Mercurial > projects > ddmd

diff dmd/Lexer.d @ 49:0aa7d1437ada

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
AttribDeclaration.oneMember Lexer.decodeUTF WithStatement.ctor StructDeclaration.syntaxCopy CtorDeclaration.syntaxCopy ConditionalStatement.syntaxCopy ProtDeclaration.syntaxCopy ArrayScopeSymbol.this TemplateDeclaration.toChars
author korDen
date Sat, 21 Aug 2010 07:53:20 +0400
parents 544b922227c7
children b7d29f613539
line wrap: on
line diff
--- a/dmd/Lexer.d	Sat Aug 21 07:39:45 2010 +0400
+++ b/dmd/Lexer.d	Sat Aug 21 07:53:20 2010 +0400
@@ -2466,9 +2466,36 @@
 		assert(false);
 	}
 
+	/********************************************
+	 * Decode UTF character.
+	 * Issue error messages for invalid sequences.
+	 * Return decoded character, advance p to last character in UTF sequence.
+	 */
     uint decodeUTF()
 	{
-		assert(false);
+		dchar u;
+		ubyte c;
+		ubyte* s = p;
+		size_t len;
+		size_t idx;
+		string msg;
+
+		c = *s;
+		assert(c & 0x80);
+
+		// Check length of remaining string up to 6 UTF-8 characters
+		for (len = 1; len < 6 && s[len]; len++) {
+			;
+		}
+
+		idx = 0;
+		msg = utf_decodeChar(cast(string)s[0..len], &idx, &u);
+		p += idx - 1;
+		if (msg)
+		{
+			error("%s", msg);
+		}
+		return u;
 	}
 
     void getDocComment(Token* t, uint lineComment)