tests/strings/unicode.cpp

   1 ///////////////////////////////////////////////////////////////////////////////
   2 // Name:        tests/strings/unicode.cpp
   3 // Purpose:     Unicode unit test
   4 // Author:      Vadim Zeitlin, Wlodzimierz ABX Skiba
   5 // Created:     2004-04-28
   6 // RCS-ID:      $Id$
   7 // Copyright:   (c) 2004 Vadim Zeitlin, Wlodzimierz Skiba
   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 #endif // WX_PRECOMP
  22
  23 // ----------------------------------------------------------------------------
  24 // local functions
  25 // ----------------------------------------------------------------------------
  26
  27 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
  28
  29 // in case wcscmp is missing
  30 static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
  31 {
  32     while (*s1 == *s2 && *s1 != 0)
  33     {
  34         s1++;
  35         s2++;
  36     }
  37     return *s1 - *s2;
  38 }
  39
  40 #endif // wxUSE_WCHAR_T && !wxUSE_UNICODE
  41
  42 // ----------------------------------------------------------------------------
  43 // test class
  44 // ----------------------------------------------------------------------------
  45
  46 class UnicodeTestCase : public CppUnit::TestCase
  47 {
  48 public:
  49     UnicodeTestCase();
  50
  51 private:
  52     CPPUNIT_TEST_SUITE( UnicodeTestCase );
  53         CPPUNIT_TEST( ToFromAscii );
  54 #if wxUSE_WCHAR_T
  55         CPPUNIT_TEST( ConstructorsWithConversion );
  56         CPPUNIT_TEST( ConversionEmpty );
  57         CPPUNIT_TEST( ConversionWithNULs );
  58         CPPUNIT_TEST( ConversionUTF7 );
  59         CPPUNIT_TEST( ConversionUTF8 );
  60         CPPUNIT_TEST( ConversionUTF16 );
  61         CPPUNIT_TEST( ConversionUTF32 );
  62 #endif // wxUSE_WCHAR_T
  63     CPPUNIT_TEST_SUITE_END();
  64
  65     void ToFromAscii();
  66 #if wxUSE_WCHAR_T
  67     void ConstructorsWithConversion();
  68     void ConversionEmpty();
  69     void ConversionWithNULs();
  70     void ConversionUTF7();
  71     void ConversionUTF8();
  72     void ConversionUTF16();
  73     void ConversionUTF32();
  74
  75     // test if converting s using the given encoding gives ws and vice versa
  76     //
  77     // if either of the first 2 arguments is NULL, the conversion is supposed
  78     // to fail
  79     void DoTestConversion(const char *s, const wchar_t *w, wxMBConv& conv);
  80 #endif // wxUSE_WCHAR_T
  81
  82
  83     DECLARE_NO_COPY_CLASS(UnicodeTestCase)
  84 };
  85
  86 // register in the unnamed registry so that these tests are run by default
  87 CPPUNIT_TEST_SUITE_REGISTRATION( UnicodeTestCase );
  88
  89 // also include in it's own registry so that these tests can be run alone
  90 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( UnicodeTestCase, "Unicode" );
  91
  92 UnicodeTestCase::UnicodeTestCase()
  93 {
  94 }
  95
  96 void UnicodeTestCase::ToFromAscii()
  97 {
  98
  99 #define TEST_TO_FROM_ASCII(txt)                              \
 100     {                                                        \
 101         static const char *msg = txt;                        \
 102         wxString s = wxString::FromAscii(msg);               \
 103         CPPUNIT_ASSERT( strcmp( s.ToAscii() , msg ) == 0 );  \
 104     }
 105
 106     TEST_TO_FROM_ASCII( "Hello, world!" );
 107     TEST_TO_FROM_ASCII( "additional \" special \t test \\ component \n :-)" );
 108 }
 109
 110 #if wxUSE_WCHAR_T
 111 void UnicodeTestCase::ConstructorsWithConversion()
