+wxFontEncoding wxFontEncFromCFStringEnc(CFStringEncoding encoding)
+{
+ wxFontEncoding enc = wxFONTENCODING_DEFAULT ;
+
+ switch( encoding)
+ {
+ case kCFStringEncodingISOLatin1 :
+ enc = wxFONTENCODING_ISO8859_1 ;
+ break ;
+ case kCFStringEncodingISOLatin2 :
+ enc = wxFONTENCODING_ISO8859_2;
+ break ;
+ case kCFStringEncodingISOLatin3 :
+ enc = wxFONTENCODING_ISO8859_3 ;
+ break ;
+ case kCFStringEncodingISOLatin4 :
+ enc = wxFONTENCODING_ISO8859_4;
+ break ;
+ case kCFStringEncodingISOLatinCyrillic :
+ enc = wxFONTENCODING_ISO8859_5;
+ break ;
+ case kCFStringEncodingISOLatinArabic :
+ enc = wxFONTENCODING_ISO8859_6;
+ break ;
+ case kCFStringEncodingISOLatinGreek :
+ enc = wxFONTENCODING_ISO8859_7;
+ break ;
+ case kCFStringEncodingISOLatinHebrew :
+ enc = wxFONTENCODING_ISO8859_8;
+ break ;
+ case kCFStringEncodingISOLatin5 :
+ enc = wxFONTENCODING_ISO8859_9;
+ break ;
+ case kCFStringEncodingISOLatin6 :
+ enc = wxFONTENCODING_ISO8859_10;
+ break ;
+ case kCFStringEncodingISOLatin7 :
+ enc = wxFONTENCODING_ISO8859_13;
+ break ;
+ case kCFStringEncodingISOLatin8 :
+ enc = wxFONTENCODING_ISO8859_14;
+ break ;
+ case kCFStringEncodingISOLatin9 :
+ enc =wxFONTENCODING_ISO8859_15 ;
+ break ;
+
+ case kCFStringEncodingKOI8_R :
+ enc = wxFONTENCODING_KOI8;
+ break ;
+
+// case :
+// enc = wxFONTENCODING_BULGARIAN;
+// break ;
+
+ case kCFStringEncodingDOSLatinUS :
+ enc = wxFONTENCODING_CP437;
+ break ;
+ case kCFStringEncodingDOSLatin1 :
+ enc = wxFONTENCODING_CP850;
+ break ;
+ case kCFStringEncodingDOSLatin2 :
+ enc =wxFONTENCODING_CP852 ;
+ break ;
+ case kCFStringEncodingDOSCyrillic :
+ enc = wxFONTENCODING_CP855;
+ break ;
+ case kCFStringEncodingDOSRussian :
+ enc = wxFONTENCODING_CP866;
+ break ;
+ case kCFStringEncodingDOSThai :
+ enc =wxFONTENCODING_CP874 ;
+ break ;
+ case kCFStringEncodingDOSJapanese :
+ enc = wxFONTENCODING_CP932;
+ break ;
+ case kCFStringEncodingDOSChineseSimplif :
+ enc = wxFONTENCODING_CP936;
+ break ;
+ case kCFStringEncodingDOSKorean :
+ enc = wxFONTENCODING_CP949;
+ break ;
+ case kCFStringEncodingDOSChineseTrad :
+ enc = wxFONTENCODING_CP950;
+ break ;
+
+ case kCFStringEncodingWindowsLatin2 :
+ enc = wxFONTENCODING_CP1250;
+ break ;
+ case kCFStringEncodingWindowsCyrillic :
+ enc = wxFONTENCODING_CP1251;
+ break ;
+ case kCFStringEncodingWindowsLatin1 :
+ enc = wxFONTENCODING_CP1252;
+ break ;
+ case kCFStringEncodingWindowsGreek :
+ enc = wxFONTENCODING_CP1253;
+ break ;
+ case kCFStringEncodingWindowsLatin5 :
+ enc = wxFONTENCODING_CP1254;
+ break ;
+ case kCFStringEncodingWindowsHebrew :
+ enc = wxFONTENCODING_CP1255;
+ break ;
+ case kCFStringEncodingWindowsArabic :
+ enc = wxFONTENCODING_CP1256;
+ break ;
+ case kCFStringEncodingWindowsBalticRim :
+ enc =wxFONTENCODING_CP1257 ;
+ break ;
+ case kCFStringEncodingEUC_JP :
+ enc = wxFONTENCODING_EUC_JP;
+ break ;
+ case kCFStringEncodingUnicode :
+ enc = wxFONTENCODING_UTF16;
+ break;
+ case kCFStringEncodingMacRoman :
+ enc = wxFONTENCODING_MACROMAN ;
+ break ;
+ case kCFStringEncodingMacJapanese :
+ enc = wxFONTENCODING_MACJAPANESE ;
+ break ;
+ case kCFStringEncodingMacChineseTrad :
+ enc = wxFONTENCODING_MACCHINESETRAD ;
+ break ;
+ case kCFStringEncodingMacKorean :
