From 1ac17806967d1bef9b10dcd43450ddb02da3141d Mon Sep 17 00:00:00 2001
From: Lu Wang <coolwanglu@gmail.com>
Date: Fri, 17 Aug 2012 15:25:10 +0800
Subject: [PATCH] add a function detecting illegal unicode in html

---
 src/util.h | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/src/util.h b/src/util.h
index 92c9c5e..1e7a740 100644
--- a/src/util.h
+++ b/src/util.h
@@ -43,6 +43,28 @@ static inline bool _tm_equal(const double * tm1, const double * tm2, int size =
     return true;
 }
 
+/*
+ * http://en.wikipedia.org/wiki/HTML_decimal_character_rendering
+ */
+static inline bool isLegalUnicode(Unicode u)
+{
+    /*
+    if((u == 9) || (u == 10) || (u == 13))
+        return true;
+        */
+
+    if(u <= 31) 
+        return false;
+
+    if((u >= 127) && (u <= 159))
+        return false;
+
+    if((u >= 0xd800) && (u <= 0xdfff))
+        return false;
+
+    return true;
+}
+
 static inline void outputUnicodes(std::ostream & out, const Unicode * u, int uLen)
 {
     for(int i = 0; i < uLen; ++i)