/////////////////////////////////////////////////////////////////////////////
-// Name: encconv.cpp
+// Name: src/common/encconv.cpp
// Purpose: wxEncodingConverter class for converting between different
// font encodings
// Author: Vaclav Slavik
// Copyright: (c) 1999 Vaclav Slavik
-// Licence: wxWindows Licence
+// Licence: wxWindows licence
/////////////////////////////////////////////////////////////////////////////
-#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
-#pragma implementation "encconv.h"
-#endif
-
// For compilers that support precompilation, includes "wx.h".
#include "wx/wxprec.h"
#pragma hdrstop
#endif
-#if wxUSE_FONTMAP
-
#include "wx/encconv.h"
#include <stdlib.h>
#include "unictabl.inc"
#endif
-#if wxUSE_WCHAR_T
- typedef wchar_t tchar;
-#else
- typedef char tchar;
-#endif
-
#ifdef __WXMAC__
- #include <ATSUnicode.h>
- #include <TextCommon.h>
- #include <TextEncodingConverter.h>
-
- #include "wx/fontutil.h"
- #include "wx/mac/private.h" // includes mac headers
+ #include "wx/osx/core/cfstring.h"
+ #include <CoreFoundation/CFStringEncodingExt.h>
wxUint16 gMacEncodings[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1][128] ;
bool gMacEncodingsInited[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1] ;
#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;
+ // 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] ;
}
wxUint8 c;
} CharsetItem;
-extern "C" int wxCMPFUNC_CONV
+extern "C"
+{
+static 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);
wxEncodingConverter::wxEncodingConverter()
{
m_Table = NULL;
- m_UnicodeInput = m_UnicodeOutput = FALSE;
- m_JustCopy = FALSE;
+ m_UnicodeInput = m_UnicodeOutput = false;
+ m_JustCopy = false;
}
bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
{
unsigned i;
- wxUint16 *in_tbl, *out_tbl = NULL;
+ const wxUint16 *in_tbl;
+ const wxUint16 *out_tbl = NULL;
- if (m_Table) {delete[] m_Table; m_Table = NULL;}
+ wxDELETEA(m_Table);
-#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;}
+ 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 '?'
+ m_Table = new wchar_t[65536];
+ for (i = 0; i < 128; i++) m_Table[i] = (wchar_t)i; // 7bit ASCII
+ for (i = 128; i < 65536; i++) m_Table[i] = (wchar_t)0;
if (method == wxCONVERT_SUBSTITUTE)
{
for (i = 0; i < encoding_unicode_fallback_count; i++)
- m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
+ m_Table[encoding_unicode_fallback[i].c] = (wchar_t) encoding_unicode_fallback[i].s;
}
for (i = 0; i < 128; i++)
- m_Table[out_tbl[i]] = (tchar)(128 + i);
+ m_Table[out_tbl[i]] = (wchar_t)(128 + i);
- m_UnicodeInput = TRUE;
+ m_UnicodeInput = true;
}
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
+ m_Table = new wchar_t[256];
+ for (i = 0; i < 128; i++) m_Table[i] = (wchar_t)i; // 7bit ASCII
if (output_enc == wxFONTENCODING_UNICODE)
{
- for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i];
- return TRUE;
+ for (i = 0; i < 128; i++) m_Table[128 + i] = (wchar_t)in_tbl[i];
+ return true;
}
else // output !Unicode
{
item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
if (item)
- m_Table[128 + i] = (tchar)item -> c;
+ m_Table[128 + i] = (wchar_t)item -> c;
else
-#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 (L'?')
-void wxEncodingConverter::Convert(const char* input, char* output) const
+inline wchar_t GetTableValue(const wchar_t *table, wchar_t value, bool& repl)
+{
+ wchar_t 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*!"));
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) const
+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*!"));
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) const
+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*!"));
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) const
+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*!"));
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;
-}
-#endif // wxUSE_WCHAR_T
+ return !replaced;
+}
wxString wxEncodingConverter::Convert(const wxString& input) const
#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! ***
// 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 */ {wxFONTENCODING_MACCYRILLIC, STOP}
{
if (platform == wxPLATFORM_CURRENT)
{
-#if defined(__WXMSW__)
+#if defined(__WINDOWS__)
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;
}
int i, clas, e ;
- wxFontEncoding *f;
+ const wxFontEncoding *f;
wxFontEncodingArray arr;
clas = 0;
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
return arr;
}
-#endif // wxUSE_FONTMAP