Lib/test/test_multibytecodec_support.py

   1 #!/usr/bin/env python
   2 #
   3 # test_multibytecodec_support.py
   4 #   Common Unittest Routines for CJK codecs
   5 #
   6
   7 import sys, codecs
   8 import unittest, re
   9 from httplib import HTTPException
  10 from test import test_support
  11 from StringIO import StringIO
  12
  13 class TestBase:
  14     encoding        = ''   # codec name
  15     codec           = None # codec tuple (with 4 elements)
  16     tstring         = ''   # string to test StreamReader
  17
  18     codectests      = None # must set. codec test tuple
  19     roundtriptest   = 1    # set if roundtrip is possible with unicode
  20     has_iso10646    = 0    # set if this encoding contains whole iso10646 map
  21     xmlcharnametest = None # string to test xmlcharrefreplace
  22     unmappedunicode = u'\udeee' # a unicode codepoint that is not mapped.
  23
  24     def setUp(self):
  25         if self.codec is None:
  26             self.codec = codecs.lookup(self.encoding)
  27         self.encode = self.codec.encode
  28         self.decode = self.codec.decode
  29         self.reader = self.codec.streamreader
  30         self.writer = self.codec.streamwriter
  31         self.incrementalencoder = self.codec.incrementalencoder
  32         self.incrementaldecoder = self.codec.incrementaldecoder
  33
  34     def test_chunkcoding(self):
  35         for native, utf8 in zip(*[StringIO(f).readlines()
  36                                   for f in self.tstring]):
  37             u = self.decode(native)[0]
  38             self.assertEqual(u, utf8.decode('utf-8'))
  39             if self.roundtriptest:
  40                 self.assertEqual(native, self.encode(u)[0])
  41
  42     def test_errorhandle(self):
  43         for source, scheme, expected in self.codectests:
  44             if type(source) == type(''):
  45                 func = self.decode
  46             else:
  47                 func = self.encode
  48             if expected:
  49                 result = func(source, scheme)[0]
  50                 self.assertEqual(result, expected)
  51             else:
  52                 self.assertRaises(UnicodeError, func, source, scheme)
  53
  54     def test_xmlcharrefreplace(self):
  55         if self.has_iso10646:
  56             return
  57
  58         s = u"\u0b13\u0b23\u0b60 nd eggs"
  59         self.assertEqual(
  60             self.encode(s, "xmlcharrefreplace")[0],
  61             "&#2835;&#2851;&#2912; nd eggs"
  62         )
  63
  64     def test_customreplace_encode(self):
  65         if self.has_iso10646:
  66             return
  67
  68         from htmlentitydefs import codepoint2name
  69
  70         def xmlcharnamereplace(exc):
  71             if not isinstance(exc, UnicodeEncodeError):
  72                 raise TypeError("don't know how to handle %r" % exc)
  73             l = []
  74             for c in exc.object[exc.start:exc.end]:
  75                 if ord(c) in codepoint2name:
  76                     l.append(u"&%s;" % codepoint2name[ord(c)])
  77                 else:
  78                     l.append(u"&#%d;" % ord(c))
  79             return (u"".join(l), exc.end)
  80
  81         codecs.register_error("test.xmlcharnamereplace", xmlcharnamereplace)
  82
  83         if self.xmlcharnametest:
  84             sin, sout = self.xmlcharnametest
  85         else:
  86             sin = u"\xab\u211c\xbb = \u2329\u1234\u232a"
  87             sout = "&laquo;&real;&raquo; = &lang;&#4660;&rang;"
  88         self.assertEqual(self.encode(sin,
  89                                     "test.xmlcharnamereplace")[0], sout)
  90
  91     def test_callback_wrong_objects(self):
  92         def myreplace(exc):
  93             return (ret, exc.end)
  94         codecs.register_error("test.cjktest", myreplace)
  95
  96         for ret in ([1, 2, 3], [], None, object(), 'string', ''):
  97             self.assertRaises(TypeError, self.encode, self.unmappedunicode,
  98                               'test.cjktest')
  99
 100     def test_callback_long_index(self):
 101         def myreplace(exc):
 102             return (u'x', long(exc.end))
 103         codecs.register_error("test.cjktest", myreplace)
 104         self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
 105                                      'test.cjktest'), ('abcdxefgh', 9))
 106
 107         def myreplace(exc):
 108             return (u'x', sys.maxint + 1)
 109         codecs.register_error("test.cjktest", myreplace)
 110         self.assertRaises(IndexError, self.encode, self.unmappedunicode,
