blob: 11a64d0e0f564e0ebf851d8a9890abd49663c8df [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
Kurt B. Kaiser0f7c25d2008-01-02 04:11:28 +0000151 def test_read_bogus(self):
Kurt B. Kaiser0a112322008-01-02 05:23:38 +0000152 # urlopen() should raise IOError for many error codes.
Kurt B. Kaiser0f7c25d2008-01-02 04:11:28 +0000153 self.fakehttp('''HTTP/1.1 401 Authentication Required
154Date: Wed, 02 Jan 2008 03:03:54 GMT
155Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e
156Connection: close
157Content-Type: text/html; charset=iso-8859-1
158''')
159 try:
160 self.assertRaises(IOError, urllib.urlopen, "http://python.org/")
161 finally:
162 self.unfakehttp()
163
guido@google.comf1509302011-03-28 13:47:01 -0700164 def test_invalid_redirect(self):
165 # urlopen() should raise IOError for many error codes.
166 self.fakehttp("""HTTP/1.1 302 Found
167Date: Wed, 02 Jan 2008 03:03:54 GMT
168Server: Apache/1.3.33 (Debian GNU/Linux) mod_ssl/2.8.22 OpenSSL/0.9.7e
169Location: file:README
170Connection: close
171Content-Type: text/html; charset=iso-8859-1
172""")
173 try:
174 self.assertRaises(IOError, urllib.urlopen, "http://python.org/")
175 finally:
176 self.unfakehttp()
177
Georg Brandlf66b6032007-03-14 08:27:52 +0000178 def test_empty_socket(self):
Kurt B. Kaiser0a112322008-01-02 05:23:38 +0000179 # urlopen() raises IOError if the underlying socket does not send any
180 # data. (#1680230)
Georg Brandlf66b6032007-03-14 08:27:52 +0000181 self.fakehttp('')
182 try:
183 self.assertRaises(IOError, urllib.urlopen, 'http://something')
184 finally:
185 self.unfakehttp()
186
Brett Cannon19691362003-04-29 05:08:06 +0000187class urlretrieve_FileTests(unittest.TestCase):
Brett Cannon74bfd702003-04-25 09:39:47 +0000188 """Test urllib.urlretrieve() on local files"""
Skip Montanaro080c9972001-01-28 21:12:22 +0000189
Brett Cannon19691362003-04-29 05:08:06 +0000190 def setUp(self):
Georg Brandl5a650a22005-08-26 08:51:34 +0000191 # Create a list of temporary files. Each item in the list is a file
192 # name (absolute path or relative to the current working directory).
193 # All files in this list will be deleted in the tearDown method. Note,
194 # this only helps to makes sure temporary files get deleted, but it
195 # does nothing about trying to close files that may still be open. It
196 # is the responsibility of the developer to properly close files even
197 # when exceptional conditions occur.
198 self.tempFiles = []
199
Brett Cannon19691362003-04-29 05:08:06 +0000200 # Create a temporary file.
Georg Brandl5a650a22005-08-26 08:51:34 +0000201 self.registerFileForCleanUp(test_support.TESTFN)
Brett Cannon19691362003-04-29 05:08:06 +0000202 self.text = 'testing urllib.urlretrieve'
Georg Brandl5a650a22005-08-26 08:51:34 +0000203 try:
204 FILE = file(test_support.TESTFN, 'wb')
205 FILE.write(self.text)
206 FILE.close()
207 finally:
208 try: FILE.close()
209 except: pass
Brett Cannon19691362003-04-29 05:08:06 +0000210
211 def tearDown(self):
Georg Brandl5a650a22005-08-26 08:51:34 +0000212 # Delete the temporary files.
213 for each in self.tempFiles:
214 try: os.remove(each)
215 except: pass
216
217 def constructLocalFileUrl(self, filePath):
218 return "file://%s" % urllib.pathname2url(os.path.abspath(filePath))
219
220 def createNewTempFile(self, data=""):
221 """Creates a new temporary file containing the specified data,
222 registers the file for deletion during the test fixture tear down, and
223 returns the absolute path of the file."""
