]> git.saurik.com Git - wxWidgets.git/blame_incremental - tests/mbconv/mbconvtest.cpp
projects / wxWidgets.git / blame_incremental
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (non-incremental) | history | HEAD
stub for wxOS2 region ctor from points.
[wxWidgets.git] / tests / mbconv / mbconvtest.cpp
... / ...
CommitLineData
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
38wchar_t u41[] = { 0x41, 0 };
39wchar_t u7f[] = { 0x7f, 0 };
40
41wchar_t u80[] = { 0x80, 0 };
42wchar_t u391[] = { 0x391, 0 };
43wchar_t u7ff[] = { 0x7ff, 0 };
44
45wchar_t u800[] = { 0x800, 0 };
46wchar_t u2620[] = { 0x2620, 0 };
47wchar_t ufffd[] = { 0xfffd, 0 };
48
49#if SIZEOF_WCHAR_T == 4
50wchar_t u10000[] = { 0x10000, 0 };
51wchar_t u1000a5[] = { 0x1000a5, 0 };
52wchar_t u10fffd[] = { 0x10fffd, 0 };
53#else
54wchar_t u10000[] = { 0xd800, 0xdc00, 0 };
55wchar_t u1000a5[] = { 0xdbc0, 0xdca5, 0 };
56wchar_t u10fffd[] = { 0xdbff, 0xdffd, 0 };
57#endif
58
59// ----------------------------------------------------------------------------
60// test class
61// ----------------------------------------------------------------------------
62
63class MBConvTestCase : public CppUnit::TestCase
64{
65public:
66 MBConvTestCase() { }
67
68private:
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( FontmapTests );
82#ifdef HAVE_WCHAR_H
83 CPPUNIT_TEST( UTF8_41 );
84 CPPUNIT_TEST( UTF8_7f );
85 CPPUNIT_TEST( UTF8_80 );
86 CPPUNIT_TEST( UTF8_c2_7f );
87 CPPUNIT_TEST( UTF8_c2_80 );
88 CPPUNIT_TEST( UTF8_ce_91 );
89 CPPUNIT_TEST( UTF8_df_bf );
90 CPPUNIT_TEST( UTF8_df_c0 );
91 CPPUNIT_TEST( UTF8_e0_a0_7f );
92 CPPUNIT_TEST( UTF8_e0_a0_80 );
93 CPPUNIT_TEST( UTF8_e2_98_a0 );
94 CPPUNIT_TEST( UTF8_ef_bf_bd );
95 CPPUNIT_TEST( UTF8_ef_bf_c0 );
96 CPPUNIT_TEST( UTF8_f0_90_80_7f );
97 CPPUNIT_TEST( UTF8_f0_90_80_80 );
98 CPPUNIT_TEST( UTF8_f4_8f_bf_bd );
99 CPPUNIT_TEST( UTF8PUA_f4_80_82_a5 );
100 CPPUNIT_TEST( UTF8Octal_backslash245 );
101#endif // HAVE_WCHAR_H
102 CPPUNIT_TEST_SUITE_END();
103
104 void WC2CP1250();
105 void UTF7Tests();
106 void UTF8Tests();
107 void UTF16LETests();
108 void UTF16BETests();
109 void UTF32LETests();
110 void UTF32BETests();
111 void CP932Tests();
112 void CP1252Tests();
113 void LibcTests();
114 void FontmapTests();
115 void IconvTests();
116
117 // verifies that the specified multibyte sequence decodes to the specified wchar_t sequence
118 void TestDecoder(
119 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
120 size_t wideChars, // the number of wide characters at wideBuffer
121 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
122 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
123 wxMBConv* converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
124 int sizeofNull // number of bytes occupied by terminating null in this encoding
125 );
126
127 // verifies that the specified wchar_t sequence encodes to the specified multibyte sequence
128 void TestEncoder(
129 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
130 size_t wideChars, // the number of wide characters at wideBuffer
131 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
132 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
133 wxMBConv* converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
134 int sizeofNull // number of bytes occupied by terminating null in this encoding
135 );
136
137#if wxUSE_UNICODE && wxUSE_STREAMS
138 // use wxTextInputStream to exercise wxMBConv interface
139 // (this reveals some bugs in certain wxMBConv subclasses)
140 void TestStreamDecoder(
141 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
142 size_t wideChars, // the number of wide characters at wideBuffer
143 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
144 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
145 wxMBConv* converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
146 );
147
148 // use wxTextOutputStream to exercise wxMBConv interface
149 // (this reveals some bugs in certain wxMBConv subclasses)
150 void TestStreamEncoder(
151 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
152 size_t wideChars, // the number of wide characters at wideBuffer
153 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
154 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
155 wxMBConv* converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
156 );
157#endif
158
159 // tests the encoding and decoding capability of an wxMBConv object
160 //
161 // decodes the utf-8 bytes into wide characters
162 // encodes the wide characters to compare against input multiBuffer
163 // decodes the multiBuffer to compare against wide characters
164 // decodes the multiBuffer into wide characters
165 void TestCoder(
166 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
167 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
168 const char* utf8Buffer, // the same character sequence as multiBuffer, encoded as UTF-8
169 size_t utf8Bytes, // the byte length of the UTF-8 encoded character sequence
