X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=filters.c;h=9b456040a8c9d78ed046398c7c15fdaf1798525a;hp=c1ed95d82af86881a670bc2b9a1c84b0d90f3985;hb=b57c5c88740a9748a2cca103bf7ab0edec9489d5;hpb=1b0d53fa23240e90b61d19d621e4462934efb482

diff --git a/filters.c b/filters.c
index c1ed95d8..9b456040 100644
--- a/filters.c
+++ b/filters.c
@@ -1,7 +1,7 @@
-const char filters_rcs[] = "$Id: filters.c,v 1.22 2001/07/18 12:29:34 oes Exp $";
+const char filters_rcs[] = "$Id: filters.c,v 1.34 2001/09/20 15:49:36 steudten Exp $";
 /*********************************************************************
  *
- * File        :  $Source: /cvsroot/ijbswa//current/filters.c,v $
+ * File        :  $Source: /cvsroot/ijbswa/current/filters.c,v $
  *
  * Purpose     :  Declares functions to parse/crunch headers and pages.
  *                Functions declared include:
@@ -38,6 +38,58 @@ const char filters_rcs[] = "$Id: filters.c,v 1.22 2001/07/18 12:29:34 oes Exp $"
  *
  * Revisions   :
  *    $Log: filters.c,v $
+ *    Revision 1.34  2001/09/20 15:49:36  steudten
+ *
+ *    Fix BUG: Change int size to size_t size in pcrs_filter_response().
+ *    See cgi.c fill_template().
+ *
+ *    Revision 1.33  2001/09/16 17:05:14  jongfoster
+ *    Removing unused #include showarg.h
+ *
+ *    Revision 1.32  2001/09/16 13:21:27  jongfoster
+ *    Changes to use new list functions.
+ *
+ *    Revision 1.31  2001/09/16 11:38:02  jongfoster
+ *    Splitting fill_template() into 2 functions:
+ *    template_load() loads the file
+ *    template_fill() performs the PCRS regexps.
+ *    This is because the CGI edit interface has a "table row"
+ *    template which is used many times in the page - this
+ *    change means it's only loaded from disk once.
+ *
+ *    Revision 1.30  2001/09/16 11:00:10  jongfoster
+ *    New function alloc_http_response, for symmetry with free_http_response
+ *
+ *    Revision 1.29  2001/09/13 23:32:40  jongfoster
+ *    Moving image data to cgi.c rather than cgi.h
+ *    Fixing a GPF under Win32 (and any other OS that protects global
+ *    constants from being written to).
+ *
+ *    Revision 1.28  2001/09/10 10:18:51  oes
+ *    Silenced compiler warnings
+ *
+ *    Revision 1.27  2001/08/05 16:06:20  jongfoster
+ *    Modifiying "struct map" so that there are now separate header and
+ *    "map_entry" structures.  This means that functions which modify a
+ *    map no longer need to return a pointer to the modified map.
+ *    Also, it no longer reverses the order of the entries (which may be
+ *    important with some advanced template substitutions).
+ *
+ *    Revision 1.26  2001/07/30 22:08:36  jongfoster
+ *    Tidying up #defines:
+ *    - All feature #defines are now of the form FEATURE_xxx
+ *    - Permanently turned off WIN_GUI_EDIT
+ *    - Permanently turned on WEBDAV and SPLIT_PROXY_ARGS
+ *
+ *    Revision 1.25  2001/07/26 10:09:46  oes
+ *    Made browser detection a little less naive
+ *
+ *    Revision 1.24  2001/07/25 17:22:51  oes
+ *    Added workaround for Netscape bug that prevents display of page when loading a component fails.
+ *
+ *    Revision 1.23  2001/07/23 13:40:12  oes
+ *    Fixed bug that caused document body to be dropped when pcrs joblist was empty.
+ *
  *    Revision 1.22  2001/07/18 12:29:34  oes
  *    - Made gif_deanimate_response respect
  *      csp->action->string[ACTION_STRING_DEANIMATE]
@@ -239,6 +291,7 @@ const char filters_rcs[] = "$Id: filters.c,v 1.22 2001/07/18 12:29:34 oes Exp $"
 #include <stdlib.h>
 #include <ctype.h>
 #include <string.h>
+#include <assert.h>
 