224
225 newFd, newFilePath = tempfile.mkstemp()
226 try:
227 self.registerFileForCleanUp(newFilePath)
228 newFile = os.fdopen(newFd, "wb")
229 newFile.write(data)
230 newFile.close()
231 finally:
232 try: newFile.close()
233 except: pass
234 return newFilePath
235
236 def registerFileForCleanUp(self, fileName):
237 self.tempFiles.append(fileName)
Brett Cannon19691362003-04-29 05:08:06 +0000238
239 def test_basic(self):
240 # Make sure that a local file just gets its own location returned and
241 # a headers value is returned.
242 result = urllib.urlretrieve("file:%s" % test_support.TESTFN)
243 self.assertEqual(result[0], test_support.TESTFN)
Ezio Melottib0f5adc2010-01-24 16:58:36 +0000244 self.assertIsInstance(result[1], mimetools.Message,
245 "did not get a mimetools.Message instance as "
246 "second returned value")
Brett Cannon19691362003-04-29 05:08:06 +0000247
248 def test_copy(self):
249 # Test that setting the filename argument works.
250 second_temp = "%s.2" % test_support.TESTFN
Georg Brandl5a650a22005-08-26 08:51:34 +0000251 self.registerFileForCleanUp(second_temp)
252 result = urllib.urlretrieve(self.constructLocalFileUrl(
253 test_support.TESTFN), second_temp)
Brett Cannon19691362003-04-29 05:08:06 +0000254 self.assertEqual(second_temp, result[0])
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000255 self.assertTrue(os.path.exists(second_temp), "copy of the file was not "
Brett Cannon19691362003-04-29 05:08:06 +0000256 "made")
257 FILE = file(second_temp, 'rb')
258 try:
259 text = FILE.read()
Brett Cannon19691362003-04-29 05:08:06 +0000260 FILE.close()
Georg Brandl5a650a22005-08-26 08:51:34 +0000261 finally:
262 try: FILE.close()
263 except: pass
Brett Cannon19691362003-04-29 05:08:06 +0000264 self.assertEqual(self.text, text)
265
266 def test_reporthook(self):
267 # Make sure that the reporthook works.
268 def hooktester(count, block_size, total_size, count_holder=[0]):
Ezio Melottib0f5adc2010-01-24 16:58:36 +0000269 self.assertIsInstance(count, int)
270 self.assertIsInstance(block_size, int)
271 self.assertIsInstance(total_size, int)
Brett Cannon19691362003-04-29 05:08:06 +0000272 self.assertEqual(count, count_holder[0])
273 count_holder[0] = count_holder[0] + 1
274 second_temp = "%s.2" % test_support.TESTFN
Georg Brandl5a650a22005-08-26 08:51:34 +0000275 self.registerFileForCleanUp(second_temp)
276 urllib.urlretrieve(self.constructLocalFileUrl(test_support.TESTFN),
277 second_temp, hooktester)
278
279 def test_reporthook_0_bytes(self):
280 # Test on zero length file. Should call reporthook only 1 time.
281 report = []
282 def hooktester(count, block_size, total_size, _report=report):
283 _report.append((count, block_size, total_size))
284 srcFileName = self.createNewTempFile()
285 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
286 test_support.TESTFN, hooktester)
287 self.assertEqual(len(report), 1)
288 self.assertEqual(report[0][2], 0)
289
290 def test_reporthook_5_bytes(self):
291 # Test on 5 byte file. Should call reporthook only 2 times (once when
292 # the "network connection" is established and once when the block is
293 # read). Since the block size is 8192 bytes, only one block read is
294 # required to read the entire file.
295 report = []
296 def hooktester(count, block_size, total_size, _report=report):
297 _report.append((count, block_size, total_size))
298 srcFileName = self.createNewTempFile("x" * 5)
299 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
300 test_support.TESTFN, hooktester)
301 self.assertEqual(len(report), 2)
302 self.assertEqual(report[0][1], 8192)
303 self.assertEqual(report[0][2], 5)
304
305 def test_reporthook_8193_bytes(self):
306 # Test on 8193 byte file. Should call reporthook only 3 times (once
307 # when the "network connection" is established, once for the next 8192
308 # bytes, and once for the last byte).
