projects / wxWidgets.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
minor icc warning fixes
[wxWidgets.git] / tests / mbconv / mbconvtest.cpp
1 ///////////////////////////////////////////////////////////////////////////////
2 // Name: tests/mbconv/main.cpp
3 // Purpose: wxMBConv unit test
4 // Author: Vadim Zeitlin, Mike Wetherell, Vince Harron
5 // Created: 14.02.04
6 // RCS-ID: $Id$
7 // Copyright: (c) 2003 TT-Solutions, (c) 2005 Mike Wetherell, Vince Harron
8 ///////////////////////////////////////////////////////////////////////////////
9
10 // ----------------------------------------------------------------------------
11 // headers
12 // ----------------------------------------------------------------------------
13
14 #include "testprec.h"
15
16 #ifdef __BORLANDC__
17 #pragma hdrstop
18 #endif
19
20 #ifndef WX_PRECOMP
21 #include "wx/wx.h"
22 #endif // WX_PRECOMP
23
24 #include "wx/strconv.h"
25 #include "wx/string.h"
26 #include "wx/txtstrm.h"
27 #include "wx/mstream.h"
28
29 #if defined wxHAVE_TCHAR_SUPPORT && !defined HAVE_WCHAR_H
30 #define HAVE_WCHAR_H
31 #endif
32
33 // ----------------------------------------------------------------------------
34 // Some wide character constants. "\uXXXX" escapes aren't supported by old
35 // compilers such as VC++ 5 and g++ 2.95.
36 // ----------------------------------------------------------------------------
37
38 wchar_t u41[] = { 0x41, 0 };
39 wchar_t u7f[] = { 0x7f, 0 };
40
41 wchar_t u80[] = { 0x80, 0 };
42 wchar_t u391[] = { 0x391, 0 };
43 wchar_t u7ff[] = { 0x7ff, 0 };
44
45 wchar_t u800[] = { 0x800, 0 };
46 wchar_t u2620[] = { 0x2620, 0 };
47 wchar_t ufffd[] = { 0xfffd, 0 };
48
49 #if SIZEOF_WCHAR_T == 4
50 wchar_t u10000[] = { 0x10000, 0 };
51 wchar_t u1000a5[] = { 0x1000a5, 0 };
52 wchar_t u10fffd[] = { 0x10fffd, 0 };
53 #else
54 wchar_t u10000[] = { 0xd800, 0xdc00, 0 };
55 wchar_t u1000a5[] = { 0xdbc0, 0xdca5, 0 };
56 wchar_t u10fffd[] = { 0xdbff, 0xdffd, 0 };
57 #endif
58
59 // ----------------------------------------------------------------------------
60 // test class
61 // ----------------------------------------------------------------------------
62
63 class MBConvTestCase : public CppUnit::TestCase
64 {
65 public:
66 MBConvTestCase() { }
67
68 private:
69 CPPUNIT_TEST_SUITE( MBConvTestCase );
70 CPPUNIT_TEST( UTF32LETests );
71 CPPUNIT_TEST( UTF32BETests );
72 CPPUNIT_TEST( WC2CP1250 );
73 CPPUNIT_TEST( UTF7Tests );
74 CPPUNIT_TEST( UTF8Tests );
75 CPPUNIT_TEST( UTF16LETests );
76 CPPUNIT_TEST( UTF16BETests );
77 CPPUNIT_TEST( CP932Tests );
78 CPPUNIT_TEST( CP1252Tests ); // depends on UTF8 Decoder functioning correctly
79 CPPUNIT_TEST( LibcTests );
80 CPPUNIT_TEST( IconvTests );
81 CPPUNIT_TEST( Latin1Tests );
82 CPPUNIT_TEST( FontmapTests );
83 CPPUNIT_TEST( BufSize );
84 #ifdef HAVE_WCHAR_H
85 CPPUNIT_TEST( UTF8_41 );
86 CPPUNIT_TEST( UTF8_7f );
87 CPPUNIT_TEST( UTF8_80 );
88 CPPUNIT_TEST( UTF8_c2_7f );
89 CPPUNIT_TEST( UTF8_c2_80 );
90 CPPUNIT_TEST( UTF8_ce_91 );
91 CPPUNIT_TEST( UTF8_df_bf );
92 CPPUNIT_TEST( UTF8_df_c0 );
93 CPPUNIT_TEST( UTF8_e0_a0_7f );
94 CPPUNIT_TEST( UTF8_e0_a0_80 );
95 CPPUNIT_TEST( UTF8_e2_98_a0 );
96 CPPUNIT_TEST( UTF8_ef_bf_bd );
97 CPPUNIT_TEST( UTF8_ef_bf_c0 );
98 CPPUNIT_TEST( UTF8_f0_90_80_7f );
99 CPPUNIT_TEST( UTF8_f0_90_80_80 );
100 CPPUNIT_TEST( UTF8_f4_8f_bf_bd );
101 CPPUNIT_TEST( UTF8PUA_f4_80_82_a5 );
102 CPPUNIT_TEST( UTF8Octal_backslash245 );
103 #endif // HAVE_WCHAR_H
104 CPPUNIT_TEST_SUITE_END();
105
106 void WC2CP1250();
107 void UTF7Tests();
108 void UTF8Tests();
109 void UTF16LETests();
110 void UTF16BETests();
111 void UTF32LETests();
112 void UTF32BETests();
113 void CP932Tests();
114 void CP1252Tests();
115 void LibcTests();
116 void FontmapTests();
117 void BufSize();
118 void IconvTests();
119 void Latin1Tests();
120
121 // verifies that the specified multibyte sequence decodes to the specified wchar_t sequence
122 void TestDecoder(
123 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
124 size_t wideChars, // the number of wide characters at wideBuffer
125 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
126 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
127 wxMBConv& converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
128 int sizeofNull // number of bytes occupied by terminating null in this encoding
129 );
130
131 // verifies that the specified wchar_t sequence encodes to the specified multibyte sequence
132 void TestEncoder(
133 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
134 size_t wideChars, // the number of wide characters at wideBuffer
135 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
136 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
137 wxMBConv& converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
138 int sizeofNull // number of bytes occupied by terminating null in this encoding
139 );
140
141 #if wxUSE_UNICODE && wxUSE_STREAMS
142 // use wxTextInputStream to exercise wxMBConv interface
143 // (this reveals some bugs in certain wxMBConv subclasses)
144 void TestStreamDecoder(
145 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
146 size_t wideChars, // the number of wide characters at wideBuffer
