Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 1 | #!/usr/bin/env python |
| 2 | # |
| 3 | # test_multibytecodec_support.py |
| 4 | # Common Unittest Routines for CJK codecs |
| 5 | # |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 6 | |
| 7 | import sys, codecs, os.path |
| 8 | import unittest |
| 9 | from test import test_support |
| 10 | from StringIO import StringIO |
| 11 | |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 12 | class TestBase: |
| 13 | encoding = '' # codec name |
| 14 | codec = None # codec tuple (with 4 elements) |
| 15 | tstring = '' # string to test StreamReader |
| 16 | |
| 17 | codectests = None # must set. codec test tuple |
| 18 | roundtriptest = 1 # set if roundtrip is possible with unicode |
| 19 | has_iso10646 = 0 # set if this encoding contains whole iso10646 map |
| 20 | xmlcharnametest = None # string to test xmlcharrefreplace |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 21 | unmappedunicode = '\udeee' # a unicode codepoint that is not mapped. |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 22 | |
| 23 | def setUp(self): |
| 24 | if self.codec is None: |
| 25 | self.codec = codecs.lookup(self.encoding) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 26 | self.encode = self.codec.encode |
| 27 | self.decode = self.codec.decode |
| 28 | self.reader = self.codec.streamreader |
| 29 | self.writer = self.codec.streamwriter |
| 30 | self.incrementalencoder = self.codec.incrementalencoder |
| 31 | self.incrementaldecoder = self.codec.incrementaldecoder |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 32 | |
| 33 | def test_chunkcoding(self): |
| 34 | for native, utf8 in zip(*[StringIO(f).readlines() |
| 35 | for f in self.tstring]): |
| 36 | u = self.decode(native)[0] |
| 37 | self.assertEqual(u, utf8.decode('utf-8')) |
| 38 | if self.roundtriptest: |
| 39 | self.assertEqual(native, self.encode(u)[0]) |
| 40 | |
| 41 | def test_errorhandle(self): |
| 42 | for source, scheme, expected in self.codectests: |
| 43 | if type(source) == type(''): |
| 44 | func = self.decode |
| 45 | else: |
| 46 | func = self.encode |
| 47 | if expected: |
| 48 | result = func(source, scheme)[0] |
| 49 | self.assertEqual(result, expected) |
| 50 | else: |
| 51 | self.assertRaises(UnicodeError, func, source, scheme) |
| 52 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 53 | def test_xmlcharrefreplace(self): |
| 54 | if self.has_iso10646: |
| 55 | return |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 56 | |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 57 | s = "\u0b13\u0b23\u0b60 nd eggs" |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 58 | self.assertEqual( |
| 59 | self.encode(s, "xmlcharrefreplace")[0], |
| 60 | "ଓଣୠ nd eggs" |
| 61 | ) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 62 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 63 | def test_customreplace_encode(self): |
| 64 | if self.has_iso10646: |
| 65 | return |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 66 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 67 | from htmlentitydefs import codepoint2name |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 68 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 69 | def xmlcharnamereplace(exc): |
| 70 | if not isinstance(exc, UnicodeEncodeError): |
| 71 | raise TypeError("don't know how to handle %r" % exc) |
| 72 | l = [] |
| 73 | for c in exc.object[exc.start:exc.end]: |
| 74 | if ord(c) in codepoint2name: |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 75 | l.append("&%s;" % codepoint2name[ord(c)]) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 76 | else: |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 77 | l.append("&#%d;" % ord(c)) |
