/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
  7import sys, codecs
  8import unittest, re
  9from test import test_support
 10from StringIO import StringIO
 11
 12class 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
 21    unmappedunicode = u'\udeee' # a unicode codepoint that is not mapped.
 22
 23    def setUp(self):
 24        if self.codec is None:
 25            self.codec = codecs.lookup(self.encoding)
 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
 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
 53    def test_xmlcharrefreplace(self):
 54        if self.has_iso10646:
 55            return
 56
 57        s = u"\u0b13\u0b23\u0b60 nd eggs"
 58        self.assertEqual(
 59            self.encode(s, "xmlcharrefreplace")[0],
 60            "ଓଣୠ nd eggs"
 61        )
 62
 63    def test_customreplace_encode(self):
 64        if self.has_iso10646:
 65            return
 66
 67        from htmlentitydefs import codepoint2name
 68
 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:
 75                    l.append(u"&%s;" % codepoint2name[ord(c)])
 76                else:
 77                    l.append(u"&#%d;" % ord(c))
 78            return (u"".join(l), exc.end)
 79
 80        codecs.register_error("test.xmlcharnamereplace", xmlcharnamereplace)
 81
 82        if self.xmlcharnametest:
 83            sin, sout = self.xmlcharnametest
 84        else:
 85            sin = u"\xab\u211c\xbb = \u2329\u1234\u232a"
 86            sout = "«ℜ» = ⟨ሴ⟩"
 87        self.assertEqual(self.encode(sin,
 88                                    "test.xmlcharnamereplace")[0], sout)
 89
 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):
101            return (u'x', long(exc.end))
102        codecs.register_error("test.cjktest", myreplace)
103        self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
104                                     'test.cjktest'), ('abcdxefgh', 9))
105
106        def myreplace(exc):
107            return (u'x', sys.maxint + 1)
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):
114            return (u'x', None)
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
123                return (u'REPLACED', 0)
124            else:
125                return (u'TERMINAL', exc.end)
126        myreplace.limit = 3
127        codecs.register_error("test.cjktest", myreplace)
128        self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
129                                     'test.cjktest'),
130                ('abcdREPLACEDabcdREPLACEDabcdREPLACEDabcdTERMINALefgh', 9))
131
132    def test_callback_forward_index(self):
133        def myreplace(exc):
134            return (u'REPLACED', exc.end + 2)
135        codecs.register_error("test.cjktest", myreplace)
136        self.assertEqual(self.encode(u'abcd' + self.unmappedunicode + u'efgh',
137                                     'test.cjktest'), ('abcdREPLACEDgh', 9))
138
139    def test_callback_index_outofbound(self):
140        def myreplace(exc):
141            return (u'TERM', 100)
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):
194            return (u'called', exc.end)
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), '')
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):
223        readfuncs = ('read', 'readline', 'readlines')
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
246if len(u'\U00012345') == 2: # ucs2 build
247    _unichr = unichr
248    def unichr(v):
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
262class 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)
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)
273
274    def test_mapping_file(self):
275        if self.mapfileurl.endswith('.xml'):
276            self._test_mapping_file_ucm()
277        else:
278            self._test_mapping_file_plain()
279
280    def _test_mapping_file_plain(self):
281        unichrs = lambda s: u''.join(map(unichr, map(eval, s.split('+'))))
282        urt_wa = {}
283
284        for line in self.open_mapping_file():
285            if not line:
286                break
287            data = line.split('#')[0].strip().split()
288            if len(data) != 2:
289                continue
290
291            csetval = eval(data[0])
292            if csetval <= 0x7F:
293                csetch = chr(csetval & 0xff)
294            elif csetval >= 0x1000000:
295                csetch = chr(csetval >> 24) + chr((csetval >> 16) & 0xff) + \
296                         chr((csetval >> 8) & 0xff) + chr(csetval & 0xff)
297            elif csetval >= 0x10000:
298                csetch = chr(csetval >> 16) + \
299                         chr((csetval >> 8) & 0xff) + chr(csetval & 0xff)
300            elif csetval >= 0x100:
301                csetch = chr(csetval >> 8) + chr(csetval & 0xff)
302            else:
303                continue
304
305            unich = unichrs(data[1])
306            if ord(unich) == 0xfffd or urt_wa.has_key(unich):
307                continue
308            urt_wa[unich] = csetch
309
310            self._testpoint(csetch, unich)
311
312    def _test_mapping_file_ucm(self):
313        ucmdata = self.open_mapping_file().read()
314        uc = re.findall('<a u="([A-F0-9]{4})" b="([0-9A-F ]+)"/>', ucmdata)
315        for uni, coded in uc:
316            unich = unichr(int(uni, 16))
317            codech = ''.join(chr(int(c, 16)) for c in coded.split())
318            self._testpoint(codech, unich)
319
320    def test_mapping_supplemental(self):
321        for mapping in self.supmaps:
322            self._testpoint(*mapping)
323
324    def _testpoint(self, csetch, unich):
325        if (csetch, unich) not in self.pass_enctest:
326            try:
327                self.assertEqual(unich.encode(self.encoding), csetch)
328            except UnicodeError, exc:
329                self.fail('Encoding failed while testing %s -> %s: %s' % (
330                            repr(unich), repr(csetch), exc.reason))
331        if (csetch, unich) not in self.pass_dectest:
332            try:
333                self.assertEqual(csetch.decode(self.encoding), unich)
334            except UnicodeError, exc:
335                self.fail('Decoding failed while testing %s -> %s: %s' % (
336                            repr(csetch), repr(unich), exc.reason))
337
338def load_teststring(encoding):
339    from test import cjkencodings_test
340    return cjkencodings_test.teststring[encoding]