147 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
148 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
149 wxMBConv& converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
150 );
151
152 // use wxTextOutputStream to exercise wxMBConv interface
153 // (this reveals some bugs in certain wxMBConv subclasses)
154 void TestStreamEncoder(
155 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
156 size_t wideChars, // the number of wide characters at wideBuffer
157 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
158 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
159 wxMBConv& converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
160 );
161 #endif
162
163 // tests the encoding and decoding capability of an wxMBConv object
164 //
165 // decodes the utf-8 bytes into wide characters
166 // encodes the wide characters to compare against input multiBuffer
167 // decodes the multiBuffer to compare against wide characters
168 // decodes the multiBuffer into wide characters
169 void TestCoder(
170 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
171 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
172 const char* utf8Buffer, // the same character sequence as multiBuffer, encoded as UTF-8
173 size_t utf8Bytes, // the byte length of the UTF-8 encoded character sequence
174 wxMBConv& converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
175 int sizeofNull // the number of bytes occupied by a terminating null in the converter's encoding
176 );
177
178 #ifdef HAVE_WCHAR_H
179 // UTF-8 tests. Test the first, last and one in the middle for sequences
180 // of each length
181 void UTF8_41() { UTF8("\x41", u41); }
182 void UTF8_7f() { UTF8("\x7f", u7f); }
183 void UTF8_80() { UTF8("\x80", NULL); }
184
185 void UTF8_c2_7f() { UTF8("\xc2\x7f", NULL); }
186 void UTF8_c2_80() { UTF8("\xc2\x80", u80); }
187 void UTF8_ce_91() { UTF8("\xce\x91", u391); }
188 void UTF8_df_bf() { UTF8("\xdf\xbf", u7ff); }
189 void UTF8_df_c0() { UTF8("\xdf\xc0", NULL); }
190
191 void UTF8_e0_a0_7f() { UTF8("\xe0\xa0\x7f", NULL); }
192 void UTF8_e0_a0_80() { UTF8("\xe0\xa0\x80", u800); }
193 void UTF8_e2_98_a0() { UTF8("\xe2\x98\xa0", u2620); }
194 void UTF8_ef_bf_bd() { UTF8("\xef\xbf\xbd", ufffd); }
195 void UTF8_ef_bf_c0() { UTF8("\xef\xbf\xc0", NULL); }
196
197 void UTF8_f0_90_80_7f() { UTF8("\xf0\x90\x80\x7f", NULL); }
198 void UTF8_f0_90_80_80() { UTF8("\xf0\x90\x80\x80", u10000); }
199 void UTF8_f4_8f_bf_bd() { UTF8("\xf4\x8f\xbf\xbd", u10fffd); }
200
201 // test 'escaping the escape characters' for the two escaping schemes
202 void UTF8PUA_f4_80_82_a5() { UTF8PUA("\xf4\x80\x82\xa5", u1000a5); }
203 void UTF8Octal_backslash245() { UTF8Octal("\\245", L"\\245"); }
204
205 // implementation for the utf-8 tests (see comments below)
206 void UTF8(const char *charSequence, const wchar_t *wideSequence);
207 void UTF8PUA(const char *charSequence, const wchar_t *wideSequence);
208 void UTF8Octal(const char *charSequence, const wchar_t *wideSequence);
209 void UTF8(const char *charSequence, const wchar_t *wideSequence, int option);
210 #endif // HAVE_WCHAR_H
211
212 DECLARE_NO_COPY_CLASS(MBConvTestCase)
213 };
214
215 // register in the unnamed registry so that these tests are run by default
216 CPPUNIT_TEST_SUITE_REGISTRATION( MBConvTestCase );
217
218 // also include in it's own registry so that these tests can be run alone
219 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( MBConvTestCase, "MBConvTestCase" );
220 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( MBConvTestCase, "MBConv" );
221
222 void MBConvTestCase::WC2CP1250()
223 {
224 static const struct Data
225 {
226 const wchar_t *wc;
227 const char *cp1250;
228 } data[] =
229 {
230 { L"hello", "hello" }, // test that it works in simplest case
231 { L"\xBD of \xBD is \xBC", NULL }, // this should fail as cp1250 doesn't have 1/2
232 };
233
234 wxCSConv cs1250(wxFONTENCODING_CP1250);
235 for ( size_t n = 0; n < WXSIZEOF(data); n++ )
236 {
237 const Data& d = data[n];
238 if (d.cp1250)
239 {
240 CPPUNIT_ASSERT( strcmp(cs1250.cWC2MB(d.wc), d.cp1250) == 0 );
241 }
242 else
243 {
244 CPPUNIT_ASSERT( (const char*)cs1250.cWC2MB(d.wc) == NULL );
245 }
246 }
247 }
248
249 // Print an unsigned character array as a C unsigned character array.
250 // NB: Please don't remove this function even though it's not used anywhere,
251 // it's very useful when debugging a failed test.
252 wxString CByteArrayFormat( const void* data, size_t len, const wxChar* name )
253 {
254 const unsigned char* bytes = (unsigned char*)data;
255 wxString result;
256
257 result.Printf( _T("static const unsigned char %s[%i] = \n{"), name, (int)len );
258
259 for ( size_t i = 0; i < len; i++ )
260 {
261 if ( i != 0 )
262 {
263 result.append( _T(",") );
264 }
265 if ((i%16)==0)
266 {
267 result.append( _T("\n ") );
268 }
269 wxString byte = wxString::Format( _T("0x%02x"), bytes[i] );
270 result.append(byte);
271 }
272 result.append( _T("\n};\n") );
273 return result;
274 }
275
276 // The following bytes represent the same string, containing Japanese and English
277 // characters, encoded in several different formats.
