Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 1 | # |
| 2 | # test_multibytecodec.py |
| 3 | # Unit test for multibytecodec itself |
| 4 | # |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 5 | |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 6 | from test import support |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 7 | from test.support import TESTFN |
Berker Peksag | 1e8ee9b | 2016-04-24 07:31:42 +0300 | [diff] [blame] | 8 | import unittest, io, codecs, sys |
Georg Brandl | 3dbca81 | 2008-07-23 16:10:53 +0000 | [diff] [blame] | 9 | import _multibytecodec |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 10 | |
| 11 | ALL_CJKENCODINGS = [ |
| 12 | # _codecs_cn |
| 13 | 'gb2312', 'gbk', 'gb18030', 'hz', |
| 14 | # _codecs_hk |
| 15 | 'big5hkscs', |
| 16 | # _codecs_jp |
| 17 | 'cp932', 'shift_jis', 'euc_jp', 'euc_jisx0213', 'shift_jisx0213', |
| 18 | 'euc_jis_2004', 'shift_jis_2004', |
| 19 | # _codecs_kr |
| 20 | 'cp949', 'euc_kr', 'johab', |
| 21 | # _codecs_tw |
| 22 | 'big5', 'cp950', |
| 23 | # _codecs_iso2022 |
| 24 | 'iso2022_jp', 'iso2022_jp_1', 'iso2022_jp_2', 'iso2022_jp_2004', |
| 25 | 'iso2022_jp_3', 'iso2022_jp_ext', 'iso2022_kr', |
| 26 | ] |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 27 | |
| 28 | class Test_MultibyteCodec(unittest.TestCase): |
| 29 | |
| 30 | def test_nullcoding(self): |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 31 | for enc in ALL_CJKENCODINGS: |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 32 | self.assertEqual(b''.decode(enc), '') |
| 33 | self.assertEqual(str(b'', enc), '') |
| 34 | self.assertEqual(''.encode(enc), b'') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 35 | |
| 36 | def test_str_decode(self): |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 37 | for enc in ALL_CJKENCODINGS: |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 38 | self.assertEqual('abcd'.encode(enc), b'abcd') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 39 | |
| 40 | def test_errorcallback_longindex(self): |
| 41 | dec = codecs.getdecoder('euc-kr') |
Christian Heimes | a37d4c6 | 2007-12-04 23:02:19 +0000 | [diff] [blame] | 42 | myreplace = lambda exc: ('', sys.maxsize+1) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 43 | codecs.register_error('test.cjktest', myreplace) |
| 44 | self.assertRaises(IndexError, dec, |
Antoine Pitrou | 44bf631 | 2009-01-22 11:59:55 +0000 | [diff] [blame] | 45 | b'apple\x92ham\x93spam', 'test.cjktest') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 46 | |
Serhiy Storchaka | a1543cd | 2015-02-21 01:19:58 +0200 | [diff] [blame] | 47 | def test_errorcallback_custom_ignore(self): |
| 48 | # Issue #23215: MemoryError with custom error handlers and multibyte codecs |
| 49 | data = 100 * "\udc00" |
| 50 | codecs.register_error("test.ignore", codecs.ignore_errors) |
| 51 | for enc in ALL_CJKENCODINGS: |
| 52 | self.assertEqual(data.encode(enc, "test.ignore"), b'') |
| 53 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 54 | def test_codingspec(self): |
| 55 | try: |
| 56 | for enc in ALL_CJKENCODINGS: |
Antoine Pitrou | 92f60ed | 2010-10-14 22:11:44 +0000 | [diff] [blame] | 57 | code = '# coding: {}\n'.format(enc) |
| 58 | exec(code) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 59 | finally: |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 60 | support.unlink(TESTFN) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 61 | |
