tests/mbconv/mbconvtest.cpp

   1 ///////////////////////////////////////////////////////////////////////////////
   2 // Name:        tests/mbconv/main.cpp
   3 // Purpose:     wxMBConv unit test
   4 // Author:      Vadim Zeitlin, Mike Wetherell, Vince Harron
   5 // Created:     14.02.04
   6 // RCS-ID:      $Id$
   7 // Copyright:   (c) 2003 TT-Solutions, (c) 2005 Mike Wetherell, Vince Harron
   8 ///////////////////////////////////////////////////////////////////////////////
   9
  10 // ----------------------------------------------------------------------------
  11 // headers
  12 // ----------------------------------------------------------------------------
  13
  14 #include "testprec.h"
  15
  16 #ifdef __BORLANDC__
  17     #pragma hdrstop
  18 #endif
  19
  20 #ifndef WX_PRECOMP
  21     #include "wx/wx.h"
  22 #endif // WX_PRECOMP
  23
  24 #include "wx/strconv.h"
  25 #include "wx/string.h"
  26 #include "wx/txtstrm.h"
  27 #include "wx/mstream.h"
  28
  29 #if defined wxHAVE_TCHAR_SUPPORT && !defined HAVE_WCHAR_H
  30     #define HAVE_WCHAR_H
  31 #endif
  32
  33 // ----------------------------------------------------------------------------
  34 // Some wide character constants. "\uXXXX" escapes aren't supported by old
  35 // compilers such as VC++ 5 and g++ 2.95.
  36 // ----------------------------------------------------------------------------
  37
  38 wchar_t u41[] = { 0x41, 0 };
  39 wchar_t u7f[] = { 0x7f, 0 };
  40
  41 wchar_t u80[] = { 0x80, 0 };
  42 wchar_t u391[] = { 0x391, 0 };
  43 wchar_t u7ff[] = { 0x7ff, 0 };
  44
  45 wchar_t u800[] = { 0x800, 0 };
  46 wchar_t u2620[] = { 0x2620, 0 };
  47 wchar_t ufffd[] = { 0xfffd, 0 };
  48
  49 #if SIZEOF_WCHAR_T == 4
  50 wchar_t u10000[] = { 0x10000, 0 };
  51 wchar_t u1000a5[] = { 0x1000a5, 0 };
  52 wchar_t u10fffd[] = { 0x10fffd, 0 };
  53 #else
  54 wchar_t u10000[] = { 0xd800, 0xdc00, 0 };
  55 wchar_t u1000a5[] = { 0xdbc0, 0xdca5, 0 };
  56 wchar_t u10fffd[] = { 0xdbff, 0xdffd, 0 };
  57 #endif
  58
  59 // ----------------------------------------------------------------------------
  60 // test class
  61 // ----------------------------------------------------------------------------
  62
  63 class MBConvTestCase : public CppUnit::TestCase
  64 {
  65 public:
  66     MBConvTestCase() { }
  67
  68 private:
  69     CPPUNIT_TEST_SUITE( MBConvTestCase );
  70         CPPUNIT_TEST( UTF32LETests );
  71         CPPUNIT_TEST( UTF32BETests );
  72         CPPUNIT_TEST( WC2CP1250 );
  73         CPPUNIT_TEST( UTF7Tests );
  74         CPPUNIT_TEST( UTF8Tests );
  75         CPPUNIT_TEST( UTF16LETests );
  76         CPPUNIT_TEST( UTF16BETests );
  77         CPPUNIT_TEST( CP932Tests );
  78         CPPUNIT_TEST( CP1252Tests ); // depends on UTF8 Decoder functioning correctly
  79         CPPUNIT_TEST( LibcTests );
  80         CPPUNIT_TEST( IconvTests );
  81         CPPUNIT_TEST( FontmapTests );
  82 #ifdef HAVE_WCHAR_H
  83         CPPUNIT_TEST( UTF8_41 );
  84         CPPUNIT_TEST( UTF8_7f );
  85         CPPUNIT_TEST( UTF8_80 );
  86         CPPUNIT_TEST( UTF8_c2_7f );
  87         CPPUNIT_TEST( UTF8_c2_80 );
  88         CPPUNIT_TEST( UTF8_ce_91 );
  89         CPPUNIT_TEST( UTF8_df_bf );
  90         CPPUNIT_TEST( UTF8_df_c0 );
  91         CPPUNIT_TEST( UTF8_e0_a0_7f );
  92         CPPUNIT_TEST( UTF8_e0_a0_80 );
  93         CPPUNIT_TEST( UTF8_e2_98_a0 );
  94         CPPUNIT_TEST( UTF8_ef_bf_bd );
  95         CPPUNIT_TEST( UTF8_ef_bf_c0 );
  96         CPPUNIT_TEST( UTF8_f0_90_80_7f );
  97         CPPUNIT_TEST( UTF8_f0_90_80_80 );
  98         CPPUNIT_TEST( UTF8_f4_8f_bf_bd );
  99         CPPUNIT_TEST( UTF8PUA_f4_80_82_a5 );
 100         CPPUNIT_TEST( UTF8Octal_backslash245 );
 101 #endif // HAVE_WCHAR_H
 102     CPPUNIT_TEST_SUITE_END();
 103
 104     void WC2CP1250();
 105     void UTF7Tests();
 106     void UTF8Tests();
 107     void UTF16LETests();
 108     void UTF16BETests();
 109     void UTF32LETests();
 110     void UTF32BETests();
 111     void CP932Tests();
 112     void CP1252Tests();
 113     void LibcTests();
 114     void FontmapTests();
 115     void IconvTests();
 116
 117     // verifies that the specified multibyte sequence decodes to the specified wchar_t sequence
 118     void TestDecoder(
 119         const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 120         size_t         wideChars,   // the number of wide characters at wideBuffer
 121         const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 122         size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 123         wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 124         int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 125         );
 126
 127     // verifies that the specified wchar_t sequence encodes to the specified multibyte sequence
 128     void TestEncoder(
 129         const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 130         size_t         wideChars,   // the number of wide characters at wideBuffer
 131         const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 132         size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 133         wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 134         int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 135         );
 136
 137 #if wxUSE_UNICODE && wxUSE_STREAMS
 138     // use wxTextInputStream to exercise wxMBConv interface
 139     // (this reveals some bugs in certain wxMBConv subclasses)
 140     void TestStreamDecoder(
 141         const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 142         size_t         wideChars,   // the number of wide characters at wideBuffer
 143         const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 144         size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 145         wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
 146         );
 147
 148     // use wxTextOutputStream to exercise wxMBConv interface
