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( 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
 212 CPPUNIT_TEST_SUITE_REGISTRATION( MBConvTestCase );
 213
 214 // also include in it's own registry so that these tests can be run alone
 215 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( MBConvTestCase, "MBConv" );
 216
 217 void 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.
 245 // NB: Please don't remove this function even though it's not used anywhere,
 246 //     it's very useful when debugging a failed test.
 247 wxString CByteArrayFormat( const void* data, size_t len, const wxChar* name )
 248 {
 249     const unsigned char* bytes = (unsigned char*)data;
 250     wxString result;
 251
 252     result.Printf( _T("const static unsigned char %s[%i] = \n{"), name, (int)len );
 253
 254     for ( size_t i = 0; i < len; i++ )
 255     {
 256         if ( i != 0 )
 257         {
 258             result.append( _T(",") );
 259         }
 260         if ((i%16)==0)
 261         {
 262             result.append( _T("\n    ") );
 263         }
 264         wxString byte = wxString::Format( _T("0x%02x"), bytes[i] );
 265         result.append(byte);
 266     }
 267     result.append( _T("\n};\n") );
 268     return result;
 269 }
 270
 271 // The following bytes represent the same string, containing Japanese and English
 272 // characters, encoded in several different formats.
 273
 274 // encoded by iconv
 275 const static unsigned char welcome_utf7_iconv[84] =
 276 {
 277     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 278     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 279     0x2e,0x20,0x20,0x2b,0x4d,0x46,0x6b,0x77,0x55,0x49,0x74,0x6d,0x57,0x39,0x38,0x77,
 280     0x61,0x35,0x62,0x37,0x69,0x6e,0x45,0x77,0x6b,0x6a,0x42,0x5a,0x4d,0x49,0x73,0x77,
 281     0x65,0x7a,0x42,0x47,0x4d,0x45,0x77,0x77,0x52,0x44,0x42,0x45,0x4d,0x47,0x63,0x77,
 282     0x57,0x54,0x41,0x43
 283 };
 284 // encoded by wxWindows (iconv can decode this successfully)
 285 const static unsigned char welcome_utf7_wx[109] =
 286 {
 287     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x2b,0x41,0x43,0x41,0x2d,0x74,0x6f,0x2b,0x41,
 288     0x43,0x41,0x2d,0x6f,0x75,0x72,0x2b,0x41,0x43,0x41,0x2d,0x63,0x79,0x62,0x65,0x72,
 289     0x2b,0x41,0x43,0x41,0x2d,0x73,0x70,0x61,0x63,0x65,0x2b,0x41,0x43,0x41,0x2d,0x66,
 290     0x6f,0x72,0x63,0x65,0x2e,0x2b,0x41,0x43,0x41,0x41,0x49,0x44,0x42,0x5a,0x4d,0x46,
 291     0x43,0x4c,0x5a,0x6c,0x76,0x66,0x4d,0x47,0x75,0x57,0x2b,0x34,0x70,0x78,0x4d,0x4a,
 292     0x49,0x77,0x57,0x54,0x43,0x4c,0x4d,0x48,0x73,0x77,0x52,0x6a,0x42,0x4d,0x4d,0x45,
 293     0x51,0x77,0x52,0x44,0x42,0x6e,0x4d,0x46,0x6b,0x77,0x41,0x67,0x2d
 294 };
 295 // encoded by iconv
 296 const static unsigned char welcome_utf8[89] =
 297 {
 298     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 299     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 300     0x2e,0x20,0x20,0xe3,0x81,0x99,0xe3,0x81,0x90,0xe8,0xad,0xa6,0xe5,0xaf,0x9f,0xe3,
 301     0x81,0xab,0xe9,0x9b,0xbb,0xe8,0xa9,0xb1,0xe3,0x82,0x92,0xe3,0x81,0x99,0xe3,0x82,
 302     0x8b,0xe3,0x81,0xbb,0xe3,0x81,0x86,0xe3,0x81,0x8c,0xe3,0x81,0x84,0xe3,0x81,0x84,
 303     0xe3,0x81,0xa7,0xe3,0x81,0x99,0xe3,0x80,0x82
 304 };
 305 // encoded by iconv
 306 const static unsigned char welcome_utf16le[106] =
 307 {
 308     0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,0x00,
 309     0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,0x00,
 310     0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,0x00,
 311     0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,0x00,
 312     0x2e,0x00,0x20,0x00,0x20,0x00,0x59,0x30,0x50,0x30,0x66,0x8b,0xdf,0x5b,0x6b,0x30,
 313     0xfb,0x96,0x71,0x8a,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,0x30,
 314     0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02,0x30
 315 };
 316 // encoded by iconv
 317 const static unsigned char welcome_utf16be[106] =
 318 {
 319     0x00,0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,
 320     0x00,0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,
 321     0x00,0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,
 322     0x00,0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,