170 wxMBConv* converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
171 int sizeofNull // the number of bytes occupied by a terminating null in the converter's encoding
172 );
173
174#ifdef HAVE_WCHAR_H
175 // UTF-8 tests. Test the first, last and one in the middle for sequences
176 // of each length
177 void UTF8_41() { UTF8("\x41", u41); }
178 void UTF8_7f() { UTF8("\x7f", u7f); }
179 void UTF8_80() { UTF8("\x80", NULL); }
180
181 void UTF8_c2_7f() { UTF8("\xc2\x7f", NULL); }
182 void UTF8_c2_80() { UTF8("\xc2\x80", u80); }
183 void UTF8_ce_91() { UTF8("\xce\x91", u391); }
184 void UTF8_df_bf() { UTF8("\xdf\xbf", u7ff); }
185 void UTF8_df_c0() { UTF8("\xdf\xc0", NULL); }
186
187 void UTF8_e0_a0_7f() { UTF8("\xe0\xa0\x7f", NULL); }
188 void UTF8_e0_a0_80() { UTF8("\xe0\xa0\x80", u800); }
189 void UTF8_e2_98_a0() { UTF8("\xe2\x98\xa0", u2620); }
190 void UTF8_ef_bf_bd() { UTF8("\xef\xbf\xbd", ufffd); }
191 void UTF8_ef_bf_c0() { UTF8("\xef\xbf\xc0", NULL); }
192
193 void UTF8_f0_90_80_7f() { UTF8("\xf0\x90\x80\x7f", NULL); }
194 void UTF8_f0_90_80_80() { UTF8("\xf0\x90\x80\x80", u10000); }
195 void UTF8_f4_8f_bf_bd() { UTF8("\xf4\x8f\xbf\xbd", u10fffd); }
196
197 // test 'escaping the escape characters' for the two escaping schemes
198 void UTF8PUA_f4_80_82_a5() { UTF8PUA("\xf4\x80\x82\xa5", u1000a5); }
199 void UTF8Octal_backslash245() { UTF8Octal("\\245", L"\\245"); }
200
201 // implementation for the utf-8 tests (see comments below)
202 void UTF8(const char *charSequence, const wchar_t *wideSequence);
203 void UTF8PUA(const char *charSequence, const wchar_t *wideSequence);
204 void UTF8Octal(const char *charSequence, const wchar_t *wideSequence);
205 void UTF8(const char *charSequence, const wchar_t *wideSequence, int option);
206#endif // HAVE_WCHAR_H
207
208 DECLARE_NO_COPY_CLASS(MBConvTestCase)
209};
210
211// register in the unnamed registry so that these tests are run by default
212CPPUNIT_TEST_SUITE_REGISTRATION( MBConvTestCase );
213
214// also include in it's own registry so that these tests can be run alone
215CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( MBConvTestCase, "MBConv" );
216
217void MBConvTestCase::WC2CP1250()
218{
219 static const struct Data
220 {
221 const wchar_t *wc;
222 const char *cp1250;
223 } data[] =
224 {
225 { L"hello", "hello" }, // test that it works in simplest case
226 { L"\xBD of \xBD is \xBC", NULL }, // this should fail as cp1250 doesn't have 1/2
227 };
228
229 wxCSConv cs1250(wxFONTENCODING_CP1250);
230 for ( size_t n = 0; n < WXSIZEOF(data); n++ )
231 {
232 const Data& d = data[n];
233 if (d.cp1250)
234 {
235 CPPUNIT_ASSERT( strcmp(cs1250.cWC2MB(d.wc), d.cp1250) == 0 );
236 }
237 else
238 {
239 CPPUNIT_ASSERT( (const char*)cs1250.cWC2MB(d.wc) == NULL );
240 }
241 }
242}
243
244// print an unsigned character array as a C unsigned character array
245wxString CByteArrayFormat( const void* data, size_t len, const wxChar* name )
246{
247 const unsigned char* bytes = (unsigned char*)data;
248 wxString result;
249
250 result.Printf( _T("const static unsigned char %s[%i] = \n{"), name, (int)len );
251
252 for ( size_t i = 0; i < len; i++ )
253 {
254 if ( i != 0 )
255 {
256 result.append( _T(",") );
257 }
258 if ((i%16)==0)
259 {
260 result.append( _T("\n ") );
261 }
262 wxString byte = wxString::Format( _T("0x%02x"), bytes[i] );
263 result.append(byte);
264 }
265 result.append( _T("\n};\n") );
266 return result;
267}
268
269// The following bytes represent the same string, containing Japanese and English
270// characters, encoded in several different formats.
271
272// encoded by iconv
273const static unsigned char welcome_utf7_iconv[84] =
274{
275 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
276 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
277 0x2e,0x20,0x20,0x2b,0x4d,0x46,0x6b,0x77,0x55,0x49,0x74,0x6d,0x57,0x39,0x38,0x77,
278 0x61,0x35,0x62,0x37,0x69,0x6e,0x45,0x77,0x6b,0x6a,0x42,0x5a,0x4d,0x49,0x73,0x77,
279 0x65,0x7a,0x42,0x47,0x4d,0x45,0x77,0x77,0x52,0x44,0x42,0x45,0x4d,0x47,0x63,0x77,
280 0x57,0x54,0x41,0x43
281};
282// encoded by wxWindows (iconv can decode this successfully)
283const static unsigned char welcome_utf7_wx[109] =
284{
285 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x2b,0x41,0x43,0x41,0x2d,0x74,0x6f,0x2b,0x41,
286 0x43,0x41,0x2d,0x6f,0x75,0x72,0x2b,0x41,0x43,0x41,0x2d,0x63,0x79,0x62,0x65,0x72,
287 0x2b,0x41,0x43,0x41,0x2d,0x73,0x70,0x61,0x63,0x65,0x2b,0x41,0x43,0x41,0x2d,0x66,
288 0x6f,0x72,0x63,0x65,0x2e,0x2b,0x41,0x43,0x41,0x41,0x49,0x44,0x42,0x5a,0x4d,0x46,
289 0x43,0x4c,0x5a,0x6c,0x76,0x66,0x4d,0x47,0x75,0x57,0x2b,0x34,0x70,0x78,0x4d,0x4a,
290 0x49,0x77,0x57,0x54,0x43,0x4c,0x4d,0x48,0x73,0x77,0x52,0x6a,0x42,0x4d,0x4d,0x45,
291 0x51,0x77,0x52,0x44,0x42,0x6e,0x4d,0x46,0x6b,0x77,0x41,0x67,0x2d
292};
293// encoded by iconv
294const static unsigned char welcome_utf8[89] =
295{
296 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
297 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
