Commit | Line | Data |
---|---|---|
920dae64 AT |
1 | #!/usr/bin/env python |
2 | # | |
3 | # test_multibytecodec_support.py | |
4 | # Common Unittest Routines for CJK codecs | |
5 | # | |
6 | # $CJKCodecs: test_multibytecodec_support.py,v 1.6 2004/06/19 06:09:55 perky Exp $ | |
7 | ||
8 | import sys, codecs, os.path | |
9 | import unittest | |
10 | from test import test_support | |
11 | from StringIO import StringIO | |
12 | ||
13 | __cjkcodecs__ = 0 # define this as 0 for python | |
14 | ||
15 | class TestBase: | |
16 | encoding = '' # codec name | |
17 | codec = None # codec tuple (with 4 elements) | |
18 | tstring = '' # string to test StreamReader | |
19 | ||
20 | codectests = None # must set. codec test tuple | |
21 | roundtriptest = 1 # set if roundtrip is possible with unicode | |
22 | has_iso10646 = 0 # set if this encoding contains whole iso10646 map | |
23 | xmlcharnametest = None # string to test xmlcharrefreplace | |
24 | ||
25 | def setUp(self): | |
26 | if self.codec is None: | |
27 | self.codec = codecs.lookup(self.encoding) | |
28 | self.encode, self.decode, self.reader, self.writer = self.codec | |
29 | ||
30 | def test_chunkcoding(self): | |
31 | for native, utf8 in zip(*[StringIO(f).readlines() | |
32 | for f in self.tstring]): | |
33 | u = self.decode(native)[0] | |
34 | self.assertEqual(u, utf8.decode('utf-8')) | |
35 | if self.roundtriptest: | |
36 | self.assertEqual(native, self.encode(u)[0]) | |
37 | ||
38 | def test_errorhandle(self): | |
39 | for source, scheme, expected in self.codectests: | |
40 | if type(source) == type(''): | |
41 | func = self.decode | |
42 | else: | |
43 | func = self.encode | |
44 | if expected: | |
45 | result = func(source, scheme)[0] | |
46 | self.assertEqual(result, expected) | |
47 | else: | |
48 | self.assertRaises(UnicodeError, func, source, scheme) | |
49 | ||
50 | if sys.hexversion >= 0x02030000: | |
51 | def test_xmlcharrefreplace(self): | |
52 | if self.has_iso10646: | |
53 | return | |
54 | ||
55 | s = u"\u0b13\u0b23\u0b60 nd eggs" | |
56 | self.assertEqual( | |
57 | self.encode(s, "xmlcharrefreplace")[0], | |
58 | "ଓଣୠ nd eggs" | |
59 | ) | |
60 | ||
61 | def test_customreplace(self): | |
62 | if self.has_iso10646: | |
63 | return | |
64 | ||
65 | import htmlentitydefs | |
66 | ||
67 | names = {} | |
68 | for (key, value) in htmlentitydefs.entitydefs.items(): | |
69 | if len(value)==1: | |
70 | names[value.decode('latin-1')] = self.decode(key)[0] | |
71 | else: | |
72 | names[unichr(int(value[2:-1]))] = self.decode(key)[0] | |
73 | ||
74 | def xmlcharnamereplace(exc): | |
75 | if not isinstance(exc, UnicodeEncodeError): | |
76 | raise TypeError("don't know how to handle %r" % exc) | |
77 | l = [] | |
78 | for c in exc.object[exc.start:exc.end]: | |
79 | try: | |
80 | l.append(u"&%s;" % names[c]) | |
81 | except KeyError: | |
82 | l.append(u"&#%d;" % ord(c)) | |
83 | return (u"".join(l), exc.end) | |
84 | ||
85 | codecs.register_error( | |
86 | "test.xmlcharnamereplace", xmlcharnamereplace) | |
87 | ||
88 | if self.xmlcharnametest: | |
89 | sin, sout = self.xmlcharnametest | |
90 | else: | |
91 | sin = u"\xab\u211c\xbb = \u2329\u1234\u232a" | |
92 | sout = "«ℜ» = ⟨ሴ⟩" | |
93 | self.assertEqual(self.encode(sin, | |
94 | "test.xmlcharnamereplace")[0], sout) | |
95 | ||
96 | def test_streamreader(self): | |
97 | UTF8Writer = codecs.getwriter('utf-8') | |
98 | for name in ["read", "readline", "readlines"]: | |
99 | for sizehint in [None, -1] + range(1, 33) + \ | |
100 | [64, 128, 256, 512, 1024]: | |
101 | istream = self.reader(StringIO(self.tstring[0])) | |
102 | ostream = UTF8Writer(StringIO()) | |
103 | func = getattr(istream, name) | |
104 | while 1: | |
105 | data = func(sizehint) | |
106 | if not data: | |
107 | break | |
108 | if name == "readlines": | |
109 | ostream.writelines(data) | |
110 | else: | |
111 | ostream.write(data) | |
112 | ||
113 | self.assertEqual(ostream.getvalue(), self.tstring[1]) | |
114 | ||
115 | def test_streamwriter(self): | |
116 | if __cjkcodecs__: | |
117 | readfuncs = ('read', 'readline', 'readlines') | |