 149     // (this reveals some bugs in certain wxMBConv subclasses)
 150     void TestStreamEncoder(
 151         const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 152         size_t         wideChars,   // the number of wide characters at wideBuffer
 153         const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 154         size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 155         wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
 156         );
 157 #endif
 158
 159     // tests the encoding and decoding capability of an wxMBConv object
 160     //
 161     // decodes the utf-8 bytes into wide characters
 162     // encodes the wide characters to compare against input multiBuffer
 163     // decodes the multiBuffer to compare against wide characters
 164     // decodes the multiBuffer into wide characters
 165     void TestCoder(
 166         const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 167         size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 168         const char*    utf8Buffer,  // the same character sequence as multiBuffer, encoded as UTF-8
 169         size_t         utf8Bytes,   // the byte length of the UTF-8 encoded character sequence
 170         wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 171         int            sizeofNull   // the number of bytes occupied by a terminating null in the converter's encoding
 172         );
 173
 174 #ifdef HAVE_WCHAR_H
 175     // UTF-8 tests. Test the first, last and one in the middle for sequences
 176     // of each length
 177     void UTF8_41() { UTF8("\x41", u41); }
 178     void UTF8_7f() { UTF8("\x7f", u7f); }
 179     void UTF8_80() { UTF8("\x80", NULL); }
 180
 181     void UTF8_c2_7f() { UTF8("\xc2\x7f", NULL); }
 182     void UTF8_c2_80() { UTF8("\xc2\x80", u80); }
 183     void UTF8_ce_91() { UTF8("\xce\x91", u391); }
 184     void UTF8_df_bf() { UTF8("\xdf\xbf", u7ff); }
 185     void UTF8_df_c0() { UTF8("\xdf\xc0", NULL); }
 186
 187     void UTF8_e0_a0_7f() { UTF8("\xe0\xa0\x7f", NULL); }
 188     void UTF8_e0_a0_80() { UTF8("\xe0\xa0\x80", u800); }
 189     void UTF8_e2_98_a0() { UTF8("\xe2\x98\xa0", u2620); }
 190     void UTF8_ef_bf_bd() { UTF8("\xef\xbf\xbd", ufffd); }
 191     void UTF8_ef_bf_c0() { UTF8("\xef\xbf\xc0", NULL); }
 192
 193     void UTF8_f0_90_80_7f() { UTF8("\xf0\x90\x80\x7f", NULL); }
 194     void UTF8_f0_90_80_80() { UTF8("\xf0\x90\x80\x80", u10000); }
 195     void UTF8_f4_8f_bf_bd() { UTF8("\xf4\x8f\xbf\xbd", u10fffd); }
 196
 197     // test 'escaping the escape characters' for the two escaping schemes
 198     void UTF8PUA_f4_80_82_a5() { UTF8PUA("\xf4\x80\x82\xa5", u1000a5); }
 199     void UTF8Octal_backslash245() { UTF8Octal("\\245", L"\\245"); }
 200
 201     // implementation for the utf-8 tests (see comments below)
 202     void UTF8(const char *charSequence, const wchar_t *wideSequence);
 203     void UTF8PUA(const char *charSequence, const wchar_t *wideSequence);
 204     void UTF8Octal(const char *charSequence, const wchar_t *wideSequence);
 205     void UTF8(const char *charSequence, const wchar_t *wideSequence, int option);
 206 #endif // HAVE_WCHAR_H
 207
 208     DECLARE_NO_COPY_CLASS(MBConvTestCase)
 209 };
 210
 211 // register in the unnamed registry so that these tests are run by default
 212 CPPUNIT_TEST_SUITE_REGISTRATION( MBConvTestCase );
 213
 214 // also include in it's own registry so that these tests can be run alone
 215 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( MBConvTestCase, "MBConv" );
 216
 217 void MBConvTestCase::WC2CP1250()
 218 {
 219     static const struct Data
 220     {
 221         const wchar_t *wc;
 222         const char *cp1250;
 223     } data[] =
 224     {
 225         { L"hello", "hello" },  // test that it works in simplest case
 226         { L"\xBD of \xBD is \xBC", NULL }, // this should fail as cp1250 doesn't have 1/2
 227     };
 228
 229     wxCSConv cs1250(wxFONTENCODING_CP1250);
 230     for ( size_t n = 0; n < WXSIZEOF(data); n++ )
 231     {
 232         const Data& d = data[n];
 233         if (d.cp1250)
 234         {
 235             CPPUNIT_ASSERT( strcmp(cs1250.cWC2MB(d.wc), d.cp1250) == 0 );
 236         }
 237         else
 238         {
 239             CPPUNIT_ASSERT( (const char*)cs1250.cWC2MB(d.wc) == NULL );
 240         }
 241     }
 242 }
 243
 244 // The following bytes represent the same string, containing Japanese and English
 245 // characters, encoded in several different formats.
 246
 247 // encoded by iconv
 248 const static unsigned char welcome_utf7_iconv[84] =
 249 {
 250     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 251     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 252     0x2e,0x20,0x20,0x2b,0x4d,0x46,0x6b,0x77,0x55,0x49,0x74,0x6d,0x57,0x39,0x38,0x77,
 253     0x61,0x35,0x62,0x37,0x69,0x6e,0x45,0x77,0x6b,0x6a,0x42,0x5a,0x4d,0x49,0x73,0x77,
 254     0x65,0x7a,0x42,0x47,0x4d,0x45,0x77,0x77,0x52,0x44,0x42,0x45,0x4d,0x47,0x63,0x77,
 255     0x57,0x54,0x41,0x43
 256 };
 257 // encoded by wxWindows (iconv can decode this successfully)
 258 const static unsigned char welcome_utf7_wx[109] =
 259 {
 260     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x2b,0x41,0x43,0x41,0x2d,0x74,0x6f,0x2b,0x41,
 261     0x43,0x41,0x2d,0x6f,0x75,0x72,0x2b,0x41,0x43,0x41,0x2d,0x63,0x79,0x62,0x65,0x72,
 262     0x2b,0x41,0x43,0x41,0x2d,0x73,0x70,0x61,0x63,0x65,0x2b,0x41,0x43,0x41,0x2d,0x66,
 263     0x6f,0x72,0x63,0x65,0x2e,0x2b,0x41,0x43,0x41,0x41,0x49,0x44,0x42,0x5a,0x4d,0x46,
 264     0x43,0x4c,0x5a,0x6c,0x76,0x66,0x4d,0x47,0x75,0x57,0x2b,0x34,0x70,0x78,0x4d,0x4a,
 265     0x49,0x77,0x57,0x54,0x43,0x4c,0x4d,0x48,0x73,0x77,0x52,0x6a,0x42,0x4d,0x4d,0x45,
 266     0x51,0x77,0x52,0x44,0x42,0x6e,0x4d,0x46,0x6b,0x77,0x41,0x67,0x2d
 267 };
 268 // encoded by iconv
 269 const static unsigned char welcome_utf8[89] =
 270 {
