[0][1], 8192)
313 self.assertEqual(report[0][2], 5)
314
315 def test_reporthook_8193_bytes(self):
316 # Test on 8193 byte file. Should call reporthook only 3 times (once
317 # when the "network connection" is established, once for the next 8192
318 # bytes, and once for the last byte).
319 report = []
320 def hooktester(count, block_size, total_size, _report=report):
321 _report.append((count, block_size, total_size))
322 srcFileName = self.createNewTempFile("x" * 8193)
323 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
324 test_support.TESTFN, hooktester)
325 self.assertEqual(len(report), 3)
326 self.assertEqual(report[0][1], 8192)
327 self.assertEqual(report[0][2], 8193)
Skip Montanaro080c9972001-01-28 21:12:22 +0000328
Brett Cannon74bfd702003-04-25 09:39:47 +0000329class QuotingTests(unittest.TestCase):
330 """Tests for urllib.quote() and urllib.quote_plus()
Tim Petersc2659cf2003-05-12 20:19:37 +0000331
Brett Cannon74bfd702003-04-25 09:39:47 +0000332 According to RFC 2396 ("Uniform Resource Identifiers), to escape a
333 character you write it as '%' + <2 character US-ASCII hex value>. The Python
334 code of ``'%' + hex(ord(<character>))[2:]`` escapes a character properly.
335 Case does not matter on the hex letters.
336
337 The various character sets specified are:
Tim Petersc2659cf2003-05-12 20:19:37 +0000338
Brett Cannon74bfd702003-04-25 09:39:47 +0000339 Reserved characters : ";/?:@&=+$,"
340 Have special meaning in URIs and must be escaped if not being used for
341 their special meaning
342 Data characters : letters, digits, and "-_.!~*'()"
343 Unreserved and do not need to be escaped; can be, though, if desired
344 Control characters : 0x00 - 0x1F, 0x7F
345 Have no use in URIs so must be escaped
346 space : 0x20
347 Must be escaped
348 Delimiters : '<>#%"'
349 Must be escaped
350 Unwise : "{}|\^[]`"
351 Must be escaped
Tim Petersc2659cf2003-05-12 20:19:37 +0000352
Brett Cannon74bfd702003-04-25 09:39:47 +0000353 """
354
355 def test_never_quote(self):
356 # Make sure quote() does not quote letters, digits, and "_,.-"
357 do_not_quote = '' .join(["ABCDEFGHIJKLMNOPQRSTUVWXYZ",
358 "abcdefghijklmnopqrstuvwxyz",
359 "0123456789",
360 "_.-"])
361 result = urllib.quote(do_not_quote)
362 self.assertEqual(do_not_quote, result,
363 "using quote(): %s != %s" % (do_not_quote, result))
364 result = urllib.quote_plus(do_not_quote)
365 self.assertEqual(do_not_quote, result,
366 "using quote_plus(): %s != %s" % (do_not_quote, result))
367
368 def test_default_safe(self):
369 # Test '/' is default value for 'safe' parameter
370 self.assertEqual(urllib.quote.func_defaults[0], '/')
371
372 def test_safe(self):
373 # Test setting 'safe' parameter does what it should do
374 quote_by_default = "<>"
375 result = urllib.quote(quote_by_default, safe=quote_by_default)
376 self.assertEqual(quote_by_default, result,
377 "using quote(): %s != %s" % (quote_by_default, result))
378 result = urllib.quote_plus(quote_by_default, safe=quote_by_default)
379 self.assertEqual(quote_by_default, result,
380 "using quote_plus(): %s != %s" %
381 (quote_by_default, result))
382
383 def test_default_quoting(self):
384 # Make sure all characters that should be quoted are by default sans
385 # space (separate test for that).
