From 1b2092fd06edfe1b752bb9e0308262d843419794 Mon Sep 17 00:00:00 2001
From: Giuseppe Scrivano <gscrivano@gnu.org>
Date: Sun, 30 May 2010 14:01:10 +0200
Subject: [PATCH] Cope better with unclosed html tags.

---
 NEWS             |  2 ++
 src/ChangeLog    |  6 ++++++
 src/html-parse.c | 22 +++++++++++++---------
 3 files changed, 21 insertions(+), 9 deletions(-)

diff --git a/NEWS b/NEWS
index 6faae62e..6f20a8b7 100644
--- a/NEWS
+++ b/NEWS
@@ -19,6 +19,8 @@ Please send GNU Wget bug reports to <bug-wget@gnu.org>.
 ** Set new cookies after an authorization failure.
 
 ** Exit with failure if -k is specified and -O is not a regular file.
+
+** Cope better with unclosed html tags.
 
 * Changes in Wget 1.12
 
diff --git a/src/ChangeLog b/src/ChangeLog
index a84ff378..e5cc8cd1 100644
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,9 @@
+2010-05-30  Giuseppe Scrivano  <gscrivano@gnu.org>
+
+	* html-parse.c (NAME_CHAR_P): Consider '<' an invalid character.
+	(advance_declaration): Close the tag if '<' is found.
+	(map_html_tags): Likewise.
+
 2010-05-27  Giuseppe Scrivano  <gscrivano@gnu.org>
 
 	* main.c (main): Exit with failure when -k is specified and -O is not
diff --git a/src/html-parse.c b/src/html-parse.c
index 070913f4..4cd86b93 100644
--- a/src/html-parse.c
+++ b/src/html-parse.c
@@ -528,13 +528,14 @@ convert_and_copy (struct pool *pool, const char *beg, const char *end, int flags
      * whitespace
      * 8-bit and control chars
      * characters that clearly cannot be part of name:
-       '=', '>', '/'.
+       '=', '<', '>', '/'.
 
    This only affects attribute and tag names; attribute values allow
    an even greater variety of characters.  */
 
 #define NAME_CHAR_P(x) ((x) > 32 && (x) < 127                           \
-                        && (x) != '=' && (x) != '>' && (x) != '/')
+                        && (x) != '=' && (x) != '<' && (x) != '>'       \
+                        && (x) != '/')
 
 #ifdef STANDALONE
 static int comment_backout_count;
@@ -619,6 +620,7 @@ advance_declaration (const char *beg, const char *end)
             case '\n':
               ch = *p++;
               break;
+            case '<':
             case '>':
               state = AC_S_DONE;
               break;
@@ -926,7 +928,7 @@ map_html_tags (const char *text, int size,
           }
       }
 
-    if (end_tag && *p != '>')
+    if (end_tag && *p != '>' && *p != '<')
       goto backout_tag;
 
     if (!name_allowed (allowed_tags, tag_name_begin, tag_name_end))
@@ -958,12 +960,12 @@ map_html_tags (const char *text, int size,
             /*              ^  */
             ADVANCE (p);
             SKIP_WS (p);
-            if (*p != '>')
+            if (*p != '<' || *p != '>')
               goto backout_tag;
           }
 
         /* Check for end of tag definition. */
-        if (*p == '>')
+        if (*p == '<' || *p == '>')
           break;
 
         /* Establish bounds of attribute name. */
@@ -978,7 +980,8 @@ map_html_tags (const char *text, int size,
 
         /* Establish bounds of attribute value. */
         SKIP_WS (p);
-        if (NAME_CHAR_P (*p) || *p == '/' || *p == '>')
+
+        if (NAME_CHAR_P (*p) || *p == '/' || *p == '<' || *p == '>')
           {
             /* Minimized attribute syntax allows `=' to be omitted.
                For example, <UL COMPACT> is a valid shorthand for <UL
@@ -1015,7 +1018,7 @@ map_html_tags (const char *text, int size,
                         newline_seen = true;
                         continue;
                       }
-                    else if (newline_seen && *p == '>')
+                    else if (newline_seen && (*p == '<' || *p == '>'))
                       break;
                     ADVANCE (p);
                   }
@@ -1040,7 +1043,7 @@ map_html_tags (const char *text, int size,
                    violated by, for instance, `%' in `width=75%'.
                    We'll be liberal and allow just about anything as
                    an attribute value.  */
-                while (!c_isspace (*p) && *p != '>')
+                while (!c_isspace (*p) && *p != '<' && *p != '>')
                   ADVANCE (p);
                 attr_value_end = p; /* <foo bar=baz qux=quix> */
                                     /*             ^          */
@@ -1138,7 +1141,8 @@ map_html_tags (const char *text, int size,
         }
 
       mapfun (&taginfo, maparg);
-      ADVANCE (p);
+      if (*p != '<')
+        ADVANCE (p);
     }
     goto look_for_tag;