interpret ~ specially only when it is the first character of the path (closes #10948...

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

diff --git a/src/common/stringops.cpp b/src/common/stringops.cpp
index 205004806242242cb4048d1fdaccfa7817721879..c69a18774fb8ba7713a09554e43d831c3a31828f 100644 (file)
--- 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
@@ -131,7 +131,16 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str, size_t len)
             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 )
             {
@@ -182,19 +191,11 @@ bool wxStringOperationsUtf8::IsValidUtf8String(const char *str, size_t len)
     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
 {
-    Utf8CharBuffer buf;
+    Utf8CharBuffer buf = { "" }; // init to avoid g++ 4.1 warning with -O2
     char *out = buf.data;
 
     value_type code = GetValue();
@@ -249,7 +250,7 @@ wxUniChar::Utf8CharBuffer wxUniChar::AsUTF8() const
 }
 
 wxUniChar
-wxStringOperationsUtf8::DecodeChar(wxStringImpl::const_iterator i)
+wxStringOperationsUtf8::DecodeNonAsciiChar(wxStringImpl::const_iterator i)
 {
     wxASSERT( IsValidUtf8LeadByte(*i) );
 
@@ -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 };
-#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 };