X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/307fd9567ea20ad2ecc1312b22063f16eabbe377..949750de631b2c2ee59c6e9080e7d38fa66c0167:/src/common/encconv.cpp?ds=sidebyside diff --git a/src/common/encconv.cpp b/src/common/encconv.cpp index b4e88adf63..80642b96f1 100644 --- a/src/common/encconv.cpp +++ b/src/common/encconv.cpp @@ -4,13 +4,9 @@ // font encodings // Author: Vaclav Slavik // Copyright: (c) 1999 Vaclav Slavik -// Licence: wxWindows Licence +// Licence: wxWindows licence ///////////////////////////////////////////////////////////////////////////// -#ifdef __GNUG__ -#pragma implementation "encconv.h" -#endif - // For compilers that support precompilation, includes "wx.h". #include "wx/wxprec.h" @@ -18,27 +14,63 @@ #pragma hdrstop #endif -#if wxUSE_FONTMAP - #include "wx/encconv.h" #include // conversion tables, generated by scripts in $(WXWIN)/misc/unictabl: -#ifdef __BORLANDC__ -#include "../common/unictabl.inc" +#if defined( __BORLANDC__ ) || defined(__DARWIN__) + #include "../common/unictabl.inc" #else -#include "unictabl.inc" + #include "unictabl.inc" #endif #if wxUSE_WCHAR_T -typedef wchar_t tchar; + typedef wchar_t tchar; #else -typedef char tchar; + typedef char tchar; +#endif + +#ifdef __WXMAC__ + #include "wx/osx/core/cfstring.h" + #include + + wxUint16 gMacEncodings[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1][128] ; + bool gMacEncodingsInited[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1] ; +#endif + +#ifdef __WXWINCE__ + #include "wx/msw/wince/missing.h" // for bsearch() #endif -static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc) +static const wxUint16* GetEncTable(wxFontEncoding enc) { +#ifdef __WXMAC__ + if( enc >= wxFONTENCODING_MACMIN && enc <= wxFONTENCODING_MACMAX ) + { + int i = enc-wxFONTENCODING_MACMIN ; + if ( gMacEncodingsInited[i] == false ) + { + // create + CFStringEncoding cfencoding = wxMacGetSystemEncFromFontEnc( enc ) ; + if( !CFStringIsEncodingAvailable( cfencoding ) ) + return NULL; + + memset( gMacEncodings[i] , 0 , 128 * 2 ); + char s[2] = { 0 , 0 }; + CFRange firstchar = CFRangeMake( 0, 1 ); + for( unsigned char c = 255 ; c >= 128 ; --c ) + { + s[0] = c ; + wxCFStringRef cfref( CFStringCreateWithCStringNoCopy( NULL, s, cfencoding , kCFAllocatorNull ) ); + CFStringGetCharacters( cfref, firstchar, (UniChar*) &gMacEncodings[i][c-128] ); + } + gMacEncodingsInited[i]=true; + } + return gMacEncodings[i] ; + } +#endif + for (int i = 0; encodings_list[i].table != NULL; i++) { if (encodings_list[i].encoding == enc) @@ -52,20 +84,19 @@ typedef struct { wxUint8 c; } CharsetItem; - - -extern "C" int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2) +extern "C" int wxCMPFUNC_CONV +CompareCharsetItems(const void *i1, const void *i2) { return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u ); } -static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl) +static CharsetItem* BuildReverseTable(const wxUint16 *tbl) { CharsetItem *rev = new CharsetItem[128]; for (int i = 0; i < 128; i++) - rev[i].c = 128 + i, rev[i].u = tbl[i]; + rev[i].c = wxUint8(128 + i), rev[i].u = tbl[i]; qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems); @@ -77,8 +108,8 @@ static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl) wxEncodingConverter::wxEncodingConverter() { m_Table = NULL; - m_UnicodeInput = m_UnicodeOutput = FALSE; - m_JustCopy = FALSE; + m_UnicodeInput = m_UnicodeOutput = false; + m_JustCopy = false; } @@ -86,27 +117,27 @@ wxEncodingConverter::wxEncodingConverter() bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method) { unsigned i; - wxUint16 *in_tbl = NULL, *out_tbl = NULL; + const wxUint16 *in_tbl; + const wxUint16 *out_tbl = NULL; if (m_Table) {delete[] m_Table; m_Table = NULL;} #if !wxUSE_WCHAR_T - if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE; + if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return false; #endif - if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;} + if (input_enc == output_enc) {m_JustCopy = true; return true;} m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE); - m_JustCopy = FALSE; + m_JustCopy = false; if (input_enc == wxFONTENCODING_UNICODE) { - if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE; + if ((out_tbl = GetEncTable(output_enc)) == NULL) return false; m_Table = new tchar[65536]; for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII - for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?'; - // FIXME - this should be character that means `unicode to charset' impossible, not '?' + for (i = 128; i < 65536; i++) m_Table[i] = (tchar)0; if (method == wxCONVERT_SUBSTITUTE) { @@ -117,17 +148,15 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e for (i = 0; i < 128; i++) m_Table[out_tbl[i]] = (tchar)(128 + i); - m_UnicodeInput = TRUE; - return TRUE; + m_UnicodeInput = true; } - - else + else // input !Unicode { - if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE; + if ((in_tbl = GetEncTable(input_enc)) == NULL) return false; if (output_enc != wxFONTENCODING_UNICODE) - if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE; + if ((out_tbl = GetEncTable(output_enc)) == NULL) return false; - m_UnicodeInput = FALSE; + m_UnicodeInput = false; m_Table = new tchar[256]; for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII @@ -135,9 +164,9 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e if (output_enc == wxFONTENCODING_UNICODE) { for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i]; - return TRUE; + return true; } - else + else // output !Unicode { CharsetItem *rev = BuildReverseTable(out_tbl); CharsetItem *item; @@ -161,14 +190,28 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e } delete[] rev; - return TRUE; } } + + return true; } +#define REPLACEMENT_CHAR ((tchar)'?') -void wxEncodingConverter::Convert(const char* input, char* output) +inline tchar GetTableValue(const tchar *table, tchar value, bool& repl) +{ + tchar r = table[value]; + if (r == 0 && value != 0) + { + r = REPLACEMENT_CHAR; + repl = true; + } + return r; +} + + +bool wxEncodingConverter::Convert(const char* input, char* output) const { wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!")); wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!")); @@ -179,20 +222,25 @@ void wxEncodingConverter::Convert(const char* input, char* output) if (m_JustCopy) { strcpy(output, input); - return; + return true; } - wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + wxCHECK_MSG(m_Table != NULL, false, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + bool replaced = false; for (i = input, o = output; *i != 0;) - *(o++) = (char)(m_Table[(wxUint8)*(i++)]); + *(o++) = (char)(GetTableValue(m_Table, (wxUint8)*(i++), replaced)); *o = 0; + + return !replaced; } #if wxUSE_WCHAR_T -void wxEncodingConverter::Convert(const char* input, wchar_t* output) +bool wxEncodingConverter::Convert(const char* input, wchar_t* output) const { wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!")); wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!")); @@ -205,19 +253,24 @@ void wxEncodingConverter::Convert(const char* input, wchar_t* output) for (i = input, o = output; *i != 0;) *(o++) = (wchar_t)(*(i++)); *o = 0; - return; + return true; } - wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + wxCHECK_MSG(m_Table != NULL, false, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + bool replaced = false; for (i = input, o = output; *i != 0;) - *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); + *(o++) = (wchar_t)(GetTableValue(m_Table, (wxUint8)*(i++), replaced)); *o = 0; + + return !replaced; } -void wxEncodingConverter::Convert(const wchar_t* input, char* output) +bool wxEncodingConverter::Convert(const wchar_t* input, char* output) const { wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!")); wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!")); @@ -230,19 +283,24 @@ void wxEncodingConverter::Convert(const wchar_t* input, char* output) for (i = input, o = output; *i != 0;) *(o++) = (char)(*(i++)); *o = 0; - return; + return true; } - wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + wxCHECK_MSG(m_Table != NULL, false, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + bool replaced = false; for (i = input, o = output; *i != 0;) - *(o++) = (char)(m_Table[(wxUint16)*(i++)]); + *(o++) = (char)(GetTableValue(m_Table, (wxUint16)*(i++), replaced)); *o = 0; + + return !replaced; } -void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) +bool wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) const { wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!")); wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!")); @@ -256,27 +314,33 @@ void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) for (i = input, o = output; *i != 0;) *(o++) = (*(i++)); *o = 0; - return; + return true; } - wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + wxCHECK_MSG(m_Table != NULL, false, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); + bool replaced = false; + for (i = input, o = output; *i != 0;) - *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); + *(o++) = (wchar_t)(GetTableValue(m_Table, (wxUint8)*(i++), replaced)); *o = 0; + + return !replaced; } #endif // wxUSE_WCHAR_T -wxString wxEncodingConverter::Convert(const wxString& input) +wxString wxEncodingConverter::Convert(const wxString& input) const { if (m_JustCopy) return input; wxString s; const wxChar *i; - wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + wxCHECK_MSG(m_Table != NULL, s, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); if (m_UnicodeInput) { @@ -304,11 +368,12 @@ wxString wxEncodingConverter::Convert(const wxString& input) #define STOP wxFONTENCODING_SYSTEM #define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/ -#define ENC_PER_PLATFORM 5 - // max no. of encodings for one language used on one platform - // Anybody thinks 5 is not enough? ;-) +#define ENC_PER_PLATFORM 3 + // max no. of encodings for one language used on one platform. + // Using maximum of everything at the current moment to not make the + // library larger than necessary. Make larger only if necessary - MR -static wxFontEncoding +static const wxFontEncoding EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = { // *** Please put more common encodings as first! *** @@ -318,7 +383,7 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP}, /* windows */ {wxFONTENCODING_CP1252, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACROMAN, STOP} }, // Central European @@ -326,7 +391,7 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_2, STOP}, /* windows */ {wxFONTENCODING_CP1250, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACCENTRALEUR, STOP} }, // Baltic @@ -342,7 +407,7 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_8, STOP}, /* windows */ {wxFONTENCODING_CP1255, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACHEBREW, STOP} }, // Greek @@ -350,7 +415,7 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_7, STOP}, /* windows */ {wxFONTENCODING_CP1253, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACGREEK, STOP} }, // Arabic @@ -358,7 +423,7 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_6, STOP}, /* windows */ {wxFONTENCODING_CP1256, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACARABIC, STOP} }, // Turkish @@ -366,15 +431,15 @@ static wxFontEncoding /* unix */ {wxFONTENCODING_ISO8859_9, STOP}, /* windows */ {wxFONTENCODING_CP1254, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACTURKISH, STOP} }, // Cyrillic { - /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP}, + /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_KOI8_U, wxFONTENCODING_ISO8859_5, STOP}, /* windows */ {wxFONTENCODING_CP1251, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACCYRILLIC, STOP} }, {{STOP},{STOP},{STOP},{STOP}} /* Terminator */ @@ -382,7 +447,14 @@ static wxFontEncoding }; - +static bool FindEncoding(const wxFontEncodingArray& arr, wxFontEncoding f) +{ + for (wxFontEncodingArray::const_iterator it = arr.begin(), en = arr.end(); + it != en; ++it) + if (*it == f) + return true; + return false; +} wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform) { @@ -392,7 +464,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e platform = wxPLATFORM_WINDOWS; #elif defined(__WXGTK__) || defined(__WXMOTIF__) platform = wxPLATFORM_UNIX; -#elif defined(__WXOS2__) +#elif defined(__WXPM__) platform = wxPLATFORM_OS2; #elif defined(__WXMAC__) platform = wxPLATFORM_MAC; @@ -400,7 +472,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e } int i, clas, e ; - wxFontEncoding *f; + const wxFontEncoding *f; wxFontEncodingArray arr; clas = 0; @@ -411,9 +483,9 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e if (EquivalentEncodings[clas][i][e] == enc) { for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) - if (*f == enc) arr.Add(enc); + if (*f == enc) arr.push_back(enc); for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) - if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f); + if (!FindEncoding(arr, *f)) arr.push_back(*f); i = NUM_OF_PLATFORMS/*hack*/; break; } clas++; @@ -427,7 +499,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) { int i, clas, e, j ; - wxFontEncoding *f; + const wxFontEncoding *f; wxFontEncodingArray arr; arr = GetPlatformEquivalents(enc); // we want them to be first items in array @@ -441,7 +513,7 @@ wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) { for (j = 0; j < NUM_OF_PLATFORMS; j++) for (f = EquivalentEncodings[clas][j]; *f != STOP; f++) - if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f); + if (!FindEncoding(arr, *f)) arr.push_back(*f); i = NUM_OF_PLATFORMS/*hack*/; break; } clas++; @@ -450,4 +522,3 @@ wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) return arr; } -#endif // wxUSE_FONTMAP