X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/c958260b3886cb83aa93c7508d0986c6433ce165..6493aacaeb7b22b9fa35c559f7753e9fec0da71f:/src/common/encconv.cpp diff --git a/src/common/encconv.cpp b/src/common/encconv.cpp index 85acc2143a..15509c9cbc 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" @@ -22,15 +18,69 @@ #include -#include "unictabl.inc" -// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl +// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl: +#if defined( __BORLANDC__ ) || defined(__DARWIN__) + #include "../common/unictabl.inc" +#else + #include "unictabl.inc" +#endif + +#if wxUSE_WCHAR_T + typedef wchar_t tchar; +#else + typedef char tchar; +#endif + +#ifdef __WXMAC__ +#ifdef __DARWIN__ +#include +#else +#include +#include +#include +#endif + #include "wx/fontutil.h" + #include "wx/mac/private.h" // includes mac headers + 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 *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 ) + { + TECObjectRef converter ; + TextEncodingBase code = wxMacGetSystemEncFromFontEnc( enc ) ; + TextEncodingBase unicode = CreateTextEncoding(kTextEncodingUnicodeDefault,0,kUnicode16BitFormat) ; + OSStatus status = TECCreateConverter(&converter,code,unicode); + char s[2] ; + s[1] = 0 ; + ByteCount byteInLen, byteOutLen ; + for( unsigned char c = 255 ; c >= 128 ; --c ) + { + s[0] = c ; + status = TECConvertText(converter, (ConstTextPtr) &s , 1, &byteInLen, + (TextPtr) &gMacEncodings[i][c-128] , 2, &byteOutLen); + } + status = TECDisposeConverter(converter); + gMacEncodingsInited[i]=true; + } + return gMacEncodings[i] ; + } +#endif + for (int i = 0; encodings_list[i].table != NULL; i++) { - if (encodings_list[i].encoding == enc) + if (encodings_list[i].encoding == enc) return encodings_list[i].table; } return NULL; @@ -41,23 +91,22 @@ typedef struct { wxUint8 c; } CharsetItem; - - -static int 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* 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); - + return rev; } @@ -66,8 +115,8 @@ static CharsetItem* BuildReverseTable(wxUint16 *tbl) wxEncodingConverter::wxEncodingConverter() { m_Table = NULL; - m_UnicodeInput = FALSE; - m_JustCopy = FALSE; + m_UnicodeInput = m_UnicodeOutput = false; + m_JustCopy = false; } @@ -75,153 +124,329 @@ 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_UNICODE - if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE; +#if !wxUSE_WCHAR_T + if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return false; #endif - if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;} - - m_JustCopy = FALSE; - + if (input_enc == output_enc) {m_JustCopy = true; return true;} + + m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE); + 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 wxChar[65536]; - for (i = 0; i < 128; i++) m_Table[i] = (wxChar)i; // 7bit ASCII - for (i = 128; i < 65536; i++) m_Table[i] = (wxChar)'?'; - // FIXME - this should be character that means `unicode to charset' impossible, not '?' + 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)0; if (method == wxCONVERT_SUBSTITUTE) { for (i = 0; i < encoding_unicode_fallback_count; i++) - m_Table[encoding_unicode_fallback[i].c] = (wxChar) encoding_unicode_fallback[i].s; + m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s; } for (i = 0; i < 128; i++) - m_Table[out_tbl[i]] = (wxChar)(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_Table = new tchar[256]; + for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII - m_UnicodeInput = FALSE; - - m_Table = new wxChar[256]; - for (i = 0; i < 128; i++) m_Table[i] = (wxChar)i; // 7bit ASCII - if (output_enc == wxFONTENCODING_UNICODE) { - for (i = 0; i < 128; i++) m_Table[128 + i] = (wxChar)in_tbl[i]; // wxChar is 2byte now - return TRUE; + for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i]; + return true; } - else + else // output !Unicode { CharsetItem *rev = BuildReverseTable(out_tbl); - CharsetItem *item, key; - - for (i = 0; i < 128; i++) + CharsetItem *item; + CharsetItem key; + + for (i = 0; i < 128; i++) { key.u = in_tbl[i]; item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems); if (item == NULL && method == wxCONVERT_SUBSTITUTE) - item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback, + item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback, encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems); if (item) - m_Table[128 + i] = (wxChar)item -> c; + m_Table[128 + i] = (tchar)item -> c; else - m_Table[128 + i] = 128 + i; // don't know => don't touch +#if wxUSE_WCHAR_T + m_Table[128 + i] = (wchar_t)(128 + i); +#else + m_Table[128 + i] = (char)(128 + i); +#endif } - + delete[] rev; - return TRUE; } } + + return true; +} + + +#define REPLACEMENT_CHAR ((tchar)'?') + +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; } -void wxEncodingConverter::Convert(const wxChar* input, wxChar* output) +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*!")); + + const char *i; + char *o; + if (m_JustCopy) { - wxStrcpy(output, input); - return; + strcpy(output, input); + return true; } + + 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)(GetTableValue(m_Table, (wxUint8)*(i++), replaced)); + *o = 0; + + return !replaced; +} + + +#if wxUSE_WCHAR_T + +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*!")); + + const char *i; + wchar_t *o; + + if (m_JustCopy) + { + for (i = input, o = output; *i != 0;) + *(o++) = (wchar_t)(*(i++)); + *o = 0; + return true; + } + + 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)(GetTableValue(m_Table, (wxUint8)*(i++), replaced)); + *o = 0; + + return !replaced; +} + + + +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*!")); + + const wchar_t *i; + char *o; + + if (m_JustCopy) + { + for (i = input, o = output; *i != 0;) + *(o++) = (char)(*(i++)); + *o = 0; + return true; + } + + 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)(GetTableValue(m_Table, (wxUint16)*(i++), replaced)); + *o = 0; + + return !replaced; +} + + + +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*!")); + + const wchar_t *i; + wchar_t *o; + + if (m_JustCopy) + { + // wcscpy() is not guaranteed to exist + for (i = input, o = output; *i != 0;) + *(o++) = (*(i++)); + *o = 0; + return true; + } + + wxCHECK_MSG(m_Table != NULL, false, + wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + bool replaced = false; - wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); - - const wxChar *i; - wxChar *o; - - if (m_UnicodeInput) - for (i = input, o = output; *i != 0; i++, o++) - *o = (wxChar)(m_Table[(wxUint16)*i]); - else - for (i = input, o = output; *i != 0; i++, o++) - *o = (wxChar)(m_Table[(wxUint8)*i]); + for (i = input, o = output; *i != 0;) + *(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; - - wxASSERT_MSG(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) + { for (i = input.c_str(); *i != 0; i++) s << (wxChar)(m_Table[(wxUint16)*i]); + } else + { for (i = input.c_str(); *i != 0; i++) s << (wxChar)(m_Table[(wxUint8)*i]); + } + return s; } + + + // Following tables describe classes of encoding equivalence. -// +// #define STOP wxFONTENCODING_SYSTEM #define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/ #define ENC_PER_PLATFORM 3 - // max no. of encodings for one language used on one platform - // Anybody thinks 3 is not enough? ;-) + // 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] = { - // West European (Latin1) + // *** Please put more common encodings as first! *** + + // Western European { /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP}, /* windows */ {wxFONTENCODING_CP1252, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACROMAN, STOP} }, - // Central European (Latin2) + // Central European { /* unix */ {wxFONTENCODING_ISO8859_2, STOP}, /* windows */ {wxFONTENCODING_CP1250, STOP}, /* os2 */ {STOP}, - /* mac */ {STOP} + /* mac */ {wxFONTENCODING_MACCENTRALEUR, STOP} + }, + + // Baltic + { + /* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP}, + /* windows */ {wxFONTENCODING_CP1257, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Hebrew + { + /* unix */ {wxFONTENCODING_ISO8859_8, STOP}, + /* windows */ {wxFONTENCODING_CP1255, STOP}, + /* os2 */ {STOP}, + /* mac */ {wxFONTENCODING_MACHEBREW, STOP} + }, + + // Greek + { + /* unix */ {wxFONTENCODING_ISO8859_7, STOP}, + /* windows */ {wxFONTENCODING_CP1253, STOP}, + /* os2 */ {STOP}, + /* mac */ {wxFONTENCODING_MACGREEK, STOP} + }, + + // Arabic + { + /* unix */ {wxFONTENCODING_ISO8859_6, STOP}, + /* windows */ {wxFONTENCODING_CP1256, STOP}, + /* os2 */ {STOP}, + /* mac */ {wxFONTENCODING_MACARABIC, STOP} + }, + + // Turkish + { + /* unix */ {wxFONTENCODING_ISO8859_9, STOP}, + /* windows */ {wxFONTENCODING_CP1254, STOP}, + /* os2 */ {STOP}, + /* mac */ {wxFONTENCODING_MACTURKISH, STOP} + }, + + // Cyrillic + { + /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_KOI8_U, wxFONTENCODING_ISO8859_5, STOP}, + /* windows */ {wxFONTENCODING_CP1251, STOP}, + /* os2 */ {STOP}, + /* mac */ {wxFONTENCODING_MACCYRILLIC, STOP} }, {{STOP},{STOP},{STOP},{STOP}} /* Terminator */ @@ -229,7 +454,14 @@ 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) { @@ -239,15 +471,15 @@ 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; #endif } - + int i, clas, e ; - wxFontEncoding *f; + const wxFontEncoding *f; wxFontEncodingArray arr; clas = 0; @@ -257,13 +489,15 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++) if (EquivalentEncodings[clas][i][e] == enc) { - for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) - if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f); - i = NUM_OF_PLATFORMS/*hack*/; break; + for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) + if (*f == enc) arr.push_back(enc); + for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) + if (!FindEncoding(arr, *f)) arr.push_back(*f); + i = NUM_OF_PLATFORMS/*hack*/; break; } clas++; } - + return arr; } @@ -272,9 +506,11 @@ 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 + clas = 0; while (EquivalentEncodings[clas][0][0] != STOP) { @@ -283,12 +519,13 @@ wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) if (EquivalentEncodings[clas][i][e] == 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); - i = NUM_OF_PLATFORMS/*hack*/; break; + for (f = EquivalentEncodings[clas][j]; *f != STOP; f++) + if (!FindEncoding(arr, *f)) arr.push_back(*f); + i = NUM_OF_PLATFORMS/*hack*/; break; } clas++; } - + return arr; } +