// headers
// ----------------------------------------------------------------------------
-#include "wx/wxprec.h"
+#include "testprec.h"
#ifdef __BORLANDC__
#pragma hdrstop
#include "wx/tokenzr.h"
-#include "wx/cppunit.h"
-
// ----------------------------------------------------------------------------
// test class
// ----------------------------------------------------------------------------
CPPUNIT_TEST( PChar );
CPPUNIT_TEST( Format );
CPPUNIT_TEST( Constructors );
+#if wxUSE_WCHAR_T
+ CPPUNIT_TEST( ConstructorsWithConversion );
+ CPPUNIT_TEST( Conversion );
+ CPPUNIT_TEST( ConversionUTF7 );
+ CPPUNIT_TEST( ConversionUTF8 );
+#endif // wxUSE_WCHAR_T
CPPUNIT_TEST( Extraction );
CPPUNIT_TEST( Find );
CPPUNIT_TEST( Tokenizer );
+ CPPUNIT_TEST( TokenizerGetPosition );
CPPUNIT_TEST( Replace );
CPPUNIT_TEST( Match );
CPPUNIT_TEST( CaseChanges );
+ CPPUNIT_TEST( Compare );
+ CPPUNIT_TEST( CompareNoCase );
CPPUNIT_TEST_SUITE_END();
void String();
void PChar();
void Format();
void Constructors();
+#if wxUSE_WCHAR_T
+ void ConstructorsWithConversion();
+ 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 Tokenizer();
+ void TokenizerGetPosition();
void Replace();
void Match();
void CaseChanges();
+ 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)
};
TEST_CTOR((start, end), _T("really"));
}
+#if wxUSE_WCHAR_T
+void StringTestCase::ConstructorsWithConversion()
+{
+ // 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 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);
+
+#if wxUSE_UNICODE
+ CPPUNIT_ASSERT( s1 == wchar );
+ CPPUNIT_ASSERT( s2 == wchar );
+#else
+ CPPUNIT_ASSERT( s1 == utf8 );
+ CPPUNIT_ASSERT( s2 == utf8 );
+#endif
+
+ wxString sub(utf8sub, wxConvUTF8); // "Dej" substring
+ wxString s3(utf8, wxConvUTF8, 4);
+ wxString s4(wchar, wxConvUTF8, 3);
+
+ 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()
+{
+#if wxUSE_UNICODE
+ wxString szTheString(L"The\0String", wxConvLibc, 10);
+ wxCharBuffer theBuffer = szTheString.mb_str();
+
+ CPPUNIT_ASSERT( memcmp(theBuffer.data(), "The\0String", 11) == 0 );
+
+ wxString szTheString2("The\0String", wxConvLocal, 10);
+ CPPUNIT_ASSERT( szTheString2.length() == 11 );
+ CPPUNIT_ASSERT( wxTmemcmp(szTheString2.c_str(), L"The\0String", 11) == 0 );
+#else
+ wxString szTheString(wxT("TheString"));
+ szTheString.insert(3, 1, '\0');
+ wxWCharBuffer theBuffer = szTheString.wc_str(wxConvLibc);
+
+ CPPUNIT_ASSERT( memcmp(theBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
+
+ wxString szLocalTheString(wxT("TheString"));
+ szLocalTheString.insert(3, 1, '\0');
+ wxWCharBuffer theLocalBuffer = szLocalTheString.wc_str(wxConvLocal);
+
+ CPPUNIT_ASSERT( memcmp(theLocalBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 );
+#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"+-" },
+#if !defined(__GNUC__) || (__GNUC__ >= 3)
+ { "+AKM-", L"\u00a3" },
+#endif
+ // 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[] =
+ {
+#if !defined(__GNUC__) || (__GNUC__ >= 3)
+ { "\xc2\xa3", L"\u00a3" },
+#endif
+ { "\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!"));
wxChar *buf, *s = NULL, *last;
if ( tkz.GetMode() == wxTOKEN_STRTOK )
- {
- buf = new wxChar[wxStrlen(str) + 1];
- wxStrcpy(buf, str);
- s = wxStrtok(buf, delims, &last);
- }
- else
- {
- buf = NULL;
- }
-
- size_t count2 = 0;
- while ( tkz.HasMoreTokens() )
- {
- wxString token = tkz.GetNextToken();
- if ( buf )
- {
- CPPUNIT_ASSERT( token == s );
- s = wxStrtok(NULL, delims, &last);
- }
- count2++;
- }
-
- CPPUNIT_ASSERT( count2 == count );
- if ( buf )
- {
- delete [] buf;
- }
+ {
+ buf = new wxChar[wxStrlen(str) + 1];
+ wxStrcpy(buf, str);
+ s = wxStrtok(buf, delims, &last);
+ }
+ else
+ {
+ buf = NULL;
+ }
+
+ size_t count2 = 0;
+ while ( tkz.HasMoreTokens() )
+ {
+ wxString token = tkz.GetNextToken();
+ if ( buf )
+ {
+ CPPUNIT_ASSERT( token == s );
+ s = wxStrtok(NULL, delims, &last);
+ }
+ count2++;
+ }
+
+ CPPUNIT_ASSERT( count2 == count );
+ if ( buf )
+ {
+ delete [] buf;
+ }
}
void StringTestCase::Tokenizer()
SingleTokenizerTest( _T("01-02/99"), _T("/-"), 3, wxTOKEN_RET_DELIMS );
}
+// call this with the string to tokenize, delimeters to use and the expected
+// positions (i.e. results of GetPosition()) after each GetNextToken() call,
+// terminate positions with 0
+static void
+DoTokenizerGetPosition(const wxChar *s, const wxChar *delims, int pos, ...)