| 78 | return ("".join(l), exc.end) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 79 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 80 | codecs.register_error("test.xmlcharnamereplace", xmlcharnamereplace) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 81 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 82 | if self.xmlcharnametest: |
| 83 | sin, sout = self.xmlcharnametest |
| 84 | else: |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 85 | sin = "\xab\u211c\xbb = \u2329\u1234\u232a" |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 86 | sout = "«ℜ» = ⟨ሴ⟩" |
| 87 | self.assertEqual(self.encode(sin, |
| 88 | "test.xmlcharnamereplace")[0], sout) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 89 | |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 90 | def test_callback_wrong_objects(self): |
| 91 | def myreplace(exc): |
| 92 | return (ret, exc.end) |
| 93 | codecs.register_error("test.cjktest", myreplace) |
| 94 | |
| 95 | for ret in ([1, 2, 3], [], None, object(), 'string', ''): |
| 96 | self.assertRaises(TypeError, self.encode, self.unmappedunicode, |
| 97 | 'test.cjktest') |
| 98 | |
| 99 | def test_callback_long_index(self): |
| 100 | def myreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 101 | return ('x', int(exc.end)) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 102 | codecs.register_error("test.cjktest", myreplace) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 103 | self.assertEqual(self.encode('abcd' + self.unmappedunicode + 'efgh', |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 104 | 'test.cjktest'), ('abcdxefgh', 9)) |
| 105 | |
| 106 | def myreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 107 | return ('x', sys.maxint + 1) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 108 | codecs.register_error("test.cjktest", myreplace) |
| 109 | self.assertRaises(IndexError, self.encode, self.unmappedunicode, |
| 110 | 'test.cjktest') |
| 111 | |
| 112 | def test_callback_None_index(self): |
| 113 | def myreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 114 | return ('x', None) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 115 | codecs.register_error("test.cjktest", myreplace) |
| 116 | self.assertRaises(TypeError, self.encode, self.unmappedunicode, |
| 117 | 'test.cjktest') |
| 118 | |
| 119 | def test_callback_backward_index(self): |
| 120 | def myreplace(exc): |
| 121 | if myreplace.limit > 0: |
| 122 | myreplace.limit -= 1 |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 123 | return ('REPLACED', 0) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 124 | else: |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 125 | return ('TERMINAL', exc.end) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 126 | myreplace.limit = 3 |
| 127 | codecs.register_error("test.cjktest", myreplace) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 128 | self.assertEqual(self.encode('abcd' + self.unmappedunicode + 'efgh', |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 129 | 'test.cjktest'), |
| 130 | ('abcdREPLACEDabcdREPLACEDabcdREPLACEDabcdTERMINALefgh', 9)) |
| 131 | |
| 132 | def test_callback_forward_index(self): |
| 133 | def myreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 134 | return ('REPLACED', exc.end + 2) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 135 | codecs.register_error("test.cjktest", myreplace) |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 136 | self.assertEqual(self.encode('abcd' + self.unmappedunicode + 'efgh', |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 137 | 'test.cjktest'), ('abcdREPLACEDgh', 9)) |
| 138 | |
| 139 | def test_callback_index_outofbound(self): |