 323     0x00,0x2e,0x00,0x20,0x00,0x20,0x30,0x59,0x30,0x50,0x8b,0x66,0x5b,0xdf,0x30,0x6b,
 324     0x96,0xfb,0x8a,0x71,0x30,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,
 325     0x30,0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02
 326 };
 327 // encoded by iconv
 328 const static unsigned char welcome_utf32le[212] =
 329 {
 330     0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
 331     0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,
 332     0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,
 333     0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
 334     0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,0x00,0x00,0x00,
 335     0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,0x00,0x00,0x00,
 336     0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,0x00,0x00,0x00,
 337     0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,
 338     0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x59,0x30,0x00,0x00,
 339     0x50,0x30,0x00,0x00,0x66,0x8b,0x00,0x00,0xdf,0x5b,0x00,0x00,0x6b,0x30,0x00,0x00,
 340     0xfb,0x96,0x00,0x00,0x71,0x8a,0x00,0x00,0x92,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
 341     0x8b,0x30,0x00,0x00,0x7b,0x30,0x00,0x00,0x46,0x30,0x00,0x00,0x4c,0x30,0x00,0x00,
 342     0x44,0x30,0x00,0x00,0x44,0x30,0x00,0x00,0x67,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
 343     0x02,0x30,0x00,0x00
 344 };
 345 // encoded by iconv
 346 const static unsigned char welcome_utf32be[212] =
 347 {
 348     0x00,0x00,0x00,0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,
 349     0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,
 350     0x00,0x00,0x00,0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,
 351     0x00,0x00,0x00,0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,
 352     0x00,0x00,0x00,0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,
 353     0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,
 354     0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,
 355     0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,
 356     0x00,0x00,0x00,0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x30,0x59,
 357     0x00,0x00,0x30,0x50,0x00,0x00,0x8b,0x66,0x00,0x00,0x5b,0xdf,0x00,0x00,0x30,0x6b,
 358     0x00,0x00,0x96,0xfb,0x00,0x00,0x8a,0x71,0x00,0x00,0x30,0x92,0x00,0x00,0x30,0x59,
 359     0x00,0x00,0x30,0x8b,0x00,0x00,0x30,0x7b,0x00,0x00,0x30,0x46,0x00,0x00,0x30,0x4c,
 360     0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x67,0x00,0x00,0x30,0x59,
 361     0x00,0x00,0x30,0x02
 362 };
 363 // encoded by iconv
 364 const static unsigned char welcome_cp932[71] =
 365 {
 366     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 367     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 368     0x2e,0x20,0x20,0x82,0xb7,0x82,0xae,0x8c,0x78,0x8e,0x40,0x82,0xc9,0x93,0x64,0x98,
 369     0x62,0x82,0xf0,0x82,0xb7,0x82,0xe9,0x82,0xd9,0x82,0xa4,0x82,0xaa,0x82,0xa2,0x82,
 370     0xa2,0x82,0xc5,0x82,0xb7,0x81,0x42
 371 };
 372
 373 #if wxBYTE_ORDER == wxBIG_ENDIAN
 374     #if SIZEOF_WCHAR_T == 2
 375         #define welcome_wchar_t welcome_utf16be
 376     #elif SIZEOF_WCHAR_T == 4
 377         #define welcome_wchar_t welcome_utf32be
 378     #endif
 379 #elif wxBYTE_ORDER == wxLITTLE_ENDIAN
 380     #if SIZEOF_WCHAR_T == 2
 381         #define welcome_wchar_t welcome_utf16le
 382     #elif SIZEOF_WCHAR_T == 4
 383         #define welcome_wchar_t welcome_utf32le
 384     #endif
 385 #endif
 386
 387 void MBConvTestCase::UTF7Tests()
 388 {
 389     TestDecoder
 390         (
 391         (const wchar_t*)welcome_wchar_t,
 392         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 393         (const char*)welcome_utf7_iconv,
 394         sizeof(welcome_utf7_iconv),
 395         &wxConvUTF7,
 396         1
 397         );
 398     TestDecoder
 399         (
 400         (const wchar_t*)welcome_wchar_t,
 401         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 402         (const char*)welcome_utf7_wx,
 403         sizeof(welcome_utf7_wx),
 404         &wxConvUTF7,
 405         1
 406         );
 407 #if 0
 408     // wxWidget's UTF-7 encoder generates different byte sequences than iconv's.
 409     // but both seem to be equally legal.
 410     // This test won't work and that's okay.