278
279 // encoded by iconv
280 static const unsigned char welcome_utf7_iconv[84] =
281 {
282 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
283 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
284 0x2e,0x20,0x20,0x2b,0x4d,0x46,0x6b,0x77,0x55,0x49,0x74,0x6d,0x57,0x39,0x38,0x77,
285 0x61,0x35,0x62,0x37,0x69,0x6e,0x45,0x77,0x6b,0x6a,0x42,0x5a,0x4d,0x49,0x73,0x77,
286 0x65,0x7a,0x42,0x47,0x4d,0x45,0x77,0x77,0x52,0x44,0x42,0x45,0x4d,0x47,0x63,0x77,
287 0x57,0x54,0x41,0x43
288 };
289 // encoded by wxWindows (iconv can decode this successfully)
290 static const unsigned char welcome_utf7_wx[109] =
291 {
292 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x2b,0x41,0x43,0x41,0x2d,0x74,0x6f,0x2b,0x41,
293 0x43,0x41,0x2d,0x6f,0x75,0x72,0x2b,0x41,0x43,0x41,0x2d,0x63,0x79,0x62,0x65,0x72,
294 0x2b,0x41,0x43,0x41,0x2d,0x73,0x70,0x61,0x63,0x65,0x2b,0x41,0x43,0x41,0x2d,0x66,
295 0x6f,0x72,0x63,0x65,0x2e,0x2b,0x41,0x43,0x41,0x41,0x49,0x44,0x42,0x5a,0x4d,0x46,
296 0x43,0x4c,0x5a,0x6c,0x76,0x66,0x4d,0x47,0x75,0x57,0x2b,0x34,0x70,0x78,0x4d,0x4a,
297 0x49,0x77,0x57,0x54,0x43,0x4c,0x4d,0x48,0x73,0x77,0x52,0x6a,0x42,0x4d,0x4d,0x45,
298 0x51,0x77,0x52,0x44,0x42,0x6e,0x4d,0x46,0x6b,0x77,0x41,0x67,0x2d
299 };
300 // encoded by iconv
301 static const unsigned char welcome_utf8[89] =
302 {
303 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
304 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
305 0x2e,0x20,0x20,0xe3,0x81,0x99,0xe3,0x81,0x90,0xe8,0xad,0xa6,0xe5,0xaf,0x9f,0xe3,
306 0x81,0xab,0xe9,0x9b,0xbb,0xe8,0xa9,0xb1,0xe3,0x82,0x92,0xe3,0x81,0x99,0xe3,0x82,
307 0x8b,0xe3,0x81,0xbb,0xe3,0x81,0x86,0xe3,0x81,0x8c,0xe3,0x81,0x84,0xe3,0x81,0x84,
308 0xe3,0x81,0xa7,0xe3,0x81,0x99,0xe3,0x80,0x82
309 };
310 // encoded by iconv
311 static const unsigned char welcome_utf16le[106] =
312 {
313 0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,0x00,
314 0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,0x00,
315 0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,0x00,
316 0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,0x00,
317 0x2e,0x00,0x20,0x00,0x20,0x00,0x59,0x30,0x50,0x30,0x66,0x8b,0xdf,0x5b,0x6b,0x30,
318 0xfb,0x96,0x71,0x8a,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,0x30,
319 0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02,0x30
320 };
321 // encoded by iconv
322 static const unsigned char welcome_utf16be[106] =
323 {
324 0x00,0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,
325 0x00,0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,
326 0x00,0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,
327 0x00,0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,
328 0x00,0x2e,0x00,0x20,0x00,0x20,0x30,0x59,0x30,0x50,0x8b,0x66,0x5b,0xdf,0x30,0x6b,
329 0x96,0xfb,0x8a,0x71,0x30,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,
330 0x30,0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02
331 };
332 // encoded by iconv
333 static const unsigned char welcome_utf32le[212] =
334 {
335 0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
336 0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,
337 0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,
338 0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
339 0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,0x00,0x00,0x00,
340 0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,0x00,0x00,0x00,
341 0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,0x00,0x00,0x00,
342 0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,
343 0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x59,0x30,0x00,0x00,
344 0x50,0x30,0x00,0x00,0x66,0x8b,0x00,0x00,0xdf,0x5b,0x00,0x00,0x6b,0x30,0x00,0x00,
345 0xfb,0x96,0x00,0x00,0x71,0x8a,0x00,0x00,0x92,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
346 0x8b,0x30,0x00,0x00,0x7b,0x30,0x00,0x00,0x46,0x30,0x00,0x00,0x4c,0x30,0x00,0x00,
347 0x44,0x30,0x00,0x00,0x44,0x30,0x00,0x00,0x67,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
348 0x02,0x30,0x00,0x00
349 };
350 // encoded by iconv
351 static const unsigned char welcome_utf32be[212] =
352 {
353 0x00,0x00,0x00,0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,
354 0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,
355 0x00,0x00,0x00,0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,
356 0x00,0x00,0x00,0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,
357 0x00,0x00,0x00,0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,
358 0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,
359 0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,
360 0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,
361 0x00,0x00,0x00,0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x30,0x59,
362 0x00,0x00,0x30,0x50,0x00,0x00,0x8b,0x66,0x00,0x00,0x5b,0xdf,0x00,0x00,0x30,0x6b,
363 0x00,0x00,0x96,0xfb,0x00,0x00,0x8a,0x71,0x00,0x00,0x30,0x92,0x00,0x00,0x30,0x59,
364 0x00,0x00,0x30,0x8b,0x00,0x00,0x30,0x7b,0x00,0x00,0x30,0x46,0x00,0x00,0x30,0x4c,
365 0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x67,0x00,0x00,0x30,0x59,
366 0x00,0x00,0x30,0x02
367 };
368 // encoded by iconv
369 static const unsigned char welcome_cp932[71] =
370 {
371 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
372 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
373 0x2e,0x20,0x20,0x82,0xb7,0x82,0xae,0x8c,0x78,0x8e,0x40,0x82,0xc9,0x93,0x64,0x98,
374 0x62,0x82,0xf0,0x82,0xb7,0x82,0xe9,0x82,0xd9,0x82,0xa4,0x82,0xaa,0x82,0xa2,0x82,
375 0xa2,0x82,0xc5,0x82,0xb7,0x81,0x42
376 };
377
378 #if wxBYTE_ORDER == wxBIG_ENDIAN
379 #if SIZEOF_WCHAR_T == 2
380 #define welcome_wchar_t welcome_utf16be
381 #elif SIZEOF_WCHAR_T == 4
382 #define welcome_wchar_t welcome_utf32be
383 #endif
384 #elif wxBYTE_ORDER == wxLITTLE_ENDIAN
385 #if SIZEOF_WCHAR_T == 2
386 #define welcome_wchar_t welcome_utf16le
387 #elif SIZEOF_WCHAR_T == 4
388 #define welcome_wchar_t welcome_utf32le
389 #endif
390 #endif
391
392 void MBConvTestCase::UTF7Tests()
393 {
394 #if 0
395 wxCSConv convUTF7(wxFONTENCODING_UTF7);
396 #else
397 wxMBConvUTF7 convUTF7;
398 #endif
399
400 TestDecoder
401 (
402 (const wchar_t*)welcome_wchar_t,
403 sizeof(welcome_wchar_t)/sizeof(wchar_t),
404 (const char*)welcome_utf7_iconv,
405 sizeof(welcome_utf7_iconv),
406 convUTF7,
407 1
408 );
409 TestDecoder
410 (
411 (const wchar_t*)welcome_wchar_t,
412 sizeof(welcome_wchar_t)/sizeof(wchar_t),
413 (const char*)welcome_utf7_wx,
414 sizeof(welcome_utf7_wx),
415 convUTF7,
416 1
417 );
418 #if 0
419 // wxWidget's UTF-7 encoder generates different byte sequences than iconv's.
