X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/02272c9c3e902dbc910b702744b750d2b978f4bb..a018a119bdcd0fd3d2ba8161c9db3e9b1c6319ff:/src/common/strconv.cpp diff --git a/src/common/strconv.cpp b/src/common/strconv.cpp index 7b39e4bb3f..011d3280b8 100644 --- a/src/common/strconv.cpp +++ b/src/common/strconv.cpp @@ -168,7 +168,7 @@ static size_t decode_utf16(const wxUint16* input, wxUint32& output) wxMBConv::~wxMBConv() { - // nothing to do here + // nothing to do here (necessary for Darwin linking probably) } const wxWCharBuffer wxMBConv::cMB2WC(const char *psz) const @@ -1121,7 +1121,7 @@ size_t wxMBConv_iconv::WC2MB(char *buf, const wchar_t *psz, size_t n) const #ifdef wxHAVE_WIN32_MB2WC // from utils.cpp -#if wxUSE_FONTMAP +#if wxUSE_FONTMAP extern WXDLLIMPEXP_BASE long wxCharsetToCodepage(const wxChar *charset); extern WXDLLIMPEXP_BASE long wxEncodingToCodepage(wxFontEncoding encoding); #endif @@ -1177,7 +1177,7 @@ public: quality approximations such as turning "1/2" symbol (U+00BD) into "1" for the code pages which don't have it and we, obviously, want to avoid this at any price - + the trouble is that this function does it _silently_, i.e. it won't even tell us whether it did or not... Win98/2000 and higher provide WC_NO_BEST_FIT_CHARS but it doesn't work for the older systems and @@ -1190,8 +1190,8 @@ public: // it doesn't work with CJK encodings (which we test for rather roughly // here...) nor with UTF-7/8 nor, of course, with Windows versions not // supporting it - BOOL usedDef wxDUMMY_INITIALIZE(false), - *pUsedDef; + BOOL usedDef wxDUMMY_INITIALIZE(false); + BOOL *pUsedDef; int flags; if ( CanUseNoBestFit() && m_CodePage < 50000 ) { @@ -1286,6 +1286,701 @@ private: #endif // wxHAVE_WIN32_MB2WC +// ============================================================================ +// Cocoa conversion classes +// ============================================================================ + +#if defined(__WXCOCOA__) + +// RN: There is no UTF-32 support in either Core Foundation or +// Cocoa. Strangely enough, internally Core Foundation uses +// UTF 32 internally quite a bit - its just not public (yet). + +#include +#include + +CFStringEncoding wxCFStringEncFromFontEnc(wxFontEncoding encoding) +{ + CFStringEncoding enc = 0 ; + if ( encoding == wxFONTENCODING_DEFAULT ) + { +#if wxUSE_GUI + encoding = wxFont::GetDefaultEncoding() ; +#else + encoding = wxLocale::GetSystemEncoding() ; +#endif + } + else switch( encoding) + { + case wxFONTENCODING_ISO8859_1 : + enc = kCFStringEncodingISOLatin1 ; + break ; + case wxFONTENCODING_ISO8859_2 : + enc = kCFStringEncodingISOLatin2; + break ; + case wxFONTENCODING_ISO8859_3 : + enc = kCFStringEncodingISOLatin3 ; + break ; + case wxFONTENCODING_ISO8859_4 : + enc = kCFStringEncodingISOLatin4; + break ; + case wxFONTENCODING_ISO8859_5 : + enc = kCFStringEncodingISOLatinCyrillic; + break ; + case wxFONTENCODING_ISO8859_6 : + enc = kCFStringEncodingISOLatinArabic; + break ; + case wxFONTENCODING_ISO8859_7 : + enc = kCFStringEncodingISOLatinGreek; + break ; + case wxFONTENCODING_ISO8859_8 : + enc = kCFStringEncodingISOLatinHebrew; + break ; + case wxFONTENCODING_ISO8859_9 : + enc = kCFStringEncodingISOLatin5; + break ; + case wxFONTENCODING_ISO8859_10 : + enc = kCFStringEncodingISOLatin6; + break ; + case wxFONTENCODING_ISO8859_11 : + enc = kCFStringEncodingISOLatinThai; + break ; + case wxFONTENCODING_ISO8859_13 : + enc = kCFStringEncodingISOLatin7; + break ; + case wxFONTENCODING_ISO8859_14 : + enc = kCFStringEncodingISOLatin8; + break ; + case wxFONTENCODING_ISO8859_15 : + enc = kCFStringEncodingISOLatin9; + break ; + + case wxFONTENCODING_KOI8 : + enc = kCFStringEncodingKOI8_R; + break ; + case wxFONTENCODING_ALTERNATIVE : // MS-DOS CP866 + enc = kCFStringEncodingDOSRussian; + break ; + +// case wxFONTENCODING_BULGARIAN : +// enc = ; +// break ; + + case wxFONTENCODING_CP437 : + enc =kCFStringEncodingDOSLatinUS ; + break ; + case wxFONTENCODING_CP850 : + enc = kCFStringEncodingDOSLatin1; + break ; + case wxFONTENCODING_CP852 : + enc = kCFStringEncodingDOSLatin2; + break ; + case wxFONTENCODING_CP855 : + enc = kCFStringEncodingDOSCyrillic; + break ; + case wxFONTENCODING_CP866 : + enc =kCFStringEncodingDOSRussian ; + break ; + case wxFONTENCODING_CP874 : + enc = kCFStringEncodingDOSThai; + break ; + case wxFONTENCODING_CP932 : + enc = kCFStringEncodingDOSJapanese; + break ; + case wxFONTENCODING_CP936 : + enc =kCFStringEncodingDOSChineseSimplif ; + break ; + case wxFONTENCODING_CP949 : + enc = kCFStringEncodingDOSKorean; + break ; + case wxFONTENCODING_CP950 : + enc = kCFStringEncodingDOSChineseTrad; + break ; + + case wxFONTENCODING_CP1250 : + enc = kCFStringEncodingWindowsLatin2; + break ; + case wxFONTENCODING_CP1251 : + enc =kCFStringEncodingWindowsCyrillic ; + break ; + case wxFONTENCODING_CP1252 : + enc =kCFStringEncodingWindowsLatin1 ; + break ; + case wxFONTENCODING_CP1253 : + enc = kCFStringEncodingWindowsGreek; + break ; + case wxFONTENCODING_CP1254 : + enc = kCFStringEncodingWindowsLatin5; + break ; + case wxFONTENCODING_CP1255 : + enc =kCFStringEncodingWindowsHebrew ; + break ; + case wxFONTENCODING_CP1256 : + enc =kCFStringEncodingWindowsArabic ; + break ; + case wxFONTENCODING_CP1257 : + enc = kCFStringEncodingWindowsBalticRim; + break ; + case wxFONTENCODING_UTF7 : + enc = kCFStringEncodingNonLossyASCII ; + break ; + case wxFONTENCODING_UTF8 : + enc = kCFStringEncodingUTF8 ; + break ; + case wxFONTENCODING_EUC_JP : + enc = kCFStringEncodingEUC_JP; + break ; + case wxFONTENCODING_UTF16 : + enc = kCFStringEncodingUnicode ; + break ; + case wxFONTENCODING_MACROMAN : + enc = kCFStringEncodingMacRoman ; + break ; + case wxFONTENCODING_MACJAPANESE : + enc = kCFStringEncodingMacJapanese ; + break ; + case wxFONTENCODING_MACCHINESETRAD : + enc = kCFStringEncodingMacChineseTrad ; + break ; + case wxFONTENCODING_MACKOREAN : + enc = kCFStringEncodingMacKorean ; + break ; + case wxFONTENCODING_MACARABIC : + enc = kCFStringEncodingMacArabic ; + break ; + case wxFONTENCODING_MACHEBREW : + enc = kCFStringEncodingMacHebrew ; + break ; + case wxFONTENCODING_MACGREEK : + enc = kCFStringEncodingMacGreek ; + break ; + case wxFONTENCODING_MACCYRILLIC : + enc = kCFStringEncodingMacCyrillic ; + break ; + case wxFONTENCODING_MACDEVANAGARI : + enc = kCFStringEncodingMacDevanagari ; + break ; + case wxFONTENCODING_MACGURMUKHI : + enc = kCFStringEncodingMacGurmukhi ; + break ; + case wxFONTENCODING_MACGUJARATI : + enc = kCFStringEncodingMacGujarati ; + break ; + case wxFONTENCODING_MACORIYA : + enc = kCFStringEncodingMacOriya ; + break ; + case wxFONTENCODING_MACBENGALI : + enc = kCFStringEncodingMacBengali ; + break ; + case wxFONTENCODING_MACTAMIL : + enc = kCFStringEncodingMacTamil ; + break ; + case wxFONTENCODING_MACTELUGU : + enc = kCFStringEncodingMacTelugu ; + break ; + case wxFONTENCODING_MACKANNADA : + enc = kCFStringEncodingMacKannada ; + break ; + case wxFONTENCODING_MACMALAJALAM : + enc = kCFStringEncodingMacMalayalam ; + break ; + case wxFONTENCODING_MACSINHALESE : + enc = kCFStringEncodingMacSinhalese ; + break ; + case wxFONTENCODING_MACBURMESE : + enc = kCFStringEncodingMacBurmese ; + break ; + case wxFONTENCODING_MACKHMER : + enc = kCFStringEncodingMacKhmer ; + break ; + case wxFONTENCODING_MACTHAI : + enc = kCFStringEncodingMacThai ; + break ; + case wxFONTENCODING_MACLAOTIAN : + enc = kCFStringEncodingMacLaotian ; + break ; + case wxFONTENCODING_MACGEORGIAN : + enc = kCFStringEncodingMacGeorgian ; + break ; + case wxFONTENCODING_MACARMENIAN : + enc = kCFStringEncodingMacArmenian ; + break ; + case wxFONTENCODING_MACCHINESESIMP : + enc = kCFStringEncodingMacChineseSimp ; + break ; + case wxFONTENCODING_MACTIBETAN : + enc = kCFStringEncodingMacTibetan ; + break ; + case wxFONTENCODING_MACMONGOLIAN : + enc = kCFStringEncodingMacMongolian ; + break ; + case wxFONTENCODING_MACETHIOPIC : + enc = kCFStringEncodingMacEthiopic ; + break ; + case wxFONTENCODING_MACCENTRALEUR : + enc = kCFStringEncodingMacCentralEurRoman ; + break ; + case wxFONTENCODING_MACVIATNAMESE : + enc = kCFStringEncodingMacVietnamese ; + break ; + case wxFONTENCODING_MACARABICEXT : + enc = kCFStringEncodingMacExtArabic ; + break ; + case wxFONTENCODING_MACSYMBOL : + enc = kCFStringEncodingMacSymbol ; + break ; + case wxFONTENCODING_MACDINGBATS : + enc = kCFStringEncodingMacDingbats ; + break ; + case wxFONTENCODING_MACTURKISH : + enc = kCFStringEncodingMacTurkish ; + break ; + case wxFONTENCODING_MACCROATIAN : + enc = kCFStringEncodingMacCroatian ; + break ; + case wxFONTENCODING_MACICELANDIC : + enc = kCFStringEncodingMacIcelandic ; + break ; + case wxFONTENCODING_MACROMANIAN : + enc = kCFStringEncodingMacRomanian ; + break ; + case wxFONTENCODING_MACCELTIC : + enc = kCFStringEncodingMacCeltic ; + break ; + case wxFONTENCODING_MACGAELIC : + enc = kCFStringEncodingMacGaelic ; + break ; +// case wxFONTENCODING_MACKEYBOARD : +// enc = kCFStringEncodingMacKeyboardGlyphs ; +// break ; + default : + // because gcc is picky + break ; + } ; + return enc ; +} + +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 = ((nBufSize - 1) << 1)+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 // ============================================================================ @@ -1302,127 +1997,127 @@ public: wxMBConv_mac(const wxChar* name) { - Init( wxMacGetSystemEncFromFontEnc(wxFontMapper::Get()->CharsetToEncoding(name, FALSE) ) ) ; + 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); - } - + 