// headers
// ----------------------------------------------------------------------------
-#include "wx/wxprec.h"
+#include "testprec.h"
#ifdef __BORLANDC__
#pragma hdrstop
#include "wx/tokenzr.h"
-#include "wx/cppunit.h"
-
// ----------------------------------------------------------------------------
// test class
// ----------------------------------------------------------------------------
CPPUNIT_TEST( Constructors );
#if wxUSE_WCHAR_T
CPPUNIT_TEST( ConstructorsWithConversion );
-#endif
CPPUNIT_TEST( Conversion );
+ CPPUNIT_TEST( ConversionUTF7 );
+ CPPUNIT_TEST( ConversionUTF8 );
+#endif // wxUSE_WCHAR_T
CPPUNIT_TEST( Extraction );
CPPUNIT_TEST( Find );
CPPUNIT_TEST( Tokenizer );
void Constructors();
#if wxUSE_WCHAR_T
void ConstructorsWithConversion();
-#endif
void Conversion();
+ void ConversionUTF7();
+ void ConversionUTF8();
+#endif // wxUSE_WCHAR_T
void Extraction();
void Find();
void SingleTokenizerTest( wxChar *str, wxChar *delims, size_t count , wxStringTokenizerMode mode );
void Compare();
void CompareNoCase();
+#if wxUSE_WCHAR_T
+ // test if converting s using the given encoding gives ws and vice versa
+ //
+ // if either of the first 2 arguments is NULL, the conversion is supposed
+ // to fail
+ void DoTestConversion(const char *s, const wchar_t *w, wxCSConv& conv);
+#endif // wxUSE_WCHAR_T
+
DECLARE_NO_COPY_CLASS(StringTestCase)
};
#if wxUSE_WCHAR_T
void StringTestCase::ConstructorsWithConversion()
{
- // Déj`a in UTF-8 and wchar_t:
- const char utf8[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0};
+ // the string "Déjà" in UTF-8 and wchar_t:
+ const unsigned char utf8Buf[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0};
const wchar_t wchar[] = {0x44,0xE9,0x6A,0xE0,0};
- const char utf8sub[] = {0x44,0xC3,0xA9,0x6A,0}; // "Dej"
+ const unsigned char utf8subBuf[] = {0x44,0xC3,0xA9,0x6A,0}; // just "Déj"
+ const char *utf8 = (char *)utf8Buf;
+ const char *utf8sub = (char *)utf8subBuf;
wxString s1(utf8, wxConvUTF8);
wxString s2(wchar, wxConvUTF8);
CPPUNIT_ASSERT( s3 == sub );
CPPUNIT_ASSERT( s4 == sub );
-}
+
+#if wxUSE_UNICODE
+ CPPUNIT_ASSERT ( wxString("\t[pl]open.format.Sformatuj dyskietkê=gfloppy %f",
+ wxConvUTF8) == wxT("") ); //should stop at pos 35
#endif
+}
void StringTestCase::Conversion()
{
CPPUNIT_ASSERT( memcmp(theBuffer.data(), "The\0String", 11) == 0 );
wxString szTheString2("The\0String", wxConvLocal, 10);
- CPPUNIT_ASSERT( wxMemcmp(szTheString2.c_str(), L"The\0String", 11) == 0 );
+ CPPUNIT_ASSERT( szTheString2.length() == 11 );
+ CPPUNIT_ASSERT( wxTmemcmp(szTheString2.c_str(), L"The\0String", 11) == 0 );
#else
-# if wxUSE_WCHAR_T
wxString szTheString(wxT("TheString"));
szTheString.insert(3, 1, '\0');
wxWCharBuffer theBuffer = szTheString.wc_str(wxConvLibc);
wxWCharBuffer theLocalBuffer = szLocalTheString.wc_str(wxConvLocal);
CPPUNIT_ASSERT( memcmp(theLocalBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
-# endif
#endif
}
+void
+StringTestCase::DoTestConversion(const char *s,
+ const wchar_t *ws,
+ wxCSConv& conv)
+{
+#if wxUSE_UNICODE
+ if ( ws )
+ {
+ wxCharBuffer buf(wxString(ws).mb_str(conv));
+
+ CPPUNIT_ASSERT( strcmp(buf, s) == 0 );
+ }
+#else // wxUSE_UNICODE
+ if ( s )
+ {
+ wxWCharBuffer wbuf(wxString(s).wc_str(conv));
+
+ if ( ws )
+ CPPUNIT_ASSERT( wcscmp(wbuf, ws) == 0 );
+ else
+ CPPUNIT_ASSERT( !*wbuf );
+ }
+#endif // wxUSE_UNICODE/!wxUSE_UNICODE
+}
+
+struct StringConversionData
+{
+ const char *str;
+ const wchar_t *wcs;
+};
+
+void StringTestCase::ConversionUTF7()
+{
+ static const StringConversionData utf7data[] =
+ {
+ { "+-", L"+" },
+ { "+--", L"+-" },
+ { "+AKM-", L"\u00a3" },
+
+ // Windows accepts invalid UTF-7 strings and so does our UTF-7
+ // conversion code -- this is wrong IMO but the way it is for now
+ //
+ // notice that converting "+" still behaves as expected because the
+ // result is just an empty string, i.e. the same as if there were an
+ // error, but converting "a+" results in "a" while it really should
+ // fail
+ { "+", NULL },
+ { "a+", L"a" },
+ };
+
+ wxCSConv conv(_T("utf-7"));
+ for ( size_t n = 0; n < WXSIZEOF(utf7data); n++ )
+ {
+ const StringConversionData& d = utf7data[n];
+ DoTestConversion(d.str, d.wcs, conv);
+ }
+}
+
+void StringTestCase::ConversionUTF8()
+{
+ static const StringConversionData utf8data[] =
+ {
+ { "\xc2\xa3", L"\u00a3" },
+ { "\xc2", NULL },
+ };
+
+ wxCSConv conv(_T("utf-8"));
+ for ( size_t n = 0; n < WXSIZEOF(utf8data); n++ )
+ {
+ const StringConversionData& d = utf8data[n];
+ DoTestConversion(d.str, d.wcs, conv);
+ }
+}
+
+#endif // wxUSE_WCHAR_T
+
+
void StringTestCase::Extraction()
{
wxString s(_T("Hello, world!"));
void StringTestCase::TokenizerGetPosition()
{
DoTokenizerGetPosition(_T("foo"), _T("_"), 3, 0);
- DoTokenizerGetPosition(_T("foo_bar"), _T("_"), 3, 7, 0);
- DoTokenizerGetPosition(_T("foo_bar_"), _T("_"), 3, 7,8, 0);
+ DoTokenizerGetPosition(_T("foo_bar"), _T("_"), 4, 7, 0);
+ DoTokenizerGetPosition(_T("foo_bar_"), _T("_"), 4, 8, 0);
}
void StringTestCase::Replace()