1. use a manifest constant wxNO_LEN instead of -1 for lengths everywhere

[wxWidgets.git] / src / common / encconv.cpp

diff --git a/src/common/encconv.cpp b/src/common/encconv.cpp
index e3327c909a317f18a071e3f4bcbd2fdc838878fe..15509c9cbc2c93128c10d775faa804bdad813f3b 100644 (file)
--- a/src/common/encconv.cpp
+++ b/src/common/encconv.cpp
@@ -4,13 +4,9 @@
 //              font encodings
 // Author:      Vaclav Slavik
 // Copyright:   (c) 1999 Vaclav Slavik
 //              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"
 
 // For compilers that support precompilation, includes "wx.h".
 #include "wx/wxprec.h"
 


@@ -18,45 +14,43 @@
   #pragma hdrstop
 #endif
 
   #pragma hdrstop
 #endif
 

-#if wxUSE_FONTMAP
-

 #include "wx/encconv.h"
 
 #include <stdlib.h>
 
 // conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
 #include "wx/encconv.h"
 
 #include <stdlib.h>
 
 // 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
 #else

-#include "unictabl.inc"
+    #include "unictabl.inc"

 #endif
 
 #if wxUSE_WCHAR_T
 #endif
 
 #if wxUSE_WCHAR_T

-typedef wchar_t tchar;
+    typedef wchar_t tchar;

 #else
 #else

-typedef char tchar;
-#endif
-
-#ifdef __WXWINCE__
-#undef LINKAGEMODE
-#define LINKAGEMODE __cdecl
+    typedef char tchar;

 #endif
 
 #ifdef __WXMAC__
 #endif
 
 #ifdef __WXMAC__

+#ifdef __DARWIN__
+#include <Carbon/Carbon.h>
+#else
+#include <ATSUnicode.h>
+#include <TextCommon.h>
+#include <TextEncodingConverter.h>
+#endif
+    #include "wx/fontutil.h"
+    #include "wx/mac/private.h"  // includes mac headers

 
 

-#include "ATSUnicode.h"
-#include "TextCommon.h"
-#include "TextEncodingConverter.h"
-
-#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] ;
+    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
 
 #endif
 

-static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
+static const wxUint16* GetEncTable(wxFontEncoding enc)

 {
 #ifdef __WXMAC__
     if( enc >= wxFONTENCODING_MACMIN && enc <= wxFONTENCODING_MACMAX )
 {
 #ifdef __WXMAC__
     if( enc >= wxFONTENCODING_MACMIN && enc <= wxFONTENCODING_MACMAX )


@@ -66,19 +60,19 @@ static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
         {
             TECObjectRef converter ;
             TextEncodingBase code = wxMacGetSystemEncFromFontEnc( enc ) ;
         {
             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( char c = 255 ; c >= 128 ; --c )
-           {
-               s[0] = c ;
+            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 = TECConvertText(converter, (ConstTextPtr) &s , 1, &byteInLen,
                 (TextPtr) &gMacEncodings[i][c-128] , 2, &byteOutLen);

-           }
+            }

             status = TECDisposeConverter(converter);
             status = TECDisposeConverter(converter);

-            gMacEncodingsInited[i]=true;       
+            gMacEncodingsInited[i]=true;

         }
         return gMacEncodings[i] ;
     }
         }
         return gMacEncodings[i] ;
     }


@@ -97,18 +91,19 @@ typedef struct {
     wxUint8  c;
 } CharsetItem;
 
     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 );
 }
 
 
 {
     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++)
 {
     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);
 
 
     qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
 


@@ -120,8 +115,8 @@ static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
 wxEncodingConverter::wxEncodingConverter()
 {
     m_Table = NULL;
 wxEncodingConverter::wxEncodingConverter()
 {
     m_Table = NULL;

-    m_UnicodeInput = m_UnicodeOutput = FALSE;
-    m_JustCopy = FALSE;
+    m_UnicodeInput = m_UnicodeOutput = false;
+    m_JustCopy = false;

 }
 
 
 }
 
 


@@ -129,27 +124,27 @@ wxEncodingConverter::wxEncodingConverter()
 bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
 {
     unsigned i;
 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;}
 
 #if !wxUSE_WCHAR_T
 
     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
 
 #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_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);

-    m_JustCopy = FALSE;
+    m_JustCopy = false;

 
     if (input_enc == wxFONTENCODING_UNICODE)
     {
 
     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
 
         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)
         {
 
         if (method == wxCONVERT_SUBSTITUTE)
         {


@@ -160,15 +155,15 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e
         for (i = 0; i < 128; i++)
             m_Table[out_tbl[i]] = (tchar)(128 + i);
 
         for (i = 0; i < 128; i++)
             m_Table[out_tbl[i]] = (tchar)(128 + i);
 