 411     TestEncoder
 412         (
 413         (const wchar_t*)welcome_wchar_t,
 414         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 415         (const char*)welcome_utf7_iconv,
 416         sizeof(welcome_utf7_iconv),
 417         &wxConvUTF7,
 418         1
 419         );
 420 #endif
 421     TestEncoder
 422         (
 423         (const wchar_t*)welcome_wchar_t,
 424         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 425         (const char*)welcome_utf7_wx,
 426         sizeof(welcome_utf7_wx),
 427         &wxConvUTF7,
 428         1
 429         );
 430 }
 431
 432 void MBConvTestCase::UTF8Tests()
 433 {
 434     TestDecoder
 435         (
 436         (const wchar_t*)welcome_wchar_t,
 437         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 438         (const char*)welcome_utf8,
 439         sizeof(welcome_utf8),
 440         &wxConvUTF8,
 441         1
 442         );
 443     TestEncoder
 444         (
 445         (const wchar_t*)welcome_wchar_t,
 446         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 447         (const char*)welcome_utf8,
 448         sizeof(welcome_utf8),
 449         &wxConvUTF8,
 450         1
 451         );
 452 }
 453
 454 void MBConvTestCase::UTF16LETests()
 455 {
 456     wxMBConvUTF16LE convUTF16LE;
 457     TestDecoder
 458         (
 459         (const wchar_t*)welcome_wchar_t,
 460         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 461         (const char*)welcome_utf16le,
 462         sizeof(welcome_utf16le),
 463         &convUTF16LE,
 464         2
 465         );
 466     TestEncoder
 467         (
 468         (const wchar_t*)welcome_wchar_t,
 469         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 470         (const char*)welcome_utf16le,
 471         sizeof(welcome_utf16le),
 472         &convUTF16LE,
 473         2
 474         );
 475 }
 476
 477 void MBConvTestCase::UTF16BETests()
 478 {
 479     wxMBConvUTF16BE convUTF16BE;
 480     TestDecoder
 481         (
 482         (const wchar_t*)welcome_wchar_t,
 483         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 484         (const char*)welcome_utf16be,
 485         sizeof(welcome_utf16be),
 486         &convUTF16BE,
 487         2
 488         );
 489     TestEncoder
 490         (
 491         (const wchar_t*)welcome_wchar_t,
 492         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 493         (const char*)welcome_utf16be,
 494         sizeof(welcome_utf16be),
 495         &convUTF16BE,
 496         2
 497         );
 498 }
 499
 500 void MBConvTestCase::UTF32LETests()
 501 {
 502     wxMBConvUTF32LE convUTF32LE;
 503     TestDecoder
 504         (
 505         (const wchar_t*)welcome_wchar_t,
 506         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 507         (const char*)welcome_utf32le,
 508         sizeof(welcome_utf32le),
 509         &convUTF32LE,
 510         4
 511         );
 512     TestEncoder
 513         (
 514         (const wchar_t*)welcome_wchar_t,
 515         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 516         (const char*)welcome_utf32le,
 517         sizeof(welcome_utf32le),
 518         &convUTF32LE,
 519         4
 520         );
 521 }
 522
 523 void MBConvTestCase::UTF32BETests()
 524 {
 525     wxMBConvUTF32BE convUTF32BE;
 526     TestDecoder
 527         (
 528         (const wchar_t*)welcome_wchar_t,
 529         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 530         (const char*)welcome_utf32be,
 531         sizeof(welcome_utf32be),
 532         &convUTF32BE,
 533         4
 534         );
 535     TestEncoder
 536         (
 537         (const wchar_t*)welcome_wchar_t,
 538         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 539         (const char*)welcome_utf32be,
 540         sizeof(welcome_utf32be),
 541         &convUTF32BE,
 542         4
 543         );
 544 }
 545
 546 void MBConvTestCase::CP932Tests()
 547 {
 548     wxCSConv convCP932( wxFONTENCODING_CP932 );
 549     TestDecoder
 550         (
 551         (const wchar_t*)welcome_wchar_t,
 552         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 553         (const char*)welcome_cp932,
 554         sizeof(welcome_cp932),
 555         &convCP932,
 556         1
 557         );
 558     TestEncoder
 559         (
 560         (const wchar_t*)welcome_wchar_t,
 561         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 562         (const char*)welcome_cp932,
 563         sizeof(welcome_cp932),
 564         &convCP932,
 565         1
 566         );
 567 }
 568
 569 // a character sequence encoded as iso8859-1 (iconv)
 570 static const unsigned char iso8859_1[251] =
 571 {
 572     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 573     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 574     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 575     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 576     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 577     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 578     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 579     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
 580     0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
 581     0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xa1,0xa2,0xa3,0xa4,
 582     0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
 583     0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
 584     0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
 585     0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
 586     0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
 587     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
 588 };
 589 // the above character sequence encoded as UTF-8 (iconv)
 590 static const unsigned char iso8859_1_utf8[379] =
 591 {
 592     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 593     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 594     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 595     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 596     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 597     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 598     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 599     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
 600     0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
 601     0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
 602     0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
 603     0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa1,0xc2,
 604     0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
 605     0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
 606     0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
 607     0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
 608     0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
