Implement the client-header-order directive

It can be used to forward client headers in a different
order than they arrived.

diff --git a/doc/source/p-config.sgml b/doc/source/p-config.sgml
index 8e171f1..c8d2c8d 100644 (file)
--- a/doc/source/p-config.sgml
+++ b/doc/source/p-config.sgml
@@ -3,7 +3,7 @@
 
  Purpose     :  Used with other docs and files only.
 
 
  Purpose     :  Used with other docs and files only.
 

- $Id: p-config.sgml,v 2.80 2012/03/19 12:56:26 fabiankeil Exp $
+ $Id: p-config.sgml,v 2.81 2012/03/19 12:56:41 fabiankeil Exp $

 
  Copyright (C) 2001-2011 Privoxy Developers http://www.privoxy.org/
  See LICENSE.
 
  Copyright (C) 2001-2011 Privoxy Developers http://www.privoxy.org/
  See LICENSE.


@@ -97,7 +97,7 @@
  Sample Configuration File for Privoxy v&p-version;
 </title>
 <para>
  Sample Configuration File for Privoxy v&p-version;
 </title>
 <para>

- $Id: p-config.sgml,v 2.80 2012/03/19 12:56:26 fabiankeil Exp $
+ $Id: p-config.sgml,v 2.81 2012/03/19 12:56:41 fabiankeil Exp $

 </para>
 <para>
 Copyright (C) 2001-2011 Privoxy Developers http://www.privoxy.org/
 </para>
 <para>
 Copyright (C) 2001-2011 Privoxy Developers http://www.privoxy.org/


@@ -3120,6 +3120,70 @@ forward-socks4, forward-socks4a and forward-socks5</title>
 </sect3>
 
 
 </sect3>
 
 

+<sect3 renderas="sect4" id="client-header-order"><title>client-header-order</title>
+<variablelist>
+ <varlistentry>
+  <term>Specifies:</term>
+  <listitem>
+   <para>
+    The order in which client headers are sorted before forwarding them.
+   </para>
+  </listitem>
+ </varlistentry>
+ <varlistentry>
+  <term>Type of value:</term>
+  <listitem>
+   <para>
+    <replaceable>Client header names delimited by spaces or tabs</replaceable>
+   </para>
+  </listitem>
+ </varlistentry>
+ <varlistentry>
+  <term>Default value:</term>
+  <listitem>
+   <para>None</para>
+  </listitem>
+ </varlistentry>
+ <varlistentry>
+  <term>Notes:</term>
+  <listitem>
+   <para>
+     By default &my-app; leaves the client headers in the order they
+     were sent by the client. Headers are modified in-place, new headers
+     are added at the end of the already existing headers.
+   </para>
+   <para>
+     The header order can be used to fingerprint client requests
+     independently of other headers like the User-Agent.
+   </para>
+   <para>
+     This directive allows to sort the headers differently to better
+     mimic a different User-Agent. Client headers will be emitted
+     in the order given, headers whose name isn't explicitly specified
+     are added at the end.
+   </para>
+   <para>
+     Note that sorting headers in an uncommon way will make fingerprinting
+     actually easier. Encrypted headers are not affected by this directive.
+   </para>
+  </listitem>
+ </varlistentry>
+</variablelist>
+<![%config-file;[<literallayout>@@#client-header-order Host \
+# User-Agent \
+# Accept \
+# Accept-Language \
+# Accept-Encoding \
+# Proxy-Connection,\
+# Referer,Cookie \
+# If-Modified-Since \
+# Cache-Control \
+# Content-Length \
+# Content-Type
+</literallayout>]]>
+</sect3>
+
+

 </sect2>
 
 <!--  ~  End section  ~  -->
 </sect2>
 
 <!--  ~  End section  ~  -->

diff --git a/loadcfg.c b/loadcfg.c
index 6c329ec..9a7a52d 100644 (file)
--- a/loadcfg.c
+++ b/loadcfg.c
@@ -1,4 +1,4 @@
-const char loadcfg_rcs[] = "$Id: loadcfg.c,v 1.128 2012/05/24 14:58:16 fabiankeil Exp $";
+const char loadcfg_rcs[] = "$Id: loadcfg.c,v 1.129 2012/06/08 15:15:11 fabiankeil Exp $";

 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/loadcfg.c,v $
 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/loadcfg.c,v $


