Ahh ... I see. I should have done a bit more digging to find where the standard tests were. I created a few new tests that could be included in that test suite -- see the attached file. Do you think that this would be sufficient? - Dan Bill Janssen wrote:
Dan Mahn <dan.mahn@digidescorp.com> wrote:
Yes, that was a good idea. I found some problems, and attached a new version. It looks more complicated than I wanted, but it is a very regular repetition, so I hope it is generally readable.
That's great, but I was hoping for more tests in lib/test/test_urllib.py.
Bill
def test_encoding(self): # Test for special character encoding given = (("\u00a0", "\u00c1"),) expect = '%3F=%3F' result = urllib.parse.urlencode(given, encoding="ASCII", errors="replace") self.assertEqual(expect, result) result = urllib.parse.urlencode(given, True, encoding="ASCII", errors="replace") self.assertEqual(expect, result) given = (("\u00a0", (1, "\u00c1")),) # ... now with default utf-8 ... given = (("\u00a0", "\u00c1"),) expect = '%C2%A0=%C3%81' result = urllib.parse.urlencode(given) self.assertEqual(expect, result) # ... now with latin-1 ... expect = '%A0=%C1' result = urllib.parse.urlencode(given, encoding="latin-1") self.assertEqual(expect, result) # ... now with sequence ... given = (("\u00a0", (1, "\u00c1")),) expect = '%3F=1&%3F=%3F' result = urllib.parse.urlencode(given, True, encoding="ASCII", errors="replace") self.assertEqual(expect, result) # ... again with default utf-8 ... given = (("\u00a0", "\u00c1"),) expect = '%C2%A0=%C3%81' result = urllib.parse.urlencode(given, True) self.assertEqual(expect, result) # ... again with latin-1 ... expect = '%A0=%C1' result = urllib.parse.urlencode(given, True, encoding="latin-1") self.assertEqual(expect, result) def test_bytes(self): # Test for encoding bytes given = ((b'\xa0\x24', b'\xc1\x24'),) expect = '%A0%24=%C1%24' result = urllib.parse.urlencode(given) self.assertEqual(expect, result) # ... now with sequence ... result = urllib.parse.urlencode(given, True) self.assertEqual(expect, result) # ... now with safe and encoding ... expect = '%A0$=%C1$' result = urllib.parse.urlencode(given, safe=":$") self.assertEqual(expect, result) result = urllib.parse.urlencode(given, safe=":$", encoding="latin-1") self.assertEqual(expect, result) # ... again with sequence ... result = urllib.parse.urlencode(given, True, safe=":$") self.assertEqual(expect, result) result = urllib.parse.urlencode(given, True, safe=":$", encoding="latin-1") self.assertEqual(expect, result) # ... now with an actual sequence ... given = ((b'\xa0\x24', (b'\xc1\x24', 0xd)),) result = urllib.parse.urlencode(given, True, safe=":$") self.assert_(expect in result, "%s not found in %s" % (expect, result)) expect2 = '%A0$=1' self.assert_(expect2 in result, "%s not found in %s" % (expect2, result))