Merge in from trunk r64802 - r68625

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

diff --git a/src/common/stringops.cpp b/src/common/stringops.cpp
index 205004806242242cb4048d1fdaccfa7817721879..599c05c1d02cb685cc85c2929328a4bf317b1ba8 100644 (file)
--- a/src/common/stringops.cpp
+++ b/src/common/stringops.cpp
@@ -34,7 +34,7 @@
 // UTF-8 sequences lengths
 // ---------------------------------------------------------------------------
 
 // 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
     // 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


@@ -131,7 +131,16 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str, size_t len)
             if ( !(b >= 0x80 && b <= 0xBF ) )
                 return false;
         }
             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 )
             {
         {
             for ( int i = 0; i < 2; ++i )
             {


@@ -182,19 +191,11 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str, size_t len)
     return true;
 }
 
     return true;
 }
 

-#ifdef __WXDEBUG__
-bool wxStringOperationsUtf8::IsValidUtf8LeadByte(unsigned char c)
-{
-    return (c <= 0x7F) || (c >= 0xC2 && c <= 0xF4);
-}
-#endif
-
-

 // 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
 {
 // 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;
 
     value_type code = GetValue();
     char *out = buf.data;
 
     value_type code = GetValue();


@@ -241,7 +242,7 @@ wxUniChar::Utf8CharBuffer wxUniChar::AsUTF8() const
     }
     else
     {
     }
     else
     {

-        wxFAIL_MSG( _T("trying to encode undefined Unicode character") );
+        wxFAIL_MSG( wxT("trying to encode undefined Unicode character") );

         out[0] = 0;
     }
 
         out[0] = 0;
     }
 


@@ -249,13 +250,13 @@ wxUniChar::Utf8CharBuffer wxUniChar::AsUTF8() const
 }
 
 wxUniChar
 }
 
 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( 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)
 
     //    Char. number range   |        UTF-8 octet sequence
     //       (hexadecimal)     |              (binary)


@@ -271,7 +272,7 @@ 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 };
 
     // 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 };
     // 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 };


@@ -280,7 +281,7 @@ 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],
     // 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") );
+                  wxT("invalid UTF-8 lead byte") );

     code = (unsigned char)*i & s_leadValueMask[len-1];
 
     // all remaining bytes, if any, are handled in the same way regardless of
     code = (unsigned char)*i & s_leadValueMask[len-1];
 
     // all remaining bytes, if any, are handled in the same way regardless of


@@ -288,7 +289,7 @@ wxStringOperationsUtf8::DecodeChar(wxStringImpl::const_iterator i)
     for ( ++i ; len > 1; --len, ++i )
     {
         wxASSERT_MSG( ((unsigned char)*i & 0xC0) == 0x80,
     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;
 
         code <<= 6;
         code |= (unsigned char)*i & 0x3F;