X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=filters.c;h=dd7ccb4409680751b223dd97049d571a0985bf65;hp=cab956f748182b0ee0c2e8d95d8b9f81b44f0a84;hb=5f390b415777db21894f1b856ab3d21c9c24f7c6;hpb=43da0ce639b1cf540198b2f63e790d9127725343 diff --git a/filters.c b/filters.c index cab956f7..dd7ccb44 100644 --- a/filters.c +++ b/filters.c @@ -1,4 +1,4 @@ -const char filters_rcs[] = "$Id: filters.c,v 1.5 2001/05/25 22:34:30 jongfoster Exp $"; +const char filters_rcs[] = "$Id: filters.c,v 1.38 2001/10/23 21:32:33 jongfoster Exp $"; /********************************************************************* * * File : $Source: /cvsroot/ijbswa/current/filters.c,v $ @@ -6,19 +6,19 @@ const char filters_rcs[] = "$Id: filters.c,v 1.5 2001/05/25 22:34:30 jongfoster * Purpose : Declares functions to parse/crunch headers and pages. * Functions declared include: * `acl_addr', `add_stats', `block_acl', `block_imageurl', - * `block_url', `url_permissions', `domaincmp', `dsplit', + * `block_url', `url_actions', `domaincmp', `dsplit', * `filter_popups', `forward_url', 'redirect_url', - * `ij_untrusted_url', `intercept_url', `re_process_buffer', + * `ij_untrusted_url', `intercept_url', `pcrs_filter_respose', * `show_proxy_args', 'ijb_send_banner', and `trust_url' * * Copyright : Written by and Copyright (C) 2001 the SourceForge * IJBSWA team. http://ijbswa.sourceforge.net * * Based on the Internet Junkbuster originally written - * by and Copyright (C) 1997 Anonymous Coders and + * by and Copyright (C) 1997 Anonymous Coders and * Junkbusters Corporation. http://www.junkbusters.com * - * This program is free software; you can redistribute it + * This program is free software; you can redistribute it * and/or modify it under the terms of the GNU General * Public License as published by the Free Software * Foundation; either version 2 of the License, or (at @@ -38,6 +38,207 @@ const char filters_rcs[] = "$Id: filters.c,v 1.5 2001/05/25 22:34:30 jongfoster * * Revisions : * $Log: filters.c,v $ + * Revision 1.38 2001/10/23 21:32:33 jongfoster + * Adding error-checking to selected functions + * + * Revision 1.37 2001/10/22 15:33:56 david__schmidt + * Special-cased OS/2 out of the Netscape-abort-on-404-in-js problem in + * filters.c. Added a FIXME in front of the offending code. I'll gladly + * put in a better/more robust fix for all parties if one is presented... + * It seems that just returning 200 instead of 404 would pretty much fix + * it for everyone, but I don't know all the history of the problem. + * + * Revision 1.36 2001/10/10 16:44:16 oes + * Added match_portlist function + * + * Revision 1.35 2001/10/07 15:41:23 oes + * Replaced 6 boolean members of csp with one bitmap (csp->flags) + * + * New function remove_chunked_transfer_coding that strips chunked + * transfer coding to plain and is called by pcrs_filter_response + * and gif_deanimate_response if neccessary + * + * Improved handling of zero-change re_filter runs + * + * pcrs_filter_response and gif_deanimate_response now remove + * chunked transfer codeing before processing the body. + * + * 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] + * - Logging cosmetics + * + * Revision 1.21 2001/07/13 13:59:53 oes + * - Introduced gif_deanimate_response which shares the + * generic content modification interface of pcrs_filter_response + * and acts as a wrapper to deanimate.c:gif_deanimate() + * - Renamed re_process_buffer to pcrs_filter_response + * - pcrs_filter_response now returns NULL on failiure + * - Removed all #ifdef PCRS + * + * Revision 1.20 2001/07/01 17:01:04 oes + * Added comments and missing return statement in is_untrusted_url() + * + * Revision 1.19 2001/06/29 21:45:41 oes + * Indentation, CRLF->LF, Tab-> Space + * + * Revision 1.18 2001/06/29 13:27:38 oes + * - Cleaned up, renamed and reorderd functions + * and improved comments + * + * - block_url: + * - Ported to CGI platform. Now delivers + * http_response or NULL + * - Unified HTML and GIF generation (moved image detection + * and GIF generation here from jcc.c:chat()) + * - Fixed HTTP status to: + * - 403 (Forbidden) for the "blocked" HTML message + * - 200 (OK) for GIF answers + * - 302 (Redirect) for redirect to GIF + * + * - trust_url: + * - Ported to CGI platform. Now delivers + * http_response or NULL + * - Separated detection of untrusted URL into + * (bool)is_untrusted_url + * - Added enforcement of untrusted requests + * + * - Moved redirect_url() from cgi.c to here + * and ported it to the CGI platform + * + * - Removed logentry from cancelled commit + * + * Revision 1.17 2001/06/09 10:55:28 jongfoster + * Changing BUFSIZ ==> BUFFER_SIZE + * + * Revision 1.16 2001/06/07 23:10:26 jongfoster + * Allowing unanchored domain patterns to back off and retry + * if they partially match. Optimized right-anchored patterns. + * Moving ACL and forward files into config file. + * Replacing struct gateway with struct forward_spec + * + * Revision 1.15 2001/06/03 19:12:00 oes + * extracted-CGI relevant stuff + * + * Revision 1.14 2001/06/01 10:30:55 oes + * Added optional left-anchoring to domaincmp + * + * Revision 1.13 2001/05/31 21:21:30 jongfoster + * Permissionsfile / actions file changes: + * - Changed "permission" to "action" throughout + * - changes to file format to allow string parameters + * - Moved helper functions to actions.c + * + * Revision 1.12 2001/05/31 17:35:20 oes + * + * - Enhanced domain part globbing with infix and prefix asterisk + * matching and optional unanchored operation + * + * Revision 1.11 2001/05/29 11:53:23 oes + * "See why" link added to "blocked" page + * + * Revision 1.10 2001/05/29 09:50:24 jongfoster + * Unified blocklist/imagelist/permissionslist. + * File format is still