1 #!/usr/bin/env python 2 # 3 # test_multibytecodec_support.py 4 # Common Unittest Routines for CJK codecs 5 # 6 7 import codecs 8 import os 9 import re 10 import sys 11 import unittest 12 from httplib import HTTPException 13 from test import test_support 14 from StringIO import StringIO 15 16 class TestBase: 17 encoding = '' # codec name 18 codec = None # codec tuple (with 4 elements) 19 tstring = '' # string to test StreamReader 20 21 codectests = None # must set. codec test tuple 22 roundtriptest = 1 # set if roundtrip is possible with unicode 23 has_iso10646 = 0 # set if this encoding contains whole iso10646 map 24 xmlcharnametest = None # string to test xmlcharrefreplace 25 unmappedunicode = u'\udeee' # a unicode codepoint that is not mapped. 26 27 def setUp(self): 28 if self.codec is None: 29 self.codec = codecs.lookup(self.encoding) 30 self.encode = self.codec.encode 31 self.decode = self.codec.decode 32 self.reader = self.codec.streamreader 33 self.writer = self.codec.streamwriter 34 self.incrementalencoder = self.codec.incrementalencoder 35 self.incrementaldecoder = self.codec.incrementaldecoder 36 37 def test_chunkcoding(self): 38 for native, utf8 in zip(*[StringIO(f).readlines() 39 for f in self.tstring]): 40 u = self.decode(native)[0] 41 self.assertEqual(u, utf8.decode('utf-8')) 42 if self.roundtriptest: 43 self.assertEqual(native, self.encode(u)[0]) 44 45 def test_errorhandle(self): 46 for source, scheme, expected in self.codectests: 47 if type(source) == type(''): 48 func = self.decode 49 else: 50 func = self.encode 51 if expected: 52 result = func(source, scheme)[0] 53 self.assertEqual(result, expected) 54 else: 55 self.assertRaises(UnicodeError, func, source, scheme) 56 57 def test_xmlcharrefreplace(self): 58 if self.has_iso10646: 59 return 60 61 s = u"\u0b13\u0b23\u0b60 nd eggs" 62 self.assertEqual( 63 self.encode(s, "xmlcharrefreplace")[0], 64 "ଓଣୠ nd eggs" 65 ) 66 67 def test_customreplace_encode(self): 68 if self.has_iso10646: 69 return 70 71 from htmlentitydefs import codepoint2name 72 73 def xmlcharnamereplace(exc): 74 if not isinstance(exc, UnicodeEncodeError): 75 raise TypeError("don't know how to handle %r" % exc) 76 l = [] 77 for c in exc.object[exc.start:exc.end]: 78 if ord(c) in codepoint2name: 79 l.append(u"&%s;" % codepoint2name[ord(c)]) 80 else: 81 l.append(u"&#%d;" % ord(c)) 82 return (u"".join(l), exc.end) 83 84 codecs.register_error("test.xmlcharnamereplace", xmlcharnamereplace) 85 86 if self.xmlcharnametest: 87 sin, sout = self.xmlcharnametest 88 else: 89 sin = u"\xab\u211c\xbb = \u2329\u1234\u232a" 90 sout = "«ℜ» = ⟨ሴ⟩" 91 self.assertEqual(self.encode(sin, 92 "test.xmlcharnamereplace")[0], sout) 93 94 def test_callback_wrong_objects(self): 95 def myreplace(exc): 96 return (ret, exc.end) 97 codecs.register_error("test.cjktest", myreplace) 98 99 for ret in ([1, 2, 3], [], None, object(), 'string', ''): 100 self.assertRaises(TypeError, self.encode, self.unmappedunicode, 101 'test.cjktest') 102 103 def test_callback_long_index(self): 104 def myreplace(exc): 105 return (u'x', long(exc.end)) 106 codecs.register_error("test.cjktest", myreplace) 107 self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh', 108 'test.cjktest'), ('abcdxefgh', 9)) 109 110 def myreplace(exc): 111 return (u'x', sys.maxint + 1) 112 codecs.register_error("test.cjktest", myreplace) 113 self.assertRaises(IndexError, self.encode, self.unmappedunicode, 114 'test.cjktest') 115 