cvs to svn

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

diff --git a/src/common/strconv.cpp b/src/common/strconv.cpp
index 2a0cb613c663d598e89a26451adedef7f99fbabc..843c84e046ce7c5b007af15ab4c07b76ca207a73 100644 (file)
--- a/src/common/strconv.cpp
+++ b/src/common/strconv.cpp
@@ -476,10 +476,10 @@ size_t wxMBConvLibc::WC2MB(char *buf, const wchar_t *psz, size_t n) const
 
 #ifdef __UNIX__
 
 
 #ifdef __UNIX__
 

-wxConvBrokenFileNames::wxConvBrokenFileNames(const wxChar *charset)
+wxConvBrokenFileNames::wxConvBrokenFileNames(const wxString& charset)

 {
 {

-    if ( !charset || wxStricmp(charset, _T("UTF-8")) == 0
-                  || wxStricmp(charset, _T("UTF8")) == 0  )
+    if ( wxStricmp(charset, _T("UTF-8")) == 0 ||
+         wxStricmp(charset, _T("UTF8")) == 0  )

         m_conv = new wxMBConvUTF8(wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
     else
         m_conv = new wxCSConv(charset);
         m_conv = new wxMBConvUTF8(wxMBConvUTF8::MAP_INVALID_UTF8_TO_PUA);
     else
         m_conv = new wxCSConv(charset);


@@ -1588,7 +1588,7 @@ wxMBConvUTF32swap::FromWChar(char *dst, size_t dstLen,
 class wxMBConv_iconv : public wxMBConv
 {
 public:
 class wxMBConv_iconv : public wxMBConv
 {
 public:

-    wxMBConv_iconv(const wxChar *name);
+    wxMBConv_iconv(const char *name);

     virtual ~wxMBConv_iconv();
 
     virtual size_t MB2WC(wchar_t *buf, const char *psz, size_t n) const;
     virtual ~wxMBConv_iconv();
 
     virtual size_t MB2WC(wchar_t *buf, const char *psz, size_t n) const;


@@ -1597,9 +1597,13 @@ public:
     // classify this encoding as explained in wxMBConv::GetMBNulLen() comment
     virtual size_t GetMBNulLen() const;
 
     // classify this encoding as explained in wxMBConv::GetMBNulLen() comment
     virtual size_t GetMBNulLen() const;
 

+#if wxUSE_UNICODE_UTF8
+    virtual bool IsUTF8() const;
+#endif
+

     virtual wxMBConv *Clone() const
     {
     virtual wxMBConv *Clone() const
     {

-        wxMBConv_iconv *p = new wxMBConv_iconv(m_name);
+        wxMBConv_iconv *p = new wxMBConv_iconv(m_name.ToAscii());

         p->m_minMBCharWidth = m_minMBCharWidth;
         return p;
     }
         p->m_minMBCharWidth = m_minMBCharWidth;
         return p;
     }


@@ -1637,7 +1641,7 @@ private:
 };
 
 // make the constructor available for unit testing
 };
 
 // make the constructor available for unit testing

-WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const wxChar* name )
+WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const char* name )

 {
     wxMBConv_iconv* result = new wxMBConv_iconv( name );
     if ( !result->IsOk() )
 {
     wxMBConv_iconv* result = new wxMBConv_iconv( name );
     if ( !result->IsOk() )


@@ -1652,15 +1656,11 @@ WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_iconv( const wxChar* name )
 wxString wxMBConv_iconv::ms_wcCharsetName;
 bool wxMBConv_iconv::ms_wcNeedsSwap = false;
 
 wxString wxMBConv_iconv::ms_wcCharsetName;
 bool wxMBConv_iconv::ms_wcNeedsSwap = false;
 

-wxMBConv_iconv::wxMBConv_iconv(const wxChar *name)
+wxMBConv_iconv::wxMBConv_iconv(const char *name)

               : m_name(name)
 {
     m_minMBCharWidth = 0;
 
               : m_name(name)
 {
     m_minMBCharWidth = 0;
 