| 140 | def myreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 141 | return ('TERM', 100) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 142 | codecs.register_error("test.cjktest", myreplace) |
| 143 | self.assertRaises(IndexError, self.encode, self.unmappedunicode, |
| 144 | 'test.cjktest') |
| 145 | |
| 146 | def test_incrementalencoder(self): |
| 147 | UTF8Reader = codecs.getreader('utf-8') |
| 148 | for sizehint in [None] + range(1, 33) + \ |
| 149 | [64, 128, 256, 512, 1024]: |
| 150 | istream = UTF8Reader(StringIO(self.tstring[1])) |
| 151 | ostream = StringIO() |
| 152 | encoder = self.incrementalencoder() |
| 153 | while 1: |
| 154 | if sizehint is not None: |
| 155 | data = istream.read(sizehint) |
| 156 | else: |
| 157 | data = istream.read() |
| 158 | |
| 159 | if not data: |
| 160 | break |
| 161 | e = encoder.encode(data) |
| 162 | ostream.write(e) |
| 163 | |
| 164 | self.assertEqual(ostream.getvalue(), self.tstring[0]) |
| 165 | |
| 166 | def test_incrementaldecoder(self): |
| 167 | UTF8Writer = codecs.getwriter('utf-8') |
| 168 | for sizehint in [None, -1] + range(1, 33) + \ |
| 169 | [64, 128, 256, 512, 1024]: |
| 170 | istream = StringIO(self.tstring[0]) |
| 171 | ostream = UTF8Writer(StringIO()) |
| 172 | decoder = self.incrementaldecoder() |
| 173 | while 1: |
| 174 | data = istream.read(sizehint) |
| 175 | if not data: |
| 176 | break |
| 177 | else: |
| 178 | u = decoder.decode(data) |
| 179 | ostream.write(u) |
| 180 | |
| 181 | self.assertEqual(ostream.getvalue(), self.tstring[1]) |
| 182 | |
| 183 | def test_incrementalencoder_error_callback(self): |
| 184 | inv = self.unmappedunicode |
| 185 | |
| 186 | e = self.incrementalencoder() |
| 187 | self.assertRaises(UnicodeEncodeError, e.encode, inv, True) |
| 188 | |
| 189 | e.errors = 'ignore' |
| 190 | self.assertEqual(e.encode(inv, True), '') |
| 191 | |
| 192 | e.reset() |
| 193 | def tempreplace(exc): |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 194 | return ('called', exc.end) |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 195 | codecs.register_error('test.incremental_error_callback', tempreplace) |
| 196 | e.errors = 'test.incremental_error_callback' |
| 197 | self.assertEqual(e.encode(inv, True), 'called') |
| 198 | |
| 199 | # again |
| 200 | e.errors = 'ignore' |
| 201 | self.assertEqual(e.encode(inv, True), '') |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 202 | |
| 203 | def test_streamreader(self): |
| 204 | UTF8Writer = codecs.getwriter('utf-8') |
| 205 | for name in ["read", "readline", "readlines"]: |
| 206 | for sizehint in [None, -1] + range(1, 33) + \ |
| 207 | [64, 128, 256, 512, 1024]: |
| 208 | istream = self.reader(StringIO(self.tstring[0])) |
| 209 | ostream = UTF8Writer(StringIO()) |
| 210 | func = getattr(istream, name) |
| 211 | while 1: |
| 212 | data = func(sizehint) |
| 213 | if not data: |
| 214 | break |
| 215 | if name == "readlines": |
| 216 | ostream.writelines(data) |
| 217 | else: |
| 218 | ostream.write(data) |
| 219 | |
| 220 | self.assertEqual(ostream.getvalue(), self.tstring[1]) |
| 221 | |
| 222 | def test_streamwriter(self): |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 223 | readfuncs = ('read', 'readline', 'readlines') |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 224 | UTF8Reader = codecs.getreader('utf-8') |
| 225 | for name in readfuncs: |
| 226 | for sizehint in [None] + range(1, 33) + \ |
| 227 | [64, 128, 256, 512, 1024]: |
| 228 | istream = UTF8Reader(StringIO(self.tstring[1])) |
| 229 | ostream = self.writer(StringIO()) |
| 230 | func = getattr(istream, name) |
| 231 | while 1: |
| 232 | if sizehint is not None: |
| 233 | data = func(sizehint) |
| 234 | else: |
| 235 | data = func() |