Georg Brandl | 3dbca81 | 2008-07-23 16:10:53 +0000 | [diff] [blame] | 62 | def test_init_segfault(self): |
| 63 | # bug #3305: this used to segfault |
| 64 | self.assertRaises(AttributeError, |
| 65 | _multibytecodec.MultibyteStreamReader, None) |
| 66 | self.assertRaises(AttributeError, |
| 67 | _multibytecodec.MultibyteStreamWriter, None) |
| 68 | |
Antoine Pitrou | 44bf631 | 2009-01-22 11:59:55 +0000 | [diff] [blame] | 69 | def test_decode_unicode(self): |
Martin Panter | 6245cb3 | 2016-04-15 02:14:19 +0000 | [diff] [blame] | 70 | # Trying to decode a unicode string should raise a TypeError |
Antoine Pitrou | 44bf631 | 2009-01-22 11:59:55 +0000 | [diff] [blame] | 71 | for enc in ALL_CJKENCODINGS: |
| 72 | self.assertRaises(TypeError, codecs.getdecoder(enc), "") |
Georg Brandl | 3dbca81 | 2008-07-23 16:10:53 +0000 | [diff] [blame] | 73 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 74 | class Test_IncrementalEncoder(unittest.TestCase): |
| 75 | |
| 76 | def test_stateless(self): |
| 77 | # cp949 encoder isn't stateful at all. |
| 78 | encoder = codecs.getincrementalencoder('cp949')() |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 79 | self.assertEqual(encoder.encode('\ud30c\uc774\uc36c \ub9c8\uc744'), |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 80 | b'\xc6\xc4\xc0\xcc\xbd\xe3 \xb8\xb6\xc0\xbb') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 81 | self.assertEqual(encoder.reset(), None) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 82 | self.assertEqual(encoder.encode('\u2606\u223c\u2606', True), |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 83 | b'\xa1\xd9\xa1\xad\xa1\xd9') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 84 | self.assertEqual(encoder.reset(), None) |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 85 | self.assertEqual(encoder.encode('', True), b'') |
| 86 | self.assertEqual(encoder.encode('', False), b'') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 87 | self.assertEqual(encoder.reset(), None) |
| 88 | |
| 89 | def test_stateful(self): |
Serhiy Storchaka | d3faf43 | 2015-01-18 11:28:37 +0200 | [diff] [blame] | 90 | # jisx0213 encoder is stateful for a few code points. eg) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 91 | # U+00E6 => A9DC |
| 92 | # U+00E6 U+0300 => ABC4 |
| 93 | # U+0300 => ABDC |
| 94 | |
| 95 | encoder = codecs.getincrementalencoder('jisx0213')() |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 96 | self.assertEqual(encoder.encode('\u00e6\u0300'), b'\xab\xc4') |
| 97 | self.assertEqual(encoder.encode('\u00e6'), b'') |
| 98 | self.assertEqual(encoder.encode('\u0300'), b'\xab\xc4') |
| 99 | self.assertEqual(encoder.encode('\u00e6', True), b'\xa9\xdc') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 100 | |
| 101 | self.assertEqual(encoder.reset(), None) |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 102 | self.assertEqual(encoder.encode('\u0300'), b'\xab\xdc') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 103 | |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 104 | self.assertEqual(encoder.encode('\u00e6'), b'') |
| 105 | self.assertEqual(encoder.encode('', True), b'\xa9\xdc') |
| 106 | self.assertEqual(encoder.encode('', True), b'') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 107 | |