420 // but both seem to be equally legal.
421 // This test won't work and that's okay.
422 TestEncoder
423 (
424 (const wchar_t*)welcome_wchar_t,
425 sizeof(welcome_wchar_t)/sizeof(wchar_t),
426 (const char*)welcome_utf7_iconv,
427 sizeof(welcome_utf7_iconv),
428 convUTF7,
429 1
430 );
431 #endif
432 TestEncoder
433 (
434 (const wchar_t*)welcome_wchar_t,
435 sizeof(welcome_wchar_t)/sizeof(wchar_t),
436 (const char*)welcome_utf7_wx,
437 sizeof(welcome_utf7_wx),
438 convUTF7,
439 1
440 );
441 }
442
443 void MBConvTestCase::UTF8Tests()
444 {
445 TestDecoder
446 (
447 (const wchar_t*)welcome_wchar_t,
448 sizeof(welcome_wchar_t)/sizeof(wchar_t),
449 (const char*)welcome_utf8,
450 sizeof(welcome_utf8),
451 wxConvUTF8,
452 1
453 );
454 TestEncoder
455 (
456 (const wchar_t*)welcome_wchar_t,
457 sizeof(welcome_wchar_t)/sizeof(wchar_t),
458 (const char*)welcome_utf8,
459 sizeof(welcome_utf8),
460 wxConvUTF8,
461 1
462 );
463 }
464
465 void MBConvTestCase::UTF16LETests()
466 {
467 wxMBConvUTF16LE convUTF16LE;
468 TestDecoder
469 (
470 (const wchar_t*)welcome_wchar_t,
471 sizeof(welcome_wchar_t)/sizeof(wchar_t),
472 (const char*)welcome_utf16le,
473 sizeof(welcome_utf16le),
474 convUTF16LE,
475 2
476 );
477 TestEncoder
478 (
479 (const wchar_t*)welcome_wchar_t,
480 sizeof(welcome_wchar_t)/sizeof(wchar_t),
481 (const char*)welcome_utf16le,
482 sizeof(welcome_utf16le),
483 convUTF16LE,
484 2
485 );
486 }
487
488 void MBConvTestCase::UTF16BETests()
489 {
490 wxMBConvUTF16BE convUTF16BE;
491 TestDecoder
492 (
493 (const wchar_t*)welcome_wchar_t,
494 sizeof(welcome_wchar_t)/sizeof(wchar_t),
495 (const char*)welcome_utf16be,
496 sizeof(welcome_utf16be),
497 convUTF16BE,
498 2
499 );
500 TestEncoder
501 (
502 (const wchar_t*)welcome_wchar_t,
503 sizeof(welcome_wchar_t)/sizeof(wchar_t),
504 (const char*)welcome_utf16be,
505 sizeof(welcome_utf16be),
506 convUTF16BE,
507 2
508 );
509 }
510
511 void MBConvTestCase::UTF32LETests()
512 {
513 wxMBConvUTF32LE convUTF32LE;
514 TestDecoder
515 (
516 (const wchar_t*)welcome_wchar_t,
517 sizeof(welcome_wchar_t)/sizeof(wchar_t),
518 (const char*)welcome_utf32le,
519 sizeof(welcome_utf32le),
520 convUTF32LE,
521 4
522 );
523 TestEncoder
524 (
525 (const wchar_t*)welcome_wchar_t,
526 sizeof(welcome_wchar_t)/sizeof(wchar_t),
527 (const char*)welcome_utf32le,
528 sizeof(welcome_utf32le),
529 convUTF32LE,
530 4
531 );
532 }
533
534 void MBConvTestCase::UTF32BETests()
535 {
536 wxMBConvUTF32BE convUTF32BE;
537 TestDecoder
538 (
539 (const wchar_t*)welcome_wchar_t,
540 sizeof(welcome_wchar_t)/sizeof(wchar_t),
541 (const char*)welcome_utf32be,
542 sizeof(welcome_utf32be),
543 convUTF32BE,
544 4
545 );
546 TestEncoder
547 (
548 (const wchar_t*)welcome_wchar_t,
549 sizeof(welcome_wchar_t)/sizeof(wchar_t),
550 (const char*)welcome_utf32be,
551 sizeof(welcome_utf32be),
552 convUTF32BE,
553 4
554 );
555 }
556
557 void MBConvTestCase::CP932Tests()
558 {
559 wxCSConv convCP932( wxFONTENCODING_CP932 );
560 TestDecoder
561 (
562 (const wchar_t*)welcome_wchar_t,
563 sizeof(welcome_wchar_t)/sizeof(wchar_t),
564 (const char*)welcome_cp932,
565 sizeof(welcome_cp932),
566 convCP932,
567 1
568 );
569 TestEncoder
570 (
571 (const wchar_t*)welcome_wchar_t,
572 sizeof(welcome_wchar_t)/sizeof(wchar_t),
573 (const char*)welcome_cp932,
574 sizeof(welcome_cp932),
575 convCP932,
576 1
577 );
578 }
579
580 // a character sequence encoded as iso8859-1 (iconv)
581 static const unsigned char iso8859_1[251] =
582 {
583 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
584 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
585 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
586 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
587 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
588 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
589 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
590 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
591 0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
592 0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xa1,0xa2,0xa3,0xa4,
593 0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
594 0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
595 0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
596 0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
597 0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
598 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
599 };
600 // the above character sequence encoded as UTF-8 (iconv)
601 static const unsigned char iso8859_1_utf8[379] =
602 {
603 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
604 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
605 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
606 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
607 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
608 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
609 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
610 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
611 0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
612 0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
613 0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
614 0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa1,0xc2,
615 0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
616 0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
617 0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
618 0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
619 0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
620 0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
621 0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
622 0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
623 0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
624 0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
625 0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
626 0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf
627 };
628
629 // a character sequence encoded as CP1252 (iconv)
630 static const unsigned char CP1252[246] =