386 should_quote = [chr(num) for num in range(32)] # For 0x00 - 0x1F
387 should_quote.append('<>#%"{}|\^[]`')
388 should_quote.append(chr(127)) # For 0x7F
389 should_quote = ''.join(should_quote)
390 for char in should_quote:
391 result = urllib.quote(char)
392 self.assertEqual(hexescape(char), result,
393 "using quote(): %s should be escaped to %s, not %s" %
394 (char, hexescape(char), result))
395 result = urllib.quote_plus(char)
396 self.assertEqual(hexescape(char), result,
397 "using quote_plus(): "
Tim Petersc2659cf2003-05-12 20:19:37 +0000398 "%s should be escapes to %s, not %s" %
Brett Cannon74bfd702003-04-25 09:39:47 +0000399 (char, hexescape(char), result))
400 del should_quote
401 partial_quote = "ab[]cd"
402 expected = "ab%5B%5Dcd"
403 result = urllib.quote(partial_quote)
404 self.assertEqual(expected, result,
405 "using quote(): %s != %s" % (expected, result))
406 self.assertEqual(expected, result,
407 "using quote_plus(): %s != %s" % (expected, result))
Senthil Kumaranc7743aa2010-07-19 17:35:50 +0000408 self.assertRaises(TypeError, urllib.quote, None)
Brett Cannon74bfd702003-04-25 09:39:47 +0000409
410 def test_quoting_space(self):
411 # Make sure quote() and quote_plus() handle spaces as specified in
412 # their unique way
413 result = urllib.quote(' ')
414 self.assertEqual(result, hexescape(' '),
415 "using quote(): %s != %s" % (result, hexescape(' ')))
416 result = urllib.quote_plus(' ')
417 self.assertEqual(result, '+',
418 "using quote_plus(): %s != +" % result)
419 given = "a b cd e f"
420 expect = given.replace(' ', hexescape(' '))
421 result = urllib.quote(given)
422 self.assertEqual(expect, result,
423 "using quote(): %s != %s" % (expect, result))
424 expect = given.replace(' ', '+')
425 result = urllib.quote_plus(given)
426 self.assertEqual(expect, result,
427 "using quote_plus(): %s != %s" % (expect, result))
428
Raymond Hettinger2bdec7b2005-09-10 14:30:09 +0000429 def test_quoting_plus(self):
430 self.assertEqual(urllib.quote_plus('alpha+beta gamma'),
431 'alpha%2Bbeta+gamma')
432 self.assertEqual(urllib.quote_plus('alpha+beta gamma', '+'),
433 'alpha+beta+gamma')
434
Brett Cannon74bfd702003-04-25 09:39:47 +0000435class UnquotingTests(unittest.TestCase):
436 """Tests for unquote() and unquote_plus()
Tim Petersc2659cf2003-05-12 20:19:37 +0000437
Brett Cannon74bfd702003-04-25 09:39:47 +0000438 See the doc string for quoting_Tests for details on quoting and such.
439
440 """
441
442 def test_unquoting(self):
443 # Make sure unquoting of all ASCII values works
444 escape_list = []
445 for num in range(128):
446 given = hexescape(chr(num))
447 expect = chr(num)
448 result = urllib.unquote(given)
449 self.assertEqual(expect, result,
450 "using unquote(): %s != %s" % (expect, result))
451 result = urllib.unquote_plus(given)
452 self.assertEqual(expect, result,
453 "using unquote_plus(): %s != %s" %
454 (expect, result))
455 escape_list.append(given)
456 escape_string = ''.join(escape_list)
457 del escape_list
458 result = urllib.unquote(escape_string)
459 self.assertEqual(result.count('%'), 1,
460 "using quote(): not all characters escaped; %s" %
461 result)
462 result = urllib.unquote(escape_string)
463 self.assertEqual(result.count('%'), 1,
464 "using unquote(): not all characters escaped: "
465 "%s" % result)
466
Senthil Kumaranf3e9b2a2010-03-18 12:14:15 +0000467 def test_unquoting_badpercent(self):
468 # Test unquoting on bad percent-escapes
469 given = '%xab'
470 expect = given
471 result = urllib.unquote(given)
472 self.assertEqual(expect, result, "using unquote(): %r != %r"
473 % (expect, result))
474 given = '%x'
475 expect = given
476 result = urllib.unquote(given)
477 self.assertEqual(expect, result, "using unquote(): %r != %r"
478 % (expect, result))
479 given = '%'
480 expect = given
481 result = urllib.unquote(given)
482 self.assertEqual(expect, result, "using unquote(): %r != %r"
483 % (expect, result))
484
485 def test_unquoting_mixed_case(self):
486 # Test unquoting on mixed-case hex digits in the percent-escapes
487 given = '%Ab%eA'
488 expect = '\xab\xea'
489 result = urllib.unquote(given)
490 self.assertEqual(expect, result, "using unquote(): %r != %r"
491 % (expect, result))
492
Brett Cannon74bfd702003-04-25 09:39:47 +0000493 def test_unquoting_parts(self):
494 # Make sure unquoting works when have non-quoted characters
495 # interspersed
496 given = 'ab%sd' % hexescape('c')
497 expect = "abcd"
498 result = urllib.unquote(given)
499 self.assertEqual(expect, result,
500 "using quote(): %s != %s" % (expect, result))
501 result = urllib.unquote_plus(given)
502 self.assertEqual(expect, result,
503 "using unquote_plus(): %s != %s" % (expect, result))
Tim Petersc2659cf2003-05-12 20:19:37 +0000504
Brett Cannon74bfd702003-04-25 09:39:47 +0000505 def test_unquoting_plus(self):
506 # Test difference between unquote() and unquote_plus()
507 given = "are+there+spaces..."