 271     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 272     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 273     0x2e,0x20,0x20,0xe3,0x81,0x99,0xe3,0x81,0x90,0xe8,0xad,0xa6,0xe5,0xaf,0x9f,0xe3,
 274     0x81,0xab,0xe9,0x9b,0xbb,0xe8,0xa9,0xb1,0xe3,0x82,0x92,0xe3,0x81,0x99,0xe3,0x82,
 275     0x8b,0xe3,0x81,0xbb,0xe3,0x81,0x86,0xe3,0x81,0x8c,0xe3,0x81,0x84,0xe3,0x81,0x84,
 276     0xe3,0x81,0xa7,0xe3,0x81,0x99,0xe3,0x80,0x82
 277 };
 278 // encoded by iconv
 279 const static unsigned char welcome_utf16le[106] =
 280 {
 281     0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,0x00,
 282     0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,0x00,
 283     0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,0x00,
 284     0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,0x00,
 285     0x2e,0x00,0x20,0x00,0x20,0x00,0x59,0x30,0x50,0x30,0x66,0x8b,0xdf,0x5b,0x6b,0x30,
 286     0xfb,0x96,0x71,0x8a,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,0x30,
 287     0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02,0x30
 288 };
 289 // encoded by iconv
 290 const static unsigned char welcome_utf16be[106] =
 291 {
 292     0x00,0x57,0x00,0x65,0x00,0x6c,0x00,0x63,0x00,0x6f,0x00,0x6d,0x00,0x65,0x00,0x20,
 293     0x00,0x74,0x00,0x6f,0x00,0x20,0x00,0x6f,0x00,0x75,0x00,0x72,0x00,0x20,0x00,0x63,
 294     0x00,0x79,0x00,0x62,0x00,0x65,0x00,0x72,0x00,0x20,0x00,0x73,0x00,0x70,0x00,0x61,
 295     0x00,0x63,0x00,0x65,0x00,0x20,0x00,0x66,0x00,0x6f,0x00,0x72,0x00,0x63,0x00,0x65,
 296     0x00,0x2e,0x00,0x20,0x00,0x20,0x30,0x59,0x30,0x50,0x8b,0x66,0x5b,0xdf,0x30,0x6b,
 297     0x96,0xfb,0x8a,0x71,0x30,0x92,0x30,0x59,0x30,0x8b,0x30,0x7b,0x30,0x46,0x30,0x4c,
 298     0x30,0x44,0x30,0x44,0x30,0x67,0x30,0x59,0x30,0x02
 299 };
 300 // encoded by iconv
 301 const static unsigned char welcome_utf32le[212] =
 302 {
 303     0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
 304     0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,
 305     0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,
 306     0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,0x00,0x00,0x00,
 307     0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,0x00,0x00,0x00,
 308     0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,0x00,0x00,0x00,
 309     0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,0x00,0x00,0x00,
 310     0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,
 311     0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x59,0x30,0x00,0x00,
 312     0x50,0x30,0x00,0x00,0x66,0x8b,0x00,0x00,0xdf,0x5b,0x00,0x00,0x6b,0x30,0x00,0x00,
 313     0xfb,0x96,0x00,0x00,0x71,0x8a,0x00,0x00,0x92,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
 314     0x8b,0x30,0x00,0x00,0x7b,0x30,0x00,0x00,0x46,0x30,0x00,0x00,0x4c,0x30,0x00,0x00,
 315     0x44,0x30,0x00,0x00,0x44,0x30,0x00,0x00,0x67,0x30,0x00,0x00,0x59,0x30,0x00,0x00,
 316     0x02,0x30,0x00,0x00
 317 };
 318 // encoded by iconv
 319 const static unsigned char welcome_utf32be[212] =
 320 {
 321     0x00,0x00,0x00,0x57,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x6c,0x00,0x00,0x00,0x63,
 322     0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x6d,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,
 323     0x00,0x00,0x00,0x74,0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x6f,
 324     0x00,0x00,0x00,0x75,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x63,
 325     0x00,0x00,0x00,0x79,0x00,0x00,0x00,0x62,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x72,
 326     0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x73,0x00,0x00,0x00,0x70,0x00,0x00,0x00,0x61,
 327     0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x66,
 328     0x00,0x00,0x00,0x6f,0x00,0x00,0x00,0x72,0x00,0x00,0x00,0x63,0x00,0x00,0x00,0x65,
 329     0x00,0x00,0x00,0x2e,0x00,0x00,0x00,0x20,0x00,0x00,0x00,0x20,0x00,0x00,0x30,0x59,
 330     0x00,0x00,0x30,0x50,0x00,0x00,0x8b,0x66,0x00,0x00,0x5b,0xdf,0x00,0x00,0x30,0x6b,
 331     0x00,0x00,0x96,0xfb,0x00,0x00,0x8a,0x71,0x00,0x00,0x30,0x92,0x00,0x00,0x30,0x59,
 332     0x00,0x00,0x30,0x8b,0x00,0x00,0x30,0x7b,0x00,0x00,0x30,0x46,0x00,0x00,0x30,0x4c,
 333     0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x44,0x00,0x00,0x30,0x67,0x00,0x00,0x30,0x59,
 334     0x00,0x00,0x30,0x02
 335 };
 336 // encoded by iconv
 337 const static unsigned char welcome_cp932[71] =
 338 {
 339     0x57,0x65,0x6c,0x63,0x6f,0x6d,0x65,0x20,0x74,0x6f,0x20,0x6f,0x75,0x72,0x20,0x63,
 340     0x79,0x62,0x65,0x72,0x20,0x73,0x70,0x61,0x63,0x65,0x20,0x66,0x6f,0x72,0x63,0x65,
 341     0x2e,0x20,0x20,0x82,0xb7,0x82,0xae,0x8c,0x78,0x8e,0x40,0x82,0xc9,0x93,0x64,0x98,
 342     0x62,0x82,0xf0,0x82,0xb7,0x82,0xe9,0x82,0xd9,0x82,0xa4,0x82,0xaa,0x82,0xa2,0x82,
 343     0xa2,0x82,0xc5,0x82,0xb7,0x81,0x42
 344 };
 345
 346 #if wxBYTE_ORDER == wxBIG_ENDIAN
 347     #if SIZEOF_WCHAR_T == 2
 348         #define welcome_wchar_t welcome_utf16be
 349     #elif SIZEOF_WCHAR_T == 4
 350         #define welcome_wchar_t welcome_utf32be
 351     #endif
 352 #elif wxBYTE_ORDER == wxLITTLE_ENDIAN
 353     #if SIZEOF_WCHAR_T == 2
 354         #define welcome_wchar_t welcome_utf16le
 355     #elif SIZEOF_WCHAR_T == 4
 356         #define welcome_wchar_t welcome_utf32le
 357     #endif
 358 #endif
 359
 360 void MBConvTestCase::UTF7Tests()
 361 {
 362     TestDecoder
 363         (
 364         (const wchar_t*)welcome_wchar_t,
 365         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 366         (const char*)welcome_utf7_iconv,
 367         sizeof(welcome_utf7_iconv),
 368         &wxConvUTF7,
 369         1
 370         );
 371     TestDecoder
 372         (
 373         (const wchar_t*)welcome_wchar_t,
 374         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 375         (const char*)welcome_utf7_wx,
 376         sizeof(welcome_utf7_wx),
 377         &wxConvUTF7,
 378         1
 379         );
 380 #if 0
 381     // wxWidget's UTF-7 encoder generates different byte sequences than iconv's.