+ enc = wxFONTENCODING_MACKOREAN ;
+ break ;
+ case kCFStringEncodingMacArabic :
+ enc =wxFONTENCODING_MACARABIC ;
+ break ;
+ case kCFStringEncodingMacHebrew :
+ enc = wxFONTENCODING_MACHEBREW ;
+ break ;
+ case kCFStringEncodingMacGreek :
+ enc = wxFONTENCODING_MACGREEK ;
+ break ;
+ case kCFStringEncodingMacCyrillic :
+ enc = wxFONTENCODING_MACCYRILLIC ;
+ break ;
+ case kCFStringEncodingMacDevanagari :
+ enc = wxFONTENCODING_MACDEVANAGARI ;
+ break ;
+ case kCFStringEncodingMacGurmukhi :
+ enc = wxFONTENCODING_MACGURMUKHI ;
+ break ;
+ case kCFStringEncodingMacGujarati :
+ enc = wxFONTENCODING_MACGUJARATI ;
+ break ;
+ case kCFStringEncodingMacOriya :
+ enc =wxFONTENCODING_MACORIYA ;
+ break ;
+ case kCFStringEncodingMacBengali :
+ enc =wxFONTENCODING_MACBENGALI ;
+ break ;
+ case kCFStringEncodingMacTamil :
+ enc = wxFONTENCODING_MACTAMIL ;
+ break ;
+ case kCFStringEncodingMacTelugu :
+ enc = wxFONTENCODING_MACTELUGU ;
+ break ;
+ case kCFStringEncodingMacKannada :
+ enc = wxFONTENCODING_MACKANNADA ;
+ break ;
+ case kCFStringEncodingMacMalayalam :
+ enc = wxFONTENCODING_MACMALAJALAM ;
+ break ;
+ case kCFStringEncodingMacSinhalese :
+ enc = wxFONTENCODING_MACSINHALESE ;
+ break ;
+ case kCFStringEncodingMacBurmese :
+ enc = wxFONTENCODING_MACBURMESE ;
+ break ;
+ case kCFStringEncodingMacKhmer :
+ enc = wxFONTENCODING_MACKHMER ;
+ break ;
+ case kCFStringEncodingMacThai :
+ enc = wxFONTENCODING_MACTHAI ;
+ break ;
+ case kCFStringEncodingMacLaotian :
+ enc = wxFONTENCODING_MACLAOTIAN ;
+ break ;
+ case kCFStringEncodingMacGeorgian :
+ enc = wxFONTENCODING_MACGEORGIAN ;
+ break ;
+ case kCFStringEncodingMacArmenian :
+ enc = wxFONTENCODING_MACARMENIAN ;
+ break ;
+ case kCFStringEncodingMacChineseSimp :
+ enc = wxFONTENCODING_MACCHINESESIMP ;
+ break ;
+ case kCFStringEncodingMacTibetan :
+ enc = wxFONTENCODING_MACTIBETAN ;
+ break ;
+ case kCFStringEncodingMacMongolian :
+ enc = wxFONTENCODING_MACMONGOLIAN ;
+ break ;
+ case kCFStringEncodingMacEthiopic :
+ enc = wxFONTENCODING_MACETHIOPIC ;
+ break ;
+ case kCFStringEncodingMacCentralEurRoman:
+ enc = wxFONTENCODING_MACCENTRALEUR ;
+ break ;
+ case kCFStringEncodingMacVietnamese:
+ enc = wxFONTENCODING_MACVIATNAMESE ;
+ break ;
+ case kCFStringEncodingMacExtArabic :
+ enc = wxFONTENCODING_MACARABICEXT ;
+ break ;
+ case kCFStringEncodingMacSymbol :
+ enc = wxFONTENCODING_MACSYMBOL ;
+ break ;
+ case kCFStringEncodingMacDingbats :
+ enc = wxFONTENCODING_MACDINGBATS ;
+ break ;
+ case kCFStringEncodingMacTurkish :
+ enc = wxFONTENCODING_MACTURKISH ;
+ break ;
+ case kCFStringEncodingMacCroatian :
+ enc = wxFONTENCODING_MACCROATIAN ;
+ break ;
+ case kCFStringEncodingMacIcelandic :
+ enc = wxFONTENCODING_MACICELANDIC ;
+ break ;
+ case kCFStringEncodingMacRomanian :
+ enc = wxFONTENCODING_MACROMANIAN ;
+ break ;
+ case kCFStringEncodingMacCeltic :
+ enc = wxFONTENCODING_MACCELTIC ;
+ break ;
+ case kCFStringEncodingMacGaelic :
+ enc = wxFONTENCODING_MACGAELIC ;
+ break ;
+// case kCFStringEncodingMacKeyboardGlyphs :
+// enc = wxFONTENCODING_MACKEYBOARD ;
+// break ;
+ } ;
+ return enc ;
+}
+
+class wxMBConv_cocoa : public wxMBConv
+{
+public:
+ wxMBConv_cocoa()
+ {
+ Init(CFStringGetSystemEncoding()) ;
+ }
+
+ wxMBConv_cocoa(const wxChar* name)