508 expect = given
509 result = urllib.unquote(given)
510 self.assertEqual(expect, result,
511 "using unquote(): %s != %s" % (expect, result))
512 expect = given.replace('+', ' ')
513 result = urllib.unquote_plus(given)
514 self.assertEqual(expect, result,
515 "using unquote_plus(): %s != %s" % (expect, result))
516
Raymond Hettinger4b0f20d2005-10-15 16:41:53 +0000517 def test_unquote_with_unicode(self):
518 r = urllib.unquote(u'br%C3%BCckner_sapporo_20050930.doc')
519 self.assertEqual(r, u'br\xc3\xbcckner_sapporo_20050930.doc')
520
Brett Cannon74bfd702003-04-25 09:39:47 +0000521class urlencode_Tests(unittest.TestCase):
522 """Tests for urlencode()"""
523
524 def help_inputtype(self, given, test_type):
525 """Helper method for testing different input types.
Tim Petersc2659cf2003-05-12 20:19:37 +0000526
Brett Cannon74bfd702003-04-25 09:39:47 +0000527 'given' must lead to only the pairs:
528 * 1st, 1
529 * 2nd, 2
530 * 3rd, 3
Tim Petersc2659cf2003-05-12 20:19:37 +0000531
Brett Cannon74bfd702003-04-25 09:39:47 +0000532 Test cannot assume anything about order. Docs make no guarantee and
533 have possible dictionary input.
Tim Petersc2659cf2003-05-12 20:19:37 +0000534
Brett Cannon74bfd702003-04-25 09:39:47 +0000535 """
536 expect_somewhere = ["1st=1", "2nd=2", "3rd=3"]
537 result = urllib.urlencode(given)
538 for expected in expect_somewhere:
Ezio Melottiaa980582010-01-23 23:04:36 +0000539 self.assertIn(expected, result,
Brett Cannon74bfd702003-04-25 09:39:47 +0000540 "testing %s: %s not found in %s" %
541 (test_type, expected, result))
542 self.assertEqual(result.count('&'), 2,
543 "testing %s: expected 2 '&'s; got %s" %
544 (test_type, result.count('&')))
545 amp_location = result.index('&')
546 on_amp_left = result[amp_location - 1]
547 on_amp_right = result[amp_location + 1]
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000548 self.assertTrue(on_amp_left.isdigit() and on_amp_right.isdigit(),
Brett Cannon74bfd702003-04-25 09:39:47 +0000549 "testing %s: '&' not located in proper place in %s" %
550 (test_type, result))
551 self.assertEqual(len(result), (5 * 3) + 2, #5 chars per thing and amps
552 "testing %s: "
553 "unexpected number of characters: %s != %s" %
554 (test_type, len(result), (5 * 3) + 2))
555
556 def test_using_mapping(self):
557 # Test passing in a mapping object as an argument.
558 self.help_inputtype({"1st":'1', "2nd":'2', "3rd":'3'},
559 "using dict as input type")
560
561 def test_using_sequence(self):
562 # Test passing in a sequence of two-item sequences as an argument.
