def test_expected_decoding_quietly_wrong(self):
"""
An expected decoding can be wrong, and not throw errors.
Perhaps not ideal, but I don't know if it's possible to catch this.
"""
text = b"Ume\xe5"
expected = "Umeĺ" # Decoding incorrectly throws no error in this case
result = to_unicode(text, ["windows_1250"])
self.assertEqual(result, expected)
def test_expected_decoding_first(self):
"""
An undecoded bytestring will try "expected" before utf8.
This is because some non-UTF8 strings can be "valid" utf8.
"""
text = b"\x1b$BEl5~ET\x1b(B"
expected = "東京都" # as opposed to '\x1b$BEl5~ET\x1b(B'
result = to_unicode(text, ["iso-2022-jp"])
self.assertEqual(result, expected)
def test_already_unicode(self):
text = 'тнιѕ ιѕ αℓяєα∂у υηι¢σ∂є'
result = to_unicode(text)
self.assertEqual(result, text)
def test_not_bytes_or_string(self):
with self.assertRaises(AttributeError):
to_unicode(None)
def test_windows_1250(self):
text = b'Miko\xb3aj Kopernik'
expected = 'Mikołaj Kopernik'
result = to_unicode(text)
self.assertEqual(result, expected)
def test_latin_1(self):
text = b'Ume\xe5'
expected = 'Umeå'
result = to_unicode(text)
self.assertEqual(result, expected)
def test_ascii(self):
text = b'This is just plain ASCII'
expected = 'This is just plain ASCII'
result = to_unicode(text)
self.assertEqual(result, expected)
def test_expected_decoding_loudly_wrong(self):
"""An expected decoding can fall back to another encoding."""
text = b"\xff\xfe\xb5\x03\xbb\x03\xbb\x03\xb7\x03\xbd\x03\xb9\x03\xba\x03\xac\x03"
expected = "ελληνικά"
result = to_unicode(text, ["iso-2022-jp", "utf16"]) # `text` is utf16
self.assertEqual(result, expected)
def test_utf8(self):
text = b"Rhoi'r ffidil yn y t\xc3\xb4"
expected = "Rhoi'r ffidil yn y tô"
result = to_unicode(text)
self.assertEqual(result, expected)
def test_latin_1(self):
text = b"Ume\xe5"
expected = "Umeå"
result = to_unicode(text)
self.assertEqual(result, expected)
