From: Tatsuo Ishii <ishii@postgresql.org>
Date: Sun, 28 May 2000 04:57:06 +0000 (+0000)
Subject: Make EUC_CN support more robust.
X-Git-Tag: REL7_0_2~17
X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=67f6b333acc70e7346078dde25fb237465a31271;p=thirdparty%2Fpostgresql.git

Make EUC_CN support more robust.
---

diff --git a/src/backend/utils/mb/wchar.c b/src/backend/utils/mb/wchar.c
index bc576a1789b..383ec476f70 100644
--- a/src/backend/utils/mb/wchar.c
+++ b/src/backend/utils/mb/wchar.c
@@ -1,7 +1,7 @@
 /*
  * conversion functions between pg_wchar and multi-byte streams.
  * Tatsuo Ishii
- * $Id: wchar.c,v 1.11 2000/04/12 17:16:06 momjian Exp $
+ * $Id: wchar.c,v 1.11.2.1 2000/05/28 04:57:06 ishii Exp $
  *
  * WIN1250 client encoding updated by Pavel Behal
  *
@@ -166,7 +166,11 @@ pg_euccn_mblen(const unsigned char *s)
 {
 	int			len;
 
-	if (*s & 0x80)
+	/* if the first byte > 0x80, then second byte must be > 0x80, also,
+	 * according to the EUC-CN standard. But in reality there are
+	 * so many broken EUC-CN data violating the rule. Sigh.
+	 */
+	if ((*s & 0x80) && (*(s+1) & 0x80))
 		len = 2;
 	else
 		len = 1;