-    // iconv operates with chars, not wxChars, but luckily it uses only ASCII
-    // names for the charsets
-    const wxCharBuffer cname(wxString(name).ToAscii());
-

     // check for charset that represents wchar_t:
     if ( ms_wcCharsetName.empty() )
     {
     // check for charset that represents wchar_t:
     if ( ms_wcCharsetName.empty() )
     {


@@ -1697,13 +1697,13 @@ wxMBConv_iconv::wxMBConv_iconv(const wxChar *name)
             wxLogTrace(TRACE_STRCONV, _T("  trying charset \"%s\""),
                        nameXE.c_str());
 
             wxLogTrace(TRACE_STRCONV, _T("  trying charset \"%s\""),
                        nameXE.c_str());
 

-            m2w = iconv_open(nameXE.ToAscii(), cname);
+            m2w = iconv_open(nameXE.ToAscii(), name);

             if ( m2w == ICONV_T_INVALID )
             {
                 // try charset w/o bytesex info (e.g. "UCS4")
                 wxLogTrace(TRACE_STRCONV, _T("  trying charset \"%s\""),
                            nameCS.c_str());
             if ( m2w == ICONV_T_INVALID )
             {
                 // try charset w/o bytesex info (e.g. "UCS4")
                 wxLogTrace(TRACE_STRCONV, _T("  trying charset \"%s\""),
                            nameCS.c_str());

-                m2w = iconv_open(nameCS.ToAscii(), cname);
+                m2w = iconv_open(nameCS.ToAscii(), name);

 
                 // and check for bytesex ourselves:
                 if ( m2w != ICONV_T_INVALID )
 
                 // and check for bytesex ourselves:
                 if ( m2w != ICONV_T_INVALID )


@@ -1746,14 +1746,14 @@ wxMBConv_iconv::wxMBConv_iconv(const wxChar *name)
 
         wxLogTrace(TRACE_STRCONV,
                    wxT("iconv wchar_t charset is \"%s\"%s"),
 
         wxLogTrace(TRACE_STRCONV,
                    wxT("iconv wchar_t charset is \"%s\"%s"),

-                   ms_wcCharsetName.empty() ? _T("<none>")
-                                            : ms_wcCharsetName.c_str(),
+                   ms_wcCharsetName.empty() ? wxString("<none>")
+                                            : ms_wcCharsetName,

                    ms_wcNeedsSwap ? _T(" (needs swap)")
                                   : _T(""));
     }
     else // we already have ms_wcCharsetName
     {
                    ms_wcNeedsSwap ? _T(" (needs swap)")
                                   : _T(""));
     }
     else // we already have ms_wcCharsetName
     {

-        m2w = iconv_open(ms_wcCharsetName.ToAscii(), cname);
+        m2w = iconv_open(ms_wcCharsetName.ToAscii(), name);

     }
 
     if ( ms_wcCharsetName.empty() )
     }
 
     if ( ms_wcCharsetName.empty() )


@@ -1762,12 +1762,12 @@ wxMBConv_iconv::wxMBConv_iconv(const wxChar *name)
     }
     else
     {
     }
     else
     {

-        w2m = iconv_open(cname, ms_wcCharsetName.ToAscii());
+        w2m = iconv_open(name, ms_wcCharsetName.ToAscii());

         if ( w2m == ICONV_T_INVALID )
         {
             wxLogTrace(TRACE_STRCONV,
                        wxT("\"%s\" -> \"%s\" works but not the converse!?"),
         if ( w2m == ICONV_T_INVALID )
         {
             wxLogTrace(TRACE_STRCONV,
                        wxT("\"%s\" -> \"%s\" works but not the converse!?"),

-                       ms_wcCharsetName.c_str(), cname.data());
+                       ms_wcCharsetName.c_str(), name);

         }
     }
 }
         }
     }
 }


@@ -1956,7 +1956,7 @@ size_t wxMBConv_iconv::GetMBNulLen() const
         wxMutexLocker lock(self->m_iconvMutex);
 #endif
 
         wxMutexLocker lock(self->m_iconvMutex);
 #endif
 

-        wchar_t *wnul = L"";
+        const wchar_t *wnul = L"";

         char buf[8]; // should be enough for NUL in any encoding
         size_t inLen = sizeof(wchar_t),
                outLen = WXSIZEOF(buf);
         char buf[8]; // should be enough for NUL in any encoding
         size_t inLen = sizeof(wchar_t),
                outLen = WXSIZEOF(buf);