+{
+ wxStringTokenizer tkz(s, delims);
+
+ CPPUNIT_ASSERT( tkz.GetPosition() == 0 );
+
+ va_list ap;
+ va_start(ap, pos);
+
+ for ( ;; )
+ {
+ if ( !pos )
+ {
+ CPPUNIT_ASSERT( !tkz.HasMoreTokens() );
+ break;
+ }
+
+ tkz.GetNextToken();
+
+ CPPUNIT_ASSERT( tkz.GetPosition() == (size_t)pos );
+
+ pos = va_arg(ap, int);
+ }
+
+ va_end(ap);
+}
+
+void StringTestCase::TokenizerGetPosition()
+{
+ DoTokenizerGetPosition(_T("foo"), _T("_"), 3, 0);
+ DoTokenizerGetPosition(_T("foo_bar"), _T("_"), 4, 7, 0);
+ DoTokenizerGetPosition(_T("foo_bar_"), _T("_"), 4, 8, 0);
+}
+
void StringTestCase::Replace()
{
#define TEST_REPLACE( original , pos , len , replacement , result ) \
CPPUNIT_ASSERT( s1l == _T("hello!") );
CPPUNIT_ASSERT( s2u == wxEmptyString );
CPPUNIT_ASSERT( s2l == wxEmptyString );
+
+#if !wxUSE_UNICODE
+ wxLocale locRu(wxLANGUAGE_RUSSIAN, 0 /* flags */);
+ if ( locRu.IsOk() )
+ {
+ // try upper casing 8bit strings
+ wxString sUpper("\xdf"),
+ sLower("\xff");
+
+ CPPUNIT_ASSERT( sUpper.Lower() == sLower );
+ CPPUNIT_ASSERT( sLower.Upper() == sUpper );
+ }
+#endif // !wxUSE_UNICODE
+}
+
+void StringTestCase::Compare()
+{
+ wxString s1 = wxT("AHH");
+ wxString eq = wxT("AHH");
+ wxString neq1 = wxT("HAH");
+ wxString neq2 = wxT("AH");
+ wxString neq3 = wxT("AHHH");
+ wxString neq4 = wxT("AhH");
+
+ CPPUNIT_ASSERT( s1 == eq );
+ CPPUNIT_ASSERT( s1 != neq1 );
+ CPPUNIT_ASSERT( s1 != neq2 );
+ CPPUNIT_ASSERT( s1 != neq3 );
+ CPPUNIT_ASSERT( s1 != neq4 );
+
+// wxString _s1 = wxT("A\0HH");
+// wxString _eq = wxT("A\0HH");
+// wxString _neq1 = wxT("H\0AH");
+// wxString _neq2 = wxT("A\0H");
+// wxString _neq3 = wxT("A\0HHH");
+// wxString _neq4 = wxT("A\0hH");
+ s1.insert(1,1,'\0');
+ eq.insert(1,1,'\0');
+ neq1.insert(1,1,'\0');
+ neq2.insert(1,1,'\0');
+ neq3.insert(1,1,'\0');
+ neq4.insert(1,1,'\0');
+
+ CPPUNIT_ASSERT( s1 == eq );
+ CPPUNIT_ASSERT( s1 != neq1 );
+ CPPUNIT_ASSERT( s1 != neq2 );
+ CPPUNIT_ASSERT( s1 != neq3 );
+ CPPUNIT_ASSERT( s1 != neq4 );
}
+
+void StringTestCase::CompareNoCase()
+{
+ wxString s1 = wxT("AHH");
+ wxString eq = wxT("AHH");
+ wxString eq2 = wxT("AhH");
+ wxString eq3 = wxT("ahh");
+ wxString neq = wxT("HAH");
+ wxString neq2 = wxT("AH");
+ wxString neq3 = wxT("AHHH");
+
+ #define CPPUNIT_CNCEQ_ASSERT(s1, s2) CPPUNIT_ASSERT( s1.CmpNoCase(s2) == 0)
+ #define CPPUNIT_CNCNEQ_ASSERT(s1, s2) CPPUNIT_ASSERT( s1.CmpNoCase(s2) != 0)
+
+ CPPUNIT_CNCEQ_ASSERT( s1, eq );
+ CPPUNIT_CNCEQ_ASSERT( s1, eq2 );
+ CPPUNIT_CNCEQ_ASSERT( s1, eq3 );
+
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq );
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq2 );
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq3 );
+
+
+// wxString _s1 = wxT("A\0HH");
+// wxString _eq = wxT("A\0HH");
+// wxString _eq2 = wxT("A\0hH");
+// wxString _eq3 = wxT("a\0hh");
+// wxString _neq = wxT("H\0AH");
+// wxString _neq2 = wxT("A\0H");
+// wxString _neq3 = wxT("A\0HHH");
+
+ s1.insert(1,1,'\0');
+ eq.insert(1,1,'\0');
+ eq2.insert(1,1,'\0');
+ eq3.insert(1,1,'\0');
+ neq.insert(1,1,'\0');
+ neq2.insert(1,1,'\0');
+ neq3.insert(1,1,'\0');
+
+ CPPUNIT_CNCEQ_ASSERT( s1, eq );
+ CPPUNIT_CNCEQ_ASSERT( s1, eq2 );
+ CPPUNIT_CNCEQ_ASSERT( s1, eq3 );
+
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq );
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq2 );
+ CPPUNIT_CNCNEQ_ASSERT( s1, neq3 );
+}
+