309 report = []
310 def hooktester(count, block_size, total_size, _report=report):
311 _report.append((count, block_size, total_size))
312 srcFileName = self.createNewTempFile("x" * 8193)
313 urllib.urlretrieve(self.constructLocalFileUrl(srcFileName),
314 test_support.TESTFN, hooktester)
315 self.assertEqual(len(report), 3)
316 self.assertEqual(report[0][1], 8192)
317 self.assertEqual(report[0][2], 8193)
Skip Montanaro080c9972001-01-28 21:12:22 +0000318
Brett Cannon74bfd702003-04-25 09:39:47 +0000319class QuotingTests(unittest.TestCase):
320 """Tests for urllib.quote() and urllib.quote_plus()
Tim Petersc2659cf2003-05-12 20:19:37 +0000321
Brett Cannon74bfd702003-04-25 09:39:47 +0000322 According to RFC 2396 ("Uniform Resource Identifiers), to escape a
323 character you write it as '%' + <2 character US-ASCII hex value>. The Python
324 code of ``'%' + hex(ord(<character>))[2:]`` escapes a character properly.
325 Case does not matter on the hex letters.
326
327 The various character sets specified are:
Tim Petersc2659cf2003-05-12 20:19:37 +0000328
Brett Cannon74bfd702003-04-25 09:39:47 +0000329 Reserved characters : ";/?:@&=+$,"
330 Have special meaning in URIs and must be escaped if not being used for
331 their special meaning
332 Data characters : letters, digits, and "-_.!~*'()"
333 Unreserved and do not need to be escaped; can be, though, if desired
334 Control characters : 0x00 - 0x1F, 0x7F
335 Have no use in URIs so must be escaped
336 space : 0x20
337 Must be escaped
338 Delimiters : '<>#%"'
339 Must be escaped
340 Unwise : "{}|\^[]`"
341 Must be escaped
Tim Petersc2659cf2003-05-12 20:19:37 +0000342
Brett Cannon74bfd702003-04-25 09:39:47 +0000343 """
344
345 def test_never_quote(self):
346 # Make sure quote() does not quote letters, digits, and "_,.-"
347 do_not_quote = '' .join(["ABCDEFGHIJKLMNOPQRSTUVWXYZ",
348 "abcdefghijklmnopqrstuvwxyz",
349 "0123456789",
350 "_.-"])
351 result = urllib.quote(do_not_quote)
352 self.assertEqual(do_not_quote, result,
353 "using quote(): %s != %s" % (do_not_quote, result))
354 result = urllib.quote_plus(do_not_quote)
355 self.assertEqual(do_not_quote, result,
356 "using quote_plus(): %s != %s" % (do_not_quote, result))
357
358 def test_default_safe(self):
359 # Test '/' is default value for 'safe' parameter
360 self.assertEqual(urllib.quote.func_defaults[0], '/')
361
362 def test_safe(self):
363 # Test setting 'safe' parameter does what it should do
364 quote_by_default = "<>"
365 result = urllib.quote(quote_by_default, safe=quote_by_default)
366 self.assertEqual(quote_by_default, result,
367 "using quote(): %s != %s" % (quote_by_default, result))
368 result = urllib.quote_plus(quote_by_default, safe=quote_by_default)
369 self.assertEqual(quote_by_default, result,
370 "using quote_plus(): %s != %s" %
371 (quote_by_default, result))
372
373 def test_default_quoting(self):
374 # Make sure all characters that should be quoted are by default sans
375 # space (separate test for that).
