X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/467175ab3f2177aa824ceb7b67934fd54ce4e8e0..76015a6bb79156d8af1b5e8b4b8e46b4e192d18f:/src/common/stringops.cpp

diff --git a/src/common/stringops.cpp b/src/common/stringops.cpp
index ac0455da53..ff44bcd182 100644
--- a/src/common/stringops.cpp
+++ b/src/common/stringops.cpp
@@ -34,7 +34,7 @@
 // UTF-8 sequences lengths
 // ---------------------------------------------------------------------------
 
-unsigned char wxStringOperationsUtf8::ms_utf8IterTable[256] = {
+const unsigned char wxStringOperationsUtf8::ms_utf8IterTable[256] = {
     // single-byte sequences (ASCII):
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,  // 00..0F
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,  // 10..1F
@@ -87,17 +87,26 @@ unsigned char wxStringOperationsUtf8::ms_utf8IterTable[256] = {
 // U+100000..U+10FFFF |  F4      |  80..8F  |  80..BF  |  80..BF  |
 // -------------------+----------+----------+----------+----------+
 
-bool wxStringOperationsUtf8::IsValidUtf8String(const char *str)
+bool wxStringOperationsUtf8::IsValidUtf8String(const char *str, size_t len)
 {
     if ( !str )
         return true; // empty string is UTF8 string
 
     const unsigned char *c = (const unsigned char*)str;
+    const unsigned char * const end = (len == wxStringImpl::npos) ? NULL : c + len;
 
-    for ( ; *c; ++c )
+    for ( ; c != end && *c; ++c )
     {
         unsigned char b = *c;
 
+        if ( end != NULL )
+        {
+            // if the string is not NULL-terminated, verify we have enough
+            // bytes in it left for current character's encoding:
+            if ( c + ms_utf8IterTable[*c] > end )
+                return false;
+        }
+
         if ( b <= 0x7F ) // 00..7F
             continue;
 
@@ -122,7 +131,16 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str)
             if ( !(b >= 0x80 && b <= 0xBF ) )
                 return false;
         }
-        else if ( b <= 0xEF ) // E1..EF
+        else if ( b == 0xED )
+        {
+            b = *(++c);
+            if ( !(b >= 0x80 && b <= 0x9F ) )
+                return false;
+            b = *(++c);
+            if ( !(b >= 0x80 && b <= 0xBF ) )
+                return false;
+        }
+        else if ( b <= 0xEF ) // E1..EC EE..EF
         {
             for ( int i = 0; i < 2; ++i )
             {
@@ -173,21 +191,14 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str)
     return true;
 }
 
-#ifdef __WXDEBUG__
-bool wxStringOperationsUtf8::IsValidUtf8LeadByte(unsigned char c)
-{
-    return (c <= 0x7F) || (c >= 0xC2 && c <= 0xF4);
-}
-#endif
-
-
-wxStringOperationsUtf8::Utf8CharBuffer
-wxStringOperationsUtf8::EncodeChar(const wxUniChar& ch)
+// NB: this is in this file and not unichar.cpp to keep all UTF-8 encoding
+//     code in single place
+wxUniChar::Utf8CharBuffer wxUniChar::AsUTF8() const
 {
-    Utf8CharBuffer buf;
+    Utf8CharBuffer buf = { "" }; // init to avoid g++ 4.1 warning with -O2
     char *out = buf.data;
 
-    wxUniChar::value_type code = ch.GetValue();
+    value_type code = GetValue();
 
     //    Char. number range   |        UTF-8 octet sequence
     //       (hexadecimal)     |              (binary)
@@ -231,7 +242,7 @@ wxStringOperationsUtf8::EncodeChar(const wxUniChar& ch)
     }
     else
     {
-        wxFAIL_MSG( _T("trying to encode undefined Unicode character") );
+        wxFAIL_MSG( wxT("trying to encode undefined Unicode character") );
         out[0] = 0;
     }
 
@@ -239,13 +250,12 @@ wxStringOperationsUtf8::EncodeChar(const wxUniChar& ch)
 }
 
 wxUniChar
-wxStringOperationsUtf8::DecodeChar(wxStringImpl::const_iterator i)
+wxStringOperationsUtf8::DecodeNonAsciiChar(wxStringImpl::const_iterator i)
 {
     wxASSERT( IsValidUtf8LeadByte(*i) );
 
-    wxUniChar::value_type code = 0;
     size_t len = GetUtf8CharLength(*i);
-    wxASSERT_MSG( len <= 4, _T("invalid UTF-8 sequence length") );
+    wxASSERT_MSG( len <= 4, wxT("invalid UTF-8 sequence length") );
 
     //    Char. number range   |        UTF-8 octet sequence
     //       (hexadecimal)     |              (binary)
@@ -261,7 +271,7 @@ wxStringOperationsUtf8::DecodeChar(wxStringImpl::const_iterator i)
 
     // mask to extract lead byte's value ('x' bits above), by sequence's length:
     static const unsigned char s_leadValueMask[4] =  { 0x7F, 0x1F, 0x0F, 0x07 };
-#ifdef __WXDEBUG__
+#if wxDEBUG_LEVEL
     // mask and value of lead byte's most significant bits, by length:
     static const unsigned char s_leadMarkerMask[4] = { 0x80, 0xE0, 0xF0, 0xF8 };
     static const unsigned char s_leadMarkerVal[4] =  { 0x00, 0xC0, 0xE0, 0xF0 };
@@ -270,15 +280,15 @@ wxStringOperationsUtf8::DecodeChar(wxStringImpl::const_iterator i)
     // extract the lead byte's value bits:
     wxASSERT_MSG( ((unsigned char)*i & s_leadMarkerMask[len-1]) ==
                   s_leadMarkerVal[len-1],
-                  _T("invalid UTF-8 lead byte") );
-    code = (unsigned char)*i & s_leadValueMask[len-1];
+                  wxT("invalid UTF-8 lead byte") );
+    wxUniChar::value_type code = (unsigned char)*i & s_leadValueMask[len-1];
 
     // all remaining bytes, if any, are handled in the same way regardless of
     // sequence's length:
     for ( ++i ; len > 1; --len, ++i )
     {
         wxASSERT_MSG( ((unsigned char)*i & 0xC0) == 0x80,
-                      _T("invalid UTF-8 byte") );
+                      wxT("invalid UTF-8 byte") );
 
         code <<= 6;
         code |= (unsigned char)*i & 0x3F;