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 ) ; + 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 ) ; + ubuf = (UniChar*) malloc( byteBufferLen + 2 ) ; #else - ubuf = (UniChar*) (buf ? buf : tbuf) ; + ubuf = (UniChar*) (buf ? buf : tbuf) ; #endif - status = TECConvertText(m_MB2WC_converter, (ConstTextPtr) psz , byteInLen, &byteInLen, - (TextPtr) ubuf , byteBufferLen, &byteOutLen); + 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 ) ; + wxMBConvUTF16BE converter ; + res = converter.MB2WC( (buf ? buf : tbuf) , (const char*)ubuf , n ) ; + free( ubuf ) ; #else - res = byteOutLen / sizeof( UniChar ) ; + res = byteOutLen / sizeof( UniChar ) ; #endif - if ( buf == NULL ) - free(tbuf) ; + if ( buf == NULL ) + free(tbuf) ; if ( buf && res < n) buf[res] = 0; - return res ; + 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 = byteInLen * 2 ; - tbuf = (char*) malloc( n ) ; - } - - ByteCount byteBufferLen = n ; - UniChar* ubuf = NULL ; + { + OSStatus status = noErr ; + ByteCount byteOutLen ; + ByteCount byteInLen = wxWcslen(psz) * SIZEOF_WCHAR_T ; + + char *tbuf = NULL ; + + if (buf == NULL) + { + // worst case + n = byteInLen * 2 ; + 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 ) ; + 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 ; + ubuf = (UniChar*) psz ; #endif - status = TECConvertText(m_WC2MB_converter, (ConstTextPtr) ubuf , byteInLen, &byteInLen, - (TextPtr) (buf ? buf : tbuf) , byteBufferLen, &byteOutLen); + status = TECConvertText(m_WC2MB_converter, (ConstTextPtr) ubuf , byteInLen, &byteInLen, + (TextPtr) (buf ? buf : tbuf) , byteBufferLen, &byteOutLen); #if SIZEOF_WCHAR_T == 4 - free( ubuf ) ; + free( ubuf ) ; #endif - if ( buf == NULL ) - free(tbuf) ; + if ( buf == NULL ) + free(tbuf) ; - size_t res = byteOutLen ; + size_t res = byteOutLen ; if ( buf && res < n) buf[res] = 0; - return res ; + 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 ; + TECObjectRef m_MB2WC_converter ; + TECObjectRef m_WC2MB_converter ; + + TextEncodingBase m_char_encoding ; + TextEncodingBase m_unicode_encoding ; }; #endif // defined(__WXMAC__) && defined(TARGET_CARBON) @@ -1626,18 +2321,32 @@ wxMBConv *wxCSConv::DoCreate() const #endif } #endif // wxHAVE_WIN32_MB2WC -#if defined(__WXMAC__) +#if defined(__WXMAC__) { - if ( m_name || ( m_encoding < wxFONTENCODING_UTF16BE ) ) - { - - wxMBConv_mac *conv = m_name ? new wxMBConv_mac(m_name) - : new wxMBConv_mac(m_encoding); - if ( conv->IsOk() ) - return conv; + if ( m_name || ( m_encoding < wxFONTENCODING_UTF16BE ) ) + { - delete conv; - } + wxMBConv_mac *conv = m_name ? new wxMBConv_mac(m_name) + : new wxMBConv_mac(m_encoding); + if ( conv->IsOk() ) + return conv; + + delete conv; + } + } +#endif +#if defined(__WXCOCOA__) + { + if ( m_name || ( m_encoding <= wxFONTENCODING_UTF16 ) ) + { + + wxMBConv_cocoa *conv = m_name ? new wxMBConv_cocoa(m_name) + : new wxMBConv_cocoa(m_encoding); + if ( conv->IsOk() ) + return conv; + + delete conv; + } } #endif // step (2) @@ -1769,7 +2478,7 @@ size_t wxCSConv::WC2MB(char *buf, const wchar_t *psz, size_t n) const { if (psz[c] > 0xFF) return (size_t)-1; - buf[c] = psz[c]; + buf[c] = (char)psz[c]; } } else