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 // test class
  25 // ----------------------------------------------------------------------------
  26
  27 class UnicodeTestCase : public CppUnit::TestCase
  28 {
  29 public:
  30     UnicodeTestCase();
  31
  32 private:
  33     CPPUNIT_TEST_SUITE( UnicodeTestCase );
  34         CPPUNIT_TEST( ToFromAscii );
  35 #if wxUSE_WCHAR_T
  36         CPPUNIT_TEST( ConstructorsWithConversion );
  37         CPPUNIT_TEST( Conversion );
  38         CPPUNIT_TEST( ConversionUTF7 );
  39         CPPUNIT_TEST( ConversionUTF8 );
  40 #endif // wxUSE_WCHAR_T
  41     CPPUNIT_TEST_SUITE_END();
  42
  43     void ToFromAscii();
  44 #if wxUSE_WCHAR_T
  45     void ConstructorsWithConversion();
  46     void Conversion();
  47     void ConversionUTF7();
  48     void ConversionUTF8();
  49
  50     // test if converting s using the given encoding gives ws and vice versa
  51     //
  52     // if either of the first 2 arguments is NULL, the conversion is supposed
  53     // to fail
  54     void DoTestConversion(const char *s, const wchar_t *w, wxCSConv& conv);
  55 #endif // wxUSE_WCHAR_T
  56
  57
  58     DECLARE_NO_COPY_CLASS(UnicodeTestCase)
  59 };
  60
  61 // register in the unnamed registry so that these tests are run by default
  62 CPPUNIT_TEST_SUITE_REGISTRATION( UnicodeTestCase );
  63
  64 // also include in it's own registry so that these tests can be run alone
  65 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( UnicodeTestCase, "UnicodeTestCase" );
  66
  67 UnicodeTestCase::UnicodeTestCase()
  68 {
  69 }
  70
  71 void UnicodeTestCase::ToFromAscii()
  72 {
  73
  74 #define TEST_TO_FROM_ASCII(txt)                              \
  75     {                                                        \
  76         static const char *msg = txt;                        \
  77         wxString s = wxString::FromAscii(msg);               \
  78         CPPUNIT_ASSERT( strcmp( s.ToAscii() , msg ) == 0 );  \
  79     }
  80
  81     TEST_TO_FROM_ASCII( "Hello, world!" );
  82     TEST_TO_FROM_ASCII( "additional \" special \t test \\ component \n :-)" );
  83 }
  84
  85 #if wxUSE_WCHAR_T
  86 void UnicodeTestCase::ConstructorsWithConversion()
  87 {
  88     // the string "Déjà" in UTF-8 and wchar_t:
  89     const unsigned char utf8Buf[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0};
  90     const wchar_t wchar[] = {0x44,0xE9,0x6A,0xE0,0};
  91     const unsigned char utf8subBuf[] = {0x44,0xC3,0xA9,0x6A,0}; // just "Déj"
  92     const char *utf8 = (char *)utf8Buf;
  93     const char *utf8sub = (char *)utf8subBuf;
  94
  95     wxString s1(utf8, wxConvUTF8);
  96     wxString s2(wchar, wxConvUTF8);
  97
  98 #if wxUSE_UNICODE
  99     CPPUNIT_ASSERT( s1 == wchar );
 100     CPPUNIT_ASSERT( s2 == wchar );
 101 #else
 102     CPPUNIT_ASSERT( s1 == utf8 );
 103     CPPUNIT_ASSERT( s2 == utf8 );
 104 #endif
 105
 106     wxString sub(utf8sub, wxConvUTF8); // "Dej" substring
 107     wxString s3(utf8, wxConvUTF8, 4);
 108     wxString s4(wchar, wxConvUTF8, 3);
 109
 110     CPPUNIT_ASSERT( s3 == sub );
 111     CPPUNIT_ASSERT( s4 == sub );
 112
 113 #if wxUSE_UNICODE
 114     CPPUNIT_ASSERT ( wxString("\t[pl]open.format.Sformatuj dyskietkê=gfloppy %f",
 115                                wxConvUTF8) == wxT("") ); //should stop at pos 35
 116 #endif
 117 }
 118
 119 void UnicodeTestCase::Conversion()
 120 {
 121 #if wxUSE_UNICODE
 122         wxString szTheString(L"The\0String", wxConvLibc, 10);
 123         wxCharBuffer theBuffer = szTheString.mb_str();
 124
 125         CPPUNIT_ASSERT( memcmp(theBuffer.data(), "The\0String", 11) == 0 );