| 108 | def test_stateful_keep_buffer(self): |
| 109 | encoder = codecs.getincrementalencoder('jisx0213')() |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 110 | self.assertEqual(encoder.encode('\u00e6'), b'') |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 111 | self.assertRaises(UnicodeEncodeError, encoder.encode, '\u0123') |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 112 | self.assertEqual(encoder.encode('\u0300\u00e6'), b'\xab\xc4') |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 113 | self.assertRaises(UnicodeEncodeError, encoder.encode, '\u0123') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 114 | self.assertEqual(encoder.reset(), None) |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 115 | self.assertEqual(encoder.encode('\u0300'), b'\xab\xdc') |
| 116 | self.assertEqual(encoder.encode('\u00e6'), b'') |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 117 | self.assertRaises(UnicodeEncodeError, encoder.encode, '\u0123') |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 118 | self.assertEqual(encoder.encode('', True), b'\xa9\xdc') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 119 | |
Christopher Thorne | ac22f6a | 2018-11-01 10:48:49 +0000 | [diff] [blame] | 120 | def test_state_methods_with_buffer_state(self): |
| 121 | # euc_jis_2004 stores state as a buffer of pending bytes |
| 122 | encoder = codecs.getincrementalencoder('euc_jis_2004')() |
| 123 | |
| 124 | initial_state = encoder.getstate() |
| 125 | self.assertEqual(encoder.encode('\u00e6\u0300'), b'\xab\xc4') |
| 126 | encoder.setstate(initial_state) |
| 127 | self.assertEqual(encoder.encode('\u00e6\u0300'), b'\xab\xc4') |
| 128 | |
| 129 | self.assertEqual(encoder.encode('\u00e6'), b'') |
| 130 | partial_state = encoder.getstate() |
| 131 | self.assertEqual(encoder.encode('\u0300'), b'\xab\xc4') |
| 132 | encoder.setstate(partial_state) |
| 133 | self.assertEqual(encoder.encode('\u0300'), b'\xab\xc4') |
| 134 | |
| 135 | def test_state_methods_with_non_buffer_state(self): |
| 136 | # iso2022_jp stores state without using a buffer |
| 137 | encoder = codecs.getincrementalencoder('iso2022_jp')() |
| 138 | |
| 139 | self.assertEqual(encoder.encode('z'), b'z') |
| 140 | en_state = encoder.getstate() |
| 141 | |
| 142 | self.assertEqual(encoder.encode('\u3042'), b'\x1b\x24\x42\x24\x22') |
| 143 | jp_state = encoder.getstate() |
| 144 | self.assertEqual(encoder.encode('z'), b'\x1b\x28\x42z') |
| 145 | |
| 146 | encoder.setstate(jp_state) |
| 147 | self.assertEqual(encoder.encode('\u3042'), b'\x24\x22') |
| 148 | |
| 149 | encoder.setstate(en_state) |
| 150 | self.assertEqual(encoder.encode('z'), b'z') |
| 151 | |
| 152 | def test_getstate_returns_expected_value(self): |
| 153 | # Note: getstate is implemented such that these state values |
| 154 | # are expected to be the same across all builds of Python, |
| 155 | # regardless of x32/64 bit, endianness and compiler. |
| 156 | |
| 157 | # euc_jis_2004 stores state as a buffer of pending bytes |
| 158 | buffer_state_encoder = codecs.getincrementalencoder('euc_jis_2004')() |
| 159 | self.assertEqual(buffer_state_encoder.getstate(), 0) |
| 160 | buffer_state_encoder.encode('\u00e6') |
| 161 | self.assertEqual(buffer_state_encoder.getstate(), |
| 162 | int.from_bytes( |
| 163 | b"\x02" |
| 164 | b"\xc3\xa6" |
| 165 | b"\x00\x00\x00\x00\x00\x00\x00\x00", |
| 166 | 'little')) |