+ {
+ Init( wxCFStringEncFromFontEnc(wxFontMapper::Get()->CharsetToEncoding(name, false) ) ) ;
+ }
+
+ wxMBConv_cocoa(wxFontEncoding encoding)
+ {
+ Init( wxCFStringEncFromFontEnc(encoding) );
+ }
+
+ ~wxMBConv_cocoa()
+ {
+ }
+
+ void Init( CFStringEncoding encoding)
+ {
+ m_char_encoding = encoding ;
+ m_unicode_encoding = kCFStringEncodingUnicode;
+ }
+
+ size_t MB2WC(wchar_t * szOut, const char * szUnConv, size_t nOutSize) const
+ {
+ wxASSERT(szUnConv);
+
+ size_t nBufSize = strlen(szUnConv) + 1;
+ size_t nRealOutSize;
+
+ UniChar* szUniCharBuffer = (UniChar*) szOut;
+ wchar_t* szConvBuffer = szOut;
+
+ if (szConvBuffer == NULL && nOutSize != 0)
+ {
+ szConvBuffer = new wchar_t[nOutSize] ;
+ }
+
+#if SIZEOF_WCHAR_T == 4
+ szUniCharBuffer = new UniChar[nOutSize];
+#endif
+
+ CFDataRef theData = CFDataCreateWithBytesNoCopy (
+ NULL, //allocator
+ (const UInt8*)szUnConv,
+ nBufSize - 1,
+ NULL //deallocator
+ );
+
+ wxASSERT(theData);
+
+ CFStringRef theString = CFStringCreateFromExternalRepresentation (
+ NULL,
+ theData,
+ m_char_encoding
+ );
+
+ wxASSERT(theString);
+
+ if (nOutSize == 0)
+ {
+ nRealOutSize = CFStringGetLength(theString) + 1;
+ CFRelease(theString);
+ return nRealOutSize - 1;
+ }
+
+ CFRange theRange = { 0, CFStringGetLength(theString) };
+
+ CFStringGetCharacters(theString, theRange, szUniCharBuffer);
+
+
+ nRealOutSize = (CFStringGetLength(theString) + 1);
+
+ CFRelease(theString);
+
+ szUniCharBuffer[nRealOutSize-1] = '\0' ;
+
+#if SIZEOF_WCHAR_T == 4
+ wxMBConvUTF16 converter ;
+ converter.MB2WC(szConvBuffer , (const char*)szUniCharBuffer , nRealOutSize ) ;
+ delete[] szUniCharBuffer;
+#endif
+ if ( szOut == NULL )
+ delete [] szConvBuffer;
+
+ return nRealOutSize ;
+ }
+
+ size_t WC2MB(char *szOut, const wchar_t *szUnConv, size_t nOutSize) const
+ {
+ size_t nBufSize = wxWcslen(szUnConv) + 1;
+ size_t nRealOutSize;
+ char* szBuffer = szOut;
+ UniChar* szUniBuffer = (UniChar*) szUnConv;
+
+ if (szOut == NULL)
+ {
+ // worst case
+ nRealOutSize = wxString::WorstEncodingCase(nBufSize - 1, *this)+1 ;
+ szBuffer = new char[ nRealOutSize ] ;
+ }
+ else
+ nRealOutSize = nOutSize;
+
+#if SIZEOF_WCHAR_T == 4
+ wxMBConvUTF16BE converter ;
+ nBufSize = converter.WC2MB( NULL , szUnConv , 0 );
+ szUniBuffer = new UniChar[ (nBufSize / sizeof(UniChar)) + 1] ;
+ converter.WC2MB( (char*) szUniBuffer , szUnConv, nBufSize + sizeof(UniChar)) ;
+ nBufSize /= sizeof(UniChar);
+ ++nBufSize;
+#endif
+
+ CFStringRef theString = CFStringCreateWithCharactersNoCopy(
+ NULL, //allocator
+ szUniBuffer,
+ nBufSize,
+ NULL //deallocator
+ );
+
+ wxASSERT(theString);
+
+ //Note that CER puts a BOM when converting to unicode
+ //so we may want to check and use getchars instead in that case
+ CFDataRef theData = CFStringCreateExternalRepresentation(
+ NULL, //allocator
+ theString,
+ m_char_encoding,
+ 0 //what to put in characters that can't be converted -
+ //0 tells CFString to return NULL if it meets such a character
+ );
+
+ if(!theData)
+ return (size_t)-1;
+
+ CFRelease(theString);
+
+ nRealOutSize = CFDataGetLength(theData);
+
+ if ( szOut == NULL )
+ delete[] szBuffer;
+
+ if(nOutSize == 0)
+ {
+//TODO: This gets flagged as a non-malloced address by the debugger...