 609     0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
 610     0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
 611     0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
 612     0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
 613     0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
 614     0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
 615     0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf
 616 };
 617
 618 // a character sequence encoded as CP1252 (iconv)
 619 static const unsigned char CP1252[246] =
 620 {
 621     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 622     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 623     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 624     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 625     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 626     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 627     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 628     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xa0,0xa1,0xa2,0xa3,0xa4,
 629     0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
 630     0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
 631     0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
 632     0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
 633     0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
 634     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff,0x8c,0x9c,0x8a,0x9a,0x9f,
 635     0x8e,0x9e,0x83,0x88,0x98,0x96,0x97,0x91,0x92,0x82,0x93,0x94,0x84,0x86,0x87,0x95,
 636     0x85,0x89,0x8b,0x9b,0x80,0x99
 637 };
 638 // the above character sequence encoded as UTF-8 (iconv)
 639 static const unsigned char CP1252_utf8[386] =
 640 {
 641     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 642     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 643     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 644     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 645     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 646     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 647     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 648     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0xa0,0xc2,0xa1,0xc2,
 649     0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
 650     0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
 651     0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
 652     0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
 653     0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
 654     0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
 655     0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
 656     0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
 657     0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
 658     0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
 659     0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
 660     0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf,0xc5,0x92,0xc5,0x93,0xc5,
 661     0xa0,0xc5,0xa1,0xc5,0xb8,0xc5,0xbd,0xc5,0xbe,0xc6,0x92,0xcb,0x86,0xcb,0x9c,0xe2,
 662     0x80,0x93,0xe2,0x80,0x94,0xe2,0x80,0x98,0xe2,0x80,0x99,0xe2,0x80,0x9a,0xe2,0x80,
 663     0x9c,0xe2,0x80,0x9d,0xe2,0x80,0x9e,0xe2,0x80,0xa0,0xe2,0x80,0xa1,0xe2,0x80,0xa2,
 664     0xe2,0x80,0xa6,0xe2,0x80,0xb0,0xe2,0x80,0xb9,0xe2,0x80,0xba,0xe2,0x82,0xac,0xe2,
 665     0x84,0xa2
 666 };
 667
 668 // a character sequence encoded as iso8859-5 (iconv)
 669 static const unsigned char iso8859_5[251] =
 670 {
 671     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 672     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 673     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 674     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 675     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 676     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 677     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 678     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
 679     0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
 680     0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xfd,0xad,0xa1,0xa2,
 681     0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,
 682     0xb4,0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,
 683     0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,
 684     0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,
 685     0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf1,0xf2,0xf3,0xf4,
 686     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfe,0xff,0xf0
 687 };
 688 // the above character sequence encoded as UTF-8 (iconv)
 689 static const unsigned char iso8859_5_utf8[380] =
 690 {
 691     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 692     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 693     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 694     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 695     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 696     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 697     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 698     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
 699     0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
 700     0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
 701     0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
 702     0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa7,0xc2,
 703     0xad,0xd0,0x81,0xd0,0x82,0xd0,0x83,0xd0,0x84,0xd0,0x85,0xd0,0x86,0xd0,0x87,0xd0,
 704     0x88,0xd0,0x89,0xd0,0x8a,0xd0,0x8b,0xd0,0x8c,0xd0,0x8e,0xd0,0x8f,0xd0,0x90,0xd0,
 705     0x91,0xd0,0x92,0xd0,0x93,0xd0,0x94,0xd0,0x95,0xd0,0x96,0xd0,0x97,0xd0,0x98,0xd0,
 706     0x99,0xd0,0x9a,0xd0,0x9b,0xd0,0x9c,0xd0,0x9d,0xd0,0x9e,0xd0,0x9f,0xd0,0xa0,0xd0,
 707     0xa1,0xd0,0xa2,0xd0,0xa3,0xd0,0xa4,0xd0,0xa5,0xd0,0xa6,0xd0,0xa7,0xd0,0xa8,0xd0,
 708     0xa9,0xd0,0xaa,0xd0,0xab,0xd0,0xac,0xd0,0xad,0xd0,0xae,0xd0,0xaf,0xd0,0xb0,0xd0,
 709     0xb1,0xd0,0xb2,0xd0,0xb3,0xd0,0xb4,0xd0,0xb5,0xd0,0xb6,0xd0,0xb7,0xd0,0xb8,0xd0,
 710     0xb9,0xd0,0xba,0xd0,0xbb,0xd0,0xbc,0xd0,0xbd,0xd0,0xbe,0xd0,0xbf,0xd1,0x80,0xd1,
 711     0x81,0xd1,0x82,0xd1,0x83,0xd1,0x84,0xd1,0x85,0xd1,0x86,0xd1,0x87,0xd1,0x88,0xd1,
 712     0x89,0xd1,0x8a,0xd1,0x8b,0xd1,0x8c,0xd1,0x8d,0xd1,0x8e,0xd1,0x8f,0xd1,0x91,0xd1,
 713     0x92,0xd1,0x93,0xd1,0x94,0xd1,0x95,0xd1,0x96,0xd1,0x97,0xd1,0x98,0xd1,0x99,0xd1,
 714     0x9a,0xd1,0x9b,0xd1,0x9c,0xd1,0x9e,0xd1,0x9f,0xe2,0x84,0x96
 715 };
 716
 717 // DecodeUTF8
 718 // decodes the specified *unterminated* UTF-8 byte array
 719 wxWCharBuffer DecodeUTF8(
 720     const void* data, // an unterminated UTF-8 encoded byte array
 721     size_t size       // the byte length of data
 722     )
 723 {
 724     // the decoder requires a null terminated buffer.