@@ -123,6 +123,7 @@ static struct file_list *current_configfile = NULL;
 #define hash_admin_address               4112573064U /* "admin-address" */
 #define hash_allow_cgi_request_crunching  258915987U /* "allow-cgi-request-crunching" */
 #define hash_buffer_limit                1881726070U /* "buffer-limit */
 #define hash_admin_address               4112573064U /* "admin-address" */
 #define hash_allow_cgi_request_crunching  258915987U /* "allow-cgi-request-crunching" */
 #define hash_buffer_limit                1881726070U /* "buffer-limit */

+#define hash_client_header_order         2701453514U /* "client-header-order" */

 #define hash_compression_level           2464423563U /* "compression-level" */
 #define hash_confdir                        1978389U /* "confdir" */
 #define hash_connection_sharing          1348841265U /* "connection-sharing" */
 #define hash_compression_level           2464423563U /* "compression-level" */
 #define hash_confdir                        1978389U /* "confdir" */
 #define hash_connection_sharing          1348841265U /* "connection-sharing" */


@@ -233,6 +234,8 @@ static void unload_configfile (void * data)
       freez(config->re_filterfile[i]);
    }
 
       freez(config->re_filterfile[i]);
    }
 

+   list_remove_all(config->ordered_client_headers);
+

    freez(config->admin_address);
    freez(config->proxy_info_url);
    freez(config->proxy_args);
    freez(config->admin_address);
    freez(config->proxy_info_url);
    freez(config->proxy_args);


@@ -316,6 +319,75 @@ static int parse_toggle_state(const char *name, const char *value)
 }
 
 
 }
 
 

+/*********************************************************************
+ *
+ * Function    :  parse_client_header_order
+ *
+ * Description :  Parse the value of the header-order directive
+ *
+ * Parameters  :
+ *          1  :  ordered_header_list:  List to insert the ordered
+ *                                      headers into.
+ *          2  :  ordered_headers:  The ordered header names separated
+ *                                  by spaces or tabs.
+ *
+ *
+ * Returns     :  N/A
+ *
+ *********************************************************************/
+static void parse_client_header_order(struct list *ordered_header_list, const char *ordered_headers)
+{
+   char *original_headers_copy;
+   char **vector;
+   int number_of_headers;
+   int i;
+
+   assert(ordered_header_list != NULL);
+   assert(ordered_headers != NULL);
+
+   if (ordered_headers == NULL)
+   {
+      log_error(LOG_LEVEL_FATAL, "header-order used without argument");
+   }
+
+   /*
+    * XXX: This estimate is guaranteed to be high enough as we
+    *      let ssplit() ignore empty fields, but also a bit wasteful.
+    *      The same hack is used in get_last_url() so it looks like
+    *      a real solution is needed.
+    */
+   size_t max_segments = strlen(ordered_headers) / 2;
+   if (max_segments == 0)
+   {
+      max_segments = 1;
+   }
+   vector = malloc_or_die(max_segments * sizeof(char *));
+
+   original_headers_copy = strdup_or_die(ordered_headers);
+
+   number_of_headers = ssplit(original_headers_copy, "\t ", vector, max_segments);
+   if (number_of_headers == -1)
+   {
+      log_error(LOG_LEVEL_FATAL, "Failed to split ordered headers");
+   }
+
+   for (i = 0; i < number_of_headers; i++)
+   {
+      if (JB_ERR_OK != enlist(ordered_header_list, vector[i]))
+      {
+         log_error(LOG_LEVEL_FATAL,
+            "Failed to enlist ordered header: %s", vector[i]);
+      }
+   }
+
+   freez(vector);
+   freez(original_headers_copy);
+
+   return;
+
+}
+
+

 /*********************************************************************
  *
  * Function    :  load_config
 /*********************************************************************
  *
  * Function    :  load_config


@@ -539,6 +611,14 @@ struct configuration_spec * load_config(void)
             config->buffer_limit = (size_t)(1024 * atoi(arg));
             break;
 
             config->buffer_limit = (size_t)(1024 * atoi(arg));
             break;
 

+/* *************************************************************************
+ * client-header-order header-1 header-2 ... header-n
+ * *************************************************************************/
+         case hash_client_header_order:
+            list_remove_all(config->ordered_client_headers);
+            parse_client_header_order(config->ordered_client_headers, arg);
+            break;
+

 /* *************************************************************************
  * confdir directory-name
  * *************************************************************************/
 /* *************************************************************************
  * confdir directory-name
  * *************************************************************************/

diff --git a/parsers.c b/parsers.c
index e29c451..09a12e4 100644 (file)
--- a/parsers.c
+++ b/parsers.c
@@ -1,4 +1,4 @@
-const char parsers_rcs[] = "$Id: parsers.c,v 1.245 2012/04/06 15:17:10 fabiankeil Exp $";
+const char parsers_rcs[] = "$Id: parsers.c,v 1.246 2012/07/23 12:40:30 fabiankeil Exp $";