376 should_quote = [chr(num) for num in range(32)] # For 0x00 - 0x1F
377 should_quote.append('<>#%"{}|\^[]`')
378 should_quote.append(chr(127)) # For 0x7F
379 should_quote = ''.join(should_quote)
380 for char in should_quote:
381 result = urllib.quote(char)
382 self.assertEqual(hexescape(char), result,
383 "using quote(): %s should be escaped to %s, not %s" %
384 (char, hexescape(char), result))
385 result = urllib.quote_plus(char)
386 self.assertEqual(hexescape(char), result,
387 "using quote_plus(): "
Tim Petersc2659cf2003-05-12 20:19:37 +0000388 "%s should be escapes to %s, not %s" %
Brett Cannon74bfd702003-04-25 09:39:47 +0000389 (char, hexescape(char), result))
390 del should_quote
391 partial_quote = "ab[]cd"
392 expected = "ab%5B%5Dcd"
393 result = urllib.quote(partial_quote)
394 self.assertEqual(expected, result,
395 "using quote(): %s != %s" % (expected, result))
396 self.assertEqual(expected, result,
397 "using quote_plus(): %s != %s" % (expected, result))
Senthil Kumaranc7743aa2010-07-19 17:35:50 +0000398 self.assertRaises(TypeError, urllib.quote, None)
Brett Cannon74bfd702003-04-25 09:39:47 +0000399
400 def test_quoting_space(self):
401 # Make sure quote() and quote_plus() handle spaces as specified in
402 # their unique way
403 result = urllib.quote(' ')
404 self.assertEqual(result, hexescape(' '),
405 "using quote(): %s != %s" % (result, hexescape(' ')))
406 result = urllib.quote_plus(' ')
407 self.assertEqual(result, '+',
408 "using quote_plus(): %s != +" % result)
409 given = "a b cd e f"
410 expect = given.replace(' ', hexescape(' '))
411 result = urllib.quote(given)
412 self.assertEqual(expect, result,
413 "using quote(): %s != %s" % (expect, result))
414 expect = given.replace(' ', '+')
415 result = urllib.quote_plus(given)
416 self.assertEqual(expect, result,
417 "using quote_plus(): %s != %s" % (expect, result))
418
Raymond Hettinger2bdec7b2005-09-10 14:30:09 +0000419 def test_quoting_plus(self):
420 self.assertEqual(urllib.quote_plus('alpha+beta gamma'),
421 'alpha%2Bbeta+gamma')
422 self.assertEqual(urllib.quote_plus('alpha+beta gamma', '+'),
423 'alpha+beta+gamma')
424
Brett Cannon74bfd702003-04-25 09:39:47 +0000425class UnquotingTests(unittest.TestCase):
426 """Tests for unquote() and unquote_plus()
Tim Petersc2659cf2003-05-12 20:19:37 +0000427
Brett Cannon74bfd702003-04-25 09:39:47 +0000428 See the doc string for quoting_Tests for details on quoting and such.
429
430 """
431
432 def test_unquoting(self):
433 # Make sure unquoting of all ASCII values works
434 escape_list = []
435 for num in range(128):
436 given = hexescape(chr(num))
437 expect = chr(num)
438 result = urllib.unquote(given)
439 self.assertEqual(expect, result,
440 "using unquote(): %s != %s" % (expect, result))
441 result = urllib.unquote_plus(given)
442 self.assertEqual(expect, result,
443 "using unquote_plus(): %s != %s" %
444 (expect, result))
445 escape_list.append(given)
446 escape_string = ''.join(escape_list)
447 del escape_list
448 result = urllib.unquote(escape_string)
449 self.assertEqual(result.count('%'), 1,
450 "using quote(): not all characters escaped; %s" %
451 result)
452 result = urllib.unquote(escape_string)
453 self.assertEqual(result.count('%'), 1,
454 "using unquote(): not all characters escaped: "
455 "%s" % result)
456
Senthil Kumaranf3e9b2a2010-03-18 12:14:15 +0000457 def test_unquoting_badpercent(self):
458 # Test unquoting on bad percent-escapes
459 given = '%xab'
460 expect = given
461 result = urllib.unquote(given)
462 self.assertEqual(expect, result, "using unquote(): %r != %r"
463 % (expect, result))
464 given = '%x'
465 expect = given
466 result = urllib.unquote(given)
467 self.assertEqual(expect, result, "using unquote(): %r != %r"
468 % (expect, result))
469 given = '%'
470 expect = given
471 result = urllib.unquote(given)
472 self.assertEqual(expect, result, "using unquote(): %r != %r"
473 % (expect, result))
474
475 def test_unquoting_mixed_case(self):
476 # Test unquoting on mixed-case hex digits in the percent-escapes
477 given = '%Ab%eA'
478 expect = '\xab\xea'
479 result = urllib.unquote(given)
480 self.assertEqual(expect, result, "using unquote(): %r != %r"
481 % (expect, result))
482
Brett Cannon74bfd702003-04-25 09:39:47 +0000483 def test_unquoting_parts(self):
484 # Make sure unquoting works when have non-quoted characters
485 # interspersed
486 given = 'ab%sd' % hexescape('c')
487 expect = "abcd"
488 result = urllib.unquote(given)
489 self.assertEqual(expect, result,
490 "using quote(): %s != %s" % (expect, result))
491 result = urllib.unquote_plus(given)
492 self.assertEqual(expect, result,
493 "using unquote_plus(): %s != %s" % (expect, result))
Tim Petersc2659cf2003-05-12 20:19:37 +0000494
Brett Cannon74bfd702003-04-25 09:39:47 +0000495 def test_unquoting_plus(self):
496 # Test difference between unquote() and unquote_plus()
497 given = "are+there+spaces..."