 725     // the input data is not null terminated.
 726     // copy to null terminated buffer
 727
 728     wxCharBuffer nullTerminated( size+1 );
 729     memcpy( nullTerminated.data(), data, size );
 730     nullTerminated.data()[size] = 0;
 731     return wxConvUTF8.cMB2WC(nullTerminated.data());
 732 }
 733
 734 // tests the encoding and decoding capability of an wxMBConv object
 735 //
 736 // decodes the utf-8 bytes into wide characters
 737 // encodes the wide characters to compare against input multiBuffer
 738 // decodes the multiBuffer to compare against wide characters
 739 // decodes the multiBuffer into wide characters
 740 void MBConvTestCase::TestCoder(
 741     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 742     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 743     const char*    utf8Buffer,  // the same character sequence as multiBuffer, encoded as UTF-8
 744     size_t         utf8Bytes,   // the byte length of the UTF-8 encoded character sequence
 745     wxMBConv*      converter,   // the wxMBConv object thta can decode multiBuffer into a wide character sequence
 746     int            sizeofNull   // the number of bytes occupied by a terminating null in the converter's encoding
 747     )
 748 {
 749     // wide character size and endian-ess varies from platform to platform
 750     // compiler support for wide character literals varies from compiler to compiler
 751     // so we should store the wide character version as UTF-8 and depend on
 752     // the UTF-8 converter's ability to decode it to platform specific wide characters
 753     // this test is invalid if the UTF-8 converter can't decode
 754     wxWCharBuffer wideBuffer((size_t)0);
 755     wideBuffer = DecodeUTF8( utf8Buffer, utf8Bytes );
 756     size_t wideChars = wxWcslen( wideBuffer.data() );
 757
 758     TestDecoder
 759         (
 760         wideBuffer.data(),
 761         wideChars,
 762         (const char*)multiBuffer,
 763         multiBytes,
 764         converter,
 765         sizeofNull
 766         );
 767     TestEncoder
 768         (
 769         wideBuffer.data(),
 770         wideChars,
 771         (const char*)multiBuffer,
 772         multiBytes,
 773         converter,
 774         sizeofNull
 775         );
 776 }
 777
 778
 779 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const wxChar* name );
 780
 781 void MBConvTestCase::FontmapTests()
 782 {
 783 #ifdef wxUSE_FONTMAP
 784     wxMBConv* converter = new_wxMBConv_wxwin( _T("CP1252") );
 785     if ( !converter )
 786     {
 787         return;
 788     }
 789     TestCoder(
 790         (const char*)CP1252,
 791         sizeof(CP1252),
 792         (const char*)CP1252_utf8,
 793         sizeof(CP1252_utf8),
 794         converter,
 795         1
 796         );
 797     delete converter;
 798 #endif
 799 }
 800
 801
 802 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const wxChar* name );
 803
 804 void MBConvTestCase::IconvTests()
 805 {
 806 #ifdef HAVE_ICONV
 807     wxMBConv* converter = new_wxMBConv_iconv( _T("CP932") );
 808     if ( !converter )
 809     {
 810         return;
 811     }
 812     TestCoder(
 813         (const char*)welcome_cp932,
 814         sizeof(welcome_cp932),
 815         (const char*)welcome_utf8,
 816         sizeof(welcome_utf8),
 817         converter,
 818         1
 819         );
 820     delete converter;
 821 #endif
 822 }
 823
 824 void MBConvTestCase::CP1252Tests()
 825 {
 826     wxCSConv convCP1252( wxFONTENCODING_CP1252 );
 827     TestCoder(
 828         (const char*)CP1252,
 829         sizeof(CP1252),
 830         (const char*)CP1252_utf8,
 831         sizeof(CP1252_utf8),
 832         &convCP1252,
 833         1
 834         );
 835 }
 836
 837 void MBConvTestCase::LibcTests()
 838 {
 839     // There isn't a locale that all systems support (except "C"), so leave
 840     // this one disabled for non-Windows systems for the moment, until
 841     // a solution can be found.
