tests/mbconv/mbconvtest.cpp

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