X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/c958260b3886cb83aa93c7508d0986c6433ce165..bf62af352e4573b5307be6ff8997c58a376a1959:/src/common/encconv.cpp diff --git a/src/common/encconv.cpp b/src/common/encconv.cpp index 85acc2143a..6404d6ef85 100644 --- a/src/common/encconv.cpp +++ b/src/common/encconv.cpp @@ -22,11 +22,20 @@ #include +// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl: +#ifdef __BORLANDC__ +#include "../common/unictabl.inc" +#else #include "unictabl.inc" -// conversion tables, generated by scripts in $(WXWIN)/misc/unictabl +#endif +#if wxUSE_WCHAR_T +typedef wchar_t tchar; +#else +typedef char tchar; +#endif -static wxUint16 *GetEncTable(wxFontEncoding enc) +static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc) { for (int i = 0; encodings_list[i].table != NULL; i++) { @@ -43,13 +52,13 @@ typedef struct { -static int CompareCharsetItems(const void *i1, const void *i2) +static int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2) { return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u ); } -static CharsetItem* BuildReverseTable(wxUint16 *tbl) +static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl) { CharsetItem *rev = new CharsetItem[128]; @@ -66,7 +75,7 @@ static CharsetItem* BuildReverseTable(wxUint16 *tbl) wxEncodingConverter::wxEncodingConverter() { m_Table = NULL; - m_UnicodeInput = FALSE; + m_UnicodeInput = m_UnicodeOutput = FALSE; m_JustCopy = FALSE; } @@ -79,31 +88,32 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e if (m_Table) {delete[] m_Table; m_Table = NULL;} -#if !wxUSE_UNICODE +#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_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE); m_JustCopy = FALSE; if (input_enc == wxFONTENCODING_UNICODE) { 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)'?'; + 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 '?' 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; @@ -117,12 +127,12 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e m_UnicodeInput = FALSE; - m_Table = new wxChar[256]; - for (i = 0; i < 128; i++) m_Table[i] = (wxChar)i; // 7bit ASCII + m_Table = new tchar[256]; + for (i = 0; i < 128; i++) m_Table[i] = (tchar)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 + for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i]; return TRUE; } else @@ -138,7 +148,7 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e 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 } @@ -150,28 +160,106 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e } -void wxEncodingConverter::Convert(const wxChar* input, wxChar* output) + +void wxEncodingConverter::Convert(const char* input, char* output) { + 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); + strcpy(output, input); return; } wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + for (i = input, o = output; *i != 0;) + *(o++) = (char)(m_Table[(wxUint8)*(i++)]); + *o = 0; +} + + +#if wxUSE_WCHAR_T + +void wxEncodingConverter::Convert(const char* input, wchar_t* output) +{ + 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; + } - const wxChar *i; - wxChar *o; + wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + for (i = input, o = output; *i != 0;) + *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); + *o = 0; +} + + + +void wxEncodingConverter::Convert(const wchar_t* input, char* output) +{ + 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; + } - 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]); + wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + for (i = input, o = output; *i != 0;) + *(o++) = (char)(m_Table[(wxUint16)*(i++)]); + *o = 0; +} + + + +void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) +{ + 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; + } + + wxASSERT_MSG(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!")); + + for (i = input, o = output; *i != 0;) + *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]); *o = 0; } +#endif // wxUSE_WCHAR_T wxString wxEncodingConverter::Convert(const wxString& input) @@ -195,20 +283,25 @@ wxString wxEncodingConverter::Convert(const wxString& input) + + + // 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 +#define ENC_PER_PLATFORM 5 // max no. of encodings for one language used on one platform - // Anybody thinks 3 is not enough? ;-) + // Anybody thinks 5 is not enough? ;-) static wxFontEncoding EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = { - // West European (Latin1) + // *** Please put more common encodings as first! *** + + // West European { /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP}, /* windows */ {wxFONTENCODING_CP1252, STOP}, @@ -216,13 +309,61 @@ static wxFontEncoding /* mac */ {STOP} }, - // Central European (Latin2) + // Central European { /* unix */ {wxFONTENCODING_ISO8859_2, STOP}, /* windows */ {wxFONTENCODING_CP1250, STOP}, /* os2 */ {STOP}, /* mac */ {STOP} }, + + // Baltic + { + /* unix */ {wxFONTENCODING_ISO8859_13, STOP}, + /* windows */ {wxFONTENCODING_CP1257, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Hebrew + { + /* unix */ {wxFONTENCODING_ISO8859_8, STOP}, + /* windows */ {wxFONTENCODING_CP1255, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Greek + { + /* unix */ {wxFONTENCODING_ISO8859_7, STOP}, + /* windows */ {wxFONTENCODING_CP1253, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Arabic + { + /* unix */ {wxFONTENCODING_ISO8859_6, STOP}, + /* windows */ {wxFONTENCODING_CP1256, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Turkish + { + /* unix */ {wxFONTENCODING_ISO8859_9, STOP}, + /* windows */ {wxFONTENCODING_CP1254, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, + + // Cyrillic + { + /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP}, + /* windows */ {wxFONTENCODING_CP1251, STOP}, + /* os2 */ {STOP}, + /* mac */ {STOP} + }, {{STOP},{STOP},{STOP},{STOP}} /* Terminator */ /* no, _not_ Arnold! */ @@ -257,7 +398,9 @@ 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++) + for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) + if (*f == enc) arr.Add(enc); + for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++) if (arr.Index(*f) == wxNOT_FOUND) arr.Add(*f); i = NUM_OF_PLATFORMS/*hack*/; break; } @@ -274,6 +417,8 @@ wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc) int i, clas, e, j ; wxFontEncoding *f; wxFontEncodingArray arr; + + arr = GetPlatformEquivalents(enc); // we want them to be first items in array clas = 0; while (EquivalentEncodings[clas][0][0] != STOP)