X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/0a19929958e22101608221753f0245dd31e76702..0e41582e529b14dab0f6da62451b133b85227176:/tests/strings/strings.cpp

diff --git a/tests/strings/strings.cpp b/tests/strings/strings.cpp
index 636ee7e993..3e25c70f48 100644
--- a/tests/strings/strings.cpp
+++ b/tests/strings/strings.cpp
@@ -11,7 +11,7 @@
 // headers
 // ----------------------------------------------------------------------------
 
-#include "wx/wxprec.h"
+#include "testprec.h"
 
 #ifdef __BORLANDC__
     #pragma hdrstop
@@ -21,10 +21,6 @@
     #include "wx/wx.h"
 #endif // WX_PRECOMP
 
-#include "wx/tokenzr.h"
-
-#include "wx/cppunit.h"
-
 // ----------------------------------------------------------------------------
 // test class
 // ----------------------------------------------------------------------------
@@ -40,23 +36,52 @@ private:
         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( Replace );
         CPPUNIT_TEST( Match );
+        CPPUNIT_TEST( CaseChanges );
+        CPPUNIT_TEST( Compare );
+        CPPUNIT_TEST( CompareNoCase );
+        CPPUNIT_TEST( ToLong );
+        CPPUNIT_TEST( ToULong );
+        CPPUNIT_TEST( ToDouble );
     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 Replace();
     void Match();
+    void CaseChanges();
+    void Compare();
+    void CompareNoCase();
+    void ToLong();
+    void ToULong();
+    void ToDouble();
+
+#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)
 };
@@ -116,6 +141,15 @@ void StringTestCase::Format()
     CPPUNIT_ASSERT( s1 == wxString::Format(_T("%03d"), 18) );
     s2.Printf(_T("Number 18: %s\n"), s1.c_str());
     CPPUNIT_ASSERT( s2 == wxString::Format(_T("Number 18: %s\n"), s1.c_str()) );
+
+    static const size_t lengths[] = { 1, 512, 1024, 1025, 2048, 4096, 4097 };
+    for ( size_t n = 0; n < WXSIZEOF(lengths); n++ )
+    {
+        const size_t len = lengths[n];
+
+        wxString s(_T('Z'), len);
+        CPPUNIT_ASSERT_EQUAL( len, wxString::Format(_T("%s"), s.c_str()).length());
+    }
 }
 
 void StringTestCase::Constructors()
@@ -136,6 +170,166 @@ void StringTestCase::Constructors()
     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
+}
+
+#if !wxUSE_UNICODE
+// in case wcscmp is missing
+//
+static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2)
+{
+    while (*s1 == *s2 && *s1 != 0)
+    {
+        s1++;
+        s2++;
+    }
+    return *s1 - *s2;
+}
+#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( wx_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"+-" },
+        //\u isn't recognized on MSVC 6
+#if !defined(_MSC_VER)
+#if !defined(__GNUC__) || (__GNUC__ >= 3)
+        { "+AKM-", L"\u00a3" },
+#endif
+#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[] =
+    {
+        //\u isn't recognized on MSVC 6
+#if !defined(_MSC_VER)
+#if !defined(__GNUC__) || (__GNUC__ >= 3)
+        { "\xc2\xa3", L"\u00a3" },
+#endif
+#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!"));
@@ -179,60 +373,6 @@ void StringTestCase::Find()
     #undef TEST_FIND
 }
 
-void StringTestCase::SingleTokenizerTest( wxChar *str, wxChar *delims, size_t count , wxStringTokenizerMode mode )
-{
-    wxStringTokenizer tkz( str, delims, mode);
-    CPPUNIT_ASSERT( tkz.CountTokens() == count );
-
-    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;                            
-    }                                             
-}
-
-void StringTestCase::Tokenizer()
-{
-    SingleTokenizerTest( _T(""),                                           _T(" "),              0, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello, world"),                               _T(" "),              2, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello,   world  "),                           _T(" "),              2, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello, world"),                               _T(","),              2, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello, world!"),                              _T(",!"),             2, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello,, world!"),                             _T(",!"),             3, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("Hello, world!"),                              _T(",!"),             3, wxTOKEN_RET_EMPTY_ALL );
-    SingleTokenizerTest( _T("username:password:uid:gid:gecos:home:shell"), _T(":"),              7, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("1 \t3\t4  6   "),                             wxDEFAULT_DELIMITERS, 4, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("1 \t3\t4  6   "),                             wxDEFAULT_DELIMITERS, 6, wxTOKEN_RET_EMPTY     );
-    SingleTokenizerTest( _T("1 \t3\t4  6   "),                             wxDEFAULT_DELIMITERS, 9, wxTOKEN_RET_EMPTY_ALL );
-    SingleTokenizerTest( _T("01/02/99"),                                   _T("/-"),             3, wxTOKEN_DEFAULT       );
-    SingleTokenizerTest( _T("01-02/99"),                                   _T("/-"),             3, wxTOKEN_RET_DELIMS    );
-}
-
 void StringTestCase::Replace()
 {
     #define TEST_REPLACE( original , pos , len , replacement , result ) \
@@ -248,6 +388,32 @@ void StringTestCase::Replace()
     TEST_REPLACE( _T("foobar"),        3, 0, _T("-"),      _T("foo-bar")       );
     TEST_REPLACE( _T("barfoo"),        0, 6, _T("foobar"), _T("foobar")        );
 