631 {
632 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
633 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
634 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
635 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
636 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
637 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
638 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
639 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xa0,0xa1,0xa2,0xa3,0xa4,
640 0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
641 0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
642 0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
643 0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
644 0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
645 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff,0x8c,0x9c,0x8a,0x9a,0x9f,
646 0x8e,0x9e,0x83,0x88,0x98,0x96,0x97,0x91,0x92,0x82,0x93,0x94,0x84,0x86,0x87,0x95,
647 0x85,0x89,0x8b,0x9b,0x80,0x99
648 };
649 // the above character sequence encoded as UTF-8 (iconv)
650 static const unsigned char CP1252_utf8[386] =
651 {
652 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
653 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
654 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
655 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
656 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
657 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
658 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
659 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0xa0,0xc2,0xa1,0xc2,
660 0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
661 0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
662 0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
663 0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
664 0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
665 0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
666 0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
667 0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
668 0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
669 0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
670 0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
671 0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf,0xc5,0x92,0xc5,0x93,0xc5,
672 0xa0,0xc5,0xa1,0xc5,0xb8,0xc5,0xbd,0xc5,0xbe,0xc6,0x92,0xcb,0x86,0xcb,0x9c,0xe2,
673 0x80,0x93,0xe2,0x80,0x94,0xe2,0x80,0x98,0xe2,0x80,0x99,0xe2,0x80,0x9a,0xe2,0x80,
674 0x9c,0xe2,0x80,0x9d,0xe2,0x80,0x9e,0xe2,0x80,0xa0,0xe2,0x80,0xa1,0xe2,0x80,0xa2,
675 0xe2,0x80,0xa6,0xe2,0x80,0xb0,0xe2,0x80,0xb9,0xe2,0x80,0xba,0xe2,0x82,0xac,0xe2,
676 0x84,0xa2
677 };
678
679 // this is unused currently so avoid warnings about this
680 #if 0
681
682 // a character sequence encoded as iso8859-5 (iconv)
683 static const unsigned char iso8859_5[251] =
684 {
685 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
686 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
687 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
688 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
689 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
690 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
691 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
692 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
693 0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
694 0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xfd,0xad,0xa1,0xa2,
695 0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,
696 0xb4,0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,
697 0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,
698 0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,
699 0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf1,0xf2,0xf3,0xf4,
700 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfe,0xff,0xf0
701 };
702 // the above character sequence encoded as UTF-8 (iconv)
703 static const unsigned char iso8859_5_utf8[380] =
704 {
705 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
706 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
707 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
708 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
709 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
710 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
711 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
712 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
713 0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
714 0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
715 0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
716 0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa7,0xc2,
717 0xad,0xd0,0x81,0xd0,0x82,0xd0,0x83,0xd0,0x84,0xd0,0x85,0xd0,0x86,0xd0,0x87,0xd0,
718 0x88,0xd0,0x89,0xd0,0x8a,0xd0,0x8b,0xd0,0x8c,0xd0,0x8e,0xd0,0x8f,0xd0,0x90,0xd0,
719 0x91,0xd0,0x92,0xd0,0x93,0xd0,0x94,0xd0,0x95,0xd0,0x96,0xd0,0x97,0xd0,0x98,0xd0,
720 0x99,0xd0,0x9a,0xd0,0x9b,0xd0,0x9c,0xd0,0x9d,0xd0,0x9e,0xd0,0x9f,0xd0,0xa0,0xd0,
721 0xa1,0xd0,0xa2,0xd0,0xa3,0xd0,0xa4,0xd0,0xa5,0xd0,0xa6,0xd0,0xa7,0xd0,0xa8,0xd0,
722 0xa9,0xd0,0xaa,0xd0,0xab,0xd0,0xac,0xd0,0xad,0xd0,0xae,0xd0,0xaf,0xd0,0xb0,0xd0,
723 0xb1,0xd0,0xb2,0xd0,0xb3,0xd0,0xb4,0xd0,0xb5,0xd0,0xb6,0xd0,0xb7,0xd0,0xb8,0xd0,
724 0xb9,0xd0,0xba,0xd0,0xbb,0xd0,0xbc,0xd0,0xbd,0xd0,0xbe,0xd0,0xbf,0xd1,0x80,0xd1,
725 0x81,0xd1,0x82,0xd1,0x83,0xd1,0x84,0xd1,0x85,0xd1,0x86,0xd1,0x87,0xd1,0x88,0xd1,
726 0x89,0xd1,0x8a,0xd1,0x8b,0xd1,0x8c,0xd1,0x8d,0xd1,0x8e,0xd1,0x8f,0xd1,0x91,0xd1,
727 0x92,0xd1,0x93,0xd1,0x94,0xd1,0x95,0xd1,0x96,0xd1,0x97,0xd1,0x98,0xd1,0x99,0xd1,
728 0x9a,0xd1,0x9b,0xd1,0x9c,0xd1,0x9e,0xd1,0x9f,0xe2,0x84,0x96
729 };
730 #endif // 0
731
732 // DecodeUTF8
733 // decodes the specified *unterminated* UTF-8 byte array
734 wxWCharBuffer DecodeUTF8(
735 const void* data, // an unterminated UTF-8 encoded byte array
736 size_t size // the byte length of data
737 )
738 {
739 // the decoder requires a null terminated buffer.
740 // the input data is not null terminated.