 842 #ifdef __WXMSW__
 843
 844 #ifdef __WXMSW__
 845     setlocale( LC_ALL, "English_United States.1252" );
 846     const unsigned char* systemMB = CP1252;
 847     size_t systemMB_size = sizeof(CP1252);
 848     const unsigned char* systemMB_utf8 = CP1252_utf8;
 849     size_t systemMB_utf8_size = sizeof(CP1252_utf8);
 850 #else
 851     setlocale( LC_ALL, "en_US.iso8859-1" );
 852     const unsigned char* systemMB = iso8859_1;
 853     size_t systemMB_size = sizeof(iso8859_1);
 854     const unsigned char* systemMB_utf8 = iso8859_1_utf8;
 855     size_t systemMB_utf8_size = sizeof(iso8859_1_utf8);
 856 #endif
 857     wxMBConvLibc convLibc;
 858     TestCoder(
 859         (const char*)systemMB,
 860         systemMB_size,
 861         (const char*)systemMB_utf8,
 862         systemMB_utf8_size,
 863         &convLibc,
 864         1
 865         );
 866
 867 #endif // __WXMSW__
 868 }
 869
 870 // verifies that the specified mb sequences decode to the specified wc sequence
 871 void MBConvTestCase::TestDecoder(
 872     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 873     size_t         wideChars,   // the number of wide characters at wideBuffer
 874     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 875     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 876     wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 877     int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 878     )
 879 {
 880     const unsigned UNINITIALIZED = 0xcd;
 881
 882     // copy the input bytes into a null terminated buffer
 883     wxCharBuffer inputCopy( multiBytes+sizeofNull );
 884     memcpy( inputCopy.data(), multiBuffer, multiBytes );
 885     memset( &inputCopy.data()[multiBytes], 0, sizeofNull );
 886
 887     // calculate the output size
 888     size_t outputWritten = converter->MB2WC
 889         (
 890         0,
 891         (const char*)inputCopy.data(),
 892         0
 893         );
 894     // make sure the correct output length was calculated
 895     CPPUNIT_ASSERT( outputWritten == wideChars );
 896
 897     // convert the string
 898     size_t guardChars = 8; // to make sure we're not overrunning the output buffer
 899     size_t nullCharacters = 1;
 900     size_t outputBufferChars = outputWritten + nullCharacters + guardChars;
 901     wxWCharBuffer outputBuffer(outputBufferChars);
 902     memset( outputBuffer.data(), UNINITIALIZED, outputBufferChars*sizeof(wchar_t) );
 903
 904     outputWritten = converter->MB2WC
 905         (
 906         outputBuffer.data(),
 907         (const char*)inputCopy.data(),
 908         outputBufferChars
 909         );
 910     // make sure the correct number of characters were outputs
 911     CPPUNIT_ASSERT( outputWritten == wideChars );
 912
 913     // make sure the characters generated are correct
 914     CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, wideBuffer, wideChars*sizeof(wchar_t) ) );
 915
 916     // the output buffer should be null terminated
 917     CPPUNIT_ASSERT(  outputBuffer[outputWritten] == 0 );
 918
 919     // make sure the rest of the output buffer is untouched
 920     for ( size_t i = (wideChars+1)*sizeof(wchar_t); i < (outputBufferChars*sizeof(wchar_t)); i++ )
 921     {
 922         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
 923     }
 924
 925 #if wxUSE_UNICODE && wxUSE_STREAMS
 926     TestStreamDecoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
 927 #endif
 928 }
 929
 930 // verifies that the specified wc sequences encodes to the specified mb sequence
 931 void MBConvTestCase::TestEncoder(
 932     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 933     size_t         wideChars,   // the number of wide characters at wideBuffer
 934     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 935     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 936     wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 937     int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 938     )
 939 {
 940     const unsigned UNINITIALIZED = 0xcd;
 941
 942     // copy the input bytes into a null terminated buffer
 943     wxWCharBuffer inputCopy( wideChars + 1 );
 944     memcpy( inputCopy.data(), wideBuffer, (wideChars*sizeof(wchar_t)) );
 945     inputCopy.data()[wideChars] = 0;
 946
 947     // calculate the output size
 948     size_t outputWritten = converter->WC2MB
 949         (
 950         0,
 951         (const wchar_t*)inputCopy.data(),
 952         0
 953         );
 954     // make sure the correct output length was calculated
 955     CPPUNIT_ASSERT( outputWritten == multiBytes );
 956
 957     // convert the string
 958     size_t guardBytes = 8; // to make sure we're not overrunning the output buffer
 959     size_t outputBufferSize = outputWritten + sizeofNull + guardBytes;
 960     wxCharBuffer outputBuffer(outputBufferSize);
 961     memset( outputBuffer.data(), UNINITIALIZED, outputBufferSize );
 962
 963     outputWritten = converter->WC2MB
 964         (
 965         outputBuffer.data(),
 966         (const wchar_t*)inputCopy.data(),
 967         outputBufferSize
 968         );
 969