under discussion, but the internal changes + * are (mostly) done. + * + * Also modified interceptor behaviour: + * - We now intercept all URLs beginning with one of the following + * prefixes (and *only* these prefixes): + * * http://i.j.b/ + * * http://ijbswa.sf.net/config/ + * * http://ijbswa.sourceforge.net/config/ + * - New interceptors "home page" - go to http://i.j.b/ to see it. + * - Internal changes so that intercepted and fast redirect pages + * are not replaced with an image. + * - Interceptors now have the option to send a binary page direct + * to the client. (i.e. ijb-send-banner uses this) + * - Implemented show-url-info interceptor. (Which is why I needed + * the above interceptors changes - a typical URL is + * "http://i.j.b/show-url-info?url=www.somesite.com/banner.gif". + * The previous mechanism would not have intercepted that, and + * if it had been intercepted then it then it would have replaced + * it with an image.) + * + * Revision 1.9 2001/05/27 22:17:04 oes + * + * - re_process_buffer no longer writes the modified buffer + * to the client, which was very ugly. It now returns the + * buffer, which it is then written by chat. + * + * - content_length now adjusts the Content-Length: header + * for modified documents rather than crunch()ing it. + * (Length info in csp->content_length, which is 0 for + * unmodified documents) + * + * - For this to work, sed() is called twice when filtering. + * + * Revision 1.8 2001/05/26 17:13:28 jongfoster + * Filled in a function comment. + * + * Revision 1.7 2001/05/26 15:26:15 jongfoster + * ACL feature now provides more security by immediately dropping + * connections from untrusted hosts. + * + * Revision 1.6 2001/05/26 00:28:36 jongfoster + * Automatic reloading of config file. + * Removed obsolete SIGHUP support (Unix) and Reload menu option (Win32). + * Most of the global variables have been moved to a new + * struct configuration_spec, accessed through csp->config->globalname + * Most of the globals remaining are used by the Win32 GUI. + * * Revision 1.5 2001/05/25 22:34:30 jongfoster * Hard tabs->Spaces * @@ -115,25 +316,33 @@ const char filters_rcs[] = "$Id: filters.c,v 1.5 2001/05/25 22:34:30 jongfoster #include #include #include +#include #ifndef _WIN32 +#ifndef __OS2__ #include +#endif /* ndef __OS2__ */ #include #else #include -#endif +#endif /* ndef _WIN32 */ + +#ifdef __OS2__ +#include +#endif /* def __OS2__ */ #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" +#include "actions.h" +#include "cgi.h" +#include "list.h" +#include "deanimate.h" #ifdef _WIN32 #include "win32.h" @@ -151,61 +360,7 @@ const char filters_h_rcs[] = FILTERS_H_VERSION; #define ijb_isdigit(__X) isdigit((int)(unsigned char)(__X)) -static const char CBLOCK[] = -#ifdef AMIGA - "HTTP/1.0 403 Request for blocked URL\n" -#else /* ifndef AMIGA */ - "HTTP/1.0 202 Request for blocked URL\n" -#endif /* ndef AMIGA */ - "Pragma: no-cache\n" - "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: Request for blocked URL\n" - "\n" - WHITEBG - "

" - BANNER - "

\n" - "

Your request for %s%s
\n" - "was blocked because it matches the following pattern " - "in the blockfile: %s\n

" -#ifdef FORCE_LOAD - "

Go there anyway.

" -#endif /* def FORCE_LOAD */ - "\n" - "\n"; - -#ifdef TRUST_FILES -static const char CTRUST[] = -#ifdef AMIGA - "HTTP/1.0 403 Request for untrusted URL\n" -#else /* ifndef AMIGA */ - "HTTP/1.0 202 Request for untrusted URL\n" -#endif /* ndef AMIGA */ - "Pragma: no-cache\n" - "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: Request for untrusted URL\n" - "\n" - WHITEBG - "
" - "" - BANNER - "" - "
" - "\n" - "\n"; -#endif /* def TRUST_FILES */ - - -#ifdef ACL_FILES +#ifdef FEATURE_ACL /********************************************************************* * * Function : block_acl @@ -214,52 +369,51 @@ static const char CTRUST[] = * Decide yes or no based on ACL file. * * Parameters : - * 1 : src = Address the browser/user agent is requesting. - * 2 : dst = The proxy or gateway address this is going to. - * 3 : csp = Current client state (buffers, headers, etc...) + * 1 : dst = The proxy or gateway address this is going to. + * Or NULL to check all possible targets. + * 2 : csp = Current client state (buffers, headers, etc...) + * Also includes the client IP address. * * Returns : 0 = FALSE (don't block) and 1 = TRUE (do block) * *********************************************************************/ -int block_acl(struct access_control_addr *src, struct access_control_addr *dst, struct client_state *csp) +int block_acl(struct access_control_addr *dst, struct client_state *csp) { - struct file_list *fl; - struct access_control_list *a, *acl; - struct access_control_addr s[1], d[1]; + struct access_control_list *acl = csp->config->acl; /* if not using an access control list, then permit the connection */ - if (((fl = csp->alist) == NULL) || ((acl = fl->f) == NULL)) + if (acl == NULL) { return(0); } /* search the list */ - for (a = acl->next ; a ; a = a->next) - { - *s = *src; - *d = *dst; - - s->addr &= a->src->mask; - d->addr &= a->dst->mask; - - if ((s->addr == a->src->addr) - && (d->addr == a->dst->addr) - && ((s->port == a->src->port) - || (s->port == 0) - || (a->src->port == 0)) - && ((d->port == a->dst->port) - || (d->port == 0) - || (a->dst->port == 0))) + while (acl != NULL) + { + if ((csp->ip_addr_long & acl->src->mask) == acl->src->addr) { - if (a->action == ACL_PERMIT) + if (dst == NULL) { - return(0); + /* Just want to check if they have any access */ + if (acl->action == ACL_PERMIT) + { + return(0); + } } - else + else if ( ((dst->addr & acl->dst->mask) == acl->dst->addr) + && ((dst->port == acl->dst->port) || (acl->dst->port == 0))) { - return(1); + if (acl->action == ACL_PERMIT) + { + return(0); + } + else + { + return(1); + } } } + acl = acl->next; } return(1); @@ -271,11 +425,11 @@ int block_acl(struct access_control_addr *src, struct access_control_addr *dst, * * Function : acl_addr * - * Description : Called from `load_aclfile'. FIXME: I can't say more. + * Description : Called from `load_aclfile' to parse an ACL address. * * Parameters : - * 1 : aspec = (what?) - * 2 : aca = (what?) + * 1 : aspec = String specifying ACL address. + * 2 : aca = struct access_control_addr to fill in. * * Returns : 0 => Ok, everything else is an error. * @@ -340,348 +494,584 @@ int acl_addr(char *aspec, struct access_control_addr *aca) return(0); } -#endif /* def ACL_FILES */ +#endif /* def FEATURE_ACL */ /********************************************************************* * - * Function : block_url + * Function : match_portlist * - * Description : Called from `chat'. Check to see if we need to block this. + * Description : Check if a given number is covered by a comma + * separated list of numbers and ranges (a,b-c,d,..) * * Parameters : - * 1 : http = http_request request to "check" for blocked - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : portlist = String with list + * 2 : port = port to check * - * Returns : NULL => unblocked, else string to HTML block description. + * Returns : 0 => no match + * 1 => match * *********************************************************************/ -char *block_url(struct http_request *http, struct client_state *csp) +int match_portlist(const char *portlist, int port) { - struct file_list *fl; - struct block_spec *b; - struct url_spec url[1]; - char *p; - int n; + char *min, *max, *next, *portlist_copy; - if (((fl = csp->blist) == NULL) || ((b = fl->f) == NULL)) + min = next = portlist_copy = strdup(portlist); + + /* + * Zero-terminate first item and remember offset for next + */ + if (NULL != (next = strchr(portlist_copy, (int) ','))) { - return(NULL); + *next++ = '\0'; } - *url = dsplit(http->host); - - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(NULL); - - for (b = b->next; b ; b = b->next) + /* + * Loop through all items, checking for match + */ + while(min) { - if ((b->url->port == 0) || (b->url->port == http->port)) + if (NULL == (max = strchr(min, (int) '-'))) { - if ((b->url->domain[0] == '\0') || (domaincmp(b->url, url) == 0)) + /* + * No dash, check for equality + */ + if (port == atoi(min)) { - if ((b->url->path == NULL) || -#ifdef REGEX - (regexec(b->url->preg, http->path, 0, NULL, 0) == 0) -#else - (strncmp(b->url->path, http->path, b->url->pathlen) == 0) -#endif - ) - { - freez(url->dbuf); - freez(url->dvec); - - if (b->reject == 0) return(NULL); - - n = strlen(CBLOCK); - n += strlen(http->hostport); - n += strlen(http->path); - n += strlen(b->url->spec); -#ifdef FORCE_LOAD - n += strlen(http->hostport); - n += strlen(http->path); -#endif /* def FORCE_LOAD */ + free(portlist_copy); + return(1); + } + } + else + { + /* + * This is a range, so check if between min and max, + * or, if max was omitted, between min and 65K + */ + *max++ = '\0'; + if(port >= atoi(min) && port <= (atoi(max) ? atoi(max) : 65535)) + { + free(portlist_copy); + return(1); + } - p = (char *)malloc(n); + } -#ifdef FORCE_LOAD - sprintf(p, CBLOCK, http->hostport, http->path, b->url->spec, http->hostport, http->path); -#else - sprintf(p, CBLOCK, http->hostport, http->path, b->url->spec); -#endif /* def FORCE_LOAD */ + /* + * Jump to next item + */ + min = next; - return(p); - } - } + /* + * Zero-terminate next item and remember offset for n+1 + */ + if ((NULL != next) && (NULL != (next = strchr(next, (int) ',')))) + { + *next++ = '\0'; } } - freez(url->dbuf); - freez(url->dvec); - return(NULL); + + free(portlist_copy); + return 0; } -#if defined(DETECT_MSIE_IMAGES) || defined(USE_IMAGE_LIST) /********************************************************************* * - * Function : block_imageurl + * Function : block_url * - * Description : Given a URL which is blocked, decide whether to - * send the "blocked" image or HTML. + * Description : Called from `chat'. Check to see if we need to block this. * * Parameters : - * 1 : http = URL to check. - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : True (nonzero) if URL is in image list, false (0) - * otherwise + * Returns : NULL => unblocked, else HTTP block response * *********************************************************************/ -int block_imageurl(struct http_request *http, struct client_state *csp) +struct http_response *block_url(struct client_state *csp) { -#ifdef DETECT_MSIE_IMAGES - if ((csp->accept_types - & (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE|ACCEPT_TYPE_MSIE_HTML)) - == (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE)) +#ifdef FEATURE_IMAGE_BLOCKING + char *p; +#endif /* def FEATURE_IMAGE_BLOCKING */ + struct http_response *rsp; + + /* + * If it's not blocked, don't block it ;-) + */ + if ((csp->action->flags & ACTION_BLOCK) == 0) { - return 1; + return NULL; } - else if ((csp->accept_types - & (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE|ACCEPT_TYPE_MSIE_HTML)) - == (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_HTML)) + + /* + * Else, prepare a response + */ + if (NULL == (rsp = alloc_http_response())) { - return 0; + return cgi_error_memory(); } -#endif -#if defined(USE_IMAGE_LIST) - return block_imageurl_using_imagelist(http, csp); -#else - /* Don't know - assume HTML */ - return 0; -#endif -} -#endif /* defined(DETECT_MSIE_IMAGES) || defined(USE_IMAGE_LIST) */ + /* + * If it's an image-url, send back an image or redirect + * as specified by the relevant +image action + */ +#ifdef FEATURE_IMAGE_BLOCKING + if (((csp->action->flags & ACTION_IMAGE_BLOCKER) != 0) + && is_imageurl(csp)) + { + /* determine HOW images should be blocked */ + p = csp->action->string[ACTION_STRING_IMAGE_BLOCKER]; + /* and handle accordingly: */ + if ((p == NULL) || (0 == strcmpic(p, "logo"))) + { + rsp->body = bindup(image_junkbuster_gif_data, image_junkbuster_gif_length); + if (rsp->body == NULL) + { + free_http_response(rsp); + return cgi_error_memory(); + } + rsp->content_length = image_junkbuster_gif_length; -#ifdef USE_IMAGE_LIST -/********************************************************************* - * - * Function : block_imageurl - * - * Description : Test if a URL is in the imagelist. - * - * Parameters : - * 1 : http = URL to check. - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : True (nonzero) if URL is in image list, false (0) - * otherwise - * - *********************************************************************/ -int block_imageurl_using_imagelist(struct http_request *http, struct client_state *csp) -{ - struct file_list *fl; - struct block_spec *b; - struct url_spec url[1]; + if (enlist_unique_header(rsp->headers, "Content-Type", "image/gif")) + { + free_http_response(rsp); + return cgi_error_memory(); + } + } - if (((fl = csp->ilist) == NULL) || ((b = fl->f) == NULL)) - { - return(0); - } + else if (0 == strcmpic(p, "blank")) + { + rsp->body = bindup(image_blank_gif_data, image_blank_gif_length); + if (rsp->body == NULL) + { + free_http_response(rsp); + return cgi_error_memory(); + } + rsp->content_length = image_blank_gif_length; - *url = dsplit(http->host); + if (enlist_unique_header(rsp->headers, "Content-Type", "image/gif")) + { + free_http_response(rsp); + return cgi_error_memory(); + } + } - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(0); + else + { + rsp->status = strdup("302 Local Redirect from Junkbuster"); + if (rsp->status == NULL) + { + free_http_response(rsp); + return cgi_error_memory(); + } - for (b = b->next; b ; b = b->next) + if (enlist_unique_header(rsp->headers, "Location", p)) + { + free_http_response(rsp); + return cgi_error_memory(); + } + } + } + else +#endif /* def FEATURE_IMAGE_BLOCKING */ + + /* + * Else, generate an HTML "blocked" message: + */ { + jb_err err; + struct map * exports; + + /* FIXME */ +#ifdef __EMX__ + /* + * The entire OS/2 community will hit the stupid Netscape bug + * (all three of us! :-) so we'll just keep ourselves out + * of this contentious debate and special-case ourselves. + * The problem is... a this point in parsing, we don't know + * what the csp->http->user_agent is (yet). So we can't use + * it to decide if we should work around the NS bug or not. + */ + rsp->status = strdup("200 Request for blocked URL"); +#else + /* + * 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"); + } +#endif /* __EMX__ */ + if (rsp->status == NULL) + { + free_http_response(rsp); + return cgi_error_memory(); + } - if ((b->url->port == 0) || (b->url->port == http->port)) + exports = default_exports(csp, NULL); + if (exports == NULL) { - /* port matches, check domain */ - if ((b->url->domain[0] == '\0') || (domaincmp(b->url, url) == 0)) - { - /* domain matches, check path */ - if ((b->url->path == NULL) || -#ifdef REGEX - (regexec(b->url->preg, http->path, 0, NULL, 0) == 0) -#else - (strncmp(b->url->path, http->path, b->url->pathlen) == 0) -#endif - ) - { - /* Matches */ - freez(url->dbuf); - freez(url->dvec); + free_http_response(rsp); + return cgi_error_memory(); + } - if (b->reject == 0) return(0); +#ifdef FEATURE_FORCE_LOAD + err = map(exports, "force-prefix", 1, FORCE_PREFIX, 1); +#else /* ifndef FEATURE_FORCE_LOAD */ + err = map_block_killer(exports, "force-support"); +#endif /* ndef FEATURE_FORCE_LOAD */ - return(1); - } - } + err = err || map(exports, "hostport", 1, csp->http->hostport, 1); + err = err || map(exports, "hostport-html", 1, html_encode(csp->http->hostport), 0); + err = err || map(exports, "path", 1, csp->http->path, 1); + err = err || map(exports, "path-html", 1, html_encode(csp->http->path), 0); + + if (err) + { + free_map(exports); + free_http_response(rsp); + return cgi_error_memory(); + } + + err = template_fill_for_cgi(csp, "blocked", exports, rsp); + if (err) + { + free_http_response(rsp); + return cgi_error_memory(); } } - freez(url->dbuf); - freez(url->dvec); - return(0); + + return finish_http_response(rsp); } -#endif /* def USE_IMAGE_LIST */ -#ifdef PCRS +#ifdef FEATURE_TRUST /********************************************************************* * - * Function : re_process_buffer + * Function : trust_url FIXME: I should be called distrust_url * - * Description : Apply all jobs from the joblist (aka. Perl regexp's) to - * the text buffer that's been accumulated in csp->iob->buf. - * Then, write the modified buffer out to the client - * (Maybe this should happen from jcc.c via flush_socket - * for better readability). + * Description : Calls is_untrusted_url to determine if the URL is trusted + * and if not, returns a HTTP 304 response with a reject message. * * Parameters : * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : N/A + * Returns : NULL => trusted, else http_response. * *********************************************************************/ -void re_process_buffer(struct client_state *csp) +struct http_response *trust_url(struct client_state *csp) { - int hits=0; - int size = csp->iob->eod - csp->iob->cur; - char *old=csp->iob->cur, *new = NULL; - pcrs_job *job, *joblist; + struct http_response *rsp; + struct map * exports; + char buf[BUFFER_SIZE]; + char *p; + struct url_spec **tl; + struct url_spec *t; + jb_err err; - struct file_list *fl; - struct re_filterfile_spec *b; + /* + * Don't bother to work on trusted URLs + */ + if (!is_untrusted_url(csp)) + { + return NULL; + } - /* Sanity first ;-) */ - if (size <= 0) + /* + * Else, prepare a response: + */ + if (NULL == (rsp = alloc_http_response())) { - return; + return cgi_error_memory(); } - if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) ) + exports = default_exports(csp, NULL); + if (exports == NULL) { - log_error(LOG_LEVEL_ERROR, "Unable to get current state of regexp filtering."); - return; + free_http_response(rsp); + return cgi_error_memory(); + } + + /* + * Export the host, port, and referrer information + */ + err = 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) + { + err = err || map(exports, "referrer", 1, csp->referrer + 9, 1); + err = err || map(exports, "referrer-html", 1, html_encode(csp->referrer + 9), 0); + } + else + { + err = err || map(exports, "referrer", 1, "unknown", 1); + err = err || map(exports, "referrer-html", 1, "unknown", 1); } - joblist = b->joblist; + if (err) + { + free_map(exports); + free_http_response(rsp); + return cgi_error_memory(); + } + /* + * Export the trust list + */ + p = strdup(""); + for (tl = csp->config->trust_list; (t = *tl) ; tl++) + { + sprintf(buf, "
  • %s
    • \n", t->spec); + string_append(&p, buf); + } + err = map(exports, "trusted-referrers", 1, p, 0); - log_error(LOG_LEVEL_RE_FILTER, "re_filtering %s%s (size %d) ...", - csp->http->hostport, csp->http->path, size); + if (err) + { + free_map(exports); + free_http_response(rsp); + return cgi_error_memory(); + } - /* Apply all jobs from the joblist */ - for (job = joblist; NULL != job; job = job->next) + /* + * Export the trust info, if available + */ + if (csp->config->trust_info->first) { - hits += pcrs_exec_substitution(job, old, size, &new, &size); - if (old != csp->iob->cur) free(old); - old=new; + struct list_entry *l; + + p = strdup(""); + for (l = csp->config->trust_info->first; l ; l = l->next) + { + sprintf(buf, "
  • %s
    \n",l->str, l->str); + string_append(&p, buf); + } + err = map(exports, "trust-info", 1, p, 0); + } + else + { + err = map_block_killer(exports, "have-trust-info"); } - log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size); + if (err) + { + free_map(exports); + free_http_response(rsp); + return cgi_error_memory(); + } + + /* + * Export the force prefix or the force conditional block killer + */ +#ifdef FEATURE_FORCE_LOAD + err = map(exports, "force-prefix", 1, FORCE_PREFIX, 1); +#else /* ifndef FEATURE_FORCE_LOAD */ + err = map_block_killer(exports, "force-support"); +#endif /* ndef FEATURE_FORCE_LOAD */ - if (write_socket(csp->cfd, old, size) != size) + if (err) { - log_error(LOG_LEVEL_ERROR, "write to client failed."); + free_map(exports); + free_http_response(rsp); + return cgi_error_memory(); } - /* fwiw, reset the iob */ - IOB_RESET(csp); - freez(new); - return; + /* + * Build the response + */ + err = template_fill_for_cgi(csp, "untrusted", exports, rsp); + if (err) + { + free_http_response(rsp); + return cgi_error_memory(); + } + return finish_http_response(rsp); } -#endif /* def PCRS */ +#endif /* def FEATURE_TRUST */ -#ifdef TRUST_FILES +#ifdef FEATURE_FAST_REDIRECTS /********************************************************************* * - * Function : trust_url + * Function : redirect_url * - * Description : Should we "trust" this URL? See "trustfile" line in config. + * Description : Checks for redirection URLs and returns a HTTP redirect + * to the destination URL, if necessary * * Parameters : - * 1 : http = http_request request for requested URL - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : NULL => trusted, else string to HTML "untrusted" description. + * Returns : NULL if URL was clean, HTTP redirect otherwise. * *********************************************************************/ -char *trust_url(struct http_request *http, struct client_state *csp) +struct http_response *redirect_url(struct client_state *csp) { - struct file_list *fl; - struct block_spec *b; - struct url_spec url[1], **tl, *t; - char *p, *h; - char *hostport, *path, *refer; - struct http_request rhttp[1]; - int n; + char *p, *q; + struct http_response *rsp; - if (((fl = csp->tlist) == NULL) || ((b = fl->f) == NULL)) + p = q = csp->http->path; + log_error(LOG_LEVEL_REDIRECTS, "checking path for redirects: %s", p); + + /* + * find the last URL encoded in the request + */ + while ((p = strstr(p, "http://"))) { - return(NULL); + q = p++; } - *url = dsplit(http->host); + /* + * if there was any, generate and return a HTTP redirect + */ + if (q != csp->http->path) + { + log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q); + + if (NULL == (rsp = alloc_http_response())) + { + return cgi_error_memory(); + } + + if ( enlist_unique_header(rsp->headers, "Location", q) + || (NULL == (rsp->status = strdup("302 Local Redirect from Junkbuster"))) ) + { + free_http_response(rsp); + return cgi_error_memory(); + } + + return finish_http_response(rsp); + } + else + { + return NULL; + } + +} +#endif /* def FEATURE_FAST_REDIRECTS */ + + +#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 FEATURE_IMAGE_DETECT_MSIE, the info from + * the browser's accept header. + * + * Parameters : + * 1 : csp = Current client state (buffers, headers, etc...) + * + * Returns : True (nonzero) if URL is an image, false (0) + * otherwise + * + *********************************************************************/ +int is_imageurl(struct client_state *csp) +{ +#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)) + { + return 1; + } + else if ((csp->accept_types + & (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_IMAGE|ACCEPT_TYPE_MSIE_HTML)) + == (ACCEPT_TYPE_IS_MSIE|ACCEPT_TYPE_MSIE_HTML)) + { + return 0; + } +#endif /* def FEATURE_IMAGE_DETECT_MSIE */ + + return ((csp->action->flags & ACTION_IMAGE) != 0); + +} +#endif /* def FEATURE_IMAGE_BLOCKING */ - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(NULL); + +#ifdef FEATURE_COOKIE_JAR +/********************************************************************* + * + * Function : is_untrusted_url + * + * Description : Should we "distrust" this URL (and block it)? + * + * Yes if it matches a line in the trustfile, or if the + * referrer matches a line starting with "+" in the + * trustfile. + * No otherwise. + * + * Parameters : + * 1 : csp = Current client state (buffers, headers, etc...) + * + * Returns : 0 => trusted, 1 => untrusted + * + *********************************************************************/ +int is_untrusted_url(struct client_state *csp) +{ + struct file_list *fl; + struct block_spec *b; + struct url_spec url[1], **tl, *t; + struct http_request rhttp[1]; + char *p, *h; + + /* + * If we don't have a trustlist, we trust everybody + */ + if (((fl = csp->tlist) == NULL) || ((b = fl->f) == NULL)) + { + return(0); + } + + + /* + * Do we trust the request URL itself? + */ + *url = dsplit(csp->http->host); + + /* if splitting the domain fails, punt */ + if (url->dbuf == NULL) return(0); memset(rhttp, '\0', sizeof(*rhttp)); for (b = b->next; b ; b = b->next) { - if ((b->url->port == 0) || (b->url->port == http->port)) + if ((b->url->port == 0) || (b->url->port == csp->http->port)) { if ((b->url->domain[0] == '\0') || (domaincmp(b->url, url) == 0)) { if ((b->url->path == NULL) || #ifdef REGEX - (regexec(b->url->preg, http->path, 0, NULL, 0) == 0) + (regexec(b->url->preg, csp->http->path, 0, NULL, 0) == 0) #else - (strncmp(b->url->path, http->path, b->url->pathlen) == 0) + (strncmp(b->url->path, csp->http->path, b->url->pathlen) == 0) #endif ) { freez(url->dbuf); freez(url->dvec); - if (b->reject == 0) return(NULL); - - hostport = url_encode(http->hostport); - path = url_encode(http->path); - - if (csp->referrer) - { - refer = url_encode(csp->referrer); - } - else - { - refer = url_encode("undefined"); - } - - n = strlen(CTRUST); - n += strlen(hostport); - n += strlen(path); - n += strlen(refer); - - p = (char *)malloc(n); - - sprintf(p, CTRUST, hostport, path, refer); - - freez(hostport); - freez(path); - freez(refer); + if (b->reject == 0) return(0); - return(p); + return(1); } } } @@ -693,7 +1083,7 @@ char *trust_url(struct http_request *http, struct client_state *csp) if ((csp->referrer == NULL)|| (strlen(csp->referrer) <= 9)) { /* no referrer was supplied */ - goto trust_url_not_trusted; + return(1); } /* forge a URL from the referrer so we can use @@ -706,19 +1096,21 @@ char *trust_url(struct http_request *http, struct client_state *csp) p = strsav(p, " HTTP/1.0"); parse_http_request(p, rhttp, csp); + freez(p); if (rhttp->cmd == NULL) { - freez(p); - goto trust_url_not_trusted; + return(1); } - freez(p); + /* + * If not, do we maybe trust its referrer? + */ *url = dsplit(rhttp->host); /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) goto trust_url_not_trusted; + if (url->dbuf == NULL) return(1); for (tl = csp->config->trust_list; (t = *tl) ; tl++) { @@ -749,9 +1141,9 @@ char *trust_url(struct http_request *http, struct client_state *csp) h = NULL; h = strsav(h, "~"); - h = strsav(h, http->hostport); + h = strsav(h, csp->http->hostport); - p = http->path; + p = csp->http->path; if ((*p++ == '/') && (*p++ == '~')) { @@ -761,168 +1153,299 @@ char *trust_url(struct http_request *http, struct client_state *csp) if ((p = strchr(p, '/'))) { *p = '\0'; - h = strsav(h, http->path); + h = strsav(h, csp->http->path); /* FIXME: p?! */ h = strsav(h, "/"); } } - free_http_request(rhttp); - fprintf(fp, "%s\n", h); freez(h); fclose(fp); } - return(NULL); + return(0); } } } } + return(1); +} +#endif /* def FEATURE_COOKIE_JAR */ + + +/********************************************************************* + * + * Function : pcrs_filter_response + * + * 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 and raise the CSP_FLAG_MODIFIED flag if appropriate. + * + * Parameters : + * 1 : csp = Current client state (buffers, headers, etc...) + * + * Returns : a pointer to the (newly allocated) modified buffer. + * or NULL in case something went wrong + * + *********************************************************************/ +char *pcrs_filter_response(struct client_state *csp) +{ + int hits=0; + size_t size; + + char *old = csp->iob->cur, *new = NULL; + pcrs_job *job; + + struct file_list *fl; + struct re_filterfile_spec *b; + + /* Sanity first */ + if (csp->iob->cur >= csp->iob->eod) + { + return(NULL); + } + size = csp->iob->eod - csp->iob->cur; -trust_url_not_trusted: - free_http_request(rhttp); + /* + * 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; + } - hostport = url_encode(http->hostport); - path = url_encode(http->path); + if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) ) + { + log_error(LOG_LEVEL_ERROR, "Unable to get current state of regexp filtering."); + return(NULL); + } - if (csp->referrer) + if ( NULL == b->joblist ) { - refer = url_encode(csp->referrer); + log_error(LOG_LEVEL_RE_FILTER, "Empty joblist. Nothing to do."); + return(NULL); } - else + + log_error(LOG_LEVEL_RE_FILTER, "re_filtering %s%s (size %d) ...", + csp->http->hostport, csp->http->path, size); + + /* Apply all jobs from the joblist */ + for (job = b->joblist; NULL != job; job = job->next) { - refer = url_encode("undefined"); + hits += pcrs_execute(job, old, size, &new, &size); + if (old != csp->iob->cur) free(old); + old=new; } - n = strlen(CTRUST); - n += strlen(hostport); - n += strlen(path); - n += strlen(refer); + log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size); - p = (char *)malloc(n); - sprintf(p, CTRUST, hostport, path, refer); + /* + * If