 111                           'test.cjktest')
 112
 113     def test_callback_None_index(self):
 114         def myreplace(exc):
 115             return (u'x', None)
 116         codecs.register_error("test.cjktest", myreplace)
 117         self.assertRaises(TypeError, self.encode, self.unmappedunicode,
 118                           'test.cjktest')
 119
 120     def test_callback_backward_index(self):
 121         def myreplace(exc):
 122             if myreplace.limit > 0:
 123                 myreplace.limit -= 1
 124                 return (u'REPLACED', 0)
 125             else:
 126                 return (u'TERMINAL', exc.end)
 127         myreplace.limit = 3
 128         codecs.register_error("test.cjktest", myreplace)
 129         self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
 130                                      'test.cjktest'),
 131                 ('abcdREPLACEDabcdREPLACEDabcdREPLACEDabcdTERMINALefgh', 9))
 132
 133     def test_callback_forward_index(self):
 134         def myreplace(exc):
 135             return (u'REPLACED', exc.end + 2)
 136         codecs.register_error("test.cjktest", myreplace)
 137         self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
 138                                      'test.cjktest'), ('abcdREPLACEDgh', 9))
 139
 140     def test_callback_index_outofbound(self):
 141         def myreplace(exc):
 142             return (u'TERM', 100)
 143         codecs.register_error("test.cjktest", myreplace)
 144         self.assertRaises(IndexError, self.encode, self.unmappedunicode,
 145                           'test.cjktest')
 146
 147     def test_incrementalencoder(self):
 148         UTF8Reader = codecs.getreader('utf-8')
 149         for sizehint in [None] + range(1, 33) + \
 150                         [64, 128, 256, 512, 1024]:
 151             istream = UTF8Reader(StringIO(self.tstring[1]))
 152             ostream = StringIO()
 153             encoder = self.incrementalencoder()
 154             while 1:
 155                 if sizehint is not None:
 156                     data = istream.read(sizehint)
 157                 else:
 158                     data = istream.read()
 159
 160                 if not data:
 161                     break
 162                 e = encoder.encode(data)
 163                 ostream.write(e)
 164
 165             self.assertEqual(ostream.getvalue(), self.tstring[0])
 166
 167     def test_incrementaldecoder(self):
 168         UTF8Writer = codecs.getwriter('utf-8')
 169         for sizehint in [None, -1] + range(1, 33) + \
 170                         [64, 128, 256, 512, 1024]:
 171             istream = StringIO(self.tstring[0])
 172             ostream = UTF8Writer(StringIO())
 173             decoder = self.incrementaldecoder()
 174             while 1:
 175                 data = istream.read(sizehint)
 176                 if not data:
 177                     break
 178                 else:
 179                     u = decoder.decode(data)
 180                     ostream.write(u)
 181
 182             self.assertEqual(ostream.getvalue(), self.tstring[1])
 183
 184     def test_incrementalencoder_error_callback(self):
 185         inv = self.unmappedunicode
 186
 187         e = self.incrementalencoder()
 188         self.assertRaises(UnicodeEncodeError, e.encode, inv, True)
 189
 190         e.errors = 'ignore'
 191         self.assertEqual(e.encode(inv, True), '')
 192
 193         e.reset()
 194         def tempreplace(exc):
 195             return (u'called', exc.end)
 196         codecs.register_error('test.incremental_error_callback', tempreplace)
 197         e.errors = 'test.incremental_error_callback'
 198         self.assertEqual(e.encode(inv, True), 'called')
 199
 200         # again
 201         e.errors = 'ignore'
 202         self.assertEqual(e.encode(inv, True), '')
 203
 204     def test_streamreader(self):
 205         UTF8Writer = codecs.getwriter('utf-8')
 206         for name in ["read", "readline", "readlines"]:
 207             for sizehint in [None, -1] + range(1, 33) + \
 208                             [64, 128, 256, 512, 1024]:
 209                 istream = self.reader(StringIO(self.tstring[0]))
 210                 ostream = UTF8Writer(StringIO())
 211                 func = getattr(istream, name)
 212                 while 1:
 213                     data = func(sizehint)
 214                     if not data:
 215                         break
 216                     if name == "readlines":
 217                         ostream.writelines(data)
 218                     else:
 219                         ostream.write(data)
 220