 #ifndef _WIN32
 #include <unistd.h>
@@ -250,12 +303,8 @@ const char filters_rcs[] = "$Id: filters.c,v 1.22 2001/07/18 12:29:34 oes Exp $"
 #include "project.h"
 #include "filters.h"
 #include "encode.h"
-#include "jcc.h"
-#include "showargs.h"
 #include "parsers.h"
 #include "ssplit.h"
-#include "gateway.h"
-#include "jbsockets.h"
 #include "errlog.h"
 #include "jbsockets.h"
 #include "miscutil.h"
@@ -280,7 +329,7 @@ const char filters_h_rcs[] = FILTERS_H_VERSION;
 #define ijb_isdigit(__X) isdigit((int)(unsigned char)(__X))
 
 
-#ifdef ACL_FILES
+#ifdef FEATURE_ACL
 /*********************************************************************
  *
  * Function    :  block_acl
@@ -414,7 +463,7 @@ int acl_addr(char *aspec, struct access_control_addr *aca)
    return(0);
 
 }
-#endif /* def ACL_FILES */
+#endif /* def FEATURE_ACL */
 
 
 /*********************************************************************
@@ -431,9 +480,10 @@ int acl_addr(char *aspec, struct access_control_addr *aca)
  *********************************************************************/
 struct http_response *block_url(struct client_state *csp)
 {
+#ifdef FEATURE_IMAGE_BLOCKING
    char *p;
+#endif /* def FEATURE_IMAGE_BLOCKING */
    struct http_response *rsp;
-   struct map *exports = NULL;
 
    /* 
     * If it's not blocked, don't block it ;-)
@@ -446,7 +496,7 @@ struct http_response *block_url(struct client_state *csp)
    /* 
     * Else, prepare a response
     */
-   if (NULL == ( rsp = (struct http_response *)zalloc(sizeof(*rsp))))
+   if (NULL == (rsp = alloc_http_response()))
    {
       return NULL;
    }
@@ -455,7 +505,7 @@ struct http_response *block_url(struct client_state *csp)
     * If it's an image-url, send back an image or redirect
     * as specified by the relevant +image action
     */
-#ifdef IMAGE_BLOCKING
+#ifdef FEATURE_IMAGE_BLOCKING
    if (((csp->action->flags & ACTION_IMAGE_BLOCKER) != 0)
         && is_imageurl(csp))
    {
@@ -465,15 +515,15 @@ struct http_response *block_url(struct client_state *csp)
       /* and handle accordingly: */
       if ((p == NULL) || (0 == strcmpic(p, "logo")))
       {
-         rsp->body = bindup(JBGIF, sizeof(JBGIF));
-         rsp->content_length = sizeof(JBGIF);
+         rsp->body = bindup(image_junkbuster_gif_data, image_junkbuster_gif_length);
+         rsp->content_length = image_junkbuster_gif_length;
          enlist_unique_header(rsp->headers, "Content-Type", "image/gif");
       }
 
       else if (0 == strcmpic(p, "blank"))
       {
-         rsp->body = bindup(BLANKGIF, sizeof(BLANKGIF));
-         rsp->content_length = sizeof(BLANKGIF);
+         rsp->body = bindup(image_blank_gif_data, image_blank_gif_length);
+         rsp->content_length = image_blank_gif_length;
          enlist_unique_header(rsp->headers, "Content-Type", "image/gif");
       }
 
@@ -484,29 +534,46 @@ struct http_response *block_url(struct client_state *csp)
       }
    }  
    else
-#endif /* def IMAGE_BLOCKING */
+#endif /* def FEATURE_IMAGE_BLOCKING */
 