| 167 | buffer_state_encoder.encode('\u0300') |
| 168 | self.assertEqual(buffer_state_encoder.getstate(), 0) |
| 169 | |
| 170 | # iso2022_jp stores state without using a buffer |
| 171 | non_buffer_state_encoder = codecs.getincrementalencoder('iso2022_jp')() |
| 172 | self.assertEqual(non_buffer_state_encoder.getstate(), |
| 173 | int.from_bytes( |
| 174 | b"\x00" |
| 175 | b"\x42\x42\x00\x00\x00\x00\x00\x00", |
| 176 | 'little')) |
| 177 | non_buffer_state_encoder.encode('\u3042') |
| 178 | self.assertEqual(non_buffer_state_encoder.getstate(), |
| 179 | int.from_bytes( |
| 180 | b"\x00" |
| 181 | b"\xc2\x42\x00\x00\x00\x00\x00\x00", |
| 182 | 'little')) |
| 183 | |
| 184 | def test_setstate_validates_input_size(self): |
| 185 | encoder = codecs.getincrementalencoder('euc_jp')() |
| 186 | pending_size_nine = int.from_bytes( |
| 187 | b"\x09" |
| 188 | b"\x00\x00\x00\x00\x00\x00\x00\x00" |
| 189 | b"\x00\x00\x00\x00\x00\x00\x00\x00", |
| 190 | 'little') |
| 191 | self.assertRaises(UnicodeError, encoder.setstate, pending_size_nine) |
| 192 | |
| 193 | def test_setstate_validates_input_bytes(self): |
| 194 | encoder = codecs.getincrementalencoder('euc_jp')() |
| 195 | invalid_utf8 = int.from_bytes( |
| 196 | b"\x01" |
| 197 | b"\xff" |
| 198 | b"\x00\x00\x00\x00\x00\x00\x00\x00", |
| 199 | 'little') |
| 200 | self.assertRaises(UnicodeDecodeError, encoder.setstate, invalid_utf8) |
| 201 | |
Hye-Shik Chang | 944820b | 2009-04-02 10:33:16 +0000 | [diff] [blame] | 202 | def test_issue5640(self): |
| 203 | encoder = codecs.getincrementalencoder('shift-jis')('backslashreplace') |
| 204 | self.assertEqual(encoder.encode('\xff'), b'\\xff') |
| 205 | self.assertEqual(encoder.encode('\n'), b'\n') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 206 | |
| 207 | class Test_IncrementalDecoder(unittest.TestCase): |
| 208 | |
| 209 | def test_dbcs(self): |
| 210 | # cp949 decoder is simple with only 1 or 2 bytes sequences. |
| 211 | decoder = codecs.getincrementaldecoder('cp949')() |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 212 | self.assertEqual(decoder.decode(b'\xc6\xc4\xc0\xcc\xbd'), |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 213 | '\ud30c\uc774') |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 214 | self.assertEqual(decoder.decode(b'\xe3 \xb8\xb6\xc0\xbb'), |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 215 | '\uc36c \ub9c8\uc744') |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 216 | self.assertEqual(decoder.decode(b''), '') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 217 | |
| 218 | def test_dbcs_keep_buffer(self): |
| 219 | decoder = codecs.getincrementaldecoder('cp949')() |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 220 | self.assertEqual(decoder.decode(b'\xc6\xc4\xc0'), '\ud30c') |
Guido van Rossum | e22905a | 2007-08-27 23:09:25 +0000 | [diff] [blame] | 221 | self.assertRaises(UnicodeDecodeError, decoder.decode, b'', True) |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 222 | self.assertEqual(decoder.decode(b'\xcc'), '\uc774') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 223 | |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 224 | self.assertEqual(decoder.decode(b'\xc6\xc4\xc0'), '\ud30c') |
Guido van Rossum | e22905a | 2007-08-27 23:09:25 +0000 | [diff] [blame] | 225 | self.assertRaises(UnicodeDecodeError, decoder.decode, |