 382     // but both seem to be equally legal.
 383     // This test won't work and that's okay.
 384     TestEncoder
 385         (
 386         (const wchar_t*)welcome_wchar_t,
 387         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 388         (const char*)welcome_utf7_iconv,
 389         sizeof(welcome_utf7_iconv),
 390         &wxConvUTF7,
 391         1
 392         );
 393 #endif
 394     TestEncoder
 395         (
 396         (const wchar_t*)welcome_wchar_t,
 397         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 398         (const char*)welcome_utf7_wx,
 399         sizeof(welcome_utf7_wx),
 400         &wxConvUTF7,
 401         1
 402         );
 403 }
 404
 405 void MBConvTestCase::UTF8Tests()
 406 {
 407     TestDecoder
 408         (
 409         (const wchar_t*)welcome_wchar_t,
 410         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 411         (const char*)welcome_utf8,
 412         sizeof(welcome_utf8),
 413         &wxConvUTF8,
 414         1
 415         );
 416     TestEncoder
 417         (
 418         (const wchar_t*)welcome_wchar_t,
 419         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 420         (const char*)welcome_utf8,
 421         sizeof(welcome_utf8),
 422         &wxConvUTF8,
 423         1
 424         );
 425 }
 426
 427 void MBConvTestCase::UTF16LETests()
 428 {
 429     wxMBConvUTF16LE convUTF16LE;
 430     TestDecoder
 431         (
 432         (const wchar_t*)welcome_wchar_t,
 433         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 434         (const char*)welcome_utf16le,
 435         sizeof(welcome_utf16le),
 436         &convUTF16LE,
 437         2
 438         );
 439     TestEncoder
 440         (
 441         (const wchar_t*)welcome_wchar_t,
 442         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 443         (const char*)welcome_utf16le,
 444         sizeof(welcome_utf16le),
 445         &convUTF16LE,
 446         2
 447         );
 448 }
 449
 450 void MBConvTestCase::UTF16BETests()
 451 {
 452     wxMBConvUTF16BE convUTF16BE;
 453     TestDecoder
 454         (
 455         (const wchar_t*)welcome_wchar_t,
 456         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 457         (const char*)welcome_utf16be,
 458         sizeof(welcome_utf16be),
 459         &convUTF16BE,
 460         2
 461         );
 462     TestEncoder
 463         (
 464         (const wchar_t*)welcome_wchar_t,
 465         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 466         (const char*)welcome_utf16be,
 467         sizeof(welcome_utf16be),
 468         &convUTF16BE,
 469         2
 470         );
 471 }
 472
 473 void MBConvTestCase::UTF32LETests()
 474 {
 475     wxMBConvUTF32LE convUTF32LE;
 476     TestDecoder
 477         (
 478         (const wchar_t*)welcome_wchar_t,
 479         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 480         (const char*)welcome_utf32le,
 481         sizeof(welcome_utf32le),
 482         &convUTF32LE,
 483         4
 484         );
 485     TestEncoder
 486         (
 487         (const wchar_t*)welcome_wchar_t,
 488         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 489         (const char*)welcome_utf32le,
 490         sizeof(welcome_utf32le),
 491         &convUTF32LE,
 492         4
 493         );
 494 }
 495
 496 void MBConvTestCase::UTF32BETests()
 497 {
 498     wxMBConvUTF32BE convUTF32BE;
 499     TestDecoder
 500         (
 501         (const wchar_t*)welcome_wchar_t,
 502         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 503         (const char*)welcome_utf32be,
 504         sizeof(welcome_utf32be),
 505         &convUTF32BE,
 506         4
 507         );
 508     TestEncoder
 509         (
 510         (const wchar_t*)welcome_wchar_t,
 511         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 512         (const char*)welcome_utf32be,
 513         sizeof(welcome_utf32be),
 514         &convUTF32BE,
 515         4
 516         );
 517 }
 518
 519 void MBConvTestCase::CP932Tests()
 520 {
 521     wxCSConv convCP932( wxFONTENCODING_CP932 );
 522     TestDecoder
 523         (
 524         (const wchar_t*)welcome_wchar_t,
 525         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 526         (const char*)welcome_cp932,
 527         sizeof(welcome_cp932),
 528         &convCP932,
 529         1
 530         );
 531     TestEncoder
 532         (
 533         (const wchar_t*)welcome_wchar_t,
 534         sizeof(welcome_wchar_t)/sizeof(wchar_t),
 535         (const char*)welcome_cp932,
 536         sizeof(welcome_cp932),
 537         &convCP932,
 538         1
 539         );
 540 }
 541
 542 // a character sequence encoded as iso8859-1 (iconv)
 543 static const unsigned char iso8859_1[251] =
 544 {
 545     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 546     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 547     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 548     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 549     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 550     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 551     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 552     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
 553     0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
 554     0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xa1,0xa2,0xa3,0xa4,
 555     0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
 556     0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
 557     0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
 558     0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
 559     0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
 560     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
 561 };
 562 // the above character sequence encoded as UTF-8 (iconv)
 563 static const unsigned char iso8859_1_utf8[379] =
 564 {
 565     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 566     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 567     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 568     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 569     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 570     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 571     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 572     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
 573     0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
 574     0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
 575     0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
 576     0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa1,0xc2,
 577     0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
 578     0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
 579     0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
 580     0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
 581     0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
 582     0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
 583     0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
 584     0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
 585     0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
 586     0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
 587     0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
 588     0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf
 589 };
 590
 591 // a character sequence encoded as CP1252 (iconv)
 592 static const unsigned char CP1252[246] =
 593 {
 594     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 595     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 596     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 597     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 598     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 599     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 600     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 601     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xa0,0xa1,0xa2,0xa3,0xa4,
 602     0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,0xb4,
 603     0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,0xc4,
 604     0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,0xd4,
 605     0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,0xe4,
 606     0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf0,0xf1,0xf2,0xf3,0xf4,
 607     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff,0x8c,0x9c,0x8a,0x9a,0x9f,
 608     0x8e,0x9e,0x83,0x88,0x98,0x96,0x97,0x91,0x92,0x82,0x93,0x94,0x84,0x86,0x87,0x95,
 609     0x85,0x89,0x8b,0x9b,0x80,0x99
 610 };
 611 // the above character sequence encoded as UTF-8 (iconv)
 612 static const unsigned char CP1252_utf8[386] =
 613 {
 614     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 615     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 616     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 617     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 618     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 619     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 620     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 621     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0xa0,0xc2,0xa1,0xc2,
 622     0xa2,0xc2,0xa3,0xc2,0xa4,0xc2,0xa5,0xc2,0xa6,0xc2,0xa7,0xc2,0xa8,0xc2,0xa9,0xc2,
 623     0xaa,0xc2,0xab,0xc2,0xac,0xc2,0xad,0xc2,0xae,0xc2,0xaf,0xc2,0xb0,0xc2,0xb1,0xc2,
 624     0xb2,0xc2,0xb3,0xc2,0xb4,0xc2,0xb5,0xc2,0xb6,0xc2,0xb7,0xc2,0xb8,0xc2,0xb9,0xc2,
 625     0xba,0xc2,0xbb,0xc2,0xbc,0xc2,0xbd,0xc2,0xbe,0xc2,0xbf,0xc3,0x80,0xc3,0x81,0xc3,
 626     0x82,0xc3,0x83,0xc3,0x84,0xc3,0x85,0xc3,0x86,0xc3,0x87,0xc3,0x88,0xc3,0x89,0xc3,
 627     0x8a,0xc3,0x8b,0xc3,0x8c,0xc3,0x8d,0xc3,0x8e,0xc3,0x8f,0xc3,0x90,0xc3,0x91,0xc3,
 628     0x92,0xc3,0x93,0xc3,0x94,0xc3,0x95,0xc3,0x96,0xc3,0x97,0xc3,0x98,0xc3,0x99,0xc3,
 629     0x9a,0xc3,0x9b,0xc3,0x9c,0xc3,0x9d,0xc3,0x9e,0xc3,0x9f,0xc3,0xa0,0xc3,0xa1,0xc3,
 630     0xa2,0xc3,0xa3,0xc3,0xa4,0xc3,0xa5,0xc3,0xa6,0xc3,0xa7,0xc3,0xa8,0xc3,0xa9,0xc3,
 631     0xaa,0xc3,0xab,0xc3,0xac,0xc3,0xad,0xc3,0xae,0xc3,0xaf,0xc3,0xb0,0xc3,0xb1,0xc3,
 632     0xb2,0xc3,0xb3,0xc3,0xb4,0xc3,0xb5,0xc3,0xb6,0xc3,0xb7,0xc3,0xb8,0xc3,0xb9,0xc3,
 633     0xba,0xc3,0xbb,0xc3,0xbc,0xc3,0xbd,0xc3,0xbe,0xc3,0xbf,0xc5,0x92,0xc5,0x93,0xc5,
 634     0xa0,0xc5,0xa1,0xc5,0xb8,0xc5,0xbd,0xc5,0xbe,0xc6,0x92,0xcb,0x86,0xcb,0x9c,0xe2,
 635     0x80,0x93,0xe2,0x80,0x94,0xe2,0x80,0x98,0xe2,0x80,0x99,0xe2,0x80,0x9a,0xe2,0x80,
 636     0x9c,0xe2,0x80,0x9d,0xe2,0x80,0x9e,0xe2,0x80,0xa0,0xe2,0x80,0xa1,0xe2,0x80,0xa2,
 637     0xe2,0x80,0xa6,0xe2,0x80,0xb0,0xe2,0x80,0xb9,0xe2,0x80,0xba,0xe2,0x82,0xac,0xe2,
 638     0x84,0xa2
 639 };
 640
 641 // a character sequence encoded as iso8859-5 (iconv)
 642 static const unsigned char iso8859_5[251] =
 643 {
 644     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 645     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 646     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 647     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 648     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 649     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 650     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 651     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0x80,0x81,0x82,0x83,0x84,
 652     0x85,0x86,0x87,0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f,0x90,0x91,0x92,0x93,0x94,
 653     0x95,0x96,0x97,0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f,0xa0,0xfd,0xad,0xa1,0xa2,
 654     0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xab,0xac,0xae,0xaf,0xb0,0xb1,0xb2,0xb3,
 655     0xb4,0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf,0xc0,0xc1,0xc2,0xc3,
 656     0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xcb,0xcc,0xcd,0xce,0xcf,0xd0,0xd1,0xd2,0xd3,
 657     0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xdb,0xdc,0xdd,0xde,0xdf,0xe0,0xe1,0xe2,0xe3,
 658     0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xeb,0xec,0xed,0xee,0xef,0xf1,0xf2,0xf3,0xf4,
 659     0xf5,0xf6,0xf7,0xf8,0xf9,0xfa,0xfb,0xfc,0xfe,0xff,0xf0
 660 };
 661 // the above character sequence encoded as UTF-8 (iconv)
 662 static const unsigned char iso8859_5_utf8[380] =
 663 {
 664     0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0x10,0x11,0x12,0x13,0x14,
 665     0x15,0x16,0x17,0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,
 666     0x25,0x26,0x27,0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x34,
 667     0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f,0x40,0x41,0x42,0x43,0x44,
 668     0x45,0x46,0x47,0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f,0x50,0x51,0x52,0x53,0x54,
 669     0x55,0x56,0x57,0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f,0x60,0x61,0x62,0x63,0x64,
 670     0x65,0x66,0x67,0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f,0x70,0x71,0x72,0x73,0x74,
 671     0x75,0x76,0x77,0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f,0xc2,0x80,0xc2,0x81,0xc2,
 672     0x82,0xc2,0x83,0xc2,0x84,0xc2,0x85,0xc2,0x86,0xc2,0x87,0xc2,0x88,0xc2,0x89,0xc2,
 673     0x8a,0xc2,0x8b,0xc2,0x8c,0xc2,0x8d,0xc2,0x8e,0xc2,0x8f,0xc2,0x90,0xc2,0x91,0xc2,
 674     0x92,0xc2,0x93,0xc2,0x94,0xc2,0x95,0xc2,0x96,0xc2,0x97,0xc2,0x98,0xc2,0x99,0xc2,
 675     0x9a,0xc2,0x9b,0xc2,0x9c,0xc2,0x9d,0xc2,0x9e,0xc2,0x9f,0xc2,0xa0,0xc2,0xa7,0xc2,
 676     0xad,0xd0,0x81,0xd0,0x82,0xd0,0x83,0xd0,0x84,0xd0,0x85,0xd0,0x86,0xd0,0x87,0xd0,
 677     0x88,0xd0,0x89,0xd0,0x8a,0xd0,0x8b,0xd0,0x8c,0xd0,0x8e,0xd0,0x8f,0xd0,0x90,0xd0,
 678     0x91,0xd0,0x92,0xd0,0x93,0xd0,0x94,0xd0,0x95,0xd0,0x96,0xd0,0x97,0xd0,0x98,0xd0,
 679     0x99,0xd0,0x9a,0xd0,0x9b,0xd0,0x9c,0xd0,0x9d,0xd0,0x9e,0xd0,0x9f,0xd0,0xa0,0xd0,
 680     0xa1,0xd0,0xa2,0xd0,0xa3,0xd0,0xa4,0xd0,0xa5,0xd0,0xa6,0xd0,0xa7,0xd0,0xa8,0xd0,
 681     0xa9,0xd0,0xaa,0xd0,0xab,0xd0,0xac,0xd0,0xad,0xd0,0xae,0xd0,0xaf,0xd0,0xb0,0xd0,
 682     0xb1,0xd0,0xb2,0xd0,0xb3,0xd0,0xb4,0xd0,0xb5,0xd0,0xb6,0xd0,0xb7,0xd0,0xb8,0xd0,
 683     0xb9,0xd0,0xba,0xd0,0xbb,0xd0,0xbc,0xd0,0xbd,0xd0,0xbe,0xd0,0xbf,0xd1,0x80,0xd1,
 684     0x81,0xd1,0x82,0xd1,0x83,0xd1,0x84,0xd1,0x85,0xd1,0x86,0xd1,0x87,0xd1,0x88,0xd1,
 685     0x89,0xd1,0x8a,0xd1,0x8b,0xd1,0x8c,0xd1,0x8d,0xd1,0x8e,0xd1,0x8f,0xd1,0x91,0xd1,
 686     0x92,0xd1,0x93,0xd1,0x94,0xd1,0x95,0xd1,0x96,0xd1,0x97,0xd1,0x98,0xd1,0x99,0xd1,
 687     0x9a,0xd1,0x9b,0xd1,0x9c,0xd1,0x9e,0xd1,0x9f,0xe2,0x84,0x96
 688 };
 689
 690 // DecodeUTF8
 691 // decodes the specified *unterminated* UTF-8 byte array
 692 wxWCharBuffer DecodeUTF8(
 693     const void* data, // an unterminated UTF-8 encoded byte array
 694     size_t size       // the byte length of data
 695     )
 696 {
 697     // the decoder requires a null terminated buffer.