    /* 
     * Else, generate an HTML "blocked" message:
     */
    {
-
-      exports = default_exports(csp, NULL);	   
-#ifdef FORCE_LOAD
-      exports = map(exports, "force-prefix", 1, FORCE_PREFIX, 1);
-#else
-      exports = map_block_killer(exports, "force-support");
-#endif /* ndef FORCE_LOAD */
-
-      exports = map(exports, "hostport", 1, csp->http->hostport, 1);
-      exports = map(exports, "hostport-html", 1, html_encode(csp->http->hostport), 0);
-      exports = map(exports, "path", 1, csp->http->path, 1);
-      exports = map(exports, "path-html", 1, html_encode(csp->http->path), 0);
-
-      rsp->body = fill_template(csp, "blocked", exports);
+      struct map * exports = default_exports(csp, NULL);
+#ifdef FEATURE_FORCE_LOAD
+      map(exports, "force-prefix", 1, FORCE_PREFIX, 1);
+#else /* ifndef FEATURE_FORCE_LOAD */
+      map_block_killer(exports, "force-support");
+#endif /* ndef FEATURE_FORCE_LOAD */
+
+      map(exports, "hostport", 1, csp->http->hostport, 1);
+      map(exports, "hostport-html", 1, html_encode(csp->http->hostport), 0);
+      map(exports, "path", 1, csp->http->path, 1);
+      map(exports, "path-html", 1, html_encode(csp->http->path), 0);
+
+      rsp->body = template_load(csp, "blocked");
+      template_fill(&rsp->body, exports);
       free_map(exports);
   
-      rsp->status = strdup("403 Request for blocked URL"); 
+      /*
+       * Workaround for stupid Netscape bug which prevents
+       * pages from being displayed if loading a referenced
+       * JavaScript or style sheet fails. So make it appear
+       * as if it succeeded.
+       */
+      if (csp->http->user_agent
+          && !strncmpic(csp->http->user_agent, "mozilla", 7)
+          && !strstr(csp->http->user_agent, "compatible")
+          && !strstr(csp->http->user_agent, "Opera"))
+      {
+         rsp->status = strdup("200 Request for blocked URL"); 
+      }
+      else
+      {
+         rsp->status = strdup("404 Request for blocked URL"); 
+      }
+
    }
 
    return(finish_http_response(rsp));
@@ -514,7 +581,7 @@ struct http_response *block_url(struct client_state *csp)
 }
 
 
-#ifdef TRUST_FILES
+#ifdef FEATURE_TRUST
 /*********************************************************************
  *
  * Function    :  trust_url FIXME: I should be called distrust_url
@@ -531,7 +598,7 @@ struct http_response *block_url(struct client_state *csp)
 struct http_response *trust_url(struct client_state *csp)
 {
    struct http_response *rsp;
-   struct map *exports = NULL;
+   struct map * exports;
    char buf[BUFFER_SIZE], *p = NULL;
    struct url_spec **tl, *t;
 
@@ -546,29 +613,30 @@ struct http_response *trust_url(struct client_state *csp)
    /* 
     * Else, prepare a response:
     */
-   if (NULL == ( rsp = (struct http_response *)zalloc(sizeof(*rsp))))
+   if (NULL == (rsp = alloc_http_response()))
    {
       return NULL;
    }
+
    exports = default_exports(csp, NULL);
 
    /* 
     * Export the host, port, and referrer information
     */
-   exports = map(exports, "hostport", 1, csp->http->hostport, 1);
-   exports = map(exports, "path", 1, csp->http->path, 1);
-   exports = map(exports, "hostport-html", 1, html_encode(csp->http->hostport), 0);
-   exports = map(exports, "path-html", 1, html_encode(csp->http->path), 0);
+   map(exports, "hostport", 1, csp->http->hostport, 1);
+   map(exports, "path", 1, csp->http->path, 1);
+   map(exports, "hostport-html", 1, html_encode(csp->http->hostport), 0);
+   map(exports, "path-html", 1, html_encode(csp->http->path), 0);
 
    if (csp->referrer && strlen(csp->referrer) > 9)
    {
-      exports = map(exports, "referrer", 1, csp->referrer + 9, 1);
-      exports = map(exports, "referrer-html", 1, html_encode(csp->referrer + 9), 0);
+      map(exports, "referrer", 1, csp->referrer + 9, 1);
+      map(exports, "referrer-html", 1, html_encode(csp->referrer + 9), 0);
    }
    else
    {
-      exports = map(exports, "referrer", 1, "unknown", 1);
-      exports = map(exports, "referrer-html", 1, "unknown", 1);
+      map(exports, "referrer", 1, "unknown", 1);
+      map(exports, "referrer-html", 1, "unknown", 1);
    }
 