116 def test_callback_None_index(self): 117 def myreplace(exc): 118 return (u'x', None) 119 codecs.register_error("test.cjktest", myreplace) 120 self.assertRaises(TypeError, self.encode, self.unmappedunicode, 121 'test.cjktest') 122 123 def test_callback_backward_index(self): 124 def myreplace(exc): 125 if myreplace.limit > 0: 126 myreplace.limit -= 1 127 return (u'REPLACED', 0) 128 else: 129 return (u'TERMINAL', exc.end) 130 myreplace.limit = 3 131 codecs.register_error("test.cjktest", myreplace) 132 self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh', 133 'test.cjktest'), 134 ('abcdREPLACEDabcdREPLACEDabcdREPLACEDabcdTERMINALefgh', 9)) 135 136 def test_callback_forward_index(self): 137 def myreplace(exc): 138 return (u'REPLACED', exc.end + 2) 139 codecs.register_error("test.cjktest", myreplace) 140 self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh', 141 'test.cjktest'), ('abcdREPLACEDgh', 9)) 142 143 def test_callback_index_outofbound(self): 144 def myreplace(exc): 145 return (u'TERM', 100) 146 codecs.register_error("test.cjktest", myreplace) 147 self.assertRaises(IndexError, self.encode, self.unmappedunicode, 148 'test.cjktest') 149 150 def test_incrementalencoder(self): 151 UTF8Reader = codecs.getreader('utf-8') 152 for sizehint in [None] + range(1, 33) + \ 153 [64, 128, 256, 512, 1024]: 154 istream = UTF8Reader(StringIO(self.tstring[1])) 155 ostream = StringIO() 156 encoder = self.incrementalencoder() 157 while 1: 158 if sizehint is not None: 159 data = istream.read(sizehint) 160 else: 161 data = istream.read() 162 163 if not data: 164 break 165 e = encoder.encode(data) 166 ostream.write(e) 167 168 self.assertEqual(ostream.getvalue(), self.tstring[0]) 169 170 def test_incrementaldecoder(self): 171 UTF8Writer = codecs.getwriter('utf-8') 172 for sizehint in [None, -1] + range(1, 33) + \ 173 [64, 128, 256, 512, 1024]: 174 istream = StringIO(self.tstring[0]) 175 ostream = UTF8Writer(StringIO()) 176 decoder = self.incrementaldecoder() 177 while 1: 178 data = istream.read(sizehint) 179 if not data: 180 break 181 else: 182 u = decoder.decode(data) 183 ostream.write(u) 184 185 self.assertEqual(ostream.getvalue(), self.tstring[1]) 186 187 def test_incrementalencoder_error_callback(self): 188 inv = self.unmappedunicode 189 190 e = self.incrementalencoder() 191 self.assertRaises(UnicodeEncodeError, e.encode, inv, True) 192 193 e.errors = 'ignore' 194 self.assertEqual(e.encode(inv, True), '') 195 196 e.reset() 197 def tempreplace(exc): 198 return (u'called', exc.end) 199 codecs.register_error('test.incremental_error_callback', tempreplace) 200 e.errors = 'test.incremental_error_callback' 201 self.assertEqual(e.encode(inv, True), 'called') 202 203 # again 204 e.errors = 'ignore' 205 self.assertEqual(e.encode(inv, True), '') 206 207 def test_streamreader(self): 208 UTF8Writer = codecs.getwriter('utf-8') 209 for name in ["read", "readline", "readlines"]: 210 for sizehint in [None, -1] + range(1, 33) + \ 211 [64, 128, 256, 512, 1024]: 212 istream = self.reader(StringIO(self.tstring[0])) 213 ostream = UTF8Writer(StringIO()) 214 func = getattr(istream, name) 215 while 1: 216 data = func(sizehint) 217 if not data: 218 break 219 if name == "readlines": 220 ostream.writelines(data) 221 else: 222 ostream.write(data) 223 224 self.assertEqual(ostream.getvalue(), self.tstring[1]) 225 226 def test_streamwriter(self): 227 readfuncs = ('read', 'readline', 'readlines') 228 UTF8Reader = codecs.getreader('utf-8') 229 for name in readfuncs: 230 for sizehint in [None] + range(1, 33) + \ 231 [64, 128, 256, 512, 1024]: 232 istream = UTF8Reader(StringIO(self.tstring[1])) 233 ostream = self.writer(StringIO()) 234 func = getattr(istream, name) 235 while 1: 236 if sizehint is not None: 237 data = func(sizehint) 238 else: 239 data = func() 240 241 if not data: 242 break 243 if name == "readlines": 244 ostream.writelines(data) 245 else: 246 ostream.write(data) 247 248 self.assertEqual(ostream.getvalue(), self.tstring[0]) 249 250 class TestBase_Mapping(unittest.TestCase): 251 pass_enctest = [] 252 pass_dectest = [] 253 supmaps = [] 254 255 def __init__(self, *args, **kw): 256 unittest.TestCase.__init__(self, *args, **kw) 257 try: 258 self.open_mapping_file().close() # test it to report the error early 259 except (IOError, HTTPException): 260 self.skipTest("Could not retrieve "+self.mapfileurl) 261 262 def open_mapping_file(self): 263 return test_support.open_urlresource(self.mapfileurl) 264 265 def test_mapping_file(self): 266 if self.mapfileurl.endswith('.xml'): 267 self._test_mapping_file_ucm() 268 else: 269 self._test_mapping_file_plain() 270 271 def _test_mapping_file_plain(self): 272 _unichr = lambda c: eval("u'\\U%08x'" % int(c, 16)) 273 unichrs = lambda s: u''.join(_unichr(c) for c in s.split('+')) 274 urt_wa = {} 275 276 with self.open_mapping_file() as f: 277 for line in f: 278 if not line: 279 break 280 data = line.split('#')[0].strip().split() 281 if len(data) != 2: 282 continue 283 284 csetval = eval(data[0]) 285 if csetval <= 0x7F: 286 csetch = chr(csetval & 0xff) 287 elif csetval >= 0x1000000: 288 csetch = chr(csetval >> 24) + chr((csetval >> 16) & 0xff) + \ 289 chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) 290 elif csetval >= 0x10000: 291 csetch = chr(csetval >> 16) + \ 292 chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) 293 elif csetval >= 0x100: 294 csetch = chr(csetval >> 8) + chr(csetval & 0xff) 295 else: 296 continue 297 298 unich = unichrs(data[1]) 299 if unich == u'\ufffd' or unich in urt_wa: 300 continue 301 urt_wa[unich] = csetch 302 303 self._testpoint(csetch, unich) 304 305 def _test_mapping_file_ucm(self): 306 with self.open_mapping_file() as f: 307 ucmdata = f.read() 308 uc = re.findall('<a u="([A-F0-9]{4})" b="([0-9A-F ]+)"/>', ucmdata) 309 for uni, coded in uc: 310 unich = unichr(int(uni, 16)) 311 codech = ''.join(chr(int(c, 16)) for c in coded.split()) 312 self._testpoint(codech, unich) 313 314 def test_mapping_supplemental(self): 315 for mapping in self.supmaps: 316 self._testpoint(*mapping) 317 318 def _testpoint(self, csetch, unich): 319 if (csetch, unich) not in self.pass_enctest: 320 try: 321 self.assertEqual(unich.encode(self.encoding), csetch) 322 except UnicodeError, exc: 323 self.fail('Encoding failed while testing %s -> %s: %s' % ( 324 repr(unich), repr(csetch), exc.reason)) 325 if (csetch, unich) not in self.pass_dectest: 326 try: 327 self.assertEqual(csetch.decode(self.encoding), unich) 328 except UnicodeError, exc: 329 self.fail('Decoding failed while testing %s -> %s: %s' % ( 330 repr(csetch), repr(unich), exc.reason)) 331 332 def load_teststring(name): 333 dir = os.path.join(os.path.dirname(__file__), 'cjkencodings') 334 with open(os.path.join(dir, name + '.txt'), 'rb') as f: 335 encoded = f.read() 336 with open(os.path.join(dir, name + '-utf8.txt'), 'rb') as f: 337 utf8 = f.read() 338 return encoded, utf8 339
