Fix UTF32Char to String conversion for invalid unicode characters

Invalid, isolated surrogate pairs, UTF32 characters should handled specially and converted to UniChars first

HTMLKit/HTMLInputStreamReader.m

@@ -66,7 +66,7 @@
 		[HTMLInputStreamReaderErrors emitParseError:HTMLStreamReaderErrorIsolatedLowSurrogate
 										 atLocation:_location
 										andCallback:_errorCallback];
-		return REPLACEMENT_CHAR;
+		return nextInputCharacter;
 	}
 
 	if (CFStringIsSurrogateHighCharacter(nextInputCharacter)) {
@@ -75,7 +75,7 @@
 			[HTMLInputStreamReaderErrors emitParseError:HTMLStreamReaderErrorIsolatedHighSurrogate
 											 atLocation:_location
 											andCallback:_errorCallback];
-			return REPLACEMENT_CHAR;
+			return nextInputCharacter;
 		}
 
 		nextInputCharacter = CFStringGetLongCharacterForSurrogatePair(nextInputCharacter, surrogateLow);

HTMLKit/HTMLTokenizerCharacters.h

@@ -175,5 +175,7 @@ NS_INLINE NSString * StringFromUniChar(unichar character)
 
 NS_INLINE NSString * StringFromUTF32Char(UTF32Char character)
 {
-	return [[NSString alloc] initWithBytes:&character length:sizeof(character) encoding:NSUTF32LittleEndianStringEncoding];
+	unichar pair[2];
+	Boolean isPair = CFStringGetSurrogatePairForLongCharacter(character, pair);
+	return [[NSString alloc] initWithCharacters:(const unichar *)&pair length:(isPair ? 2 : 1)];
 }