298 0x2e,0x20,0x20,0xe3,0x81,0x99,0xe3,0x81,0x90,0xe8,0xad,0xa6,0xe5,0xaf,0x9f,0xe3,
299 0x81,0xab,0xe9,0x9b,0xbb,0xe8,0xa9,0xb1,0xe3,0x82,0x92,0xe3,0x81,0x99,0xe3,0x82,
300 0x8b,0xe3,0x81,0xbb,0xe3,0x81,0x86,0xe3,0x81,0x8c,0xe3,0x81,0x84,0xe3,0x81,0x84,
301 0xe3,0x81,0xa7,0xe3,0x81,0x99,0xe3,0x80,0x82
302};
303// encoded by iconv
304const static unsigned char welcome_utf16le[106] =
305{
306 0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,0x00,
307 0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,0x00,
308 0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,0x00,
309 0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,0x00,
310 0x2e,0x00,0x20,0x00,0x20,0x00,0x59,0x30,0x50,0x30,0x66,0x8b,0xdf,0x5b,0x6b,0x30,
311 0xfb,0x96,0x71,0x8a,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,0x30,
312 0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02,0x30
313};
314// encoded by iconv
315const static unsigned char welcome_utf16be[106] =
316{
317 0x00,0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,
318 0x00,0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,
319 0x00,0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,
320 0x00,0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,
321 0x00,0x2e,0x00,0x20,0x00,0x20,0x30,0x59,0x30,0x50,0x8b,0x66,0x5b,0xdf,0x30,0x6b,
322 0x96,0xfb,0x8a,0x71,0x30,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,
323 0x30,0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02
324};
325// encoded by iconv
326const static unsigned char welcome_utf32le[212] =
327{
328 0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
329 0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,
330 0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,
331 0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
332 0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,0x00,0x00,0x00,
333 0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,0x00,0x00,0x00,
334 0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,0x00,0x00,0x00,
335 0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,
336 0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x59,0x30,0x00,0x00,
337 0x50,0x30,0x00,0x00,0x66,0x8b,0x00,0x00,0xdf,0x5b,0x00,0x00,0x6b,0x30,0x00,0x00,
338 0xfb,0x96,0x00,0x00,0x71,0x8a,0x00,0x00,0x92,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
339 0x8b,0x30,0x00,0x00,0x7b,0x30,0x00,0x00,0x46,0x30,0x00,0x00,0x4c,0x30,0x00,0x00,
340 0x44,0x30,0x00,0x00,0x44,0x30,0x00,0x00,0x67,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
341 0x02,0x30,0x00,0x00
342};
343// encoded by iconv
344const static unsigned char welcome_utf32be[212] =
345{
346 0x00,0x00,0x00,0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,
347 0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,
348 0x00,0x00,0x00,0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,
349 0x00,0x00,0x00,0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,
350 0x00,0x00,0x00,0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,
351 0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,
352 0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,
353 0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,
354 0x00,0x00,0x00,0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x30,0x59,
355 0x00,0x00,0x30,0x50,0x00,0x00,0x8b,0x66,0x00,0x00,0x5b,0xdf,0x00,0x00,0x30,0x6b,
356 0x00,0x00,0x96,0xfb,0x00,0x00,0x8a,0x71,0x00,0x00,0x30,0x92,0x00,0x00,0x30,0x59,
357 0x00,0x00,0x30,0x8b,0x00,0x00,0x30,0x7b,0x00,0x00,0x30,0x46,0x00,0x00,0x30,0x4c,
358 0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x67,0x00,0x00,0x30,0x59,
359 0x00,0x00,0x30,0x02
360};
361// encoded by iconv
362const static unsigned char welcome_cp932[71] =
363{
364 0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
365 0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
366 0x2e,0x20,0x20,0x82,0xb7,0x82,0xae,0x8c,0x78,0x8e,0x40,0x82,0xc9,0x93,0x64,0x98,
367 0x62,0x82,0xf0,0x82,0xb7,0x82,0xe9,0x82,0xd9,0x82,0xa4,0x82,0xaa,0x82,0xa2,0x82,
368 0xa2,0x82,0xc5,0x82,0xb7,0x81,0x42
369};
370
371#if wxBYTE_ORDER == wxBIG_ENDIAN
372 #if SIZEOF_WCHAR_T == 2
373 #define welcome_wchar_t welcome_utf16be
374 #elif SIZEOF_WCHAR_T == 4
375 #define welcome_wchar_t welcome_utf32be
376 #endif
377#elif wxBYTE_ORDER == wxLITTLE_ENDIAN
378 #if SIZEOF_WCHAR_T == 2
379 #define welcome_wchar_t welcome_utf16le
380 #elif SIZEOF_WCHAR_T == 4
381 #define welcome_wchar_t welcome_utf32le
382 #endif
383#endif
384
385void MBConvTestCase::UTF7Tests()
386{
387 TestDecoder
388 (
389 (const wchar_t*)welcome_wchar_t,
390 sizeof(welcome_wchar_t)/sizeof(wchar_t),
391 (const char*)welcome_utf7_iconv,
392 sizeof(welcome_utf7_iconv),
393 &wxConvUTF7,
394 1
395 );
396 TestDecoder
397 (
398 (const wchar_t*)welcome_wchar_t,
399 sizeof(welcome_wchar_t)/sizeof(wchar_t),
400 (const char*)welcome_utf7_wx,
401 sizeof(welcome_utf7_wx),
402 &wxConvUTF7,
403 1
404 );
405#if 0
406 // wxWidget's UTF-7 encoder generates different byte sequences than iconv's.