-        m_UnicodeInput = TRUE;
+        m_UnicodeInput = true;

     }
     else // input !Unicode
     {
     }
     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 (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 tchar[256];
         for (i = 0; i < 128; i++)  m_Table[i] = (tchar)i; // 7bit ASCII


@@ -176,10 +171,8 @@ 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];
         if (output_enc == wxFONTENCODING_UNICODE)
         {
             for (i = 0; i < 128; i++)  m_Table[128 + i] = (tchar)in_tbl[i];

-            return TRUE;
+            return true;

         }
         }

-        // FIXME: write a substitute for bsearch
-#ifndef __WXWINCE__

         else // output !Unicode
         {
             CharsetItem *rev = BuildReverseTable(out_tbl);
         else // output !Unicode
         {
             CharsetItem *rev = BuildReverseTable(out_tbl);


@@ -205,15 +198,27 @@ bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_e
 
             delete[] rev;
         }
 
             delete[] rev;
         }

-#endif // !__WXWINCE__

     }
 
     }
 

-    return TRUE;
+    return true;

 }
 
 
 }
 
 

+#define REPLACEMENT_CHAR  ((tchar)'?')

 
 

-void wxEncodingConverter::Convert(const char* input, char* output) const
+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*!"));
 {
     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*!"));


@@ -224,20 +229,25 @@ void wxEncodingConverter::Convert(const char* input, char* output) const
     if (m_JustCopy)
     {
         strcpy(output, input);
     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;)
 
     for (i = input, o = output; *i != 0;)

-        *(o++) = (char)(m_Table[(wxUint8)*(i++)]);
+        *(o++) = (char)(GetTableValue(m_Table, (wxUint8)*(i++), replaced));

     *o = 0;
     *o = 0;

+
+    return !replaced;

 }
 
 
 #if wxUSE_WCHAR_T
 
 }
 
 
 #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*!"));
 {
     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*!"));


@@ -250,19 +260,24 @@ void wxEncodingConverter::Convert(const char* input, wchar_t* output) const
         for (i = input, o = output; *i != 0;)
             *(o++) = (wchar_t)(*(i++));
         *o = 0;
         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;)
 
     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;
     *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*!"));
 {
     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*!"));


@@ -275,19 +290,24 @@ void wxEncodingConverter::Convert(const wchar_t* input, char* output) const
         for (i = input, o = output; *i != 0;)
             *(o++) = (char)(*(i++));
         *o = 0;
         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;)
 
     for (i = input, o = output; *i != 0;)

-        *(o++) = (char)(m_Table[(wxUint16)*(i++)]);
+        *(o++) = (char)(GetTableValue(m_Table, (wxUint16)*(i++), replaced));

     *o = 0;
     *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*!"));
 {
     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*!"));


@@ -301,14 +321,19 @@ void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) const
         for (i = input, o = output; *i != 0;)
             *(o++) = (*(i++));
         *o = 0;
         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;)
     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;
     *o = 0;

+
+    return !replaced;

 }
 
 #endif // wxUSE_WCHAR_T
 }
 
 #endif // wxUSE_WCHAR_T


@@ -350,11 +375,12 @@ wxString wxEncodingConverter::Convert(const wxString& input) const
 #define STOP wxFONTENCODING_SYSTEM
 
 #define NUM_OF_PLATFORMS  4 /*must conform to enum wxPLATFORM_XXXX !!!*/
 #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! ***
     EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
 
     // *** Please put more common encodings as first! ***


@@ -417,7 +443,7 @@ static wxFontEncoding
 
     // Cyrillic
     {
 
     // 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}
         /* windows */ {wxFONTENCODING_CP1251, STOP},
         /* os2     */ {STOP},
         /* mac     */ {wxFONTENCODING_MACCYRILLIC, STOP}


@@ -445,7 +471,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e
         platform = wxPLATFORM_WINDOWS;
 #elif defined(__WXGTK__) || defined(__WXMOTIF__)
         platform = wxPLATFORM_UNIX;
         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;
         platform = wxPLATFORM_OS2;
 #elif defined(__WXMAC__)
         platform = wxPLATFORM_MAC;


@@ -453,7 +479,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e
     }
 
     int i, clas, e ;
     }
 
     int i, clas, e ;

-    wxFontEncoding *f;
+    const wxFontEncoding *f;

     wxFontEncodingArray arr;
 
     clas = 0;
     wxFontEncodingArray arr;
 
     clas = 0;


@@ -480,7 +506,7 @@ wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding e
 wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
 {
     int i, clas, e, j ;
 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
     wxFontEncodingArray arr;
 
     arr = GetPlatformEquivalents(enc); // we want them to be first items in array


@@ -503,4 +529,3 @@ wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
     return arr;
 }
 
     return arr;
 }
 

-#endif // wxUSE_FONTMAP