118 | else: | |
119 | # standard utf8 codec has broken readline and readlines. | |
120 | readfuncs = ('read',) | |
121 | UTF8Reader = codecs.getreader('utf-8') | |
122 | for name in readfuncs: | |
123 | for sizehint in [None] + range(1, 33) + \ | |
124 | [64, 128, 256, 512, 1024]: | |
125 | istream = UTF8Reader(StringIO(self.tstring[1])) | |
126 | ostream = self.writer(StringIO()) | |
127 | func = getattr(istream, name) | |
128 | while 1: | |
129 | if sizehint is not None: | |
130 | data = func(sizehint) | |
131 | else: | |
132 | data = func() | |
133 | ||
134 | if not data: | |
135 | break | |
136 | if name == "readlines": | |
137 | ostream.writelines(data) | |
138 | else: | |
139 | ostream.write(data) | |
140 | ||
141 | self.assertEqual(ostream.getvalue(), self.tstring[0]) | |
142 | ||
143 | if len(u'\U00012345') == 2: # ucs2 build | |
144 | _unichr = unichr | |
145 | def unichr(v): | |
146 | if v >= 0x10000: | |
147 | return _unichr(0xd800 + ((v - 0x10000) >> 10)) + \ | |
148 | _unichr(0xdc00 + ((v - 0x10000) & 0x3ff)) | |
149 | else: | |
150 | return _unichr(v) | |
151 | _ord = ord | |
152 | def ord(c): | |
153 | if len(c) == 2: | |
154 | return 0x10000 + ((_ord(c[0]) - 0xd800) << 10) + \ | |
155 | (ord(c[1]) - 0xdc00) | |
156 | else: | |
157 | return _ord(c) | |
158 | ||
159 | class TestBase_Mapping(unittest.TestCase): | |
160 | pass_enctest = [] | |
161 | pass_dectest = [] | |
162 | supmaps = [] | |
163 | ||
164 | def __init__(self, *args, **kw): | |
165 | unittest.TestCase.__init__(self, *args, **kw) | |
166 | if not os.path.exists(self.mapfilename): | |
167 | raise test_support.TestSkipped('%s not found, download from %s' % | |
168 | (self.mapfilename, self.mapfileurl)) | |
169 | ||
170 | def test_mapping_file(self): | |
171 | unichrs = lambda s: u''.join(map(unichr, map(eval, s.split('+')))) | |
172 | urt_wa = {} | |
173 | ||
174 | for line in open(self.mapfilename): | |
175 | if not line: | |
176 | break | |
177 | data = line.split('#')[0].strip().split() | |
178 | if len(data) != 2: | |
179 | continue | |
180 | ||
181 | csetval = eval(data[0]) | |
182 | if csetval <= 0x7F: | |
183 | csetch = chr(csetval & 0xff) | |
184 | elif csetval >= 0x1000000: | |
185 | csetch = chr(csetval >> 24) + chr((csetval >> 16) & 0xff) + \ | |
186 | chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) | |
187 | elif csetval >= 0x10000: | |
188 | csetch = chr(csetval >> 16) + \ | |
189 | chr((csetval >> 8) & 0xff) + chr(csetval & 0xff) | |
190 | elif csetval >= 0x100: | |
191 | csetch = chr(csetval >> 8) + chr(csetval & 0xff) | |
192 | else: | |
193 | continue | |
194 | ||
195 | unich = unichrs(data[1]) | |
196 | if ord(unich) == 0xfffd or urt_wa.has_key(unich): | |
197 | continue | |
198 | urt_wa[unich] = csetch | |
199 | ||
200 | self._testpoint(csetch, unich) | |
201 | ||
202 | def test_mapping_supplemental(self): | |
203 | for mapping in self.supmaps: | |
204 | self._testpoint(*mapping) | |
205 | ||
206 | def _testpoint(self, csetch, unich): | |
207 | if (csetch, unich) not in self.pass_enctest: | |
208 | self.assertEqual(unich.encode(self.encoding), csetch) | |
209 | if (csetch, unich) not in self.pass_dectest: | |
210 | self.assertEqual(unicode(csetch, self.encoding), unich) | |
211 | ||
212 | def load_teststring(encoding): | |
213 | if __cjkcodecs__: | |
214 | etxt = open(os.path.join('sampletexts', encoding) + '.txt').read() | |
215 | utxt = open(os.path.join('sampletexts', encoding) + '.utf8').read() | |
216 | return (etxt, utxt) | |
217 | else: | |
218 | from test import cjkencodings_test | |
219 | return cjkencodings_test.teststring[encoding] | |
220 | ||
221 | def register_skip_expected(*cases): | |
222 | for case in cases: # len(cases) must be 1 at least. | |
223 | for path in [os.path.curdir, os.path.pardir]: | |
224 | fn = os.path.join(path, case.mapfilename) | |
225 | if os.path.exists(fn): | |
226 | case.mapfilename = fn | |
227 | break | |
228 | else: | |
229 | sys.modules[case.__module__].skip_expected = True | |
230 | break | |
231 | else: | |
232 | sys.modules[case.__module__].skip_expected = False |