 970     // make sure the correct number of characters were output
 971     CPPUNIT_ASSERT( outputWritten == multiBytes );
 972
 973     // make sure the characters generated are correct
 974     CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, multiBuffer, multiBytes ) );
 975
 976     size_t i;
 977
 978     // the output buffer should be null terminated
 979     for ( i = multiBytes; i < multiBytes + sizeofNull; i++ )
 980     {
 981         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == 0 );
 982     }
 983
 984     // make sure the rest of the output buffer is untouched
 985     for ( i = multiBytes + sizeofNull; i < outputBufferSize; i++ )
 986     {
 987         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
 988     }
 989
 990 #if wxUSE_UNICODE && wxUSE_STREAMS
 991     TestStreamEncoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
 992 #endif
 993 }
 994
 995 #if wxUSE_UNICODE && wxUSE_STREAMS
 996 // use wxTextInputStream to exercise wxMBConv interface
 997 // (this reveals some bugs in certain wxMBConv subclasses)
 998 void MBConvTestCase::TestStreamDecoder(
 999                                        const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
1000                                        size_t         wideChars,   // the number of wide characters at wideBuffer
1001                                        const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1002                                        size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
1003                                        wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
1004                                        )
1005 {
1006     // this isn't meant to test wxMemoryInputStream or wxTextInputStream
1007     // it's meant to test the way wxTextInputStream uses wxMBConv
1008     // (which has exposed some problems with wxMBConv)
1009     wxMemoryInputStream memoryInputStream( multiBuffer, multiBytes );
1010     wxTextInputStream textInputStream( memoryInputStream, wxT(""), *converter );
1011     for ( size_t i = 0; i < wideChars; i++ )
1012     {
1013         wxChar wc = textInputStream.GetChar();
1014         CPPUNIT_ASSERT( wc == wideBuffer[i] );
1015     }
1016     CPPUNIT_ASSERT( 0 == textInputStream.GetChar() );
1017     CPPUNIT_ASSERT( memoryInputStream.Eof() );
1018 }
1019 #endif
1020
1021 #if wxUSE_UNICODE && wxUSE_STREAMS
1022 // use wxTextInputStream to exercise wxMBConv interface
1023 // (this reveals some bugs in certain wxMBConv subclasses)
1024 void MBConvTestCase::TestStreamEncoder(
1025     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
1026     size_t         wideChars,   // the number of wide characters at wideBuffer
1027     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1028     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
1029     wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
1030     )
1031 {
1032     // this isn't meant to test wxMemoryOutputStream or wxTextOutputStream
1033     // it's meant to test the way wxTextOutputStream uses wxMBConv
1034     // (which has exposed some problems with wxMBConv)
1035     wxMemoryOutputStream memoryOutputStream;
1036     // wxEOL_UNIX will pass \n \r unchanged
1037     wxTextOutputStream textOutputStream( memoryOutputStream, wxEOL_UNIX, *converter );
1038     for ( size_t i = 0; i < wideChars; i++ )
1039     {
1040         textOutputStream.PutChar( wideBuffer[i] );
1041     }
1042     CPPUNIT_ASSERT_EQUAL( (wxFileOffset)multiBytes, memoryOutputStream.TellO() );
1043     wxCharBuffer copy( memoryOutputStream.TellO() );
1044     memoryOutputStream.CopyTo( copy.data(), memoryOutputStream.TellO());
1045     CPPUNIT_ASSERT_EQUAL( 0, memcmp( copy.data(), multiBuffer, multiBytes ) );
1046 }
1047 #endif
1048
1049
1050 // ----------------------------------------------------------------------------
1051 // UTF-8 tests
1052 // ----------------------------------------------------------------------------
1053
1054 #ifdef HAVE_WCHAR_H
1055
1056 // Check that 'charSequence' translates to 'wideSequence' and back.
1057 // Invalid sequences can be tested by giving NULL for 'wideSequence'. Even
1058 // invalid sequences should roundtrip when an option is given and this is
1059 // checked.
1060 //
1061 void MBConvTestCase::UTF8(const char *charSequence,
1062                           const wchar_t *wideSequence)
1063 {
1064     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1065     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1066     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1067 }
1068
1069 // Use this alternative when 'charSequence' contains a PUA character. Such
1070 // sequences should still roundtrip ok, and this is checked.
1071 //
1072 void MBConvTestCase::UTF8PUA(const char *charSequence,
1073                              const wchar_t *wideSequence)
1074 {
1075     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1076     UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1077     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1078 }
1079
1080 // Use this alternative when 'charSequence' contains an octal escape sequence.