@@ -1975,6 +1975,14 @@ size_t wxMBConv_iconv::GetMBNulLen() const
     return m_minMBCharWidth;
 }
 
     return m_minMBCharWidth;
 }
 

+#if wxUSE_UNICODE_UTF8
+bool wxMBConv_iconv::IsUTF8() const
+{
+    return wxStricmp(m_name, "UTF-8") == 0 ||
+           wxStricmp(m_name, "UTF8") == 0;
+}
+#endif
+

 #endif // HAVE_ICONV
 
 
 #endif // HAVE_ICONV
 
 


@@ -1986,7 +1994,7 @@ size_t wxMBConv_iconv::GetMBNulLen() const
 
 // from utils.cpp
 #if wxUSE_FONTMAP
 
 // from utils.cpp
 #if wxUSE_FONTMAP

-extern WXDLLIMPEXP_BASE long wxCharsetToCodepage(const wxChar *charset);
+extern WXDLLIMPEXP_BASE long wxCharsetToCodepage(const char *charset);

 extern WXDLLIMPEXP_BASE long wxEncodingToCodepage(wxFontEncoding encoding);
 #endif
 
 extern WXDLLIMPEXP_BASE long wxEncodingToCodepage(wxFontEncoding encoding);
 #endif
 


@@ -2007,7 +2015,7 @@ public:
     }
 
 #if wxUSE_FONTMAP
     }
 
 #if wxUSE_FONTMAP

-    wxMBConv_win32(const wxChar* name)
+    wxMBConv_win32(const char* name)

     {
         m_CodePage = wxCharsetToCodepage(name);
         m_minMBCharWidth = 0;
     {
         m_CodePage = wxCharsetToCodepage(name);
         m_minMBCharWidth = 0;


@@ -2035,12 +2043,12 @@ public:
         //     http://msdn.microsoft.com/library/en-us/intl/unicode_17si.asp
         if ( m_CodePage == CP_UTF8 )
         {
         //     http://msdn.microsoft.com/library/en-us/intl/unicode_17si.asp
         if ( m_CodePage == CP_UTF8 )
         {

-            return wxConvUTF8.MB2WC(buf, psz, n);
+            return wxMBConvUTF8().MB2WC(buf, psz, n);

         }
 
         if ( m_CodePage == CP_UTF7 )
         {
         }
 
         if ( m_CodePage == CP_UTF7 )
         {

-            return wxConvUTF7.MB2WC(buf, psz, n);
+            return wxMBConvUTF7().MB2WC(buf, psz, n);

         }
 
         int flags = 0;
         }
 
         int flags = 0;


@@ -2730,7 +2738,7 @@ public:
     }
 
 #if wxUSE_FONTMAP
     }
 
 #if wxUSE_FONTMAP

-    wxMBConv_mac(const wxChar* name)
+    wxMBConv_mac(const char* name)

     {
         Init( wxMacGetSystemEncFromFontEnc( wxFontMapperBase::Get()->CharsetToEncoding(name, false) ) );
     }
     {
         Init( wxMacGetSystemEncFromFontEnc( wxFontMapperBase::Get()->CharsetToEncoding(name, false) ) );
     }


@@ -3095,7 +3103,7 @@ private:
 public:
     // temporarily just use wxEncodingConverter stuff,
     // so that it works while a better implementation is built
 public:
     // temporarily just use wxEncodingConverter stuff,
     // so that it works while a better implementation is built

-    wxMBConv_wxwin(const wxChar* name)
+    wxMBConv_wxwin(const char* name)

     {
         if (name)
             m_enc = wxFontMapperBase::Get()->CharsetToEncoding(name, false);
     {
         if (name)
             m_enc = wxFontMapperBase::Get()->CharsetToEncoding(name, false);


@@ -3168,7 +3176,7 @@ private:
 };
 