| 226 | b'\xcc\xbd', True) |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 227 | self.assertEqual(decoder.decode(b'\xcc'), '\uc774') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 228 | |
| 229 | def test_iso2022(self): |
| 230 | decoder = codecs.getincrementaldecoder('iso2022-jp')() |
Guido van Rossum | e22905a | 2007-08-27 23:09:25 +0000 | [diff] [blame] | 231 | ESC = b'\x1b' |
| 232 | self.assertEqual(decoder.decode(ESC + b'('), '') |
| 233 | self.assertEqual(decoder.decode(b'B', True), '') |
| 234 | self.assertEqual(decoder.decode(ESC + b'$'), '') |
| 235 | self.assertEqual(decoder.decode(b'B@$'), '\u4e16') |
| 236 | self.assertEqual(decoder.decode(b'@$@'), '\u4e16') |
| 237 | self.assertEqual(decoder.decode(b'$', True), '\u4e16') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 238 | self.assertEqual(decoder.reset(), None) |
Guido van Rossum | e22905a | 2007-08-27 23:09:25 +0000 | [diff] [blame] | 239 | self.assertEqual(decoder.decode(b'@$'), '@$') |
| 240 | self.assertEqual(decoder.decode(ESC + b'$'), '') |
| 241 | self.assertRaises(UnicodeDecodeError, decoder.decode, b'', True) |
| 242 | self.assertEqual(decoder.decode(b'B@$'), '\u4e16') |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 243 | |
Antoine Pitrou | 44bf631 | 2009-01-22 11:59:55 +0000 | [diff] [blame] | 244 | def test_decode_unicode(self): |
Martin Panter | 6245cb3 | 2016-04-15 02:14:19 +0000 | [diff] [blame] | 245 | # Trying to decode a unicode string should raise a TypeError |
Antoine Pitrou | 44bf631 | 2009-01-22 11:59:55 +0000 | [diff] [blame] | 246 | for enc in ALL_CJKENCODINGS: |
| 247 | decoder = codecs.getincrementaldecoder(enc)() |
| 248 | self.assertRaises(TypeError, decoder.decode, "") |
| 249 | |
Christopher Thorne | ac22f6a | 2018-11-01 10:48:49 +0000 | [diff] [blame] | 250 | def test_state_methods(self): |
| 251 | decoder = codecs.getincrementaldecoder('euc_jp')() |
| 252 | |
| 253 | # Decode a complete input sequence |
| 254 | self.assertEqual(decoder.decode(b'\xa4\xa6'), '\u3046') |
| 255 | pending1, _ = decoder.getstate() |
| 256 | self.assertEqual(pending1, b'') |
| 257 | |
| 258 | # Decode first half of a partial input sequence |
| 259 | self.assertEqual(decoder.decode(b'\xa4'), '') |
| 260 | pending2, flags2 = decoder.getstate() |
| 261 | self.assertEqual(pending2, b'\xa4') |
| 262 | |
| 263 | # Decode second half of a partial input sequence |
| 264 | self.assertEqual(decoder.decode(b'\xa6'), '\u3046') |
| 265 | pending3, _ = decoder.getstate() |
| 266 | self.assertEqual(pending3, b'') |
| 267 | |
| 268 | # Jump back and decode second half of partial input sequence again |
| 269 | decoder.setstate((pending2, flags2)) |
| 270 | self.assertEqual(decoder.decode(b'\xa6'), '\u3046') |
| 271 | pending4, _ = decoder.getstate() |
| 272 | self.assertEqual(pending4, b'') |
| 273 | |
Christopher Thorne | 488c0a6 | 2018-11-02 03:29:40 +0000 | [diff] [blame] | 274 | # Ensure state values are preserved correctly |
| 275 | decoder.setstate((b'abc', 123456789)) |
| 276 | self.assertEqual(decoder.getstate(), (b'abc', 123456789)) |
| 277 | |
Christopher Thorne | ac22f6a | 2018-11-01 10:48:49 +0000 | [diff] [blame] | 278 | def test_setstate_validates_input(self): |
| 279 | decoder = codecs.getincrementaldecoder('euc_jp')() |
| 280 | self.assertRaises(TypeError, decoder.setstate, 123) |