    /*
@@ -579,50 +647,51 @@ struct http_response *trust_url(struct client_state *csp)
       sprintf(buf, "<li>%s</li>\n", t->spec);
       p = strsav(p, buf);
    }
-   exports = map(exports, "trusted-referrers", 1, p, 0);
+   map(exports, "trusted-referrers", 1, p, 0);
    p = NULL;
 
    /*
     * Export the trust info, if available
     */
-   if (csp->config->trust_info->next)
+   if (csp->config->trust_info->first)
    {
-      struct list *l;
+      struct list_entry *l;
 
-      for (l = csp->config->trust_info->next; l ; l = l->next)
+      for (l = csp->config->trust_info->first; l ; l = l->next)
       {
          sprintf(buf, "<li> <a href=%s>%s</a><br>\n",l->str, l->str);
          p = strsav(p, buf);
       }
-      exports = map(exports, "trust-info", 1, p, 0);
+      map(exports, "trust-info", 1, p, 0);
    }
    else
    {
-      exports = map_block_killer(exports, "have-trust-info");
+      map_block_killer(exports, "have-trust-info");
    }
    
    /*
     * Export the force prefix or the force conditional block killer
     */
-#ifdef FORCE_LOAD
-   exports = map(exports, "force-prefix", 1, FORCE_PREFIX, 1);
-#else
-   exports = map_block_killer(exports, "force-support");
-#endif /* ndef FORCE_LOAD */
+#ifdef FEATURE_FORCE_LOAD
+   map(exports, "force-prefix", 1, FORCE_PREFIX, 1);
+#else /* ifndef FEATURE_FORCE_LOAD */
+   map_block_killer(exports, "force-support");
+#endif /* ndef FEATURE_FORCE_LOAD */
 
    /*
     * Build the response
     */
-   rsp->body = fill_template(csp, "untrusted", exports);
+   rsp->body = template_load(csp, "untrusted");
+   template_fill(&rsp->body, exports);
    free_map(exports);
 
    return(finish_http_response(rsp));
 
 }
-#endif /* def TRUST_FILES */
+#endif /* def FEATURE_TRUST */
 
 
-#ifdef FAST_REDIRECTS
+#ifdef FEATURE_FAST_REDIRECTS
 /*********************************************************************
  *
  * Function    :  redirect_url
@@ -647,7 +716,7 @@ struct http_response *redirect_url(struct client_state *csp)
    /* 
     * find the last URL encoded in the request
     */
-   while (p = strstr(p, "http://"))
+   while ((p = strstr(p, "http://")))
    {
       q = p++;
    }
@@ -659,7 +728,7 @@ struct http_response *redirect_url(struct client_state *csp)
    {
       log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q);
 
-      if (NULL == ( rsp = zalloc(sizeof(*rsp))))
+      if (NULL == (rsp = alloc_http_response()))
       {
          return NULL;
       }
@@ -675,18 +744,18 @@ struct http_response *redirect_url(struct client_state *csp)
    }
 
 }
-#endif /* def FAST_REDIRECTS */
+#endif /* def FEATURE_FAST_REDIRECTS */
 
 
-#ifdef IMAGE_BLOCKING
+#ifdef FEATURE_IMAGE_BLOCKING
 /*********************************************************************
  *
  * Function    :  is_imageurl
  *
  * Description :  Given a URL, decide whether it is an image or not,
  *                using either the info from a previous +image action
- *                or, #ifdef DETECT_MSIE_IMAGES, the info from the
- *                browser's accept header.
+ *                or, #ifdef FEATURE_IMAGE_DETECT_MSIE, the info from
+ *                the browser's accept header.
  *                
  * Parameters  :
  *          1  :  csp = Current client state (buffers, headers, etc...)
@@ -697,7 +766,7 @@ struct http_response *redirect_url(struct client_state *csp)
  *********************************************************************/
 int is_imageurl(struct client_state *csp)
 {
-#ifdef DETECT_MSIE_IMAGES
+#ifdef FEATURE_IMAGE_DETECT_MSIE
    if ((csp->accept_types 
        & (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE|ACCEPT_TYPE_MSIE_HTML))
        == (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE))
@@ -710,15 +779,15 @@ int is_imageurl(struct client_state *csp)
    {
       return 0;
    }
-#endif
+#endif /* def FEATURE_IMAGE_DETECT_MSIE */
 