498 expect = given
499 result = urllib.unquote(given)
500 self.assertEqual(expect, result,
501 "using unquote(): %s != %s" % (expect, result))
502 expect = given.replace('+', ' ')
503 result = urllib.unquote_plus(given)
504 self.assertEqual(expect, result,
505 "using unquote_plus(): %s != %s" % (expect, result))
506
Raymond Hettinger4b0f20d2005-10-15 16:41:53 +0000507 def test_unquote_with_unicode(self):
508 r = urllib.unquote(u'br%C3%BCckner_sapporo_20050930.doc')
509 self.assertEqual(r, u'br\xc3\xbcckner_sapporo_20050930.doc')
510
Brett Cannon74bfd702003-04-25 09:39:47 +0000511class urlencode_Tests(unittest.TestCase):
512 """Tests for urlencode()"""
513
514 def help_inputtype(self, given, test_type):
515 """Helper method for testing different input types.
Tim Petersc2659cf2003-05-12 20:19:37 +0000516
Brett Cannon74bfd702003-04-25 09:39:47 +0000517 'given' must lead to only the pairs:
518 * 1st, 1
519 * 2nd, 2
520 * 3rd, 3
Tim Petersc2659cf2003-05-12 20:19:37 +0000521
Brett Cannon74bfd702003-04-25 09:39:47 +0000522 Test cannot assume anything about order. Docs make no guarantee and
523 have possible dictionary input.
Tim Petersc2659cf2003-05-12 20:19:37 +0000524
Brett Cannon74bfd702003-04-25 09:39:47 +0000525 """
526 expect_somewhere = ["1st=1", "2nd=2", "3rd=3"]
527 result = urllib.urlencode(given)
528 for expected in expect_somewhere:
Ezio Melottiaa980582010-01-23 23:04:36 +0000529 self.assertIn(expected, result,
Brett Cannon74bfd702003-04-25 09:39:47 +0000530 "testing %s: %s not found in %s" %
531 (test_type, expected, result))
532 self.assertEqual(result.count('&'), 2,
533 "testing %s: expected 2 '&'s; got %s" %
534 (test_type, result.count('&')))
535 amp_location = result.index('&')
536 on_amp_left = result[amp_location - 1]
537 on_amp_right = result[amp_location + 1]
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000538 self.assertTrue(on_amp_left.isdigit() and on_amp_right.isdigit(),
Brett Cannon74bfd702003-04-25 09:39:47 +0000539 "testing %s: '&' not located in proper place in %s" %
540 (test_type, result))
541 self.assertEqual(len(result), (5 * 3) + 2, #5 chars per thing and amps
542 "testing %s: "
543 "unexpected number of characters: %s != %s" %
544 (test_type, len(result), (5 * 3) + 2))
545
546 def test_using_mapping(self):
547 # Test passing in a mapping object as an argument.
548 self.help_inputtype({"1st":'1', "2nd":'2', "3rd":'3'},
549 "using dict as input type")
550
551 def test_using_sequence(self):
552 # Test passing in a sequence of two-item sequences as an argument.