407 // but both seem to be equally legal.
408 // This test won't work and that's okay.
409 TestEncoder
410 (
411 (const wchar_t*)welcome_wchar_t,
412 sizeof(welcome_wchar_t)/sizeof(wchar_t),
413 (const char*)welcome_utf7_iconv,
414 sizeof(welcome_utf7_iconv),
415 &wxConvUTF7,
416 1
417 );
418#endif
419 TestEncoder
420 (
421 (const wchar_t*)welcome_wchar_t,
422 sizeof(welcome_wchar_t)/sizeof(wchar_t),
423 (const char*)welcome_utf7_wx,
424 sizeof(welcome_utf7_wx),
425 &wxConvUTF7,
426 1
427 );
428}
429
430void MBConvTestCase::UTF8Tests()
431{
432 TestDecoder
433 (
434 (const wchar_t*)welcome_wchar_t,
435 sizeof(welcome_wchar_t)/sizeof(wchar_t),
436 (const char*)welcome_utf8,
437 sizeof(welcome_utf8),
438 &wxConvUTF8,
439 1
440 );
441 TestEncoder
442 (
443 (const wchar_t*)welcome_wchar_t,
444 sizeof(welcome_wchar_t)/sizeof(wchar_t),
445 (const char*)welcome_utf8,
446 sizeof(welcome_utf8),
447 &wxConvUTF8,
448 1
449 );
450}
451
452void MBConvTestCase::UTF16LETests()
453{
454 wxMBConvUTF16LE convUTF16LE;
455 TestDecoder
456 (
457 (const wchar_t*)welcome_wchar_t,
458 sizeof(welcome_wchar_t)/sizeof(wchar_t),
459 (const char*)welcome_utf16le,
460 sizeof(welcome_utf16le),
461 &convUTF16LE,
462 2
463 );
464 TestEncoder
465 (
466 (const wchar_t*)welcome_wchar_t,
467 sizeof(welcome_wchar_t)/sizeof(wchar_t),
468 (const char*)welcome_utf16le,
469 sizeof(welcome_utf16le),
470 &convUTF16LE,
471 2
472 );
473}
474
475void MBConvTestCase::UTF16BETests()
476{
477 wxMBConvUTF16BE convUTF16BE;
478 TestDecoder
479 (
480 (const wchar_t*)welcome_wchar_t,
481 sizeof(welcome_wchar_t)/sizeof(wchar_t),
482 (const char*)welcome_utf16be,
483 sizeof(welcome_utf16be),
484 &convUTF16BE,
485 2
486 );
487 TestEncoder
488 (
489 (const wchar_t*)welcome_wchar_t,
490 sizeof(welcome_wchar_t)/sizeof(wchar_t),
491 (const char*)welcome_utf16be,
492 sizeof(welcome_utf16be),
493 &convUTF16BE,
494 2
495 );
496}
497
498void MBConvTestCase::UTF32LETests()
499{
500 wxMBConvUTF32LE convUTF32LE;
501 TestDecoder
502 (
503 (const wchar_t*)welcome_wchar_t,
504 sizeof(welcome_wchar_t)/sizeof(wchar_t),
505 (const char*)welcome_utf32le,
506 sizeof(welcome_utf32le),
507 &convUTF32LE,
508 4
509 );
510 TestEncoder
511 (
512 (const wchar_t*)welcome_wchar_t,
513 sizeof(welcome_wchar_t)/sizeof(wchar_t),
514 (const char*)welcome_utf32le,
515 sizeof(welcome_utf32le),
516 &convUTF32LE,
517 4
518 );
519}
520
521void MBConvTestCase::UTF32BETests()
522{
523 wxMBConvUTF32BE convUTF32BE;
524 TestDecoder
525 (
526 (const wchar_t*)welcome_wchar_t,
527 sizeof(welcome_wchar_t)/sizeof(wchar_t),
528 (const char*)welcome_utf32be,
529 sizeof(welcome_utf32be),
530 &convUTF32BE,
531 4
532 );
533 TestEncoder
534 (
535 (const wchar_t*)welcome_wchar_t,
536 sizeof(welcome_wchar_t)/sizeof(wchar_t),
537 (const char*)welcome_utf32be,
538 sizeof(welcome_utf32be),
539 &convUTF32BE,
540 4
541 );
542}
543
544void MBConvTestCase::CP932Tests()
545{
546 wxCSConv convCP932( wxFONTENCODING_CP932 );
547 TestDecoder
548 (
549 (const wchar_t*)welcome_wchar_t,
550 sizeof(welcome_wchar_t)/sizeof(wchar_t),
551 (const char*)welcome_cp932,
552 sizeof(welcome_cp932),
553 &convCP932,
554 1
555 );
556 TestEncoder
557 (
558 (const wchar_t*)welcome_wchar_t,
559 sizeof(welcome_wchar_t)/sizeof(wchar_t),
560 (const char*)welcome_cp932,
561 sizeof(welcome_cp932),
562 &convCP932,
563 1
564 );
565}
566
567// a character sequence encoded as iso8859-1 (iconv)
568static const unsigned char iso8859_1[251] =
569{
570 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
571 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
572 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
573 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
574 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
575 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
576 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
577 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
578 0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
579 0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xa1,0xa2,0xa3,0xa4,
580 0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
581 0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
582 0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
583 0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
584 0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
585 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
586};
587// the above character sequence encoded as UTF-8 (iconv)
588static const unsigned char iso8859_1_utf8[379] =
589{
590 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
591 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
592 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
593 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
594 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
595 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
596 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
597 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
598 0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
599 0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
600 0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
601 0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa1,0xc2,
602 0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
603 0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
604 0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
605 0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
606 0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
607 0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
608 0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
609 0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
610 0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
611 0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
612 0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
613 0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf
614};
615
616// a character sequence encoded as CP1252 (iconv)