 // make the constructors available for unit testing
 };
 
 // make the constructors available for unit testing

-WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const wxChar* name )
+WXDLLIMPEXP_BASE wxMBConv* new_wxMBConv_wxwin( const char* name )

 {
     wxMBConv_wxwin* result = new wxMBConv_wxwin( name );
     if ( !result->IsOk() )
 {
     wxMBConv_wxwin* result = new wxMBConv_wxwin( name );
     if ( !result->IsOk() )


@@ -3193,13 +3201,13 @@ void wxCSConv::Init()
     m_deferred = true;
 }
 
     m_deferred = true;
 }
 

-wxCSConv::wxCSConv(const wxChar *charset)
+wxCSConv::wxCSConv(const wxString& charset)

 {
     Init();
 
 {
     Init();
 

-    if ( charset )
+    if ( !charset.empty() )

     {
     {

-        SetName(charset);
+        SetName(charset.ToAscii());

     }
 
 #if wxUSE_FONTMAP
     }
 
 #if wxUSE_FONTMAP


@@ -3256,11 +3264,11 @@ void wxCSConv::Clear()
     m_convReal = NULL;
 }
 
     m_convReal = NULL;
 }
 

-void wxCSConv::SetName(const wxChar *charset)
+void wxCSConv::SetName(const char *charset)

 {
     if (charset)
     {
 {
     if (charset)
     {

-        m_name = wxStrdup(charset);
+        m_name = strdup(charset);

         m_deferred = true;
     }
 }
         m_deferred = true;
     }
 }


@@ -3279,7 +3287,7 @@ wxMBConv *wxCSConv::DoCreate() const
     wxLogTrace(TRACE_STRCONV,
                wxT("creating conversion for %s"),
                (m_name ? m_name
     wxLogTrace(TRACE_STRCONV,
                wxT("creating conversion for %s"),
                (m_name ? m_name

-                       : (const wxChar*)wxFontMapperBase::GetEncodingName(m_encoding).c_str()));
+                       : (const char*)wxFontMapperBase::GetEncodingName(m_encoding).mb_str()));

 #endif // wxUSE_FONTMAP
 
     // check for the special case of ASCII or ISO8859-1 charset: as we have
 #endif // wxUSE_FONTMAP
 
     // check for the special case of ASCII or ISO8859-1 charset: as we have


@@ -3306,14 +3314,13 @@ wxMBConv *wxCSConv::DoCreate() const
     if ( m_name )
 #endif // !wxUSE_FONTMAP
     {
     if ( m_name )
 #endif // !wxUSE_FONTMAP
     {

-        wxString name(m_name);

 #if wxUSE_FONTMAP
         wxFontEncoding encoding(m_encoding);
 #endif
 
 #if wxUSE_FONTMAP
         wxFontEncoding encoding(m_encoding);
 #endif
 

-        if ( !name.empty() )
+        if ( m_name )

         {
         {

-            wxMBConv_iconv *conv = new wxMBConv_iconv(name);
+            wxMBConv_iconv *conv = new wxMBConv_iconv(m_name);

             if ( conv->IsOk() )
                 return conv;
 
             if ( conv->IsOk() )
                 return conv;
 


@@ -3321,7 +3328,7 @@ wxMBConv *wxCSConv::DoCreate() const
 
 #if wxUSE_FONTMAP
             encoding =
 
 #if wxUSE_FONTMAP
             encoding =

-                wxFontMapperBase::Get()->CharsetToEncoding(name, false);
+                wxFontMapperBase::Get()->CharsetToEncoding(m_name, false);

 #endif // wxUSE_FONTMAP
         }
 #if wxUSE_FONTMAP
 #endif // wxUSE_FONTMAP
         }
 #if wxUSE_FONTMAP


@@ -3332,7 +3339,7 @@ wxMBConv *wxCSConv::DoCreate() const
                 if ( it->second.empty() )
                     return NULL;
 
                 if ( it->second.empty() )
                     return NULL;
 