| 236 | |
| 237 | if not data: |
| 238 | break |
| 239 | if name == "readlines": |
| 240 | ostream.writelines(data) |
| 241 | else: |
| 242 | ostream.write(data) |
| 243 | |
| 244 | self.assertEqual(ostream.getvalue(), self.tstring[0]) |
| 245 | |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 246 | if len('\U00012345') == 2: # ucs2 build |
Guido van Rossum | 84fc66d | 2007-05-03 17:18:26 +0000 | [diff] [blame] | 247 | _unichr = chr |
| 248 | def chr(v): |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 249 | if v >= 0x10000: |
| 250 | return _unichr(0xd800 + ((v - 0x10000) >> 10)) + \ |
| 251 | _unichr(0xdc00 + ((v - 0x10000) & 0x3ff)) |
| 252 | else: |
| 253 | return _unichr(v) |
| 254 | _ord = ord |
| 255 | def ord(c): |
| 256 | if len(c) == 2: |
| 257 | return 0x10000 + ((_ord(c[0]) - 0xd800) << 10) + \ |
| 258 | (ord(c[1]) - 0xdc00) |
| 259 | else: |
| 260 | return _ord(c) |
| 261 | |
| 262 | class TestBase_Mapping(unittest.TestCase): |
| 263 | pass_enctest = [] |
| 264 | pass_dectest = [] |
| 265 | supmaps = [] |
| 266 | |
| 267 | def __init__(self, *args, **kw): |
| 268 | unittest.TestCase.__init__(self, *args, **kw) |
Hye-Shik Chang | aaa2f1d | 2005-12-10 17:44:27 +0000 | [diff] [blame] | 269 | self.open_mapping_file() # test it to report the error early |
| 270 | |
| 271 | def open_mapping_file(self): |
| 272 | return test_support.open_urlresource(self.mapfileurl) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 273 | |
| 274 | def test_mapping_file(self): |
Guido van Rossum | 84fc66d | 2007-05-03 17:18:26 +0000 | [diff] [blame] | 275 | unichrs = lambda s: ''.join(map(chr, map(eval, s.split('+')))) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 276 | urt_wa = {} |
| 277 | |
Hye-Shik Chang | aaa2f1d | 2005-12-10 17:44:27 +0000 | [diff] [blame] | 278 | for line in self.open_mapping_file(): |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 279 | if not line: |
| 280 | break |
| 281 | data = line.split('#')[0].strip().split() |
| 282 | if len(data) != 2: |
| 283 | continue |
| 284 | |
| 285 | csetval = eval(data[0]) |
| 286 | if csetval <= 0x7F: |
| 287 | csetch = chr(csetval & 0xff) |
| 288 | elif csetval >= 0x1000000: |
| 289 | csetch = chr(csetval >> 24) + chr((csetval >> 16) & 0xff) + \ |
| 290 | chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) |
| 291 | elif csetval >= 0x10000: |
| 292 | csetch = chr(csetval >> 16) + \ |
| 293 | chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) |
| 294 | elif csetval >= 0x100: |
| 295 | csetch = chr(csetval >> 8) + chr(csetval & 0xff) |
| 296 | else: |
| 297 | continue |
| 298 | |
| 299 | unich = unichrs(data[1]) |
Guido van Rossum | e2b70bc | 2006-08-18 22:13:04 +0000 | [diff] [blame] | 300 | if ord(unich) == 0xfffd or unich in urt_wa: |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 301 | continue |
| 302 | urt_wa[unich] = csetch |
| 303 | |
| 304 | self._testpoint(csetch, unich) |
| 305 | |
| 306 | def test_mapping_supplemental(self): |
| 307 | for mapping in self.supmaps: |
| 308 | self._testpoint(*mapping) |
| 309 | |
| 310 | def _testpoint(self, csetch, unich): |
| 311 | if (csetch, unich) not in self.pass_enctest: |
| 312 | self.assertEqual(unich.encode(self.encoding), csetch) |
| 313 | if (csetch, unich) not in self.pass_dectest: |
Guido van Rossum | ef87d6e | 2007-05-02 19:09:54 +0000 | [diff] [blame] | 314 | self.assertEqual(str(csetch, self.encoding), unich) |
Hye-Shik Chang | 3e2a306 | 2004-01-17 14:29:29 +0000 | [diff] [blame] | 315 | |
| 316 | def load_teststring(encoding): |
Thomas Wouters | 49fd7fa | 2006-04-21 10:40:58 +0000 | [diff] [blame] | 317 | from test import cjkencodings_test |
| 318 | return cjkencodings_test.teststring[encoding] |