there were no hits, destroy our copy and let + * chat() use the original in csp->iob + */ + if (!hits) + { + free(new); + return(NULL); + } - freez(hostport); - freez(path); - freez(refer); + csp->flags |= CSP_FLAG_MODIFIED; + csp->content_length = size; + IOB_RESET(csp); - return(p); + return(new); } -#endif /* def TRUST_FILES */ /********************************************************************* * - * Function : intercept_url + * Function : gif_deanimate_response * - * Description : checks the URL `basename' against a list of URLs to - * snarf. If it matches, it calls the associated function - * which returns an HTML page to send back to the client. - * Right now, we snarf: - * "show-proxy-args", and - * "ij-untrusted-url" (optional w/TRUST_FILES) + * Description : Deanimate the GIF image that has been accumulated in + * csp->iob->buf, set csp->content_length to the modified + * size and raise the CSP_FLAG_MODIFIED flag. * * Parameters : - * 1 : http = http_request request, check `basename's of blocklist - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : NULL for no recognized URLs, or an HTML description page. + * Returns : a pointer to the (newly allocated) modified buffer. + * or NULL in case something went wrong. * *********************************************************************/ -char *intercept_url(struct http_request *http, struct client_state *csp) +char *gif_deanimate_response(struct client_state *csp) { - char *basename; - const struct interceptors *v; - - basename = strrchr(http->path, '/'); - - if (basename == NULL) return(NULL); - - basename ++; /* first char past the last slash */ + struct binbuffer *in, *out; + char *p; + int size = csp->iob->eod - csp->iob->cur; - if (*basename) + /* + * If the body has a "chunked" transfer-encoding, + * get rid of it first, adjusting size and iob->eod + */ + if (csp->flags & CSP_FLAG_CHUNKED) { - for (v = intercept_patterns; v->str; v++) + log_error(LOG_LEVEL_DEANIMATE, "Need to de-chunk first"); + if (0 == (size = remove_chunked_transfer_coding(csp->iob->cur, size))) { - if (strncmp(basename, v->str, v->len) == 0) - { - return((v->interceptor)(http, csp)); - } + 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))) ) + { + log_error(LOG_LEVEL_DEANIMATE, "failed! (no mem)"); + return NULL; } - return(NULL); + in->buffer = csp->iob->cur; + in->size = size; + + if (gif_deanimate(in, out, strncmp("last", csp->action->string[ACTION_STRING_DEANIMATE], 4))) + { + log_error(LOG_LEVEL_DEANIMATE, "failed! (gif parsing)"); + free(in); + buf_free(out); + return(NULL); + } + else + { + 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); + return(p); + } } -#ifdef FAST_REDIRECTS + /********************************************************************* * - * Function : redirect_url + * Function : remove_chunked_transfer_coding * - * Description : Checks for redirection URLs and returns a HTTP redirect - * to the destination URL. + * Description : In-situ remove the "chunked" transfer coding as defined + * in rfc2616 from a buffer. * * Parameters : - * 1 : http = http_request request, check `basename's of blocklist - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : buffer = Pointer to the text buffer + * 2 : size = Number of bytes to be processed * - * Returns : NULL if URL was clean, HTTP redirect otherwise. + * 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 * *********************************************************************/ -char *redirect_url(struct http_request *http, struct client_state *csp) +int remove_chunked_transfer_coding(char *buffer, const size_t size) { - char *p, *q; + size_t newsize = 0; + unsigned int chunksize = 0; + char *from_p, *to_p; - p = q = csp->http->path; - log_error(LOG_LEVEL_REDIRECTS, "checking path: %s", p); + assert(buffer); + from_p = to_p = buffer; - /* find the last URL encoded in the request */ - while (p = strstr(p, "http://")) + if (sscanf(buffer, "%x", &chunksize) != 1) { - q = p++; + log_error(LOG_LEVEL_ERROR, "Invalid first chunksize while stripping \"chunked\" transfer coding"); + return(0); } - /* if there was any, generate and return a HTTP redirect */ - if (q != csp->http->path) + while (chunksize > 0) { - log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q); + 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; - p = (char *)malloc(strlen(HTTP_REDIRECT_TEMPLATE) + strlen(q)); - sprintf(p, HTTP_REDIRECT_TEMPLATE, q); - return(p); - } - else - { - return(NULL); + 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); + } -#endif /* def FAST_REDIRECTS */ + /********************************************************************* * - * Function : url_permissions + * Function : url_actions * - * Description : Gets the permissions for this URL. + * Description : Gets the actions for this URL. * * Parameters : * 1 : http = http_request request for blocked URLs * 2 : csp = Current client state (buffers, headers, etc...) * - * Returns : permissions bitmask specifiying what this URL can do. - * If not on list, will be default_permissions. + * Returns : N/A * *********************************************************************/ -int url_permissions(struct http_request *http, struct client_state *csp) +void url_actions(struct http_request *http, + struct client_state *csp) { struct file_list *fl; - struct permissions_spec *b; + struct url_actions *b; + + init_current_action(csp->action); + + if (((fl = csp->actions_list) == NULL) || ((b = fl->f) == NULL)) + { + return; + } + + apply_url_actions(csp->action, http, b); + +} + + +/********************************************************************* + * + * Function : apply_url_actions + * + * Description : Applies a list of URL