 112 {
 113     // the string "Déjà" in UTF-8 and wchar_t:
 114     const unsigned char utf8Buf[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0};
 115     const wchar_t wchar[] = {0x44,0xE9,0x6A,0xE0,0};
 116     const unsigned char utf8subBuf[] = {0x44,0xC3,0xA9,0x6A,0}; // just "Déj"
 117     const char *utf8 = (char *)utf8Buf;
 118     const char *utf8sub = (char *)utf8subBuf;
 119
 120     wxString s1(utf8, wxConvUTF8);
 121     wxString s2(wchar, wxConvUTF8);
 122
 123 #if wxUSE_UNICODE
 124     CPPUNIT_ASSERT( s1 == wchar );
 125     CPPUNIT_ASSERT( s2 == wchar );
 126 #else
 127     CPPUNIT_ASSERT( s1 == utf8 );
 128     CPPUNIT_ASSERT( s2 == utf8 );
 129 #endif
 130
 131     wxString sub(utf8sub, wxConvUTF8); // "Dej" substring
 132     wxString s3(utf8, wxConvUTF8, 4);
 133     wxString s4(wchar, wxConvUTF8, 3);
 134
 135     CPPUNIT_ASSERT( s3 == sub );
 136     CPPUNIT_ASSERT( s4 == sub );
 137
 138 #if wxUSE_UNICODE
 139     CPPUNIT_ASSERT ( wxString("\t[pl]open.format.Sformatuj dyskietkê=gfloppy %f",
 140                                wxConvUTF8) == wxT("") ); //should stop at pos 35
 141 #endif
 142 }
 143
 144 void UnicodeTestCase::ConversionEmpty()
 145 {
 146     size_t len;
 147
 148 #if wxUSE_UNICODE
 149     wxCharBuffer buf = wxConvLibc.cWC2MB(L"", 0, &len);
 150 #else // !wxUSE_UNICODE
 151     wxWCharBuffer wbuf = wxConvLibc.cMB2WC("", 0, &len);
 152 #endif // wxUSE_UNICODE/!wxUSE_UNICODE
 153
 154     CPPUNIT_ASSERT_EQUAL(0u, len);
 155 }
 156
 157 void UnicodeTestCase::ConversionWithNULs()
 158 {
 159 #if wxUSE_UNICODE
 160     static const size_t lenNulString = 10;
 161
 162     wxString szTheString(L"The\0String", wxConvLibc, lenNulString);
 163     wxCharBuffer theBuffer = szTheString.mb_str();
 164
 165     CPPUNIT_ASSERT( memcmp(theBuffer.data(), "The\0String",
 166                     lenNulString + 1) == 0 );
 167
 168     wxString szTheString2("The\0String", wxConvLocal, lenNulString);
 169     CPPUNIT_ASSERT_EQUAL( lenNulString, szTheString2.length() );
 170     CPPUNIT_ASSERT( wxTmemcmp(szTheString2.c_str(), L"The\0String",
 171                     lenNulString + 1) == 0 );
 172 #else // !wxUSE_UNICODE
 173     wxString szTheString(wxT("TheString"));
 174     szTheString.insert(3, 1, '\0');
 175     wxWCharBuffer theBuffer = szTheString.wc_str(wxConvLibc);
 176
 177     CPPUNIT_ASSERT( memcmp(theBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
 178
 179     wxString szLocalTheString(wxT("TheString"));
 180     szLocalTheString.insert(3, 1, '\0');
 181     wxWCharBuffer theLocalBuffer = szLocalTheString.wc_str(wxConvLocal);
 182
 183     CPPUNIT_ASSERT( memcmp(theLocalBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
 184 #endif // wxUSE_UNICODE/!wxUSE_UNICODE
 185 }
 186
 187 void
 188 UnicodeTestCase::DoTestConversion(const char *s,
 189                                   const wchar_t *ws,
 190                                   wxMBConv& conv)
 191 {
 192 #if wxUSE_UNICODE
 193     if ( ws )
 194     {
 195         wxCharBuffer buf = conv.cWC2MB(ws, (size_t)-1, NULL);
 196
 197         CPPUNIT_ASSERT( strcmp(buf, s) == 0 );
 198     }
 199 #else // wxUSE_UNICODE
 200     if ( s )
 201     {
 202         wxWCharBuffer wbuf = conv.cMB2WC(s, (size_t)-1, NULL);
 203
 204         if ( ws )
 205         {
 206             CPPUNIT_ASSERT( wbuf.data() );
 207             CPPUNIT_ASSERT( wx_wcscmp(wbuf, ws) == 0 );
 208         }
 209         else // conversion is supposed to fail
