[svn] Fix for bug #20296: User:pass@ given in Referer header.

diff --git a/ChangeLog b/ChangeLog
index 1b00c34b3ea285dcf0dac2e83ea2d9d61c2de9e6..2ced0876febb94b0dacb5d2cb57c50a5f8f43e9c 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,6 +1,7 @@
 2007-07-29  Micah Cowan  <micah@cowan.name>
 
-       * NEWS: No more auth before challenge.
+       * NEWS: No more auth before challenge. No more auth info in
+       Referer.
 
 2007-07-09  Micah Cowan  <micah@cowan.name>
 

diff --git a/NEWS b/NEWS
index e5f6878d563b90cee01c2ea7a1db02d21c082459..fe0968385d6e798d1e2b03863b4c853fb2c6db82 100644 (file)
--- a/NEWS
+++ b/NEWS
@@ -7,6 +7,9 @@ Please send GNU Wget bug reports to <bug-wget@gnu.org>.
 \f
 * Changes in Wget 1.11.
 
+** Authentication information is no longer sent as part of the Referer
+header in recursive fetches.
+
 ** No authentication credentials are sent until a challenge is issued,
 for improved security. Authentication handling is still not
 RFC-compliant, as once a Basic challenge has been received, it will

diff --git a/src/ChangeLog b/src/ChangeLog
index c576f1170a12ed43eb5fed6b3ce2e34d33ed0d05..852469800035395ad0d33a0193703abdb4bc34a3 100644 (file)
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,12 @@
+2007-07-29  Micah Cowan  <micah@cowan.name>
+
+       * url.h, url.c (url_string): Replaced bool arg of the url_string
+       function with enum url_auth_mode, with added option to
+       completely remove user/pass auth information.
+       * http.c, ftp.c, url.c, recur.c: Adapted call to url_string
+       function to fit new usage.
+       * recur.c (retrieve_tree): Remove auth info from Referer header.
+
 2007-07-28  Micah Cowan  <micah@cowan.name>
 
        * options.h, init.c, retr.c, main.c: renamed opt maxredirect

diff --git a/src/ftp.c b/src/ftp.c
index 2e32c1f03d04b45e9f723706d8e8d7ac1567925d..02d1d85c71f40d045b1ff445f98f1c6d15374bef 100644 (file)
--- a/src/ftp.c
+++ b/src/ftp.c
@@ -1153,7 +1153,7 @@ ftp_loop_internal (struct url *u, struct fileinfo *f, ccon *con)
       /* Print fetch message, if opt.verbose.  */
       if (opt.verbose)
        {
-         char *hurl = url_string (u, true);
+         char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
          char tmp[256];
          strcpy (tmp, "        ");
          if (count > 1)
@@ -1234,7 +1234,7 @@ ftp_loop_internal (struct url *u, struct fileinfo *f, ccon *con)
          /* Need to hide the password from the URL.  The `if' is here
              so that we don't do the needless allocation every
              time. */
-         char *hurl = url_string (u, true);
+         char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
          logprintf (LOG_NONVERBOSE, "%s URL: %s [%s] -> \"%s\" [%d]\n",
                     tms, hurl, number_to_static_string (len), locf, count);
          xfree (hurl);

diff --git a/src/http.c b/src/http.c
index af024bb92a51d83429b4e5a73c1e388a08a376d9..e6f266a307a8ce668517d80e0fc5fba048cad70b 100644 (file)
--- a/src/http.c
+++ b/src/http.c
@@ -2359,7 +2359,7 @@ Spider mode enabled. Check if remote file exists.\n"));
       /* Print fetch message, if opt.verbose.  */
       if (opt.verbose)
         {
-          char *hurl = url_string (u, true);
+          char *hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
           
           if (count > 1) 
             {
@@ -2483,7 +2483,7 @@ Spider mode enabled. Check if remote file exists.\n"));
           if (!opt.verbose)
             {
               /* #### Ugly ugly ugly! */
-              hurl = url_string (u, true);
+              hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
               logprintf (LOG_NONVERBOSE, "%s:\n", hurl);
             }
           /* Maybe we should always keep track of broken links, not just in
@@ -2492,7 +2492,7 @@ Spider mode enabled. Check if remote file exists.\n"));
             {
               /* #### Again: ugly ugly ugly! */
               if (!hurl) 
-                hurl = url_string (u, true);
+                hurl = url_string (u, URL_AUTH_HIDE_PASSWD);
               nonexisting_url (hurl);
               logprintf (LOG_NOTQUIET, _("\
 Remote file does not exist -- broken link!!!\n"));

diff --git a/src/recur.c b/src/recur.c
index 53cd39cdc618311ecba790588478d1e3b4963fb7..9ecb7d8efb8bfd8bfe33271e0a1d2a8e7c8b283c 100644 (file)
--- a/src/recur.c
+++ b/src/recur.c
@@ -324,8 +324,14 @@ retrieve_tree (const char *start_url)
            {
              struct urlpos *child = children;
              struct url *url_parsed = url_parsed = url_parse (url, NULL);
+              char *referer_url = url;
+              bool strip_auth = url_parsed->user;
              assert (url_parsed != NULL);
 