+//#if SIZEOF_WCHAR_T == 4
+// delete[] szUniBuffer;
+//#endif
+ CFRelease(theData);
+ return nRealOutSize - 1;
+ }
+
+ CFRange theRange = {0, CFDataGetLength(theData) };
+ CFDataGetBytes(theData, theRange, (UInt8*) szBuffer);
+
+ CFRelease(theData);
+
+//TODO: This gets flagged as a non-malloced address by the debugger...
+//#if SIZEOF_WCHAR_T == 4
+// delete[] szUniBuffer;
+//#endif
+ return nRealOutSize - 1;
+ }
+
+ bool IsOk() const
+ {
+ //TODO: check for invalid en/de/coding
+ return true;
+ }
+
+private:
+ CFStringEncoding m_char_encoding ;
+ CFStringEncoding m_unicode_encoding ;
+};
+
+#endif // defined(__WXCOCOA__)
+
+// ============================================================================
+// Mac conversion classes
+// ============================================================================
+
+#if defined(__WXMAC__) && defined(TARGET_CARBON)
+
+class wxMBConv_mac : public wxMBConv
+{
+public:
+ wxMBConv_mac()
+ {
+ Init(CFStringGetSystemEncoding()) ;
+ }
+
+ wxMBConv_mac(const wxChar* name)
+ {
+ Init( wxMacGetSystemEncFromFontEnc(wxFontMapper::Get()->CharsetToEncoding(name, false) ) ) ;
+ }
+
+ wxMBConv_mac(wxFontEncoding encoding)
+ {
+ Init( wxMacGetSystemEncFromFontEnc(encoding) );
+ }
+
+ ~wxMBConv_mac()
+ {
+ OSStatus status = noErr ;
+ status = TECDisposeConverter(m_MB2WC_converter);
+ status = TECDisposeConverter(m_WC2MB_converter);
+ }
+
+
+ void Init( TextEncodingBase encoding)
+ {
+ OSStatus status = noErr ;
+ m_char_encoding = encoding ;
+ m_unicode_encoding = CreateTextEncoding(kTextEncodingUnicodeDefault,0,kUnicode16BitFormat) ;
+
+ status = TECCreateConverter(&m_MB2WC_converter,
+ m_char_encoding,
+ m_unicode_encoding);
+ status = TECCreateConverter(&m_WC2MB_converter,
+ m_unicode_encoding,
+ m_char_encoding);
+ }
+
+ size_t MB2WC(wchar_t *buf, const char *psz, size_t n) const
+ {
+ OSStatus status = noErr ;
+ ByteCount byteOutLen ;
+ ByteCount byteInLen = strlen(psz) ;
+ wchar_t *tbuf = NULL ;
+ UniChar* ubuf = NULL ;
+ size_t res = 0 ;
+
+ if (buf == NULL)
+ {
+ n = byteInLen ;
+ tbuf = (wchar_t*) malloc( n * SIZEOF_WCHAR_T) ;
+ }
+ ByteCount byteBufferLen = n * sizeof( UniChar ) ;
+#if SIZEOF_WCHAR_T == 4
+ ubuf = (UniChar*) malloc( byteBufferLen + 2 ) ;
+#else
+ ubuf = (UniChar*) (buf ? buf : tbuf) ;
+#endif
+ status = TECConvertText(m_MB2WC_converter, (ConstTextPtr) psz , byteInLen, &byteInLen,
+ (TextPtr) ubuf , byteBufferLen, &byteOutLen);
+#if SIZEOF_WCHAR_T == 4
+ // we have to terminate here, because n might be larger for the trailing zero, and if UniChar
+ // is not properly terminated we get random characters at the end
+ ubuf[byteOutLen / sizeof( UniChar ) ] = 0 ;
+ wxMBConvUTF16BE converter ;
+ res = converter.MB2WC( (buf ? buf : tbuf) , (const char*)ubuf , n ) ;
+ free( ubuf ) ;
+#else
+ res = byteOutLen / sizeof( UniChar ) ;
