tests/mbconv/mbconvtest.cpp

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