gconvert: Error out if g_escape_uri_string() would overflow

If the string to escape contains a very large number of unacceptable
characters (which would need escaping), the calculation of the length of
the escaped string could overflow, leading to a potential write off the
end of the newly allocated string.

In addition to that, the number of unacceptable characters was counted
in a signed integer, which would overflow to become negative, making it
easier for an attacker to craft an input string which would cause an
out-of-bounds write.

Fix that by validating the allocation length, and using an unsigned
integer to count the number of unacceptable characters.

Spotted by treeplus. Thanks to the Sovereign Tech Resilience programme
from the Sovereign Tech Agency. ID: #YWH-PGM9867-134

Signed-off-by: Philip Withnall <pwithnall@gnome.org>

Fixes: #3827

Backport 2.86: Changed the translatable error message to re-use an
existing translatable string, to avoid adding new translatable strings
to a stable branch. The re-used string doesn’t perfectly match the
error, but it’s good enough given that no users will ever see it.

glib/gconvert.c

@@ -1337,7 +1337,8 @@ static const gchar hex[] = "0123456789ABCDEF";
  * escape something else, please read RFC-2396 */
 static gchar *
 g_escape_uri_string (const gchar         *string,
-		     UnsafeCharacterSet mask)
+                     UnsafeCharacterSet   mask,
+                     GError             **error)
 {
 #define ACCEPTABLE(a) ((a)>=32 && (a)<128 && (acceptable[(a)-32] & use_mask))
 
@@ -1345,7 +1346,7 @@ g_escape_uri_string (const gchar *string,
   gchar *q;
   gchar *result;
   int c;
-  gint unacceptable;
+  size_t unacceptable;
   UnsafeCharacterSet use_mask;
   
   g_return_val_if_fail (mask == UNSAFE_ALL
@@ -1363,6 +1364,13 @@ g_escape_uri_string (const gchar *string,
 	unacceptable++;
     }
 
+  if (unacceptable >= (G_MAXSIZE - (p - string)) / 2)
+    {
+      g_set_error_literal (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
+                           _("Invalid hostname"));
+      return NULL;
+    }
+
   result = g_malloc (p - string + unacceptable * 2 + 1);
   
   use_mask = mask;
@@ -1388,11 +1396,12 @@ g_escape_uri_string (const gchar *string,
 
 static gchar *
 g_escape_file_uri (const gchar  *hostname,
-		   const gchar *pathname)
+                   const gchar  *pathname,
+                   GError      **error)
 {
   char *escaped_hostname = NULL;
-  char *escaped_path;
+  char *escaped_path = NULL;
-  char *res;
+  char *res = NULL;
 
 #ifdef G_OS_WIN32
   char *p, *backslash;
@@ -1413,10 +1422,14 @@ g_escape_file_uri (const gchar *hostname,
 
   if (hostname && *hostname != '\0')
     {
-      escaped_hostname = g_escape_uri_string (hostname, UNSAFE_HOST);
+      escaped_hostname = g_escape_uri_string (hostname, UNSAFE_HOST, error);
+      if (escaped_hostname == NULL)
+        goto out;
     }
 
-  escaped_path = g_escape_uri_string (pathname, UNSAFE_PATH);
+  escaped_path = g_escape_uri_string (pathname, UNSAFE_PATH, error);
+  if (escaped_path == NULL)
+    goto out;
 
   res = g_strconcat ("file://",
 		     (escaped_hostname) ? escaped_hostname : "",
@@ -1424,6 +1437,7 @@ g_escape_file_uri (const gchar *hostname,
 		     escaped_path,
 		     NULL);
 
+out:
 #ifdef G_OS_WIN32
   g_free ((char *) pathname);
 #endif
@@ -1757,7 +1771,7 @@ g_filename_to_uri (const gchar *filename,
     hostname = NULL;
 #endif
 
-  escaped_uri = g_escape_file_uri (hostname, filename);
+  escaped_uri = g_escape_file_uri (hostname, filename, error);
 
   return escaped_uri;
 }