617static const unsigned char CP1252[246] =
618{
619 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
620 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
621 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
622 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
623 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
624 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
625 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
626 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xa0,0xa1,0xa2,0xa3,0xa4,
627 0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
628 0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
629 0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
630 0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
631 0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
632 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff,0x8c,0x9c,0x8a,0x9a,0x9f,
633 0x8e,0x9e,0x83,0x88,0x98,0x96,0x97,0x91,0x92,0x82,0x93,0x94,0x84,0x86,0x87,0x95,
634 0x85,0x89,0x8b,0x9b,0x80,0x99
635};
636// the above character sequence encoded as UTF-8 (iconv)
637static const unsigned char CP1252_utf8[386] =
638{
639 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
640 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
641 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
642 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
643 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
644 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
645 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
646 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0xa0,0xc2,0xa1,0xc2,
647 0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
648 0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
649 0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
650 0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
651 0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
652 0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
653 0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
654 0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
655 0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
656 0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
657 0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
658 0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf,0xc5,0x92,0xc5,0x93,0xc5,
659 0xa0,0xc5,0xa1,0xc5,0xb8,0xc5,0xbd,0xc5,0xbe,0xc6,0x92,0xcb,0x86,0xcb,0x9c,0xe2,
660 0x80,0x93,0xe2,0x80,0x94,0xe2,0x80,0x98,0xe2,0x80,0x99,0xe2,0x80,0x9a,0xe2,0x80,
661 0x9c,0xe2,0x80,0x9d,0xe2,0x80,0x9e,0xe2,0x80,0xa0,0xe2,0x80,0xa1,0xe2,0x80,0xa2,
662 0xe2,0x80,0xa6,0xe2,0x80,0xb0,0xe2,0x80,0xb9,0xe2,0x80,0xba,0xe2,0x82,0xac,0xe2,
663 0x84,0xa2
664};
665
666// a character sequence encoded as iso8859-5 (iconv)
667static const unsigned char iso8859_5[251] =
668{
669 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
670 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
671 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
672 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
673 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
674 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
675 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
676 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
677 0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
678 0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xfd,0xad,0xa1,0xa2,
679 0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,
680 0xb4,0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,
681 0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,
682 0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,
683 0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf1,0xf2,0xf3,0xf4,
684 0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfe,0xff,0xf0
685};
686// the above character sequence encoded as UTF-8 (iconv)
687static const unsigned char iso8859_5_utf8[380] =
688{
689 0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
690 0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
691 0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
692 0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
693 0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
694 0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
695 0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
696 0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
697 0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
698 0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
699 0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
700 0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa7,0xc2,
701 0xad,0xd0,0x81,0xd0,0x82,0xd0,0x83,0xd0,0x84,0xd0,0x85,0xd0,0x86,0xd0,0x87,0xd0,
702 0x88,0xd0,0x89,0xd0,0x8a,0xd0,0x8b,0xd0,0x8c,0xd0,0x8e,0xd0,0x8f,0xd0,0x90,0xd0,
703 0x91,0xd0,0x92,0xd0,0x93,0xd0,0x94,0xd0,0x95,0xd0,0x96,0xd0,0x97,0xd0,0x98,0xd0,
704 0x99,0xd0,0x9a,0xd0,0x9b,0xd0,0x9c,0xd0,0x9d,0xd0,0x9e,0xd0,0x9f,0xd0,0xa0,0xd0,
705 0xa1,0xd0,0xa2,0xd0,0xa3,0xd0,0xa4,0xd0,0xa5,0xd0,0xa6,0xd0,0xa7,0xd0,0xa8,0xd0,
706 0xa9,0xd0,0xaa,0xd0,0xab,0xd0,0xac,0xd0,0xad,0xd0,0xae,0xd0,0xaf,0xd0,0xb0,0xd0,
707 0xb1,0xd0,0xb2,0xd0,0xb3,0xd0,0xb4,0xd0,0xb5,0xd0,0xb6,0xd0,0xb7,0xd0,0xb8,0xd0,
708 0xb9,0xd0,0xba,0xd0,0xbb,0xd0,0xbc,0xd0,0xbd,0xd0,0xbe,0xd0,0xbf,0xd1,0x80,0xd1,
709 0x81,0xd1,0x82,0xd1,0x83,0xd1,0x84,0xd1,0x85,0xd1,0x86,0xd1,0x87,0xd1,0x88,0xd1,
710 0x89,0xd1,0x8a,0xd1,0x8b,0xd1,0x8c,0xd1,0x8d,0xd1,0x8e,0xd1,0x8f,0xd1,0x91,0xd1,
711 0x92,0xd1,0x93,0xd1,0x94,0xd1,0x95,0xd1,0x96,0xd1,0x97,0xd1,0x98,0xd1,0x99,0xd1,
712 0x9a,0xd1,0x9b,0xd1,0x9c,0xd1,0x9e,0xd1,0x9f,0xe2,0x84,0x96
713};
714
715// DecodeUTF8
716// decodes the specified *unterminated* UTF-8 byte array
717wxWCharBuffer DecodeUTF8(
718 const void* data, // an unterminated UTF-8 encoded byte array
719 size_t size // the byte length of data
720 )
721{
722 // the decoder requires a null terminated buffer.
723 // the input data is not null terminated.