553 self.help_inputtype([('1st', '1'), ('2nd', '2'), ('3rd', '3')],
554 "using sequence of two-item tuples as input")
555
556 def test_quoting(self):
557 # Make sure keys and values are quoted using quote_plus()
558 given = {"&":"="}
559 expect = "%s=%s" % (hexescape('&'), hexescape('='))
560 result = urllib.urlencode(given)
561 self.assertEqual(expect, result)
562 given = {"key name":"A bunch of pluses"}
563 expect = "key+name=A+bunch+of+pluses"
564 result = urllib.urlencode(given)
565 self.assertEqual(expect, result)
566
567 def test_doseq(self):
568 # Test that passing True for 'doseq' parameter works correctly
569 given = {'sequence':['1', '2', '3']}
570 expect = "sequence=%s" % urllib.quote_plus(str(['1', '2', '3']))
571 result = urllib.urlencode(given)
572 self.assertEqual(expect, result)
573 result = urllib.urlencode(given, True)
574 for value in given["sequence"]:
575 expect = "sequence=%s" % value
Ezio Melottiaa980582010-01-23 23:04:36 +0000576 self.assertIn(expect, result)
Brett Cannon74bfd702003-04-25 09:39:47 +0000577 self.assertEqual(result.count('&'), 2,
578 "Expected 2 '&'s, got %s" % result.count('&'))
579
580class Pathname_Tests(unittest.TestCase):
581 """Test pathname2url() and url2pathname()"""
582
583 def test_basic(self):
584 # Make sure simple tests pass
585 expected_path = os.path.join("parts", "of", "a", "path")
586 expected_url = "parts/of/a/path"
587 result = urllib.pathname2url(expected_path)
588 self.assertEqual(expected_url, result,
589 "pathname2url() failed; %s != %s" %
590 (result, expected_url))
591 result = urllib.url2pathname(expected_url)
592 self.assertEqual(expected_path, result,
593 "url2pathame() failed; %s != %s" %
594 (result, expected_path))
595
596 def test_quoting(self):
597 # Test automatic quoting and unquoting works for pathnam2url() and
598 # url2pathname() respectively
599 given = os.path.join("needs", "quot=ing", "here")
600 expect = "needs/%s/here" % urllib.quote("quot=ing")
601 result = urllib.pathname2url(given)
602 self.assertEqual(expect, result,
603 "pathname2url() failed; %s != %s" %
604 (expect, result))
605 expect = given
606 result = urllib.url2pathname(result)
607 self.assertEqual(expect, result,
608 "url2pathname() failed; %s != %s" %
609 (expect, result))
610 given = os.path.join("make sure", "using_quote")
611 expect = "%s/using_quote" % urllib.quote("make sure")
612 result = urllib.pathname2url(given)
613 self.assertEqual(expect, result,
614 "pathname2url() failed; %s != %s" %
615 (expect, result))
616 given = "make+sure/using_unquote"
617 expect = os.path.join("make+sure", "using_unquote")
618 result = urllib.url2pathname(given)
619 self.assertEqual(expect, result,
620 "url2pathname() failed; %s != %s" %
621 (expect, result))
Tim Petersc2659cf2003-05-12 20:19:37 +0000622
Senthil Kumaran5e95e762009-03-30 21:51:50 +0000623class Utility_Tests(unittest.TestCase):
624 """Testcase to test the various utility functions in the urllib."""
625
626 def test_splitpasswd(self):
627 """Some of the password examples are not sensible, but it is added to
628 confirming to RFC2617 and addressing issue4675.
629 """
630 self.assertEqual(('user', 'ab'),urllib.splitpasswd('user:ab'))
631 self.assertEqual(('user', 'a\nb'),urllib.splitpasswd('user:a\nb'))
632 self.assertEqual(('user', 'a\tb'),urllib.splitpasswd('user:a\tb'))
633 self.assertEqual(('user', 'a\rb'),urllib.splitpasswd('user:a\rb'))
634 self.assertEqual(('user', 'a\fb'),urllib.splitpasswd('user:a\fb'))
635 self.assertEqual(('user', 'a\vb'),urllib.splitpasswd('user:a\vb'))
636 self.assertEqual(('user', 'a:b'),urllib.splitpasswd('user:a:b'))
637
638
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000639class URLopener_Tests(unittest.TestCase):
640 """Testcase to test the open method of URLopener class."""
641
642 def test_quoted_open(self):
643 class DummyURLopener(urllib.URLopener):
644 def open_spam(self, url):
645 return url
646
647 self.assertEqual(DummyURLopener().open(
648 'spam://example/ /'),'//example/%20/')
649
Senthil Kumaran18d5a692010-02-20 22:05:34 +0000650 # test the safe characters are not quoted by urlopen
651 self.assertEqual(DummyURLopener().open(
652 "spam://c:|windows%/:=&?~#+!$,;'@()*[]|/path/"),
653 "//c:|windows%/:=&?~#+!$,;'@()*[]|/path/")
654
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000655
Facundo Batistad9880d02007-05-25 04:20:22 +0000656# Just commented them out.