-                wxMBConv_iconv *conv = new wxMBConv_iconv(it->second);
+                wxMBConv_iconv *conv = new wxMBConv_iconv(it->second.ToAscii());

                 if ( conv->IsOk() )
                     return conv;
 
                 if ( conv->IsOk() )
                     return conv;
 


@@ -3340,15 +3347,19 @@ wxMBConv *wxCSConv::DoCreate() const
             }
 
             const wxChar** names = wxFontMapperBase::GetAllEncodingNames(encoding);
             }
 
             const wxChar** names = wxFontMapperBase::GetAllEncodingNames(encoding);

-            // CS : in case this does not return valid names (eg for MacRoman) encoding
-            // got a 'failure' entry in the cache all the same, although it just has to
-            // be created using a different method, so only store failed iconv creation
-            // attempts (or perhaps we shoulnd't do this at all ?)
+            // CS : in case this does not return valid names (eg for MacRoman)
+            // encoding got a 'failure' entry in the cache all the same,
+            // although it just has to be created using a different method, so
+            // only store failed iconv creation attempts (or perhaps we
+            // shoulnd't do this at all ?)

             if ( names[0] != NULL )
             {
                 for ( ; *names; ++names )
                 {
             if ( names[0] != NULL )
             {
                 for ( ; *names; ++names )
                 {

-                    wxMBConv_iconv *conv = new wxMBConv_iconv(*names);
+                    // FIXME-UTF8: wxFontMapperBase::GetAllEncodingNames()
+                    //             will need changes that will obsolete this
+                    wxString name(*names);
+                    wxMBConv_iconv *conv = new wxMBConv_iconv(name.ToAscii());

                     if ( conv->IsOk() )
                     {
                         gs_nameCache[encoding] = *names;
                     if ( conv->IsOk() )
                     {
                         gs_nameCache[encoding] = *names;


@@ -3482,9 +3493,9 @@ wxMBConv *wxCSConv::DoCreate() const
                    m_name ? m_name
                       :
 #if wxUSE_FONTMAP
                    m_name ? m_name
                       :
 #if wxUSE_FONTMAP

-                         (const wxChar*)wxFontMapperBase::GetEncodingDescription(m_encoding).c_str()
+                         (const char*)wxFontMapperBase::GetEncodingDescription(m_encoding).ToAscii()

 #else // !wxUSE_FONTMAP
 #else // !wxUSE_FONTMAP

-                         (const wxChar*)wxString::Format(_("encoding %i"), m_encoding).c_str()
+                         (const char*)wxString::Format(_("encoding %i"), m_encoding).ToAscii()

 #endif // wxUSE_FONTMAP/!wxUSE_FONTMAP
               );
 
 #endif // wxUSE_FONTMAP/!wxUSE_FONTMAP
               );
 


@@ -3613,9 +3624,25 @@ size_t wxCSConv::GetMBNulLen() const
         return m_convReal->GetMBNulLen();
     }
 
         return m_convReal->GetMBNulLen();
     }
 

+    // otherwise, we are ISO-8859-1

     return 1;
 }
 
     return 1;
 }
 

+#if wxUSE_UNICODE_UTF8
+bool wxCSConv::IsUTF8() const
+{
+    CreateConvIfNeeded();
+
+    if ( m_convReal )
+    {
+        return m_convReal->IsUTF8();
+    }
+
+    // otherwise, we are ISO-8859-1
+    return false;
+}
+#endif
+

 
 #if wxUSE_UNICODE
 
 
 #if wxUSE_UNICODE
 


@@ -3626,7 +3653,7 @@ wxWCharBuffer wxSafeConvertMB2WX(const char *s)
 
     wxWCharBuffer wbuf(wxConvLibc.cMB2WX(s));
     if ( !wbuf )
 
     wxWCharBuffer wbuf(wxConvLibc.cMB2WX(s));
     if ( !wbuf )

-        wbuf = wxConvUTF8.cMB2WX(s);
+        wbuf = wxMBConvUTF8().cMB2WX(s);

     if ( !wbuf )
         wbuf = wxConvISO8859_1.cMB2WX(s);
 
     if ( !wbuf )
         wbuf = wxConvISO8859_1.cMB2WX(s);