741 // copy to null terminated buffer
742
743 wxCharBuffer nullTerminated( size+1 );
744 memcpy( nullTerminated.data(), data, size );
745 nullTerminated.data()[size] = 0;
746 return wxConvUTF8.cMB2WC(nullTerminated.data());
747 }
748
749 // tests the encoding and decoding capability of an wxMBConv object
750 //
751 // decodes the utf-8 bytes into wide characters
752 // encodes the wide characters to compare against input multiBuffer
753 // decodes the multiBuffer to compare against wide characters
754 // decodes the multiBuffer into wide characters
755 void MBConvTestCase::TestCoder(
756 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
757 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
758 const char* utf8Buffer, // the same character sequence as multiBuffer, encoded as UTF-8
759 size_t utf8Bytes, // the byte length of the UTF-8 encoded character sequence
760 wxMBConv& converter, // the wxMBConv object thta can decode multiBuffer into a wide character sequence
761 int sizeofNull // the number of bytes occupied by a terminating null in the converter's encoding
762 )
763 {
764 // wide character size and endian-ess varies from platform to platform
765 // compiler support for wide character literals varies from compiler to compiler
766 // so we should store the wide character version as UTF-8 and depend on
767 // the UTF-8 converter's ability to decode it to platform specific wide characters
768 // this test is invalid if the UTF-8 converter can't decode
769 wxWCharBuffer wideBuffer((size_t)0);
770 wideBuffer = DecodeUTF8( utf8Buffer, utf8Bytes );
771 size_t wideChars = wxWcslen( wideBuffer.data() );
772
773 TestDecoder
774 (
775 wideBuffer.data(),
776 wideChars,
777 (const char*)multiBuffer,
778 multiBytes,
779 converter,
780 sizeofNull
781 );
782 TestEncoder
783 (
784 wideBuffer.data(),
785 wideChars,
786 (const char*)multiBuffer,
787 multiBytes,
788 converter,
789 sizeofNull
790 );
791 }
792
793
794 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const char* name );
795
796 void MBConvTestCase::FontmapTests()
797 {
798 #ifdef wxUSE_FONTMAP
799 wxMBConv* converter = new_wxMBConv_wxwin("CP1252");
800 if ( !converter )
801 {
802 return;
803 }
804 TestCoder(
805 (const char*)CP1252,
806 sizeof(CP1252),
807 (const char*)CP1252_utf8,
808 sizeof(CP1252_utf8),
809 *converter,
810 1
811 );
812 delete converter;
813 #endif
814 }
815
816 void MBConvTestCase::BufSize()
817 {
818 wxCSConv conv1251(_T("CP1251"));
819 CPPUNIT_ASSERT( conv1251.IsOk() );
820 const char *cp1251text =
821 "\313\301\326\305\324\323\321 \325\304\301\336\316\331\315";
822
823 const size_t lenW = conv1251.MB2WC(NULL, cp1251text, 0);
824 CPPUNIT_ASSERT_EQUAL( strlen(cp1251text), lenW );
825 wxWCharBuffer wbuf(lenW + 1); // allocates lenW + 2 characters
826 wbuf.data()[lenW + 1] = L'!';
827
828 // lenW is not enough because it's the length and we need the size
829 CPPUNIT_ASSERT_EQUAL(
830 wxCONV_FAILED, conv1251.MB2WC(wbuf.data(), cp1251text, lenW) );
831
832 // lenW+1 is just fine
833 CPPUNIT_ASSERT(
834 conv1251.MB2WC(wbuf.data(), cp1251text, lenW + 1) != wxCONV_FAILED );
835
836 // of course, greater values work too
837 CPPUNIT_ASSERT(
838 conv1251.MB2WC(wbuf.data(), cp1251text, lenW + 2) != wxCONV_FAILED );
839
840 // but they shouldn't write more stuff to the buffer
841 CPPUNIT_ASSERT_EQUAL( L'!', wbuf[lenW + 1] );
842
843
844 // test in the other direction too, using an encoding with multibyte NUL
845 wxCSConv convUTF16(_T("UTF-16LE"));
846 CPPUNIT_ASSERT( convUTF16.IsOk() );
847 const wchar_t *utf16text = L"Hello";
848
849 const size_t lenMB = convUTF16.WC2MB(NULL, utf16text, 0);
850 CPPUNIT_ASSERT_EQUAL( wcslen(utf16text)*2, lenMB );
851 wxCharBuffer buf(lenMB + 2); // it only adds 1 for NUL on its own, we need 2
852 // for NUL and an extra one for the guard byte
853 buf.data()[lenMB + 2] = '?';
854
855 CPPUNIT_ASSERT_EQUAL(
856 wxCONV_FAILED, convUTF16.WC2MB(buf.data(), utf16text, lenMB) );
857 CPPUNIT_ASSERT_EQUAL(
858 wxCONV_FAILED, convUTF16.WC2MB(buf.data(), utf16text, lenMB + 1) );
859 CPPUNIT_ASSERT(
860 convUTF16.WC2MB(buf.data(), utf16text, lenMB + 2) != wxCONV_FAILED );
861 CPPUNIT_ASSERT(
862 convUTF16.WC2MB(buf.data(), utf16text, lenMB + 3) != wxCONV_FAILED );
863 CPPUNIT_ASSERT_EQUAL( '?', buf[lenMB + 2] );
864 }
865
866
867 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const char* name );
868
869 void MBConvTestCase::IconvTests()
870 {
871 #ifdef HAVE_ICONV
872 wxMBConv* converter = new_wxMBConv_iconv("CP932");
873 if ( !converter )
874 {
875 return;
876 }
877 TestCoder(
878 (const char*)welcome_cp932,
879 sizeof(welcome_cp932),
880 (const char*)welcome_utf8,
881 sizeof(welcome_utf8),
882 *converter,
883 1
884 );
885 delete converter;
886 #endif
887 }
888
889 void MBConvTestCase::Latin1Tests()
890 {
891 TestCoder(
892 (const char*)iso8859_1,
893 sizeof(iso8859_1),
894 (const char*)iso8859_1_utf8,
895 sizeof(iso8859_1_utf8),
896 wxConvISO8859_1,
897 1
898 );
899
900 static const char nulstr[] = "foo\0bar\0";
901 static const size_t mbLen = WXSIZEOF(nulstr) - 1;
902 size_t wcLen;
903 wxConvISO8859_1.cMB2WC(nulstr, mbLen, &wcLen);
904 CPPUNIT_ASSERT_EQUAL( mbLen, wcLen );
905 }
906
907 void MBConvTestCase::CP1252Tests()
908 {
909 wxCSConv convCP1252( wxFONTENCODING_CP1252 );
910 TestCoder(
911 (const char*)CP1252,
912 sizeof(CP1252),
913 (const char*)CP1252_utf8,
914 sizeof(CP1252_utf8),
915 convCP1252,
916 1
917 );
918 }
919
920 void MBConvTestCase::LibcTests()
921 {
922 // There isn't a locale that all systems support (except "C"), so leave
923 // this one disabled for non-Windows systems for the moment, until
924 // a solution can be found.