563 self.help_inputtype([('1st', '1'), ('2nd', '2'), ('3rd', '3')],
564 "using sequence of two-item tuples as input")
565
566 def test_quoting(self):
567 # Make sure keys and values are quoted using quote_plus()
568 given = {"&":"="}
569 expect = "%s=%s" % (hexescape('&'), hexescape('='))
570 result = urllib.urlencode(given)
571 self.assertEqual(expect, result)
572 given = {"key name":"A bunch of pluses"}
573 expect = "key+name=A+bunch+of+pluses"
574 result = urllib.urlencode(given)
575 self.assertEqual(expect, result)
576
577 def test_doseq(self):
578 # Test that passing True for 'doseq' parameter works correctly
579 given = {'sequence':['1', '2', '3']}
580 expect = "sequence=%s" % urllib.quote_plus(str(['1', '2', '3']))
581 result = urllib.urlencode(given)
582 self.assertEqual(expect, result)
583 result = urllib.urlencode(given, True)
584 for value in given["sequence"]:
585 expect = "sequence=%s" % value
Ezio Melottiaa980582010-01-23 23:04:36 +0000586 self.assertIn(expect, result)
Brett Cannon74bfd702003-04-25 09:39:47 +0000587 self.assertEqual(result.count('&'), 2,
588 "Expected 2 '&'s, got %s" % result.count('&'))
589
590class Pathname_Tests(unittest.TestCase):
591 """Test pathname2url() and url2pathname()"""
592
593 def test_basic(self):
594 # Make sure simple tests pass
595 expected_path = os.path.join("parts", "of", "a", "path")
596 expected_url = "parts/of/a/path"
597 result = urllib.pathname2url(expected_path)
598 self.assertEqual(expected_url, result,
599 "pathname2url() failed; %s != %s" %
600 (result, expected_url))
601 result = urllib.url2pathname(expected_url)
602 self.assertEqual(expected_path, result,
603 "url2pathame() failed; %s != %s" %
604 (result, expected_path))
605
606 def test_quoting(self):
607 # Test automatic quoting and unquoting works for pathnam2url() and
608 # url2pathname() respectively
609 given = os.path.join("needs", "quot=ing", "here")
610 expect = "needs/%s/here" % urllib.quote("quot=ing")
611 result = urllib.pathname2url(given)
612 self.assertEqual(expect, result,
613 "pathname2url() failed; %s != %s" %
614 (expect, result))
615 expect = given
616 result = urllib.url2pathname(result)
617 self.assertEqual(expect, result,
618 "url2pathname() failed; %s != %s" %
619 (expect, result))
620 given = os.path.join("make sure", "using_quote")
621 expect = "%s/using_quote" % urllib.quote("make sure")
622 result = urllib.pathname2url(given)
623 self.assertEqual(expect, result,
624 "pathname2url() failed; %s != %s" %
625 (expect, result))
626 given = "make+sure/using_unquote"
627 expect = os.path.join("make+sure", "using_unquote")
628 result = urllib.url2pathname(given)
629 self.assertEqual(expect, result,
630 "url2pathname() failed; %s != %s" %
631 (expect, result))
Tim Petersc2659cf2003-05-12 20:19:37 +0000632
Senthil Kumaran5e95e762009-03-30 21:51:50 +0000633class Utility_Tests(unittest.TestCase):
634 """Testcase to test the various utility functions in the urllib."""
635
636 def test_splitpasswd(self):
637 """Some of the password examples are not sensible, but it is added to
638 confirming to RFC2617 and addressing issue4675.
639 """
640 self.assertEqual(('user', 'ab'),urllib.splitpasswd('user:ab'))
641 self.assertEqual(('user', 'a\nb'),urllib.splitpasswd('user:a\nb'))
642 self.assertEqual(('user', 'a\tb'),urllib.splitpasswd('user:a\tb'))
643 self.assertEqual(('user', 'a\rb'),urllib.splitpasswd('user:a\rb'))
644 self.assertEqual(('user', 'a\fb'),urllib.splitpasswd('user:a\fb'))
645 self.assertEqual(('user', 'a\vb'),urllib.splitpasswd('user:a\vb'))
646 self.assertEqual(('user', 'a:b'),urllib.splitpasswd('user:a:b'))
647
648
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000649class URLopener_Tests(unittest.TestCase):
650 """Testcase to test the open method of URLopener class."""