724 // copy to null terminated buffer
725
726 wxCharBuffer nullTerminated( size+1 );
727 memcpy( nullTerminated.data(), data, size );
728 nullTerminated.data()[size] = 0;
729 return wxConvUTF8.cMB2WC(nullTerminated.data());
730}
731
732// tests the encoding and decoding capability of an wxMBConv object
733//
734// decodes the utf-8 bytes into wide characters
735// encodes the wide characters to compare against input multiBuffer
736// decodes the multiBuffer to compare against wide characters
737// decodes the multiBuffer into wide characters
738void MBConvTestCase::TestCoder(
739 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
740 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
741 const char* utf8Buffer, // the same character sequence as multiBuffer, encoded as UTF-8
742 size_t utf8Bytes, // the byte length of the UTF-8 encoded character sequence
743 wxMBConv* converter, // the wxMBConv object thta can decode multiBuffer into a wide character sequence
744 int sizeofNull // the number of bytes occupied by a terminating null in the converter's encoding
745 )
746{
747 // wide character size and endian-ess varies from platform to platform
748 // compiler support for wide character literals varies from compiler to compiler
749 // so we should store the wide character version as UTF-8 and depend on
750 // the UTF-8 converter's ability to decode it to platform specific wide characters
751 // this test is invalid if the UTF-8 converter can't decode
752 wxWCharBuffer wideBuffer((size_t)0);
753 wideBuffer = DecodeUTF8( utf8Buffer, utf8Bytes );
754 size_t wideChars = wxWcslen( wideBuffer.data() );
755
756 TestDecoder
757 (
758 wideBuffer.data(),
759 wideChars,
760 (const char*)multiBuffer,
761 multiBytes,
762 converter,
763 sizeofNull
764 );
765 TestEncoder
766 (
767 wideBuffer.data(),
768 wideChars,
769 (const char*)multiBuffer,
770 multiBytes,
771 converter,
772 sizeofNull
773 );
774}
775
776
777WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const wxChar* name );
778
779void MBConvTestCase::FontmapTests()
780{
781#ifdef wxUSE_FONTMAP
782 wxMBConv* converter = new_wxMBConv_wxwin( _T("CP1252") );
783 if ( !converter )
784 {
785 return;
786 }
787 TestCoder(
788 (const char*)CP1252,
789 sizeof(CP1252),
790 (const char*)CP1252_utf8,
791 sizeof(CP1252_utf8),
792 converter,
793 1
794 );
795 delete converter;
796#endif
797}
798
799
800WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const wxChar* name );
801
802void MBConvTestCase::IconvTests()
803{
804#ifdef HAVE_ICONV
805 wxMBConv* converter = new_wxMBConv_iconv( _T("CP932") );
806 if ( !converter )
807 {
808 return;
809 }
810 TestCoder(
811 (const char*)welcome_cp932,
812 sizeof(welcome_cp932),
813 (const char*)welcome_utf8,
814 sizeof(welcome_utf8),
815 converter,
816 1
817 );
818 delete converter;
819#endif
820}
821
822void MBConvTestCase::CP1252Tests()
823{
824 wxCSConv convCP1252( wxFONTENCODING_CP1252 );
825 TestCoder(
826 (const char*)CP1252,
827 sizeof(CP1252),
828 (const char*)CP1252_utf8,
829 sizeof(CP1252_utf8),
830 &convCP1252,
831 1
832 );
833}
834
835void MBConvTestCase::LibcTests()
836{
837 // There isn't a locale that all systems support (except "C"), so leave
838 // this one disabled for non-Windows systems for the moment, until
839 // a solution can be found.
840#ifdef __WXMSW__
841
842#ifdef __WXMSW__
843 setlocale( LC_ALL, "English_United States.1252" );
844 const unsigned char* systemMB = CP1252;
845 size_t systemMB_size = sizeof(CP1252);
846 const unsigned char* systemMB_utf8 = CP1252_utf8;
847 size_t systemMB_utf8_size = sizeof(CP1252_utf8);
848#else
849 setlocale( LC_ALL, "en_US.iso8859-1" );
850 const unsigned char* systemMB = iso8859_1;
851 size_t systemMB_size = sizeof(iso8859_1);
852 const unsigned char* systemMB_utf8 = iso8859_1_utf8;
853 size_t systemMB_utf8_size = sizeof(iso8859_1_utf8);
854#endif
855 wxMBConvLibc convLibc;
856 TestCoder(
857 (const char*)systemMB,
858 systemMB_size,
859 (const char*)systemMB_utf8,
860 systemMB_utf8_size,
861 &convLibc,
862 1
863 );
864
865#endif // __WXMSW__
866}
867
868// verifies that the specified mb sequences decode to the specified wc sequence
869void MBConvTestCase::TestDecoder(
870 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
871 size_t wideChars, // the number of wide characters at wideBuffer
872 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
873 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
874 wxMBConv* converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
875 int sizeofNull // number of bytes occupied by terminating null in this encoding
876 )
877{
878 const unsigned UNINITIALIZED = 0xcd;
879
880 // copy the input bytes into a null terminated buffer
881 wxCharBuffer inputCopy( multiBytes+sizeofNull );
882 memcpy( inputCopy.data(), multiBuffer, multiBytes );
883 memset( &inputCopy.data()[multiBytes], 0, sizeofNull );
884
885 // calculate the output size
886 size_t outputWritten = converter->MB2WC
887 (
888 0,
889 (const char*)inputCopy.data(),
890 0
891 );
892 // make sure the correct output length was calculated
893 CPPUNIT_ASSERT( outputWritten == wideChars );
894
895 // convert the string
896 size_t guardChars = 8; // to make sure we're not overrunning the output buffer
897 size_t nullCharacters = 1;
898 size_t outputBufferChars = outputWritten + nullCharacters + guardChars;
899 wxWCharBuffer outputBuffer(outputBufferChars);
900 memset( outputBuffer.data(), UNINITIALIZED, outputBufferChars*sizeof(wchar_t) );
901
902 outputWritten = converter->MB2WC
903 (
904 outputBuffer.data(),
905 (const char*)inputCopy.data(),
906 outputBufferChars
907 );
908 // make sure the correct number of characters were outputs
909 CPPUNIT_ASSERT( outputWritten == wideChars );
910
911 // make sure the characters generated are correct