1081 // Such sequences should still roundtrip ok, and this is checked.
1082 //
1083 void MBConvTestCase::UTF8Octal(const char *charSequence,
1084                                const wchar_t *wideSequence)
1085 {
1086     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1087     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1088     UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1089 }
1090
1091 // in case wcscpy is missing
1092 //
1093 static wchar_t *wx_wcscpy(wchar_t *dest, const wchar_t *src)
1094 {
1095     wchar_t *d = dest;
1096     while ((*d++ = *src++) != 0)
1097         ;
1098     return dest;
1099 }
1100
1101 // in case wcscat is missing
1102 //
1103 static wchar_t *wx_wcscat(wchar_t *dest, const wchar_t *src)
1104 {
1105     wchar_t *d = dest;
1106     while (*d)
1107         d++;
1108     while ((*d++ = *src++) != 0)
1109         ;
1110     return dest;
1111 }
1112
1113 // in case wcscmp is missing
1114 //
1115 static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
1116 {
1117     while (*s1 == *s2 && *s1 != 0)
1118     {
1119         s1++;
1120         s2++;
1121     }
1122     return *s1 - *s2;
1123 }
1124
1125 // in case wcslen is missing
1126 //
1127 static size_t wx_wcslen(const wchar_t *s)
1128 {
1129     const wchar_t *t = s;
1130     while (*t != 0)
1131         t++;
1132     return t - s;
1133 }
1134
1135 // include the option in the error messages so it's possible to see which
1136 // test failed
1137 #define UTF8ASSERT(expr) CPPUNIT_ASSERT_MESSAGE(#expr + errmsg,  expr)
1138
1139 // The test implementation
1140 //
1141 void MBConvTestCase::UTF8(const char *charSequence,
1142                           const wchar_t *wideSequence,
1143                           int option)
1144 {
1145     const size_t BUFSIZE = 128;
1146     wxASSERT(strlen(charSequence) * 3 + 10 < BUFSIZE);
1147     char bytes[BUFSIZE];
1148
1149     // include the option in the error messages so it's possible to see
1150     // which test failed
1151     sprintf(bytes, " (with option == %d)", option);
1152     std::string errmsg(bytes);
1153
1154     // put the charSequence at the start, middle and end of a string
1155     strcpy(bytes, charSequence);
1156     strcat(bytes, "ABC");
1157     strcat(bytes, charSequence);
1158     strcat(bytes, "XYZ");
1159     strcat(bytes, charSequence);
1160
1161     // translate it into wide characters
1162     wxMBConvUTF8 utf8(option);
1163     wchar_t widechars[BUFSIZE];
1164     size_t lenResult = utf8.MB2WC(NULL, bytes, 0);
1165     size_t result = utf8.MB2WC(widechars, bytes, BUFSIZE);
1166     UTF8ASSERT(result == lenResult);
1167
1168     // check we got the expected result
1169     if (wideSequence) {
1170         UTF8ASSERT(result != (size_t)-1);
1171         wxASSERT(result < BUFSIZE);
1172
1173         wchar_t expected[BUFSIZE];
1174         wx_wcscpy(expected, wideSequence);
1175         wx_wcscat(expected, L"ABC");
1176         wx_wcscat(expected, wideSequence);
1177         wx_wcscat(expected, L"XYZ");
1178         wx_wcscat(expected, wideSequence);
1179
1180         UTF8ASSERT(wx_wcscmp(widechars, expected) == 0);
1181         UTF8ASSERT(wx_wcslen(widechars) == result);
1182     }
1183     else {
1184         // If 'wideSequence' is NULL, then the result is expected to be
1185         // invalid.  Normally that is as far as we can go, but if there is an
1186         // option then the conversion should succeed anyway, and it should be
1187         // possible to translate back to the original
1188         if (!option) {
1189             UTF8ASSERT(result == (size_t)-1);
1190             return;
1191         }
1192         else {
1193             UTF8ASSERT(result != (size_t)-1);
1194         }
1195     }
1196
1197     // translate it back and check we get the original
1198     char bytesAgain[BUFSIZE];
1199     size_t lenResultAgain = utf8.WC2MB(NULL, widechars, 0);
1200     size_t resultAgain = utf8.WC2MB(bytesAgain, widechars, BUFSIZE);
1201     UTF8ASSERT(resultAgain == lenResultAgain);
1202     UTF8ASSERT(resultAgain != (size_t)-1);
1203     wxASSERT(resultAgain < BUFSIZE);
1204
1205     UTF8ASSERT(strcmp(bytes, bytesAgain) == 0);
1206     UTF8ASSERT(strlen(bytesAgain) == resultAgain);
1207 }
1208
1209 #endif // HAVE_WCHAR_H