925 #ifdef __WXMSW__
926
927 #ifdef __WXMSW__
928 setlocale( LC_ALL, "English_United States.1252" );
929 const unsigned char* systemMB = CP1252;
930 size_t systemMB_size = sizeof(CP1252);
931 const unsigned char* systemMB_utf8 = CP1252_utf8;
932 size_t systemMB_utf8_size = sizeof(CP1252_utf8);
933 #else
934 setlocale( LC_ALL, "en_US.iso8859-1" );
935 const unsigned char* systemMB = iso8859_1;
936 size_t systemMB_size = sizeof(iso8859_1);
937 const unsigned char* systemMB_utf8 = iso8859_1_utf8;
938 size_t systemMB_utf8_size = sizeof(iso8859_1_utf8);
939 #endif
940 wxMBConvLibc convLibc;
941 TestCoder(
942 (const char*)systemMB,
943 systemMB_size,
944 (const char*)systemMB_utf8,
945 systemMB_utf8_size,
946 convLibc,
947 1
948 );
949
950 #endif // __WXMSW__
951 }
952
953 // verifies that the specified mb sequences decode to the specified wc sequence
954 void MBConvTestCase::TestDecoder(
955 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
956 size_t wideChars, // the number of wide characters at wideBuffer
957 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
958 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
959 wxMBConv& converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
960 int sizeofNull // number of bytes occupied by terminating null in this encoding
961 )
962 {
963 const unsigned UNINITIALIZED = 0xcd;
964
965 // copy the input bytes into a null terminated buffer
966 wxCharBuffer inputCopy( multiBytes+sizeofNull );
967 memcpy( inputCopy.data(), multiBuffer, multiBytes );
968 memset( &inputCopy.data()[multiBytes], 0, sizeofNull );
969
970 // calculate the output size
971 size_t outputWritten = converter.MB2WC
972 (
973 0,
974 (const char*)inputCopy.data(),
975 0
976 );
977 // make sure the correct output length was calculated
978 CPPUNIT_ASSERT_EQUAL( wideChars, outputWritten );
979
980 // convert the string
981 size_t guardChars = 8; // to make sure we're not overrunning the output buffer
982 size_t nullCharacters = 1;
983 size_t outputBufferChars = outputWritten + nullCharacters + guardChars;
984 wxWCharBuffer outputBuffer(outputBufferChars);
985 memset( outputBuffer.data(), UNINITIALIZED, outputBufferChars*sizeof(wchar_t) );
986
987 outputWritten = converter.MB2WC
988 (
989 outputBuffer.data(),
990 (const char*)inputCopy.data(),
991 outputBufferChars
992 );
993 // make sure the correct number of characters were outputs
994 CPPUNIT_ASSERT_EQUAL( wideChars, outputWritten );
995
996 // make sure the characters generated are correct
997 CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, wideBuffer, wideChars*sizeof(wchar_t) ) );
998
999 // the output buffer should be null terminated
1000 CPPUNIT_ASSERT( outputBuffer[outputWritten] == 0 );
1001
1002 // make sure the rest of the output buffer is untouched
1003 for ( size_t i = (wideChars+1)*sizeof(wchar_t); i < (outputBufferChars*sizeof(wchar_t)); i++ )
1004 {
1005 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
1006 }
1007
1008 #if wxUSE_UNICODE && wxUSE_STREAMS
1009 TestStreamDecoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
1010 #endif
1011 }
1012
1013 // verifies that the specified wc sequences encodes to the specified mb sequence
1014 void MBConvTestCase::TestEncoder(
1015 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
1016 size_t wideChars, // the number of wide characters at wideBuffer
1017 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1018 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
1019 wxMBConv& converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
1020 int sizeofNull // number of bytes occupied by terminating null in this encoding
1021 )
1022 {
1023 const unsigned UNINITIALIZED = 0xcd;
1024
1025 // copy the input bytes into a null terminated buffer
1026 wxWCharBuffer inputCopy( wideChars + 1 );
1027 memcpy( inputCopy.data(), wideBuffer, (wideChars*sizeof(wchar_t)) );
1028 inputCopy.data()[wideChars] = 0;
1029
1030 // calculate the output size
1031 size_t outputWritten = converter.WC2MB
1032 (
1033 0,
1034 (const wchar_t*)inputCopy.data(),
1035 0
1036 );
1037 // make sure the correct output length was calculated
1038 CPPUNIT_ASSERT_EQUAL( multiBytes, outputWritten );
1039
1040 // convert the string
1041 size_t guardBytes = 8; // to make sure we're not overrunning the output buffer
1042 size_t outputBufferSize = outputWritten + sizeofNull + guardBytes;
1043 wxCharBuffer outputBuffer(outputBufferSize);
1044 memset( outputBuffer.data(), UNINITIALIZED, outputBufferSize );
1045
1046 outputWritten = converter.WC2MB
1047 (
1048 outputBuffer.data(),
1049 (const wchar_t*)inputCopy.data(),
1050 outputBufferSize
1051 );
1052
1053 // make sure the correct number of characters were output
1054 CPPUNIT_ASSERT_EQUAL( multiBytes, outputWritten );
1055
1056 // make sure the characters generated are correct
1057 CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, multiBuffer, multiBytes ) );
1058
1059 size_t i;
1060
1061 // the output buffer should be null terminated
1062 for ( i = multiBytes; i < multiBytes + sizeofNull; i++ )
1063 {
1064 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == 0 );
1065 }
1066
1067 // make sure the rest of the output buffer is untouched
1068 for ( i = multiBytes + sizeofNull; i < outputBufferSize; i++ )
1069 {
1070 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
1071 }
1072
1073 #if wxUSE_UNICODE && wxUSE_STREAMS
1074 TestStreamEncoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
1075 #endif
1076 }
1077
1078 #if wxUSE_UNICODE && wxUSE_STREAMS
1079 // use wxTextInputStream to exercise wxMBConv interface
1080 // (this reveals some bugs in certain wxMBConv subclasses)
1081 void MBConvTestCase::TestStreamDecoder(
1082 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
1083 size_t wideChars, // the number of wide characters at wideBuffer
1084 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1085 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
1086 wxMBConv& converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
1087 )
1088 {
1089 // this isn't meant to test wxMemoryInputStream or wxTextInputStream
1090 // it's meant to test the way wxTextInputStream uses wxMBConv
1091 // (which has exposed some problems with wxMBConv)
1092 wxMemoryInputStream memoryInputStream( multiBuffer, multiBytes );
1093 wxTextInputStream textInputStream( memoryInputStream, wxT(""), converter );
1094 for ( size_t i = 0; i < wideChars; i++ )
1095 {
1096 wxChar wc = textInputStream.GetChar();
1097 CPPUNIT_ASSERT_EQUAL_MESSAGE(
1098 std::string(wxString::Format("At position %lu", (unsigned long)i)),
1099 wideBuffer[i],
1100 wc
1101 );
1102 }
1103 CPPUNIT_ASSERT( 0 == textInputStream.GetChar() );
1104 CPPUNIT_ASSERT( memoryInputStream.Eof() );
1105 }
1106 #endif
1107
1108 #if wxUSE_UNICODE && wxUSE_STREAMS
1109 // use wxTextInputStream to exercise wxMBConv interface
1110 // (this reveals some bugs in certain wxMBConv subclasses)
1111 void MBConvTestCase::TestStreamEncoder(
1112 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
1113 size_t wideChars, // the number of wide characters at wideBuffer
1114 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1115 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
1116 wxMBConv& converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
1117 )
1118 {
1119 // this isn't meant to test wxMemoryOutputStream or wxTextOutputStream
1120 // it's meant to test the way wxTextOutputStream uses wxMBConv
1121 // (which has exposed some problems with wxMBConv)
1122 wxMemoryOutputStream memoryOutputStream;
1123 // wxEOL_UNIX will pass \n \r unchanged
1124 wxTextOutputStream textOutputStream( memoryOutputStream, wxEOL_UNIX, converter );
1125 for ( size_t i = 0; i < wideChars; i++ )
1126 {
1127 textOutputStream.PutChar( wideBuffer[i] );
1128 }
1129
1130 textOutputStream.Flush();
1131
1132 CPPUNIT_ASSERT_EQUAL( multiBytes, size_t(memoryOutputStream.TellO()) );
1133 wxCharBuffer copy( memoryOutputStream.TellO() );
1134 memoryOutputStream.CopyTo( copy.data(), memoryOutputStream.TellO());
1135 CPPUNIT_ASSERT_EQUAL( 0, memcmp( copy.data(), multiBuffer, multiBytes ) );
1136 }
1137 #endif
1138
1139
1140 // ----------------------------------------------------------------------------
1141 // UTF-8 tests
1142 // ----------------------------------------------------------------------------
1143
1144 #ifdef HAVE_WCHAR_H
1145
1146 // Check that 'charSequence' translates to 'wideSequence' and back.