 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/parsers.c,v $
 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/parsers.c,v $


@@ -1025,6 +1025,79 @@ static jb_err scan_headers(struct client_state *csp)
 }
 
 
 }
 
 

+/*********************************************************************
+ *
+ * Function    :  enforce_header_order
+ *
+ * Description :  Enforces a given header order.
+ *
+ * Parameters  :
+ *          1  :  headers         = List of headers to order.
+ *          2  :  ordered_headers = List of ordered header names.
+ *
+ * Returns     :  N/A
+ *
+ *********************************************************************/
+static void enforce_header_order(struct list *headers, const struct list *ordered_headers)
+{
+   struct list_entry *sorted_header;
+   struct list new_headers[1];
+   struct list_entry *header;
+
+   init_list(new_headers);
+
+   /* The request line is always the first "header" */
+
+   assert(NULL != headers->first->str);
+   enlist(new_headers, headers->first->str);
+   freez(headers->first->str)
+
+   /* Enlist the specified headers in the given order */
+
+   for (sorted_header = ordered_headers->first; sorted_header != NULL;
+        sorted_header = sorted_header->next)
+   {
+      const size_t sorted_header_length = strlen(sorted_header->str);
+      for (header = headers->first; header != NULL; header = header->next)
+      {
+         /* Header enlisted in previous run? -> ignore */
+         if (header->str == NULL) continue;
+
+         if (0 == strncmpic(sorted_header->str, header->str, sorted_header_length)
+            && (header->str[sorted_header_length] == ':'))
+         {
+            log_error(LOG_LEVEL_HEADER, "Enlisting sorted header %s", header->str);
+            if (JB_ERR_OK != enlist(new_headers, header->str))
+            {
+               log_error(LOG_LEVEL_HEADER, "Failed to enlist %s", header->str);
+            }
+            freez(header->str);
+         }
+      }
+   }
+
+   /* Enlist the rest of the headers behind the ordered ones */
+   for (header = headers->first; header != NULL; header = header->next)
+   {
+      /* Header enlisted in previous run? -> ignore */
+      if (header->str == NULL) continue;
+
+      log_error(LOG_LEVEL_HEADER,
+         "Enlisting left-over header %s", header->str);
+      if (JB_ERR_OK != enlist(new_headers, header->str))
+      {
+         log_error(LOG_LEVEL_HEADER, "Failed to enlist %s", header->str);
+      }
+      freez(header->str);
+   }
+
+   list_remove_all(headers);
+   list_duplicate(headers, new_headers);
+   list_remove_all(new_headers);
+
+   return;
+}
+

 /*********************************************************************
  *
  * Function    :  sed
 /*********************************************************************
  *
  * Function    :  sed


@@ -1091,6 +1164,11 @@ jb_err sed(struct client_state *csp, int filter_server_headers)
       f++;
    }
 
       f++;
    }
 

+   if (!filter_server_headers && !list_is_empty(csp->config->ordered_client_headers))
+   {
+      enforce_header_order(csp->headers, csp->config->ordered_client_headers);
+   }
+

    return err;
 }
 
    return err;
 }
 

diff --git a/project.h b/project.h
index 398dc1a..98c46af 100644 (file)
--- a/project.h
+++ b/project.h
@@ -1,7 +1,7 @@
 #ifndef PROJECT_H_INCLUDED
 #define PROJECT_H_INCLUDED
 /** Version string. */
 #ifndef PROJECT_H_INCLUDED
 #define PROJECT_H_INCLUDED
 /** Version string. */

-#define PROJECT_H_VERSION "$Id: project.h,v 1.176 2012/06/19 12:50:22 fabiankeil Exp $"
+#define PROJECT_H_VERSION "$Id: project.h,v 1.177 2012/07/23 12:43:42 fabiankeil Exp $"

 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/project.h,v $
 /*********************************************************************
  *
  * File        :  $Source: /cvsroot/ijbswa/current/project.h,v $


@@ -1261,6 +1261,9 @@ struct configuration_spec
    /** The short names of the pcre filter files. */
    const char *re_filterfile_short[MAX_AF_FILES];
 
    /** The short names of the pcre filter files. */
    const char *re_filterfile_short[MAX_AF_FILES];
 

+   /**< List of ordered client header names. */
+   struct list ordered_client_headers[1];
+

    /** The hostname to show on CGI pages, or NULL to use the real one. */
    const char *hostname;
 
    /** The hostname to show on CGI pages, or NULL to use the real one. */
    const char *hostname;