657# Can't really tell why keep failing in windows and sparc.
Ezio Melottic2077b02011-03-16 12:34:31 +0200658# Everywhere else they work ok, but on those machines, sometimes
Facundo Batistad9880d02007-05-25 04:20:22 +0000659# fail in one of the tests, sometimes in other. I have a linux, and
660# the tests go ok.
661# If anybody has one of the problematic enviroments, please help!
662# . Facundo
663#
664# def server(evt):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000665# import socket, time
Facundo Batistad9880d02007-05-25 04:20:22 +0000666# serv = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
667# serv.settimeout(3)
668# serv.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
669# serv.bind(("", 9093))
670# serv.listen(5)
671# try:
672# conn, addr = serv.accept()
673# conn.send("1 Hola mundo\n")
674# cantdata = 0
675# while cantdata < 13:
676# data = conn.recv(13-cantdata)
677# cantdata += len(data)
678# time.sleep(.3)
679# conn.send("2 No more lines\n")
680# conn.close()
681# except socket.timeout:
682# pass
683# finally:
684# serv.close()
685# evt.set()
686#
687# class FTPWrapperTests(unittest.TestCase):
688#
689# def setUp(self):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000690# import ftplib, time, threading
Facundo Batistad9880d02007-05-25 04:20:22 +0000691# ftplib.FTP.port = 9093
692# self.evt = threading.Event()
693# threading.Thread(target=server, args=(self.evt,)).start()
694# time.sleep(.1)
695#
696# def tearDown(self):
697# self.evt.wait()
698#
699# def testBasic(self):
700# # connects
701# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000702# ftp.close()
Facundo Batistad9880d02007-05-25 04:20:22 +0000703#
704# def testTimeoutNone(self):
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000705# # global default timeout is ignored
706# import socket
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000707# self.assertTrue(socket.getdefaulttimeout() is None)
Facundo Batistad9880d02007-05-25 04:20:22 +0000708# socket.setdefaulttimeout(30)
709# try:
710# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
711# finally:
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000712# socket.setdefaulttimeout(None)
Facundo Batistad9880d02007-05-25 04:20:22 +0000713# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000714# ftp.close()
Facundo Batistad9880d02007-05-25 04:20:22 +0000715#
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000716# def testTimeoutDefault(self):
717# # global default timeout is used
718# import socket
Benjamin Peterson5c8da862009-06-30 22:57:08 +0000719# self.assertTrue(socket.getdefaulttimeout() is None)
Facundo Batista4f1b1ed2008-05-29 16:39:26 +0000720# socket.setdefaulttimeout(30)
721# try:
722# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [])
723# finally:
724# socket.setdefaulttimeout(None)
725# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
726# ftp.close()
727#
728# def testTimeoutValue(self):
729# ftp = urllib.ftpwrapper("myuser", "mypass", "localhost", 9093, [],
730# timeout=30)
731# self.assertEqual(ftp.ftp.sock.gettimeout(), 30)
732# ftp.close()
Facundo Batista711a54e2007-05-24 17:50:54 +0000733
Skip Montanaro080c9972001-01-28 21:12:22 +0000734
735
Brett Cannon74bfd702003-04-25 09:39:47 +0000736def test_main():
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000737 import warnings
Brett Cannon672237d2008-09-09 00:49:16 +0000738 with warnings.catch_warnings():
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000739 warnings.filterwarnings('ignore', ".*urllib\.urlopen.*Python 3.0",
740 DeprecationWarning)
741 test_support.run_unittest(
742 urlopen_FileTests,
743 urlopen_HttpTests,
744 urlretrieve_FileTests,
Benjamin Peterson2c7470d2008-09-21 21:27:51 +0000745 ProxyTests,
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000746 QuotingTests,
747 UnquotingTests,
748 urlencode_Tests,
749 Pathname_Tests,
Senthil Kumaran5e95e762009-03-30 21:51:50 +0000750 Utility_Tests,
Senthil Kumaran7c2867f2009-04-21 03:24:19 +0000751 URLopener_Tests,
Brett Cannon8bb8fa52008-07-02 01:57:08 +0000752 #FTPWrapperTests,
753 )
Brett Cannon74bfd702003-04-25 09:39:47 +0000754
755
756
757if __name__ == '__main__':
758 test_main()