    return ((csp->action->flags & ACTION_IMAGE) != 0);
 
 }
-#endif /* def IMAGE_BLOCKING */
+#endif /* def FEATURE_IMAGE_BLOCKING */
 
 
-#ifdef TRUST_FILES
+#ifdef FEATURE_COOKIE_JAR
 /*********************************************************************
  *
  * Function    :  is_untrusted_url
@@ -880,7 +949,7 @@ int is_untrusted_url(struct client_state *csp)
    }
    return(1);
 }
-#endif /* def TRUST_FILES */
+#endif /* def FEATURE_COOKIE_JAR */
 
 
 /*********************************************************************
@@ -890,7 +959,7 @@ int is_untrusted_url(struct client_state *csp)
  * Description :  Apply all the pcrs jobs from the joblist (re_filterfile)
  *                to the text buffer that's been accumulated in 
  *                csp->iob->buf and set csp->content_length to the modified
- *                size.
+ *                size and raise the CSP_FLAG_MODIFIED flag if appropriate.
  *
  * Parameters  :
  *          1  :  csp = Current client state (buffers, headers, etc...)
@@ -902,7 +971,7 @@ int is_untrusted_url(struct client_state *csp)
 char *pcrs_filter_response(struct client_state *csp)
 {
    int hits=0;
-   int size = csp->iob->eod - csp->iob->cur;
+   size_t size;
 
    char *old = csp->iob->cur, *new = NULL;
    pcrs_job *job;
@@ -910,11 +979,27 @@ char *pcrs_filter_response(struct client_state *csp)
    struct file_list *fl;
    struct re_filterfile_spec *b;
 
-   /* Sanity first ;-) */
-   if (size <= 0)
+   /* Sanity first */
+   if (csp->iob->cur >= csp->iob->eod)
    {
       return(NULL);
    }
+   size = csp->iob->eod - csp->iob->cur;
+
+   /*
+    * If the body has a "chunked" transfer-encoding,
+    * get rid of it first, adjusting size and iob->eod
+    */
+   if (csp->flags & CSP_FLAG_CHUNKED)
+   {
+      log_error(LOG_LEVEL_RE_FILTER, "Need to de-chunk first");
+      if (0 == (size = remove_chunked_transfer_coding(csp->iob->cur, size)))
+      {
+         return(NULL);
+      }
+      csp->iob->eod = csp->iob->cur + size;
+      csp->flags |= CSP_FLAG_MODIFIED;
+   }
 
    if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) )
    {
@@ -941,10 +1026,20 @@ char *pcrs_filter_response(struct client_state *csp)
 
    log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size);
 
-   csp->content_length = size;
+   /* 
+    * If there were no hits, destroy our copy and let
+    * chat() use the original in csp->iob
+    */
+   if (!hits)
+   {
+      free(new);
+      return(NULL);
+   }
 
-   /* fwiw, reset the iob */
+   csp->flags |= CSP_FLAG_MODIFIED;
+   csp->content_length = size;
    IOB_RESET(csp);
+
    return(new);
 
 }
@@ -955,8 +1050,8 @@ char *pcrs_filter_response(struct client_state *csp)
  * Function    :  gif_deanimate_response
  *
  * Description :  Deanimate the GIF image that has been accumulated in 
- *                csp->iob->buf and set csp->content_length to the modified
- *                size.
+ *                csp->iob->buf, set csp->content_length to the modified
+ *                size and raise the CSP_FLAG_MODIFIED flag.
  *
  * Parameters  :
  *          1  :  csp = Current client state (buffers, headers, etc...)
@@ -971,6 +1066,21 @@ char *gif_deanimate_response(struct client_state *csp)
    char *p;
    int size = csp->iob->eod - csp->iob->cur;
 