1147 // Invalid sequences can be tested by giving NULL for 'wideSequence'. Even
1148 // invalid sequences should roundtrip when an option is given and this is
1149 // checked.
1150 //
1151 void MBConvTestCase::UTF8(const char *charSequence,
1152 const wchar_t *wideSequence)
1153 {
1154 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1155 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1156 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1157 }
1158
1159 // Use this alternative when 'charSequence' contains a PUA character. Such
1160 // sequences should still roundtrip ok, and this is checked.
1161 //
1162 void MBConvTestCase::UTF8PUA(const char *charSequence,
1163 const wchar_t *wideSequence)
1164 {
1165 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1166 UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1167 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1168 }
1169
1170 // Use this alternative when 'charSequence' contains an octal escape sequence.
1171 // Such sequences should still roundtrip ok, and this is checked.
1172 //
1173 void MBConvTestCase::UTF8Octal(const char *charSequence,
1174 const wchar_t *wideSequence)
1175 {
1176 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1177 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1178 UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1179 }
1180
1181 // in case wcscpy is missing
1182 //
1183 static wchar_t *wx_wcscpy(wchar_t *dest, const wchar_t *src)
1184 {
1185 wchar_t *d = dest;
1186 while ((*d++ = *src++) != 0)
1187 ;
1188 return dest;
1189 }
1190
1191 // in case wcscat is missing
1192 //
1193 static wchar_t *wx_wcscat(wchar_t *dest, const wchar_t *src)
1194 {
1195 wchar_t *d = dest;
1196 while (*d)
1197 d++;
1198 while ((*d++ = *src++) != 0)
1199 ;
1200 return dest;
1201 }
1202
1203 // in case wcscmp is missing
1204 //
1205 static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
1206 {
1207 while (*s1 == *s2 && *s1 != 0)
1208 {
1209 s1++;
1210 s2++;
1211 }
1212 return *s1 - *s2;
1213 }
1214
1215 // in case wcslen is missing
1216 //
1217 static size_t wx_wcslen(const wchar_t *s)
1218 {
1219 const wchar_t *t = s;
1220 while (*t != 0)
1221 t++;
1222 return t - s;
1223 }
1224
1225 // include the option in the error messages so it's possible to see which
1226 // test failed
1227 #define UTF8ASSERT(expr) CPPUNIT_ASSERT_MESSAGE(#expr + errmsg, expr)
1228
1229 // The test implementation
1230 //
1231 void MBConvTestCase::UTF8(const char *charSequence,
1232 const wchar_t *wideSequence,
1233 int option)
1234 {
1235 const size_t BUFSIZE = 128;
1236 wxASSERT(strlen(charSequence) * 3 + 10 < BUFSIZE);
1237 char bytes[BUFSIZE];
1238
1239 // include the option in the error messages so it's possible to see
1240 // which test failed
1241 sprintf(bytes, " (with option == %d)", option);
1242 std::string errmsg(bytes);
1243
1244 // put the charSequence at the start, middle and end of a string
1245 strcpy(bytes, charSequence);
1246 strcat(bytes, "ABC");
1247 strcat(bytes, charSequence);
1248 strcat(bytes, "XYZ");
1249 strcat(bytes, charSequence);
1250
1251 // translate it into wide characters
1252 wxMBConvUTF8 utf8(option);
1253 wchar_t widechars[BUFSIZE];
1254 size_t lenResult = utf8.MB2WC(NULL, bytes, 0);
1255 size_t result = utf8.MB2WC(widechars, bytes, BUFSIZE);
1256 UTF8ASSERT(result == lenResult);
1257
1258 // check we got the expected result
1259 if (wideSequence) {
1260 UTF8ASSERT(result != (size_t)-1);
1261 wxASSERT(result < BUFSIZE);
1262
1263 wchar_t expected[BUFSIZE];
1264 wx_wcscpy(expected, wideSequence);
1265 wx_wcscat(expected, L"ABC");
1266 wx_wcscat(expected, wideSequence);
1267 wx_wcscat(expected, L"XYZ");
1268 wx_wcscat(expected, wideSequence);
1269
1270 UTF8ASSERT(wx_wcscmp(widechars, expected) == 0);
1271 UTF8ASSERT(wx_wcslen(widechars) == result);
1272 }
1273 else {
1274 // If 'wideSequence' is NULL, then the result is expected to be
1275 // invalid. Normally that is as far as we can go, but if there is an
1276 // option then the conversion should succeed anyway, and it should be
1277 // possible to translate back to the original
1278 if (!option) {
1279 UTF8ASSERT(result == (size_t)-1);
1280 return;
1281 }
1282 else {
1283 UTF8ASSERT(result != (size_t)-1);
1284 }
1285 }
1286
1287 // translate it back and check we get the original
1288 char bytesAgain[BUFSIZE];
1289 size_t lenResultAgain = utf8.WC2MB(NULL, widechars, 0);
1290 size_t resultAgain = utf8.WC2MB(bytesAgain, widechars, BUFSIZE);
1291 UTF8ASSERT(resultAgain == lenResultAgain);
1292 UTF8ASSERT(resultAgain != (size_t)-1);
1293 wxASSERT(resultAgain < BUFSIZE);
1294
1295 UTF8ASSERT(strcmp(bytes, bytesAgain) == 0);
1296 UTF8ASSERT(strlen(bytesAgain) == resultAgain);
1297 }
1298
1299 #endif // HAVE_WCHAR_H
wxWidgets (Impactor)