 210         {
 211             CPPUNIT_ASSERT_EQUAL( (wchar_t *)NULL, wbuf.data() );
 212         }
 213     }
 214 #endif // wxUSE_UNICODE/!wxUSE_UNICODE
 215 }
 216
 217 struct StringConversionData
 218 {
 219     const char *str;
 220     const wchar_t *wcs;
 221 };
 222
 223 void UnicodeTestCase::ConversionUTF7()
 224 {
 225     static const StringConversionData utf7data[] =
 226     {
 227         { "+-", L"+" },
 228         { "+--", L"+-" },
 229
 230 #ifdef wxHAVE_U_ESCAPE
 231         { "+AKM-", L"\u00a3" },
 232 #endif // wxHAVE_U_ESCAPE
 233
 234         // the following are invalid UTF-7 sequences
 235         { "+", NULL },
 236         { "+~", NULL },
 237         { "a+", NULL },
 238     };
 239
 240     wxCSConv conv(_T("utf-7"));
 241     for ( size_t n = 0; n < WXSIZEOF(utf7data); n++ )
 242     {
 243         const StringConversionData& d = utf7data[n];
 244         DoTestConversion(d.str, d.wcs, conv);
 245         DoTestConversion(d.str, d.wcs, wxConvUTF7);
 246     }
 247 }
 248
 249 void UnicodeTestCase::ConversionUTF8()
 250 {
 251     static const StringConversionData utf8data[] =
 252     {
 253 #ifdef wxHAVE_U_ESCAPE
 254         { "\xc2\xa3", L"\u00a3" },
 255 #endif
 256         { "\xc2", NULL },
 257     };
 258
 259     wxCSConv conv(_T("utf-8"));
 260     for ( size_t n = 0; n < WXSIZEOF(utf8data); n++ )
 261     {
 262         const StringConversionData& d = utf8data[n];
 263         DoTestConversion(d.str, d.wcs, conv);
 264         DoTestConversion(d.str, d.wcs, wxConvUTF8);
 265     }
 266 }
 267
 268 void UnicodeTestCase::ConversionUTF16()
 269 {
 270     static const StringConversionData utf16data[] =
 271     {
 272 #ifdef wxHAVE_U_ESCAPE
 273         { "\x04\x1f\x04\x40\x04\x38\x04\x32\x04\x35\x04\x42\0\0",
 274           L"\u041f\u0440\u0438\u0432\u0435\u0442" },
 275         { "\x01\0\0b\x01\0\0a\x01\0\0r\0\0", L"\u0100b\u0100a\u0100r" },
 276 #endif
 277         { "\0f\0o\0o\0\0", L"foo" },
 278     };
 279
 280     wxCSConv conv(wxFONTENCODING_UTF16BE);
 281     for ( size_t n = 0; n < WXSIZEOF(utf16data); n++ )
 282     {
 283         const StringConversionData& d = utf16data[n];
 284         DoTestConversion(d.str, d.wcs, conv);
 285     }
 286
 287     // special case: this string has consecutive NULs inside it which don't
 288     // terminate the string, this exposed a bug in our conversion code which
 289     // got confused in this case
 290     size_t len;
 291     wxWCharBuffer wbuf(conv.cMB2WC("\x01\0\0B\0C" /* A macron BC */, 6, &len));
 292     CPPUNIT_ASSERT_EQUAL( (size_t)3, len );
 293 }
 294
 295 void UnicodeTestCase::ConversionUTF32()
 296 {
 297     static const StringConversionData utf32data[] =
 298     {
 299 #ifdef wxHAVE_U_ESCAPE
 300         {
 301             "\0\0\x04\x1f\0\0\x04\x40\0\0\x04\x38\0\0\x04\x32\0\0\x04\x35\0\0\x04\x42\0\0\0\0",
 302           L"\u041f\u0440\u0438\u0432\u0435\u0442" },
 303 #endif
 304         { "\0\0\0f\0\0\0o\0\0\0o\0\0\0\0", L"foo" },
 305     };
 306
 307     wxCSConv conv(wxFONTENCODING_UTF32BE);
 308     for ( size_t n = 0; n < WXSIZEOF(utf32data); n++ )
 309     {
 310         const StringConversionData& d = utf32data[n];
 311         DoTestConversion(d.str, d.wcs, conv);
 312     }
 313
 314     size_t len;
 315     wxWCharBuffer wbuf(conv.cMB2WC("\0\0\x01\0\0\0\0B\0\0\0C" /* A macron BC */,
 316                                    12, &len));
 317     CPPUNIT_ASSERT_EQUAL( (size_t)3, len );
 318 }
 319
 320 #endif // wxUSE_WCHAR_T
 321