 221                 self.assertEqual(ostream.getvalue(), self.tstring[1])
 222
 223     def test_streamwriter(self):
 224         readfuncs = ('read', 'readline', 'readlines')
 225         UTF8Reader = codecs.getreader('utf-8')
 226         for name in readfuncs:
 227             for sizehint in [None] + range(1, 33) + \
 228                             [64, 128, 256, 512, 1024]:
 229                 istream = UTF8Reader(StringIO(self.tstring[1]))
 230                 ostream = self.writer(StringIO())
 231                 func = getattr(istream, name)
 232                 while 1:
 233                     if sizehint is not None:
 234                         data = func(sizehint)
 235                     else:
 236                         data = func()
 237
 238                     if not data:
 239                         break
 240                     if name == "readlines":
 241                         ostream.writelines(data)
 242                     else:
 243                         ostream.write(data)
 244
 245                 self.assertEqual(ostream.getvalue(), self.tstring[0])
 246
 247 class TestBase_Mapping(unittest.TestCase):
 248     pass_enctest = []
 249     pass_dectest = []
 250     supmaps = []
 251
 252     def __init__(self, *args, **kw):
 253         unittest.TestCase.__init__(self, *args, **kw)
 254         try:
 255             self.open_mapping_file() # test it to report the error early
 256         except (IOError, HTTPException):
 257             self.skipTest("Could not retrieve "+self.mapfileurl)
 258
 259     def open_mapping_file(self):
 260         return test_support.open_urlresource(self.mapfileurl)
 261
 262     def test_mapping_file(self):
 263         if self.mapfileurl.endswith('.xml'):
 264             self._test_mapping_file_ucm()
 265         else:
 266             self._test_mapping_file_plain()
 267
 268     def _test_mapping_file_plain(self):
 269         _unichr = lambda c: eval("u'\\U%08x'" % int(c, 16))
 270         unichrs = lambda s: u''.join(_unichr(c) for c in s.split('+'))
 271         urt_wa = {}
 272
 273         for line in self.open_mapping_file():
 274             if not line:
 275                 break
 276             data = line.split('#')[0].strip().split()
 277             if len(data) != 2:
 278                 continue
 279
 280             csetval = eval(data[0])
 281             if csetval <= 0x7F:
 282                 csetch = chr(csetval & 0xff)
 283             elif csetval >= 0x1000000:
 284                 csetch = chr(csetval >> 24) + chr((csetval >> 16) & 0xff) + \
 285                          chr((csetval >> 8) & 0xff) + chr(csetval & 0xff)
 286             elif csetval >= 0x10000:
 287                 csetch = chr(csetval >> 16) + \
 288                          chr((csetval >> 8) & 0xff) + chr(csetval & 0xff)
 289             elif csetval >= 0x100:
 290                 csetch = chr(csetval >> 8) + chr(csetval & 0xff)
 291             else:
 292                 continue
 293
 294             unich = unichrs(data[1])
 295             if unich == u'\ufffd' or unich in urt_wa:
 296                 continue
 297             urt_wa[unich] = csetch
 298
 299             self._testpoint(csetch, unich)
 300
 301     def _test_mapping_file_ucm(self):
 302         ucmdata = self.open_mapping_file().read()
 303         uc = re.findall('<a u="([A-F0-9]{4})" b="([0-9A-F ]+)"/>', ucmdata)
 304         for uni, coded in uc:
 305             unich = unichr(int(uni, 16))
 306             codech = ''.join(chr(int(c, 16)) for c in coded.split())
 307             self._testpoint(codech, unich)
 308
 309     def test_mapping_supplemental(self):
 310         for mapping in self.supmaps:
 311             self._testpoint(*mapping)
 312
 313     def _testpoint(self, csetch, unich):
 314         if (csetch, unich) not in self.pass_enctest:
 315             try:
 316                 self.assertEqual(unich.encode(self.encoding), csetch)
 317             except UnicodeError, exc:
 318                 self.fail('Encoding failed while testing %s -> %s: %s' % (
 319                             repr(unich), repr(csetch), exc.reason))
 320         if (csetch, unich) not in self.pass_dectest:
 321             try:
 322                 self.assertEqual(csetch.decode(self.encoding), unich)
 323             except UnicodeError, exc:
 324                 self.fail('Decoding failed while testing %s -> %s: %s' % (
 325                             repr(csetch), repr(unich), exc.reason))
 326
 327 def load_teststring(encoding):
 328     from test import cjkencodings_test
 329     return cjkencodings_test.teststring[encoding]