| 281 | self.assertRaises(TypeError, decoder.setstate, ("invalid", 0)) |
| 282 | self.assertRaises(TypeError, decoder.setstate, (b"1234", "invalid")) |
| 283 | self.assertRaises(UnicodeError, decoder.setstate, (b"123456789", 0)) |
| 284 | |
Guido van Rossum | cd16bf6 | 2007-06-13 18:07:49 +0000 | [diff] [blame] | 285 | class Test_StreamReader(unittest.TestCase): |
| 286 | def test_bug1728403(self): |
| 287 | try: |
Guido van Rossum | a222e4c | 2007-07-09 09:18:12 +0000 | [diff] [blame] | 288 | f = open(TESTFN, 'wb') |
| 289 | try: |
| 290 | f.write(b'\xa1') |
| 291 | finally: |
| 292 | f.close() |
Guido van Rossum | cd16bf6 | 2007-06-13 18:07:49 +0000 | [diff] [blame] | 293 | f = codecs.open(TESTFN, encoding='cp949') |
Guido van Rossum | a222e4c | 2007-07-09 09:18:12 +0000 | [diff] [blame] | 294 | try: |
| 295 | self.assertRaises(UnicodeDecodeError, f.read, 2) |
| 296 | finally: |
| 297 | f.close() |
Guido van Rossum | cd16bf6 | 2007-06-13 18:07:49 +0000 | [diff] [blame] | 298 | finally: |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 299 | support.unlink(TESTFN) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 300 | |
| 301 | class Test_StreamWriter(unittest.TestCase): |
Victor Stinner | eae41af | 2013-04-11 22:41:11 +0200 | [diff] [blame] | 302 | def test_gb18030(self): |
| 303 | s= io.BytesIO() |
| 304 | c = codecs.getwriter('gb18030')(s) |
| 305 | c.write('123') |
| 306 | self.assertEqual(s.getvalue(), b'123') |
| 307 | c.write('\U00012345') |
| 308 | self.assertEqual(s.getvalue(), b'123\x907\x959') |
| 309 | c.write('\uac00\u00ac') |
| 310 | self.assertEqual(s.getvalue(), |
| 311 | b'123\x907\x959\x827\xcf5\x810\x851') |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 312 | |
Victor Stinner | eae41af | 2013-04-11 22:41:11 +0200 | [diff] [blame] | 313 | def test_utf_8(self): |
| 314 | s= io.BytesIO() |
| 315 | c = codecs.getwriter('utf-8')(s) |
| 316 | c.write('123') |
| 317 | self.assertEqual(s.getvalue(), b'123') |
| 318 | c.write('\U00012345') |
| 319 | self.assertEqual(s.getvalue(), b'123\xf0\x92\x8d\x85') |
| 320 | c.write('\uac00\u00ac') |
| 321 | self.assertEqual(s.getvalue(), |
| 322 | b'123\xf0\x92\x8d\x85' |
| 323 | b'\xea\xb0\x80\xc2\xac') |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 324 | |
Hye-Shik Chang | 2bb146f | 2004-07-18 03:06:29 +0000 | [diff] [blame] | 325 | def test_streamwriter_strwrite(self): |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 326 | s = io.BytesIO() |
Hye-Shik Chang | 2bb146f | 2004-07-18 03:06:29 +0000 | [diff] [blame] | 327 | wr = codecs.getwriter('gb18030')(s) |
| 328 | wr.write('abcd') |
Guido van Rossum | 024da5c | 2007-05-17 23:59:11 +0000 | [diff] [blame] | 329 | self.assertEqual(s.getvalue(), b'abcd') |
Hye-Shik Chang | 2bb146f | 2004-07-18 03:06:29 +0000 | [diff] [blame] | 330 | |
Hye-Shik Chang | abb903f | 2006-03-13 10:20:08 +0000 | [diff] [blame] | 331 | class Test_ISO2022(unittest.TestCase): |
| 332 | def test_g2(self): |
Guido van Rossum | e22905a | 2007-08-27 23:09:25 +0000 | [diff] [blame] | 333 | iso2022jp2 = b'\x1b(B:hu4:unit\x1b.A\x1bNi de famille' |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 334 | uni = ':hu4:unit\xe9 de famille' |
Hye-Shik Chang | abb903f | 2006-03-13 10:20:08 +0000 | [diff] [blame] | 335 | self.assertEqual(iso2022jp2.decode('iso2022-jp-2'), uni) |
| 336 | |
Thomas Wouters | 89f507f | 2006-12-13 04:49:30 +0000 | [diff] [blame] | 337 | def test_iso2022_jp_g0(self): |