 698     // the input data is not null terminated.
 699     // copy to null terminated buffer
 700
 701     wxCharBuffer nullTerminated( size+1 );
 702     memcpy( nullTerminated.data(), data, size );
 703     nullTerminated.data()[size] = 0;
 704     return wxConvUTF8.cMB2WC(nullTerminated.data());
 705 }
 706
 707 // tests the encoding and decoding capability of an wxMBConv object
 708 //
 709 // decodes the utf-8 bytes into wide characters
 710 // encodes the wide characters to compare against input multiBuffer
 711 // decodes the multiBuffer to compare against wide characters
 712 // decodes the multiBuffer into wide characters
 713 void MBConvTestCase::TestCoder(
 714     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 715     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 716     const char*    utf8Buffer,  // the same character sequence as multiBuffer, encoded as UTF-8
 717     size_t         utf8Bytes,   // the byte length of the UTF-8 encoded character sequence
 718     wxMBConv*      converter,   // the wxMBConv object thta can decode multiBuffer into a wide character sequence
 719     int            sizeofNull   // the number of bytes occupied by a terminating null in the converter's encoding
 720     )
 721 {
 722     // wide character size and endian-ess varies from platform to platform
 723     // compiler support for wide character literals varies from compiler to compiler
 724     // so we should store the wide character version as UTF-8 and depend on
 725     // the UTF-8 converter's ability to decode it to platform specific wide characters
 726     // this test is invalid if the UTF-8 converter can't decode
 727     wxWCharBuffer wideBuffer((size_t)0);
 728     wideBuffer = DecodeUTF8( utf8Buffer, utf8Bytes );
 729     size_t wideChars = wxWcslen( wideBuffer.data() );
 730
 731     TestDecoder
 732         (
 733         wideBuffer.data(),
 734         wideChars,
 735         (const char*)multiBuffer,
 736         multiBytes,
 737         converter,
 738         sizeofNull
 739         );
 740     TestEncoder
 741         (
 742         wideBuffer.data(),
 743         wideChars,
 744         (const char*)multiBuffer,
 745         multiBytes,
 746         converter,
 747         sizeofNull
 748         );
 749 }
 750
 751
 752 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const wxChar* name );
 753
 754 void MBConvTestCase::FontmapTests()
 755 {
 756 #ifdef wxUSE_FONTMAP
 757     wxMBConv* converter = new_wxMBConv_wxwin( _T("CP1252") );
 758     if ( !converter )
 759     {
 760         return;
 761     }
 762     TestCoder(
 763         (const char*)CP1252,
 764         sizeof(CP1252),
 765         (const char*)CP1252_utf8,
 766         sizeof(CP1252_utf8),
 767         converter,
 768         1
 769         );
 770     delete converter;
 771 #endif
 772 }
 773
 774
 775 WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const wxChar* name );
 776
 777 void MBConvTestCase::IconvTests()
 778 {
 779 #ifdef HAVE_ICONV
 780     wxMBConv* converter = new_wxMBConv_iconv( _T("CP932") );
 781     if ( !converter )
 782     {
 783         return;
 784     }
 785     TestCoder(
 786         (const char*)welcome_cp932,
 787         sizeof(welcome_cp932),
 788         (const char*)welcome_utf8,
 789         sizeof(welcome_utf8),
 790         converter,
 791         1
 792         );
 793     delete converter;
 794 #endif
 795 }
 796
 797 void MBConvTestCase::CP1252Tests()
 798 {
 799     wxCSConv convCP1252( wxFONTENCODING_CP1252 );
 800     TestCoder(
 801         (const char*)CP1252,
 802         sizeof(CP1252),
 803         (const char*)CP1252_utf8,
 804         sizeof(CP1252_utf8),
 805         &convCP1252,
 806         1
 807         );
 808 }
 809
 810 void MBConvTestCase::LibcTests()
 811 {
 812     // There isn't a locale that all systems support (except "C"), so leave
 813     // this one disabled for non-Windows systems for the moment, until
 814     // a solution can be found.
