diff -ruN postgresql-7.5-old/src/backend/utils/mb/wchar.c postgresql-7.5/src/backend/utils/mb/wchar.c
--- postgresql-7.5-old/src/backend/utils/mb/wchar.c     2004-08-06 22:44:26.000000000 +1000
+++ postgresql-7.5/src/backend/utils/mb/wchar.c         2004-08-07 01:19:39.000000000 +1000
@@ -406,8 +406,14 @@
 		len = 1;
 	else if ((*s & 0xe0) == 0xc0)
 		len = 2;
-	else if ((*s & 0xe0) == 0xe0)
-		len = 3;
+        else if ((*s & 0xf0) == 0xe0)
+                len = 3;
+        else if ((*s & 0xf8) == 0xf0)
+                len = 4;
+        else if ((*s & 0xfc) == 0xf8)
+                len = 5;
+        else if ((*s & 0xfe) == 0xfc)
+                len = 6;
 	return (len);
 }
 
@@ -801,6 +801,36 @@
 
 #ifndef FRONTEND
 
+/* --------------------------------------------------------------------- */
+
+/*
+ * Utility routine to tell whether a sequence of bytes is legal UTF-8.
+ */
+  
+static unsigned char isLegalUTF8(const unsigned char *source, int srclen) {
+    int length = pg_utf_mblen(source);
+    if(length > srclen || length > 4) return false;
+    unsigned char a;
+    const unsigned char *srcptr = source+length;
+    switch (length) {
+	default: return false;
+	/* Everything else falls through when "true"... */
+	case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+	case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+	case 2: if ((a = (*--srcptr)) > 0xBF) return false;
+	switch (*source) {
+	    /* no fall-through in this inner switch */
+	    case 0xE0: if (a < 0xA0) return false; break;
+	    case 0xF0: if (a < 0x90) return false; break;
+	    case 0xF4: if (a > 0x8F) return false; break;
+	    default:  if (a < 0x80) return false;
+	}
+	case 1: if (*source >= 0x80 && *source < 0xC2) return false;
+	if (*source > 0xF4) return false;
+    }
+    return true;
+}
+
 /*
  * Verify mbstr to make sure that it has a valid character sequence.
  * mbstr is not necessarily NULL terminated; length of mbstr is
@@ -825,14 +863,16 @@
 	while (len > 0 && *mbstr)
 	{
 		/* special UTF-8 check */
-		if (encoding == PG_UTF8 && (*mbstr & 0xf8) == 0xf0)
+		if (encoding == PG_UTF8 && !isLegalUTF8(mbstr,len))
 		{
 			if (noError)
 				return false;
 			ereport(ERROR,
 					(errcode(ERRCODE_CHARACTER_NOT_IN_REPERTOIRE),
-			 errmsg("Unicode characters greater than or equal to 0x10000 are not supported")));
+			 errmsg("Invalid UNICODE byte sequence detected")));
 		}
+		if (encoding == PG_UTF8)
+		    return true;
 
 		l = pg_mblen(mbstr);