use different strings for wchar_t and UTF-8-based Unicode builds

[wxWidgets.git] / include / wx / string.h

diff --git a/include/wx/string.h b/include/wx/string.h
index aaa1581d9d3ab97fb644ce2832f90ec763a0330e..0a72bfe3206dacfb33d94b37044935577a9cfc1f 100644 (file)
--- a/include/wx/string.h
+++ b/include/wx/string.h
@@ -1268,7 +1268,7 @@ public:
 
     // conversion to/from UTF-8:
 #if wxUSE_UNICODE_UTF8
-    static wxString FromUTF8(const char *utf8)
+    static wxString FromUTF8Unchecked(const char *utf8)
     {
       if ( !utf8 )
           return wxEmptyString;
@@ -1276,16 +1276,35 @@ public:
       wxASSERT( wxStringOperations::IsValidUtf8String(utf8) );
       return FromImpl(wxStringImpl(utf8));
     }
-    static wxString FromUTF8(const char *utf8, size_t len)
+    static wxString FromUTF8Unchecked(const char *utf8, size_t len)
     {
       if ( !utf8 )
           return wxEmptyString;
       if ( len == npos )
-          return FromUTF8(utf8);
+          return FromUTF8Unchecked(utf8);
 
       wxASSERT( wxStringOperations::IsValidUtf8String(utf8, len) );
       return FromImpl(wxStringImpl(utf8, len));
     }
+
+    static wxString FromUTF8(const char *utf8)
+    {
+        if ( !utf8 || !wxStringOperations::IsValidUtf8String(utf8) )
+            return "";
+
+        return FromImpl(wxStringImpl(utf8));
+    }
+    static wxString FromUTF8(const char *utf8, size_t len)
+    {
+        if ( len == npos )
+            return FromUTF8(utf8);
+
+        if ( !utf8 || !wxStringOperations::IsValidUtf8String(utf8, len) )
+            return "";
+
+        return FromImpl(wxStringImpl(utf8, len));
+    }
+
     const char* utf8_str() const { return wx_str(); }
     const char* ToUTF8() const { return wx_str(); }
 
@@ -1293,10 +1312,15 @@ public:
     // internal UTF-8 representation
     size_t utf8_length() const { return m_impl.length(); }
 #elif wxUSE_UNICODE_WCHAR
-    static wxString FromUTF8(const char *utf8)
-      { return wxString(utf8, wxMBConvUTF8()); }
-    static wxString FromUTF8(const char *utf8, size_t len)
+    static wxString FromUTF8(const char *utf8, size_t len = npos)
       { return wxString(utf8, wxMBConvUTF8(), len); }
+    static wxString FromUTF8Unchecked(const char *utf8, size_t len = npos)
+    {
+        const wxString s(utf8, wxMBConvUTF8(), len);
+        wxASSERT_MSG( !utf8 || !*utf8 || !s.empty(),
+                      "string must be valid UTF-8" );
+        return s;
+    }
     const wxCharBuffer utf8_str() const { return mb_str(wxMBConvUTF8()); }
     const wxCharBuffer ToUTF8() const { return utf8_str(); }
 #else // ANSI
@@ -1304,9 +1328,20 @@ public:
       { return wxString(wxMBConvUTF8().cMB2WC(utf8)); }
     static wxString FromUTF8(const char *utf8, size_t len)
     {
-      size_t wlen;
-      wxWCharBuffer buf(wxMBConvUTF8().cMB2WC(utf8, len == npos ? wxNO_LEN : len, &wlen));
-      return wxString(buf.data(), wlen);
+        size_t wlen;
+        wxWCharBuffer buf(wxMBConvUTF8().cMB2WC(utf8, len == npos ? wxNO_LEN : len, &wlen));
+        return wxString(buf.data(), wlen);
+    }
+    static wxString FromUTF8Unchecked(const char *utf8, size_t len = npos)
+    {
+        size_t wlen;
+        wxWCharBuffer buf(wxMBConvUTF8().cMB2WC(utf8,
+                                                len == npos ? wxNO_LEN : len,
+                                                &wlen));
+        wxASSERT_MSG( !utf8 || !*utf8 || wlen,
+                      "string must be valid UTF-8" );
+
+        return wxString(buf.data(), wlen);
     }
     const wxCharBuffer utf8_str() const
       { return wxMBConvUTF8().cWC2MB(wc_str()); }
@@ -1351,7 +1386,7 @@ public:
     const wxWX2MBbuf mbc_str() const { return mb_str(*wxConvCurrent); }
 
 #if wxUSE_UNICODE_WCHAR
-    const wxChar* wc_str() const { return wx_str(); }
+    const wchar_t* wc_str() const { return wx_str(); }
 #elif wxUSE_UNICODE_UTF8
     const wxWCharBuffer wc_str() const;
 #endif
@@ -1369,7 +1404,7 @@ public:
     const wxChar* mb_str() const { return wx_str(); }
 
     // for compatibility with wxUSE_UNICODE version
-    const wxChar* mb_str(const wxMBConv& WXUNUSED(conv)) const { return wx_str(); }
+    const char* mb_str(const wxMBConv& WXUNUSED(conv)) const { return wx_str(); }
 
     const wxWX2MBbuf mbc_str() const { return mb_str(); }
 
@@ -1379,6 +1414,15 @@ public:
     const wxCharBuffer fn_str() const { return wxConvFile.cWC2WX( wc_str( wxConvLibc ) ); }
 #endif // Unicode/ANSI
 
+#if wxUSE_UNICODE_UTF8
+    const wxWCharBuffer t_str() const { return wc_str(); }
+#elif wxUSE_UNICODE_WCHAR
+    const wchar_t* t_str() const { return wx_str(); }
+#else
+    const char* t_str() const { return wx_str(); }
+#endif 
+
+
   // overloaded assignment
     // from another wxString
   wxString& operator=(const wxString& stringSrc)
@@ -2895,7 +2939,7 @@ public:
                 len = lenWanted - 1;
             }
 
-            wxTmemcpy(m_buf.data(), buf, len + 1);
+            memcpy(m_buf.data(), buf, (len + 1)*sizeof(CharType));
         }
         //else: conversion failed, this can happen when trying to get Unicode
         //      string contents into a char string