+   /*
+    * If the body has a "chunked" transfer-encoding,
+    * get rid of it first, adjusting size and iob->eod
+    */
+   if (csp->flags & CSP_FLAG_CHUNKED)
+   {
+      log_error(LOG_LEVEL_DEANIMATE, "Need to de-chunk first");
+      if (0 == (size = remove_chunked_transfer_coding(csp->iob->cur, size)))
+      {
+         return(NULL);
+      }
+      csp->iob->eod = csp->iob->cur + size;
+      csp->flags |= CSP_FLAG_MODIFIED;
+   }
+
    if (  (NULL == (in =  (struct binbuffer *)zalloc(sizeof *in )))
       || (NULL == (out = (struct binbuffer *)zalloc(sizeof *out))) )
    {
@@ -992,6 +1102,7 @@ char *gif_deanimate_response(struct client_state *csp)
    {
       log_error(LOG_LEVEL_DEANIMATE, "Success! GIF shrunk from %d bytes to %d.", size, out->offset);
       csp->content_length = out->offset;
+      csp->flags |= CSP_FLAG_MODIFIED;
       p = out->buffer;
       free(in);
       free(out);
@@ -1001,6 +1112,65 @@ char *gif_deanimate_response(struct client_state *csp)
 }
 
 
+/*********************************************************************
+ *
+ * Function    :  remove_chunked_transfer_coding
+ *
+ * Description :  In-situ remove the "chunked" transfer coding as defined
+ *                in rfc2616 from a buffer.
+ *
+ * Parameters  :
+ *          1  :  buffer = Pointer to the text buffer
+ *          2  :  size = Number of bytes to be processed
+ *
+ * Returns     :  The new size, i.e. the number of bytes from buffer which
+ *                are occupied by the stripped body, or 0 in case something
+ *                went wrong
+ *                
+ *********************************************************************/
+int remove_chunked_transfer_coding(char *buffer, const size_t size)
+{
+   size_t newsize = 0;
+   unsigned int chunksize = 0;
+   char *from_p, *to_p;
+
+   assert(buffer);
+   from_p = to_p = buffer;
+
+   if (sscanf(buffer, "%x", &chunksize) != 1)
+   {
+      log_error(LOG_LEVEL_ERROR, "Invalid first chunksize while stripping \"chunked\" transfer coding");
+      return(0);
+   }
+
+   while (chunksize > 0)
+   {
+      if (NULL == (from_p = strstr(from_p, "\r\n")))
+      {
+         log_error(LOG_LEVEL_ERROR, "Parse error while stripping \"chunked\" transfer coding");
+         return(0);
+      }
+      newsize += chunksize;
+      from_p += 2;
+      
+      memmove(to_p, from_p, (size_t) chunksize);
+      to_p = buffer + newsize;
+      from_p += chunksize + 2;
+
+      if (sscanf(from_p, "%x", &chunksize) != 1)
+      {
+         log_error(LOG_LEVEL_ERROR, "Parse error while stripping \"chunked\" transfer coding");
+         return(0);
+      }
+   }
+
+   /* FIXME: Should this get its own loglevel? */
+   log_error(LOG_LEVEL_RE_FILTER, "De-chunking successful. Shrunk from %d to %d\n", size, newsize);
+   return(newsize);
+
+}
+
+
 /*********************************************************************
  *
  * Function    :  url_actions
@@ -1107,7 +1277,7 @@ void apply_url_actions(struct current_action_spec *action,
 const struct forward_spec * forward_url(struct http_request *http,
                                         struct client_state *csp)
 {
-   static const struct forward_spec fwd_default[1] = { 0 }; /* All zeroes */
+   static const struct forward_spec fwd_default[1] = { FORWARD_SPEC_INITIALIZER };
    struct forward_spec *fwd = csp->config->forward;
    struct url_spec url[1];