912 CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, wideBuffer, wideChars*sizeof(wchar_t) ) );
913
914 // the output buffer should be null terminated
915 CPPUNIT_ASSERT( outputBuffer[outputWritten] == 0 );
916
917 // make sure the rest of the output buffer is untouched
918 for ( size_t i = (wideChars+1)*sizeof(wchar_t); i < (outputBufferChars*sizeof(wchar_t)); i++ )
919 {
920 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
921 }
922
923#if wxUSE_UNICODE && wxUSE_STREAMS
924 TestStreamDecoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
925#endif
926}
927
928// verifies that the specified wc sequences encodes to the specified mb sequence
929void MBConvTestCase::TestEncoder(
930 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
931 size_t wideChars, // the number of wide characters at wideBuffer
932 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
933 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
934 wxMBConv* converter, // the wxMBConv object that can decode multiBuffer into a wide character sequence
935 int sizeofNull // number of bytes occupied by terminating null in this encoding
936 )
937{
938 const unsigned UNINITIALIZED = 0xcd;
939
940 // copy the input bytes into a null terminated buffer
941 wxWCharBuffer inputCopy( wideChars + 1 );
942 memcpy( inputCopy.data(), wideBuffer, (wideChars*sizeof(wchar_t)) );
943 inputCopy.data()[wideChars] = 0;
944
945 // calculate the output size
946 size_t outputWritten = converter->WC2MB
947 (
948 0,
949 (const wchar_t*)inputCopy.data(),
950 0
951 );
952 // make sure the correct output length was calculated
953 CPPUNIT_ASSERT( outputWritten == multiBytes );
954
955 // convert the string
956 size_t guardBytes = 8; // to make sure we're not overrunning the output buffer
957 size_t outputBufferSize = outputWritten + sizeofNull + guardBytes;
958 wxCharBuffer outputBuffer(outputBufferSize);
959 memset( outputBuffer.data(), UNINITIALIZED, outputBufferSize );
960
961 outputWritten = converter->WC2MB
962 (
963 outputBuffer.data(),
964 (const wchar_t*)inputCopy.data(),
965 outputBufferSize
966 );
967
968 // make sure the correct number of characters were output
969 CPPUNIT_ASSERT( outputWritten == multiBytes );
970
971 // make sure the characters generated are correct
972 CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, multiBuffer, multiBytes ) );
973
974 size_t i;
975
976 // the output buffer should be null terminated
977 for ( i = multiBytes; i < multiBytes + sizeofNull; i++ )
978 {
979 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == 0 );
980 }
981
982 // make sure the rest of the output buffer is untouched
983 for ( i = multiBytes + sizeofNull; i < outputBufferSize; i++ )
984 {
985 CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
986 }
987
988#if wxUSE_UNICODE && wxUSE_STREAMS
989 TestStreamEncoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
990#endif
991}
992
993#if wxUSE_UNICODE && wxUSE_STREAMS
994// use wxTextInputStream to exercise wxMBConv interface
995// (this reveals some bugs in certain wxMBConv subclasses)
996void MBConvTestCase::TestStreamDecoder(
997 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
998 size_t wideChars, // the number of wide characters at wideBuffer
999 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1000 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
1001 wxMBConv* converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
1002 )
1003{
1004 // this isn't meant to test wxMemoryInputStream or wxTextInputStream
1005 // it's meant to test the way wxTextInputStream uses wxMBConv
1006 // (which has exposed some problems with wxMBConv)
1007 wxMemoryInputStream memoryInputStream( multiBuffer, multiBytes );
1008 wxTextInputStream textInputStream( memoryInputStream, wxT(""), *converter );
1009 for ( size_t i = 0; i < wideChars; i++ )
1010 {
1011 wxChar wc = textInputStream.GetChar();
1012 CPPUNIT_ASSERT( wc == wideBuffer[i] );
1013 }
1014 CPPUNIT_ASSERT( 0 == textInputStream.GetChar() );
1015 CPPUNIT_ASSERT( memoryInputStream.Eof() );
1016}
1017#endif
1018
1019#if wxUSE_UNICODE && wxUSE_STREAMS
1020// use wxTextInputStream to exercise wxMBConv interface
1021// (this reveals some bugs in certain wxMBConv subclasses)
1022void MBConvTestCase::TestStreamEncoder(
1023 const wchar_t* wideBuffer, // the same character sequence as multiBuffer, encoded as wchar_t
1024 size_t wideChars, // the number of wide characters at wideBuffer
1025 const char* multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1026 size_t multiBytes, // the byte length of the multibyte character sequence that can be decoded by "converter"
1027 wxMBConv* converter // the wxMBConv object that can decode multiBuffer into a wide character sequence
1028 )
1029{
1030 // this isn't meant to test wxMemoryOutputStream or wxTextOutputStream
1031 // it's meant to test the way wxTextOutputStream uses wxMBConv
1032 // (which has exposed some problems with wxMBConv)
1033 wxMemoryOutputStream memoryOutputStream;
1034 // wxEOL_UNIX will pass \n \r unchanged
1035 wxTextOutputStream textOutputStream( memoryOutputStream, wxEOL_UNIX, *converter );
1036 for ( size_t i = 0; i < wideChars; i++ )
1037 {
1038 textOutputStream.PutChar( wideBuffer[i] );
1039 }
1040 CPPUNIT_ASSERT( memoryOutputStream.TellO() == multiBytes );
1041 wxCharBuffer copy( memoryOutputStream.TellO() );
1042 memoryOutputStream.CopyTo( copy.data(), memoryOutputStream.TellO());
1043 CPPUNIT_ASSERT( 0 == memcmp( copy.data(), multiBuffer, multiBytes ) );
1044}
1045#endif
1046
1047
1048// ----------------------------------------------------------------------------
1049// UTF-8 tests
1050// ----------------------------------------------------------------------------
1051
1052#ifdef HAVE_WCHAR_H
1053
1054// Check that 'charSequence' translates to 'wideSequence' and back.