+    
+    #define TEST_NULLCHARREPLACE( o , olen, pos , len , replacement , r, rlen ) \
+        { \
+            wxString s(o,olen); \
+            s.replace( pos , len , replacement ); \
+            CPPUNIT_ASSERT( s == wxString(r,rlen) ); \
+        }
+    
+    TEST_NULLCHARREPLACE( _T("null\0char"), 9, 5, 1, _T("d"), 
+                          _T("null\0dhar"), 9 );   
+
+    #define TEST_WXREPLACE( o , olen, olds, news, all, r, rlen ) \
+        { \
+            wxString s(o,olen); \
+            s.Replace( olds, news, all ); \
+            CPPUNIT_ASSERT( s == wxString(r,rlen) ); \
+        }
+    
+    TEST_WXREPLACE( _T("null\0char"), 9, _T("c"), _T("de"), true,
+                          _T("null\0dehar"), 10 );   
+
+    TEST_WXREPLACE( _T("null\0dehar"), 10, _T("de"), _T("c"), true,
+                          _T("null\0char"), 9 );   
+
+    #undef TEST_WXREPLACE
+    #undef TEST_NULLCHARREPLACE
     #undef TEST_REPLACE
 }
 
@@ -271,3 +437,210 @@ void StringTestCase::Match()
     #undef TEST_MATCH
 }
 
+
+void StringTestCase::CaseChanges()
+{
+    wxString s1(_T("Hello!"));
+    wxString s1u(s1);
+    wxString s1l(s1);
+    s1u.MakeUpper();
+    s1l.MakeLower();
+    wxString s2u, s2l;
+    s2u.MakeUpper();
+    s2l.MakeLower();
+
+    CPPUNIT_ASSERT( s1u == _T("HELLO!") );
+    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
+        const wchar_t capital_ya[] = { 0x42f, 0 },
+                      small_ya[]   = { 0x44f, 0 };
+
+        wxString sUpper(wxConvLibc.cWC2MB(capital_ya)),
+                 sLower(wxConvLibc.cWC2MB(small_ya));
+
+        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 );
+}
+
+void StringTestCase::ToLong()
+{
+    long l;
+    static const struct ToLongData
+    {
+        const wxChar *str;
+        long value;
+        bool ok;
+    } longData[] =
+    {
+        { _T("1"), 1, true },
+        { _T("0"), 0, true },
+        { _T("a"), 0, false },
+        { _T("12345"), 12345, true },
+        { _T("-1"), -1, true },
+        { _T("--1"), 0, false },
+    };
+
+    size_t n;
+    for ( n = 0; n < WXSIZEOF(longData); n++ )
+    {
+        const ToLongData& ld = longData[n];
+        CPPUNIT_ASSERT_EQUAL( ld.ok, wxString(ld.str).ToLong(&l) );
+        if ( ld.ok )
+            CPPUNIT_ASSERT_EQUAL( ld.value, l );
+    }
+}
+
+void StringTestCase::ToULong()
+{
+    unsigned long ul;
+    static const struct ToULongData
+    {
+        const wxChar *str;
+        unsigned long value;
+        bool ok;
+    } ulongData[] =
+    {
+        { _T("1"), 1, true },
+        { _T("0"), 0, true },
+        { _T("a"), 0, false },
+        { _T("12345"), 12345, true },
+        // this is surprizing but consistent with strtoul() behaviour
+        { _T("-1"), ULONG_MAX, true },
+    };
+
+    size_t n;
+    for ( n = 0; n < WXSIZEOF(ulongData); n++ )
+    {
+        const ToULongData& uld = ulongData[n];
+        CPPUNIT_ASSERT_EQUAL( uld.ok, wxString(uld.str).ToULong(&ul) );
+        if ( uld.ok )
+            CPPUNIT_ASSERT_EQUAL( uld.value, ul );
+    }
+}
+
+void StringTestCase::ToDouble()
+{
+    double d;
+    static const struct ToDoubleData
+    {
+        const wxChar *str;
+        double value;
+        bool ok;
+    } doubleData[] =
+    {
+        { _T("1"), 1, true },
+        { _T("1.23"), 1.23, true },
+        { _T(".1"), .1, true },
+        { _T("1."), 1, true },
+        { _T("1.."), 0, false },
+        { _T("0"), 0, true },
+        { _T("a"), 0, false },
+        { _T("12345"), 12345, true },
+        { _T("-1"), -1, true },
+        { _T("--1"), 0, false },
+    };
+
+    // we need to use decimal point, not comma or whatever is its value for the
+    // current locale
+    wxSetlocale(LC_ALL, _T("C"));
+
+    size_t n;
+    for ( n = 0; n < WXSIZEOF(doubleData); n++ )
+    {
+        const ToDoubleData& ld = doubleData[n];
+        CPPUNIT_ASSERT_EQUAL( ld.ok, wxString(ld.str).ToDouble(&d) );
+        if ( ld.ok )
+            CPPUNIT_ASSERT_EQUAL( ld.value, d );
+    }
+}