+              /* Strip auth info if present */
+              if (strip_auth)
+                referer_url = url_string (url_parsed, URL_AUTH_HIDE);
+
              for (; child; child = child->next)
                {
                  if (child->ignore_when_downloading)
@@ -336,7 +342,7 @@ retrieve_tree (const char *start_url)
                                        blacklist))
                    {
                      url_enqueue (queue, xstrdup (child->url->url),
-                                  xstrdup (url), depth + 1,
+                                  xstrdup (referer_url), depth + 1,
                                   child->link_expect_html);
                      /* We blacklist the URL we have enqueued, because we
                         don't want to enqueue (and hence download) the
@@ -345,6 +351,8 @@ retrieve_tree (const char *start_url)
                    }
                }
 
+              if (strip_auth)
+                xfree (referer_url);
              url_free (url_parsed);
              free_urlpos (children);
            }
@@ -428,7 +436,7 @@ download_child_p (const struct urlpos *upos, struct url *parent, int depth,
     {
       if (opt.spider) 
        {
-          char *referrer = url_string (parent, true);
+          char *referrer = url_string (parent, URL_AUTH_HIDE_PASSWD);
           DEBUGP (("download_child_p: parent->url is: `%s'\n", parent->url));
           visited_url (url, referrer);
          xfree (referrer);
@@ -628,3 +636,5 @@ descend_redirect_p (const char *redirected, const char *original, int depth,
 
   return success;
 }
+
+/* vim:set sts=2 sw=2 cino+={s: */

diff --git a/src/url.c b/src/url.c
index 650c634736ed41c034f1b3bd16bd6365fad7dbc9..e95d572ff05377e5b518aa4353675e882e6acd2a 100644 (file)
--- a/src/url.c
+++ b/src/url.c
@@ -855,7 +855,7 @@ url_parse (const char *url, int *error)
       /* If we suspect that a transformation has rendered what
          url_string might return different from URL_ENCODED, rebuild
          u->url using url_string.  */
-      u->url = url_string (u, false);
+      u->url = url_string (u, URL_AUTH_SHOW);
 
       if (url_encoded != url)
         xfree ((char *) url_encoded);
@@ -1071,7 +1071,7 @@ sync_path (struct url *u)
 
   /* Regenerate u->url as well.  */
   xfree (u->url);
-  u->url = url_string (u, false);
+  u->url = url_string (u, URL_AUTH_SHOW);
 }
 
 /* Mutators.  Code in ftp.c insists on changing u->dir and u->file.
@@ -1814,7 +1814,7 @@ uri_merge (const char *base, const char *link)
    the URL will be quoted.  */
 
 char *
-url_string (const struct url *url, bool hide_password)
+url_string (const struct url *url, enum url_auth_mode auth_mode)
 {
   int size;
   char *result, *p;
@@ -1831,13 +1831,16 @@ url_string (const struct url *url, bool hide_password)
   /* Make sure the user name and password are quoted. */
   if (url->user)
     {
-      quoted_user = url_escape_allow_passthrough (url->user);
-      if (url->passwd)
+      if (auth_mode != URL_AUTH_HIDE)
         {
-          if (hide_password)
-            quoted_passwd = HIDDEN_PASSWORD;
-          else
-            quoted_passwd = url_escape_allow_passthrough (url->passwd);
+          quoted_user = url_escape_allow_passthrough (url->user);
+          if (url->passwd)
+            {
+              if (auth_mode = URL_AUTH_HIDE_PASSWD)
+                quoted_passwd = HIDDEN_PASSWORD;
+              else
+                quoted_passwd = url_escape_allow_passthrough (url->passwd);
+            }
         }
     }
 
@@ -1899,7 +1902,8 @@ url_string (const struct url *url, bool hide_password)
 
   if (quoted_user && quoted_user != url->user)
     xfree (quoted_user);
-  if (quoted_passwd && !hide_password && quoted_passwd != url->passwd)
+  if (quoted_passwd && auth_mode == URL_AUTH_SHOW
+      && quoted_passwd != url->passwd)
     xfree (quoted_passwd);
   if (quoted_host != url->host)
     xfree (quoted_host);

diff --git a/src/url.h b/src/url.h
index f4b6b7e19a13b5eb121a51dcbfab9dad7404e4ec..80e3e3c13ae538b80635e791296ccd1b2f42d47a 100644 (file)
--- a/src/url.h
+++ b/src/url.h
@@ -34,6 +34,14 @@ so, delete this exception statement from your version.  */
 #define DEFAULT_FTP_PORT 21
 #define DEFAULT_HTTPS_PORT 443
 
+/* Specifies how, or whether, user auth information should be included
+ * in URLs regenerated from URL parse structures. */
+enum url_auth_mode {
+  URL_AUTH_SHOW,
+  URL_AUTH_HIDE_PASSWD,
+  URL_AUTH_HIDE
+};
+
 /* Note: the ordering here is related to the order of elements in
    `supported_schemes' in url.c.  */
 
@@ -86,7 +94,7 @@ bool url_has_scheme (const char *);
 int scheme_default_port (enum url_scheme);
 void scheme_disable (enum url_scheme);
 
-char *url_string (const struct url *, bool);
+char *url_string (const struct url *, enum url_auth_mode);
 char *url_file_name (const struct url *);
 
 char *uri_merge (const char *, const char *);