 126
 127         wxString szTheString2("The\0String", wxConvLocal, 10);
 128         CPPUNIT_ASSERT( szTheString2.length() == 11 );
 129         CPPUNIT_ASSERT( wxTmemcmp(szTheString2.c_str(), L"The\0String", 11) == 0 );
 130 #else
 131         wxString szTheString(wxT("TheString"));
 132         szTheString.insert(3, 1, '\0');
 133         wxWCharBuffer theBuffer = szTheString.wc_str(wxConvLibc);
 134
 135         CPPUNIT_ASSERT( memcmp(theBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
 136
 137         wxString szLocalTheString(wxT("TheString"));
 138         szLocalTheString.insert(3, 1, '\0');
 139         wxWCharBuffer theLocalBuffer = szLocalTheString.wc_str(wxConvLocal);
 140
 141         CPPUNIT_ASSERT( memcmp(theLocalBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
 142 #endif
 143 }
 144
 145 #if !wxUSE_UNICODE
 146 // in case wcscmp is missing
 147 //
 148 static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
 149 {
 150     while (*s1 == *s2 && *s1 != 0)
 151     {
 152         s1++;
 153         s2++;
 154     }
 155     return *s1 - *s2;
 156 }
 157 #endif
 158
 159 void
 160 UnicodeTestCase::DoTestConversion(const char *s,
 161                                  const wchar_t *ws,
 162                                  wxCSConv& conv)
 163 {
 164 #if wxUSE_UNICODE
 165     if ( ws )
 166     {
 167         wxCharBuffer buf(wxString(ws).mb_str(conv));
 168
 169         CPPUNIT_ASSERT( strcmp(buf, s) == 0 );
 170     }
 171 #else // wxUSE_UNICODE
 172     if ( s )
 173     {
 174         wxWCharBuffer wbuf(wxString(s).wc_str(conv));
 175
 176         if ( ws )
 177             CPPUNIT_ASSERT( wx_wcscmp(wbuf, ws) == 0 );
 178         else
 179             CPPUNIT_ASSERT( !*wbuf );
 180     }
 181 #endif // wxUSE_UNICODE/!wxUSE_UNICODE
 182 }
 183
 184 struct StringConversionData
 185 {
 186     const char *str;
 187     const wchar_t *wcs;
 188 };
 189
 190 void UnicodeTestCase::ConversionUTF7()
 191 {
 192     static const StringConversionData utf7data[] =
 193     {
 194         { "+-", L"+" },
 195         { "+--", L"+-" },
 196         //\u isn't recognized on MSVC 6
 197 #if !defined(_MSC_VER)
 198 #if !defined(__GNUC__) || (__GNUC__ >= 3)
 199         { "+AKM-", L"\u00a3" },
 200 #endif
 201 #endif
 202         // Windows accepts invalid UTF-7 strings and so does our UTF-7
 203         // conversion code -- this is wrong IMO but the way it is for now
 204         //
 205         // notice that converting "+" still behaves as expected because the
 206         // result is just an empty string, i.e. the same as if there were an
 207         // error, but converting "a+" results in "a" while it really should
 208         // fail
 209         { "+", NULL },
 210         { "a+", L"a" },
 211     };
 212
 213     wxCSConv conv(_T("utf-7"));
 214     for ( size_t n = 0; n < WXSIZEOF(utf7data); n++ )
 215     {
 216         const StringConversionData& d = utf7data[n];
 217         DoTestConversion(d.str, d.wcs, conv);
 218     }
 219 }
 220
 221 void UnicodeTestCase::ConversionUTF8()
 222 {
 223     static const StringConversionData utf8data[] =
 224     {
 225         //\u isn't recognized on MSVC 6
 226 #if !defined(_MSC_VER)
 227 #if !defined(__GNUC__) || (__GNUC__ >= 3)
 228         { "\xc2\xa3", L"\u00a3" },
 229 #endif
 230 #endif
 231         { "\xc2", NULL },
 232     };
 233
 234     wxCSConv conv(_T("utf-8"));
 235     for ( size_t n = 0; n < WXSIZEOF(utf8data); n++ )
 236     {
 237         const StringConversionData& d = utf8data[n];
 238         DoTestConversion(d.str, d.wcs, conv);
 239     }
 240 }
 241
 242 #endif // wxUSE_WCHAR_T
 243