 815 #ifdef __WXMSW__
 816
 817 #ifdef __WXMSW__
 818     setlocale( LC_ALL, "English_United States.1252" );
 819     const unsigned char* systemMB = CP1252;
 820     size_t systemMB_size = sizeof(CP1252);
 821     const unsigned char* systemMB_utf8 = CP1252_utf8;
 822     size_t systemMB_utf8_size = sizeof(CP1252_utf8);
 823 #else
 824     setlocale( LC_ALL, "en_US.iso8859-1" );
 825     const unsigned char* systemMB = iso8859_1;
 826     size_t systemMB_size = sizeof(iso8859_1);
 827     const unsigned char* systemMB_utf8 = iso8859_1_utf8;
 828     size_t systemMB_utf8_size = sizeof(iso8859_1_utf8);
 829 #endif
 830     wxMBConvLibc convLibc;
 831     TestCoder(
 832         (const char*)systemMB,
 833         systemMB_size,
 834         (const char*)systemMB_utf8,
 835         systemMB_utf8_size,
 836         &convLibc,
 837         1
 838         );
 839
 840 #endif // __WXMSW__
 841 }
 842
 843 // verifies that the specified mb sequences decode to the specified wc sequence
 844 void MBConvTestCase::TestDecoder(
 845     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 846     size_t         wideChars,   // the number of wide characters at wideBuffer
 847     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 848     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 849     wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 850     int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 851     )
 852 {
 853     const unsigned UNINITIALIZED = 0xcd;
 854
 855     // copy the input bytes into a null terminated buffer
 856     wxCharBuffer inputCopy( multiBytes+sizeofNull );
 857     memcpy( inputCopy.data(), multiBuffer, multiBytes );
 858     memset( &inputCopy.data()[multiBytes], 0, sizeofNull );
 859
 860     // calculate the output size
 861     size_t outputWritten = converter->MB2WC
 862         (
 863         0,
 864         (const char*)inputCopy.data(),
 865         0
 866         );
 867     // make sure the correct output length was calculated
 868     CPPUNIT_ASSERT( outputWritten == wideChars );
 869
 870     // convert the string
 871     size_t guardChars = 8; // to make sure we're not overrunning the output buffer
 872     size_t nullCharacters = 1;
 873     size_t outputBufferChars = outputWritten + nullCharacters + guardChars;
 874     wxWCharBuffer outputBuffer(outputBufferChars);
 875     memset( outputBuffer.data(), UNINITIALIZED, outputBufferChars*sizeof(wchar_t) );
 876
 877     outputWritten = converter->MB2WC
 878         (
 879         outputBuffer.data(),
 880         (const char*)inputCopy.data(),
 881         outputBufferChars
 882         );
 883     // make sure the correct number of characters were outputs
 884     CPPUNIT_ASSERT( outputWritten == wideChars );
 885
 886     // make sure the characters generated are correct
 887     CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, wideBuffer, wideChars*sizeof(wchar_t) ) );
 888
 889     // the output buffer should be null terminated
 890     CPPUNIT_ASSERT(  outputBuffer[outputWritten] == 0 );
 891
 892     // make sure the rest of the output buffer is untouched
 893     for ( size_t i = (wideChars+1)*sizeof(wchar_t); i < (outputBufferChars*sizeof(wchar_t)); i++ )
 894     {
 895         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
 896     }
 897
 898 #if wxUSE_UNICODE && wxUSE_STREAMS
 899     TestStreamDecoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
 900 #endif
 901 }
 902
 903 // verifies that the specified wc sequences encodes to the specified mb sequence
 904 void MBConvTestCase::TestEncoder(
 905     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 906     size_t         wideChars,   // the number of wide characters at wideBuffer
 907     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 908     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 909     wxMBConv*      converter,   // the wxMBConv object that can decode multiBuffer into a wide character sequence
 910     int            sizeofNull   // number of bytes occupied by terminating null in this encoding
 911     )
 912 {
 913     const unsigned UNINITIALIZED = 0xcd;
 914
 915     // copy the input bytes into a null terminated buffer
 916     wxWCharBuffer inputCopy( wideChars + 1 );
 917     memcpy( inputCopy.data(), wideBuffer, (wideChars*sizeof(wchar_t)) );
 918     inputCopy.data()[wideChars] = 0;
 919
 920     // calculate the output size
 921     size_t outputWritten = converter->WC2MB
 922         (
 923         0,
 924         (const wchar_t*)inputCopy.data(),
 925         0
 926         );
 927     // make sure the correct output length was calculated
 928     CPPUNIT_ASSERT( outputWritten == multiBytes );
 929
 930     // convert the string
 931     size_t guardBytes = 8; // to make sure we're not overrunning the output buffer
 932     size_t outputBufferSize = outputWritten + sizeofNull + guardBytes;
 933     wxCharBuffer outputBuffer(outputBufferSize);
 934     memset( outputBuffer.data(), UNINITIALIZED, outputBufferSize );
 935
 936     outputWritten = converter->WC2MB
 937         (
 938         outputBuffer.data(),
 939         (const wchar_t*)inputCopy.data(),
 940         outputBufferSize
 941         );
 942
 943     // make sure the correct number of characters were output
 944     CPPUNIT_ASSERT( outputWritten == multiBytes );
 945
 946     // make sure the characters generated are correct
 947     CPPUNIT_ASSERT( 0 == memcmp( outputBuffer, multiBuffer, multiBytes ) );
 948
 949     size_t i;
 950
 951     // the output buffer should be null terminated
 952     for ( i = multiBytes; i < multiBytes + sizeofNull; i++ )
 953     {
 954         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == 0 );
 955     }
 956
 957     // make sure the rest of the output buffer is untouched
 958     for ( i = multiBytes + sizeofNull; i < outputBufferSize; i++ )
 959     {
 960         CPPUNIT_ASSERT( ((unsigned char*)outputBuffer.data())[i] == UNINITIALIZED );
 961     }
 962
 963 #if wxUSE_UNICODE && wxUSE_STREAMS
 964     TestStreamEncoder( wideBuffer, wideChars, multiBuffer, multiBytes, converter );
 965 #endif
 966 }
 967
 968 #if wxUSE_UNICODE && wxUSE_STREAMS
 969 // use wxTextInputStream to exercise wxMBConv interface
 970 // (this reveals some bugs in certain wxMBConv subclasses)
 971 void MBConvTestCase::TestStreamDecoder(
 972                                        const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 973                                        size_t         wideChars,   // the number of wide characters at wideBuffer
 974                                        const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
 975                                        size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
 976                                        wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
 977                                        )
 978 {
 979     // this isn't meant to test wxMemoryInputStream or wxTextInputStream
 980     // it's meant to test the way wxTextInputStream uses wxMBConv
 981     // (which has exposed some problems with wxMBConv)
 982     wxMemoryInputStream memoryInputStream( multiBuffer, multiBytes );
 983     wxTextInputStream textInputStream( memoryInputStream, wxT(""), *converter );
 984     for ( size_t i = 0; i < wideChars; i++ )
 985     {
 986         wxChar wc = textInputStream.GetChar();
 987         CPPUNIT_ASSERT( wc == wideBuffer[i] );
 988     }
 989     CPPUNIT_ASSERT( 0 == textInputStream.GetChar() );
 990     CPPUNIT_ASSERT( memoryInputStream.Eof() );
 991 }
 992 #endif
 993
 994 #if wxUSE_UNICODE && wxUSE_STREAMS
 995 // use wxTextInputStream to exercise wxMBConv interface
 996 // (this reveals some bugs in certain wxMBConv subclasses)
 997 void MBConvTestCase::TestStreamEncoder(
 998     const wchar_t* wideBuffer,  // the same character sequence as multiBuffer, encoded as wchar_t
 999     size_t         wideChars,   // the number of wide characters at wideBuffer
1000     const char*    multiBuffer, // a multibyte encoded character sequence that can be decoded by "converter"
1001     size_t         multiBytes,  // the byte length of the multibyte character sequence that can be decoded by "converter"
1002     wxMBConv*      converter    // the wxMBConv object that can decode multiBuffer into a wide character sequence
1003     )
1004 {
1005     // this isn't meant to test wxMemoryOutputStream or wxTextOutputStream
1006     // it's meant to test the way wxTextOutputStream uses wxMBConv
1007     // (which has exposed some problems with wxMBConv)
1008     wxMemoryOutputStream memoryOutputStream;
1009     // wxEOL_UNIX will pass \n \r unchanged
1010     wxTextOutputStream textOutputStream( memoryOutputStream, wxEOL_UNIX, *converter );
1011     for ( size_t i = 0; i < wideChars; i++ )
1012     {
1013         textOutputStream.PutChar( wideBuffer[i] );
1014     }
1015     CPPUNIT_ASSERT_EQUAL( (wxFileOffset)multiBytes, memoryOutputStream.TellO() );
1016     wxCharBuffer copy( memoryOutputStream.TellO() );
1017     memoryOutputStream.CopyTo( copy.data(), memoryOutputStream.TellO());
1018     CPPUNIT_ASSERT_EQUAL( 0, memcmp( copy.data(), multiBuffer, multiBytes ) );
1019 }
1020 #endif
1021
1022
1023 // ----------------------------------------------------------------------------
1024 // UTF-8 tests
1025 // ----------------------------------------------------------------------------
1026
1027 #ifdef HAVE_WCHAR_H
1028
1029 // Check that 'charSequence' translates to 'wideSequence' and back.
