From 3f0a0cb644438d4d8e3294cd0b1245d0edb0c6c6 Mon Sep 17 00:00:00 2001
From: Sebastian Pipping <sebastian@pipping.org>
Date: Tue, 8 Feb 2022 04:32:20 +0100
Subject: [PATCH] lib: Add missing validation of encoding (CVE-2022-25235)

---
 lib/xmltok_impl.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/lib/xmltok_impl.c b/lib/xmltok_impl.c
index 0430591b4..64a3b2c15 100644
--- a/lib/xmltok_impl.c
+++ b/lib/xmltok_impl.c
@@ -69,7 +69,7 @@
   case BT_LEAD##n:                                                             \
     if (end - ptr < n)                                                         \
       return XML_TOK_PARTIAL_CHAR;                                             \
-    if (! IS_NAME_CHAR(enc, ptr, n)) {                                         \
+    if (IS_INVALID_CHAR(enc, ptr, n) || ! IS_NAME_CHAR(enc, ptr, n)) {         \
       *nextTokPtr = ptr;                                                       \
       return XML_TOK_INVALID;                                                  \
     }                                                                          \
@@ -98,7 +98,7 @@
   case BT_LEAD##n:                                                             \
     if (end - ptr < n)                                                         \
       return XML_TOK_PARTIAL_CHAR;                                             \
-    if (! IS_NMSTRT_CHAR(enc, ptr, n)) {                                       \
+    if (IS_INVALID_CHAR(enc, ptr, n) || ! IS_NMSTRT_CHAR(enc, ptr, n)) {       \
       *nextTokPtr = ptr;                                                       \
       return XML_TOK_INVALID;                                                  \
     }                                                                          \
@@ -1142,6 +1142,10 @@ PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
   case BT_LEAD##n:                                                             \
     if (end - ptr < n)                                                         \
       return XML_TOK_PARTIAL_CHAR;                                             \
+    if (IS_INVALID_CHAR(enc, ptr, n)) {                                        \
+      *nextTokPtr = ptr;                                                       \
+      return XML_TOK_INVALID;                                                  \
+    }                                                                          \
     if (IS_NMSTRT_CHAR(enc, ptr, n)) {                                         \
       ptr += n;                                                                \
       tok = XML_TOK_NAME;                                                      \