Ezio Melotti | b58e0bd | 2010-01-23 15:40:09 +0000 | [diff] [blame] | 338 | self.assertNotIn(b'\x0e', '\N{SOFT HYPHEN}'.encode('iso-2022-jp-2')) |
Thomas Wouters | 89f507f | 2006-12-13 04:49:30 +0000 | [diff] [blame] | 339 | for encoding in ('iso-2022-jp-2004', 'iso-2022-jp-3'): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 340 | e = '\u3406'.encode(encoding) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 341 | self.assertFalse(any(x > 0x80 for x in e)) |
Thomas Wouters | 89f507f | 2006-12-13 04:49:30 +0000 | [diff] [blame] | 342 | |
| 343 | def test_bug1572832(self): |
Guido van Rossum | 805365e | 2007-05-07 22:24:25 +0000 | [diff] [blame] | 344 | for x in range(0x10000, 0x110000): |
Thomas Wouters | 89f507f | 2006-12-13 04:49:30 +0000 | [diff] [blame] | 345 | # Any ISO 2022 codec will cause the segfault |
Ezio Melotti | a9860ae | 2011-10-04 19:06:00 +0300 | [diff] [blame] | 346 | chr(x).encode('iso_2022_jp', 'ignore') |
Thomas Wouters | 89f507f | 2006-12-13 04:49:30 +0000 | [diff] [blame] | 347 | |
Victor Stinner | 6bcbef7 | 2011-05-24 22:17:55 +0200 | [diff] [blame] | 348 | class TestStateful(unittest.TestCase): |
| 349 | text = '\u4E16\u4E16' |
| 350 | encoding = 'iso-2022-jp' |
| 351 | expected = b'\x1b$B@$@$' |
Victor Stinner | e15dce3 | 2011-05-30 22:56:00 +0200 | [diff] [blame] | 352 | reset = b'\x1b(B' |
| 353 | expected_reset = expected + reset |
Victor Stinner | 6bcbef7 | 2011-05-24 22:17:55 +0200 | [diff] [blame] | 354 | |
| 355 | def test_encode(self): |
| 356 | self.assertEqual(self.text.encode(self.encoding), self.expected_reset) |
| 357 | |
| 358 | def test_incrementalencoder(self): |
| 359 | encoder = codecs.getincrementalencoder(self.encoding)() |
| 360 | output = b''.join( |
| 361 | encoder.encode(char) |
| 362 | for char in self.text) |
| 363 | self.assertEqual(output, self.expected) |
Victor Stinner | e15dce3 | 2011-05-30 22:56:00 +0200 | [diff] [blame] | 364 | self.assertEqual(encoder.encode('', final=True), self.reset) |
| 365 | self.assertEqual(encoder.encode('', final=True), b'') |
Victor Stinner | 6bcbef7 | 2011-05-24 22:17:55 +0200 | [diff] [blame] | 366 | |
| 367 | def test_incrementalencoder_final(self): |
| 368 | encoder = codecs.getincrementalencoder(self.encoding)() |
| 369 | last_index = len(self.text) - 1 |
| 370 | output = b''.join( |
| 371 | encoder.encode(char, index == last_index) |
| 372 | for index, char in enumerate(self.text)) |
| 373 | self.assertEqual(output, self.expected_reset) |
Victor Stinner | e15dce3 | 2011-05-30 22:56:00 +0200 | [diff] [blame] | 374 | self.assertEqual(encoder.encode('', final=True), b'') |
Victor Stinner | 6bcbef7 | 2011-05-24 22:17:55 +0200 | [diff] [blame] | 375 | |
| 376 | class TestHZStateful(TestStateful): |
| 377 | text = '\u804a\u804a' |
| 378 | encoding = 'hz' |
| 379 | expected = b'~{ADAD' |
Victor Stinner | e15dce3 | 2011-05-30 22:56:00 +0200 | [diff] [blame] | 380 | reset = b'~}' |
| 381 | expected_reset = expected + reset |
Victor Stinner | 6bcbef7 | 2011-05-24 22:17:55 +0200 | [diff] [blame] | 382 | |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 383 | def test_main(): |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 384 | support.run_unittest(__name__) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 385 | |
| 386 | if __name__ == "__main__": |
| 387 | test_main() |