+#endif
+ if ( buf == NULL )
+ free(tbuf) ;
+
+ if ( buf && res < n)
+ buf[res] = 0;
+
+ return res ;
+ }
+
+ size_t WC2MB(char *buf, const wchar_t *psz, size_t n) const
+ {
+ OSStatus status = noErr ;
+ ByteCount byteOutLen ;
+ ByteCount byteInLen = wxWcslen(psz) * SIZEOF_WCHAR_T ;
+
+ char *tbuf = NULL ;
+
+ if (buf == NULL)
+ {
+ // worst case
+ n = wxString::WorstEncodingCase(byteInLen / SIZEOF_WCHAR_T, *this) + SIZEOF_WCHAR_T;
+ tbuf = (char*) malloc( n ) ;
+ }
+
+ ByteCount byteBufferLen = n ;
+ UniChar* ubuf = NULL ;
+#if SIZEOF_WCHAR_T == 4
+ wxMBConvUTF16BE converter ;
+ size_t unicharlen = converter.WC2MB( NULL , psz , 0 ) ;
+ byteInLen = unicharlen ;
+ ubuf = (UniChar*) malloc( byteInLen + 2 ) ;
+ converter.WC2MB( (char*) ubuf , psz, unicharlen + 2 ) ;
+#else
+ ubuf = (UniChar*) psz ;
+#endif
+ status = TECConvertText(m_WC2MB_converter, (ConstTextPtr) ubuf , byteInLen, &byteInLen,
+ (TextPtr) (buf ? buf : tbuf) , byteBufferLen, &byteOutLen);
+#if SIZEOF_WCHAR_T == 4
+ free( ubuf ) ;
+#endif
+ if ( buf == NULL )
+ free(tbuf) ;
+
+ size_t res = byteOutLen ;
+ if ( buf && res < n)
+ buf[res] = 0;
+
+ return res ;
+ }
+
+ bool IsOk() const
+ { return m_MB2WC_converter != NULL && m_WC2MB_converter != NULL ; }
+
+private:
+ TECObjectRef m_MB2WC_converter ;
+ TECObjectRef m_WC2MB_converter ;
+
+ TextEncodingBase m_char_encoding ;
+ TextEncodingBase m_unicode_encoding ;
+};
+
+#endif // defined(__WXMAC__) && defined(TARGET_CARBON)
+
+// ============================================================================
+// wxEncodingConverter based conversion classes
+// ============================================================================
+
+#if wxUSE_FONTMAP
+
+class wxMBConv_wxwin : public wxMBConv
+{
+private:
+ void Init()
+ {
+ m_ok = m2w.Init(m_enc, wxFONTENCODING_UNICODE) &&
+ w2m.Init(wxFONTENCODING_UNICODE, m_enc);
+ }
+
+public:
+ // temporarily just use wxEncodingConverter stuff,
+ // so that it works while a better implementation is built
+ wxMBConv_wxwin(const wxChar* name)
+ {
+ if (name)
+ m_enc = wxFontMapper::Get()->CharsetToEncoding(name, false);
+ else
+ m_enc = wxFONTENCODING_SYSTEM;
+
+ Init();
+ }
+
+ wxMBConv_wxwin(wxFontEncoding enc)
+ {
+ m_enc = enc;
+
+ Init();
+ }
+
+ size_t MB2WC(wchar_t *buf, const char *psz, size_t WXUNUSED(n)) const
+ {
+ size_t inbuf = strlen(psz);
+ if (buf)
+ m2w.Convert(psz,buf);
+ return inbuf;
+ }
+
+ size_t WC2MB(char *buf, const wchar_t *psz, size_t WXUNUSED(n)) const
+ {
+ const size_t inbuf = wxWcslen(psz);
+ if (buf)
+ w2m.Convert(psz,buf);
+
+ return inbuf;
+ }
+
+ bool IsOk() const { return m_ok; }
+
+public:
+ wxFontEncoding m_enc;
+ wxEncodingConverter m2w, w2m;
+
+ // were we initialized successfully?
+ bool m_ok;
+
+ DECLARE_NO_COPY_CLASS(wxMBConv_wxwin)
+};
+
+#endif // wxUSE_FONTMAP
+