651
652 def test_quoted_open(self):
653 class DummyURLopener(urllib.URLopener):
654 def open_spam(self, url):
655 return url
656
657 self.assertEqual(DummyURLopener().open(
658 'spam://example/ /'),'//example/%20/')
659
Senthil Kumaran18d5a692010-02-20 22:05:34 +0000660 # test the safe characters are not quoted by urlopen
661 self.assertEqual(DummyURLopener().open(
662 "spam://c:|windows%/:=&?~#+!$,;'@()*[]|/path/"),
663 "//c:|windows%/:=&?~#+!$,;'@()*[]|/path/")
664
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000665
Facundo Batistad9880d02007-05-25 04:20:22 +0000666# Just commented them out.
667# Can't really tell why keep failing in windows and sparc.
Ezio Melottic2077b02011-03-16 12:34:31 +0200668# Everywhere else they work ok, but on those machines, sometimes
Facundo Batistad9880d02007-05-25 04:20:22 +0000669# fail in one of the tests, sometimes in other. I have a linux, and
670# the tests go ok.
671# If anybody has one of the problematic enviroments, please help!
672# . Facundo
673#
674# def server(evt):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000675# import socket, time
Facundo Batistad9880d02007-05-25 04:20:22 +0000676# serv = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
677# serv.settimeout(3)
678# serv.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
679# serv.bind(("", 9093))
680# serv.listen(5)
681# try:
682# conn, addr = serv.accept()
683# conn.send("1 Hola mundo\n")
684# cantdata = 0
685# while cantdata < 13:
686# data = conn.recv(13-cantdata)
687# cantdata += len(data)
688# time.sleep(.3)
689# conn.send("2 No more lines\n")
690# conn.close()
691# except socket.timeout:
692# pass
693# finally:
694# serv.close()
695# evt.set()
696#
697# class FTPWrapperTests(unittest.TestCase):
698#
699# def setUp(self):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000700# import ftplib, time, threading
Facundo Batistad9880d02007-05-25 04:20:22 +0000701# ftplib.FTP.port = 9093
702# self.evt = threading.Event()
703# threading.Thread(target=server, args=(self.evt,)).start()
704# time.sleep(.1)
705#
706# def tearDown(self):
707# self.evt.wait()
708#
709# def testBasic(self):
710# # connects
711# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000712# ftp.close()
Facundo Batistad9880d02007-05-25 04:20:22 +0000713#
714# def testTimeoutNone(self):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000715# # global default timeout is ignored
716# import socket
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000717# self.assertTrue(socket.getdefaulttimeout() is None)
Facundo Batistad9880d02007-05-25 04:20:22 +0000718# socket.setdefaulttimeout(30)
719# try:
720# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
721# finally:
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000722# socket.setdefaulttimeout(None)
Facundo Batistad9880d02007-05-25 04:20:22 +0000723# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000724# ftp.close()
Facundo Batistad9880d02007-05-25 04:20:22 +0000725#
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000726# def testTimeoutDefault(self):
727# # global default timeout is used
728# import socket
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000729# self.assertTrue(socket.getdefaulttimeout() is None)
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000730# socket.setdefaulttimeout(30)
731# try:
732# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
733# finally:
734# socket.setdefaulttimeout(None)
735# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
736# ftp.close()
737#
738# def testTimeoutValue(self):
739# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [],
740# timeout=30)
741# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
742# ftp.close()
Facundo Batista711a54e2007-05-24 17:50:54 +0000743
Skip Montanaro080c9972001-01-28 21:12:22 +0000744
745
Brett Cannon74bfd702003-04-25 09:39:47 +0000746def test_main():
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000747 import warnings
Brett Cannon672237d2008-09-09 00:49:16 +0000748 with warnings.catch_warnings():
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000749 warnings.filterwarnings('ignore', ".*urllib\.urlopen.*Python 3.0",
750 DeprecationWarning)
751 test_support.run_unittest(
752 urlopen_FileTests,
753 urlopen_HttpTests,
754 urlretrieve_FileTests,
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000755 ProxyTests,
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000756 QuotingTests,
757 UnquotingTests,
758 urlencode_Tests,
759 Pathname_Tests,
Senthil Kumaran5e95e762009-03-30 21:51:50 +0000760 Utility_Tests,
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000761 URLopener_Tests,
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000762 #FTPWrapperTests,
763 )
Brett Cannon74bfd702003-04-25 09:39:47 +0000764
765
766
767if __name__ == '__main__':
768 test_main()