1055// Invalid sequences can be tested by giving NULL for 'wideSequence'. Even
1056// invalid sequences should roundtrip when an option is given and this is
1057// checked.
1058//
1059void MBConvTestCase::UTF8(const char *charSequence,
1060 const wchar_t *wideSequence)
1061{
1062 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1063 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1064 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1065}
1066
1067// Use this alternative when 'charSequence' contains a PUA character. Such
1068// sequences should still roundtrip ok, and this is checked.
1069//
1070void MBConvTestCase::UTF8PUA(const char *charSequence,
1071 const wchar_t *wideSequence)
1072{
1073 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1074 UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1075 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1076}
1077
1078// Use this alternative when 'charSequence' contains an octal escape sequence.
1079// Such sequences should still roundtrip ok, and this is checked.
1080//
1081void MBConvTestCase::UTF8Octal(const char *charSequence,
1082 const wchar_t *wideSequence)
1083{
1084 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1085 UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1086 UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1087}
1088
1089// in case wcscpy is missing
1090//
1091static wchar_t *wx_wcscpy(wchar_t *dest, const wchar_t *src)
1092{
1093 wchar_t *d = dest;
1094 while ((*d++ = *src++) != 0)
1095 ;
1096 return dest;
1097}
1098
1099// in case wcscat is missing
1100//
1101static wchar_t *wx_wcscat(wchar_t *dest, const wchar_t *src)
1102{
1103 wchar_t *d = dest;
1104 while (*d)
1105 d++;
1106 while ((*d++ = *src++) != 0)
1107 ;
1108 return dest;
1109}
1110
1111// in case wcscmp is missing
1112//
1113static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
1114{
1115 while (*s1 == *s2 && *s1 != 0)
1116 {
1117 s1++;
1118 s2++;
1119 }
1120 return *s1 - *s2;
1121}
1122
1123// in case wcslen is missing
1124//
1125static size_t wx_wcslen(const wchar_t *s)
1126{
1127 const wchar_t *t = s;
1128 while (*t != 0)
1129 t++;
1130 return t - s;
1131}
1132
1133// include the option in the error messages so it's possible to see which
1134// test failed
1135#define UTF8ASSERT(expr) CPPUNIT_ASSERT_MESSAGE(#expr + errmsg, expr)
1136
1137// The test implementation
1138//
1139void MBConvTestCase::UTF8(const char *charSequence,
1140 const wchar_t *wideSequence,
1141 int option)
1142{
1143 const size_t BUFSIZE = 128;
1144 wxASSERT(strlen(charSequence) * 3 + 10 < BUFSIZE);
1145 char bytes[BUFSIZE];
1146
1147 // include the option in the error messages so it's possible to see
1148 // which test failed
1149 sprintf(bytes, " (with option == %d)", option);
1150 std::string errmsg(bytes);
1151
1152 // put the charSequence at the start, middle and end of a string
1153 strcpy(bytes, charSequence);
1154 strcat(bytes, "ABC");
1155 strcat(bytes, charSequence);
1156 strcat(bytes, "XYZ");
1157 strcat(bytes, charSequence);
1158
1159 // translate it into wide characters
1160 wxMBConvUTF8 utf8(option);
1161 wchar_t widechars[BUFSIZE];
1162 size_t lenResult = utf8.MB2WC(NULL, bytes, 0);
1163 size_t result = utf8.MB2WC(widechars, bytes, BUFSIZE);
1164 UTF8ASSERT(result == lenResult);
1165
1166 // check we got the expected result
1167 if (wideSequence) {
1168 UTF8ASSERT(result != (size_t)-1);
1169 wxASSERT(result < BUFSIZE);
1170
1171 wchar_t expected[BUFSIZE];
1172 wx_wcscpy(expected, wideSequence);
1173 wx_wcscat(expected, L"ABC");
1174 wx_wcscat(expected, wideSequence);
1175 wx_wcscat(expected, L"XYZ");
1176 wx_wcscat(expected, wideSequence);
1177
1178 UTF8ASSERT(wx_wcscmp(widechars, expected) == 0);
1179 UTF8ASSERT(wx_wcslen(widechars) == result);
1180 }
1181 else {
1182 // If 'wideSequence' is NULL, then the result is expected to be
1183 // invalid. Normally that is as far as we can go, but if there is an
1184 // option then the conversion should succeed anyway, and it should be
1185 // possible to translate back to the original
1186 if (!option) {
1187 UTF8ASSERT(result == (size_t)-1);
1188 return;
1189 }
1190 else {
1191 UTF8ASSERT(result != (size_t)-1);
1192 }
1193 }
1194
1195 // translate it back and check we get the original
1196 char bytesAgain[BUFSIZE];
1197 size_t lenResultAgain = utf8.WC2MB(NULL, widechars, 0);
1198 size_t resultAgain = utf8.WC2MB(bytesAgain, widechars, BUFSIZE);
1199 UTF8ASSERT(resultAgain == lenResultAgain);
1200 UTF8ASSERT(resultAgain != (size_t)-1);
1201 wxASSERT(resultAgain < BUFSIZE);
1202
1203 UTF8ASSERT(strcmp(bytes, bytesAgain) == 0);
1204 UTF8ASSERT(strlen(bytesAgain) == resultAgain);
1205}
1206
1207#endif // HAVE_WCHAR_H
wxWidgets (Impactor)