1030 // Invalid sequences can be tested by giving NULL for 'wideSequence'. Even
1031 // invalid sequences should roundtrip when an option is given and this is
1032 // checked.
1033 //
1034 void MBConvTestCase::UTF8(const char *charSequence,
1035                           const wchar_t *wideSequence)
1036 {
1037     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1038     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1039     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1040 }
1041
1042 // Use this alternative when 'charSequence' contains a PUA character. Such
1043 // sequences should still roundtrip ok, and this is checked.
1044 //
1045 void MBConvTestCase::UTF8PUA(const char *charSequence,
1046                              const wchar_t *wideSequence)
1047 {
1048     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1049     UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1050     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1051 }
1052
1053 // Use this alternative when 'charSequence' contains an octal escape sequence.
1054 // Such sequences should still roundtrip ok, and this is checked.
1055 //
1056 void MBConvTestCase::UTF8Octal(const char *charSequence,
1057                                const wchar_t *wideSequence)
1058 {
1059     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_NOT);
1060     UTF8(charSequence, wideSequence, wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
1061     UTF8(charSequence, NULL, wxMBConvUTF8::MAP_INVALID_UTF8_TO_OCTAL);
1062 }
1063
1064 // in case wcscpy is missing
1065 //
1066 static wchar_t *wx_wcscpy(wchar_t *dest, const wchar_t *src)
1067 {
1068     wchar_t *d = dest;
1069     while ((*d++ = *src++) != 0)
1070         ;
1071     return dest;
1072 }
1073
1074 // in case wcscat is missing
1075 //
1076 static wchar_t *wx_wcscat(wchar_t *dest, const wchar_t *src)
1077 {
1078     wchar_t *d = dest;
1079     while (*d)
1080         d++;
1081     while ((*d++ = *src++) != 0)
1082         ;
1083     return dest;
1084 }
1085
1086 // in case wcscmp is missing
1087 //
1088 static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
1089 {
1090     while (*s1 == *s2 && *s1 != 0)
1091     {
1092         s1++;
1093         s2++;
1094     }
1095     return *s1 - *s2;
1096 }
1097
1098 // in case wcslen is missing
1099 //
1100 static size_t wx_wcslen(const wchar_t *s)
1101 {
1102     const wchar_t *t = s;
1103     while (*t != 0)
1104         t++;
1105     return t - s;
1106 }
1107
1108 // include the option in the error messages so it's possible to see which
1109 // test failed
1110 #define UTF8ASSERT(expr) CPPUNIT_ASSERT_MESSAGE(#expr + errmsg,  expr)
1111
1112 // The test implementation
1113 //
1114 void MBConvTestCase::UTF8(const char *charSequence,
1115                           const wchar_t *wideSequence,
1116                           int option)
1117 {
1118     const size_t BUFSIZE = 128;
1119     wxASSERT(strlen(charSequence) * 3 + 10 < BUFSIZE);
1120     char bytes[BUFSIZE];
1121
1122     // include the option in the error messages so it's possible to see
1123     // which test failed
1124     sprintf(bytes, " (with option == %d)", option);
1125     std::string errmsg(bytes);
1126
1127     // put the charSequence at the start, middle and end of a string
1128     strcpy(bytes, charSequence);
1129     strcat(bytes, "ABC");
1130     strcat(bytes, charSequence);
1131     strcat(bytes, "XYZ");
1132     strcat(bytes, charSequence);
1133
1134     // translate it into wide characters
1135     wxMBConvUTF8 utf8(option);
1136     wchar_t widechars[BUFSIZE];
1137     size_t lenResult = utf8.MB2WC(NULL, bytes, 0);
1138     size_t result = utf8.MB2WC(widechars, bytes, BUFSIZE);
1139     UTF8ASSERT(result == lenResult);
1140
1141     // check we got the expected result
1142     if (wideSequence) {
1143         UTF8ASSERT(result != (size_t)-1);
1144         wxASSERT(result < BUFSIZE);
1145
1146         wchar_t expected[BUFSIZE];
1147         wx_wcscpy(expected, wideSequence);
1148         wx_wcscat(expected, L"ABC");
1149         wx_wcscat(expected, wideSequence);
1150         wx_wcscat(expected, L"XYZ");
1151         wx_wcscat(expected, wideSequence);
1152
1153         UTF8ASSERT(wx_wcscmp(widechars, expected) == 0);
1154         UTF8ASSERT(wx_wcslen(widechars) == result);
1155     }
1156     else {
1157         // If 'wideSequence' is NULL, then the result is expected to be
1158         // invalid.  Normally that is as far as we can go, but if there is an
1159         // option then the conversion should succeed anyway, and it should be
1160         // possible to translate back to the original
1161         if (!option) {
1162             UTF8ASSERT(result == (size_t)-1);
1163             return;
1164         }
1165         else {
1166             UTF8ASSERT(result != (size_t)-1);
1167         }
1168     }
1169
1170     // translate it back and check we get the original
1171     char bytesAgain[BUFSIZE];
1172     size_t lenResultAgain = utf8.WC2MB(NULL, widechars, 0);
1173     size_t resultAgain = utf8.WC2MB(bytesAgain, widechars, BUFSIZE);
1174     UTF8ASSERT(resultAgain == lenResultAgain);
1175     UTF8ASSERT(resultAgain != (size_t)-1);
1176     wxASSERT(resultAgain < BUFSIZE);
1177
1178     UTF8ASSERT(strcmp(bytes, bytesAgain) == 0);
1179     UTF8ASSERT(strlen(bytesAgain) == resultAgain);
1180 }
1181
1182 #endif // HAVE_WCHAR_H