actions. + * + * Parameters : + * 1 : action = Destination. + * 2 : http = Current URL + * 3 : b = list of URL actions to apply + * + * Returns : N/A + * + *********************************************************************/ +void apply_url_actions(struct current_action_spec *action, + struct http_request *http, + struct url_actions *b) +{ struct url_spec url[1]; - if (((fl = csp->permissions_list) == NULL) || ((b = fl->f) == NULL)) + if (b == NULL) { - return(csp->config->default_permissions); + /* Should never happen */ + return; } *url = dsplit(http->host); @@ -930,7 +1453,7 @@ int url_permissions(struct http_request *http, struct client_state *csp) /* if splitting the domain fails, punt */ if (url->dbuf == NULL) { - return(csp->config->default_permissions); + return; } for (b = b->next; NULL != b; b = b->next) @@ -947,9 +1470,7 @@ int url_permissions(struct http_request *http, struct client_state *csp) #endif ) { - freez(url->dbuf); - freez(url->dvec); - return(b->permissions); + merge_current_action(action, b->action); } } } @@ -957,8 +1478,6 @@ int url_permissions(struct http_request *http, struct client_state *csp) freez(url->dbuf); freez(url->dvec); - return(csp->config->default_permissions); - } @@ -972,51 +1491,55 @@ int url_permissions(struct http_request *http, struct client_state *csp) * 1 : http = http_request request for current URL * 2 : csp = Current client state (buffers, headers, etc...) * - * Returns : Return gw_default for no forward match, - * else a gateway pointer to a specific forwarding proxy. + * Returns : Pointer to forwarding information. * *********************************************************************/ -const struct gateway *forward_url(struct http_request *http, struct client_state *csp) +const struct forward_spec * forward_url(struct http_request *http, + struct client_state *csp) { - struct file_list *fl; - struct forward_spec *b; + static const struct forward_spec fwd_default[1] = { FORWARD_SPEC_INITIALIZER }; + struct forward_spec *fwd = csp->config->forward; struct url_spec url[1]; - if (((fl = csp->flist) == NULL) || ((b = fl->f) == NULL)) + if (fwd == NULL) { - return(gw_default); + return(fwd_default); } *url = dsplit(http->host); /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(gw_default); + if (url->dbuf == NULL) + { + return(fwd_default); + } - for (b = b->next; b ; b = b->next) + while (fwd != NULL) { - if ((b->url->port == 0) || (b->url->port == http->port)) + if ((fwd->url->port == 0) || (fwd->url->port == http->port)) { - if ((b->url->domain[0] == '\0') || (domaincmp(b->url, url) == 0)) + if ((fwd->url->domain[0] == '\0') || (domaincmp(fwd->url, url) == 0)) { - if ((b->url->path == NULL) || + if ((fwd->url->path == NULL) || #ifdef REGEX - (regexec(b->url->preg, http->path, 0, NULL, 0) == 0) + (regexec(fwd->url->preg, http->path, 0, NULL, 0) == 0) #else - (strncmp(b->url->path, http->path, b->url->pathlen) == 0) + (strncmp(fwd->url->path, http->path, fwd->url->pathlen) == 0) #endif ) { freez(url->dbuf); freez(url->dvec); - return(b->gw); + return(fwd); } } } + fwd = fwd->next; } freez(url->dbuf); freez(url->dvec); - return(gw_default); + return(fwd_default); } @@ -1033,598 +1556,195 @@ const struct gateway *forward_url(struct http_request *http, struct client_state * 1 : domain = a URL address * * Returns : url_spec structure populated with dbuf, dcnt and dvec. + * On error, the dbuf field will be set to NULL. (As + * will all the others, but you don't need to check + * them). + * + * FIXME: Returning a structure is horribly inefficient, please can + * this structure take a (struct url_spec * dest) + * pointer instead? * *********************************************************************/ struct url_spec dsplit(char *domain) { struct url_spec ret[1]; - char *v[BUFSIZ]; + char *v[BUFFER_SIZE]; int size; char *p; memset(ret, '\0', sizeof(*ret)); - if ((p = strrchr(domain, '.'))) + if (domain[strlen(domain) - 1] == '.') { - if (*(++p) == '\0') - { - ret->toplevel = 1; - } + ret->unanchored |= ANCHOR_RIGHT; + } + + if (domain[0] == '.') + { + ret->unanchored |= ANCHOR_LEFT; } ret->dbuf = strdup(domain); + if (NULL == ret->dbuf) + { + return *ret; + } /* map to lower case */ - for (p = ret->dbuf; *p ; p++) *p = tolower(*p); + for (p = ret->dbuf; *p ; p++) + { + *p = tolower((int)(unsigned char)*p); + } /* split the domain name into components */ ret->dcnt = ssplit(ret->dbuf, ".", v, SZ(v), 1, 1); - if (ret->dcnt <= 0) + if (ret->dcnt < 0) { + free(ret->dbuf); memset(ret, '\0', sizeof(ret)); - return(*ret); + return *ret; } - - /* save a copy of the pointers in dvec */ - size = ret->dcnt * sizeof(*ret->dvec); - - if ((ret->dvec = (char **)malloc(size))) + else if (ret->dcnt == 0) { - memcpy(ret->dvec, v, size); + return *ret; } - return(*ret); - -} - - -/********************************************************************* - * - * Function : domaincmp - * - * Description : Compare domain names. - * domaincmp("a.b.c" , "a.b.c") => 0 (MATCH) - * domaincmp("a*.b.c", "a.b.c") => 0 (MATCH) - * domaincmp("b.c" , "a.b.c") => 0 (MATCH) - * domaincmp("" , "a.b.c") => 0 (MATCH) - * - * Parameters : - * 1 : pattern = a domain that may contain a '*' as a wildcard. - * 2 : fqdn = domain name against which the patterns are compared. - * - * Returns : 0 => domains are equivalent, else no match. - * - *********************************************************************/ -int domaincmp(struct url_spec *pattern, struct url_spec *fqdn) -{ - char **pv, **fv; /* vectors */ - int pn, fn; /* counters */ - char *p, *f; /* chars */ - - pv = pattern->dvec; - pn = pattern->dcnt; - - fv = fqdn->dvec; - fn = fqdn->dcnt; + /* save a copy of the pointers in dvec */ + size = ret->dcnt * sizeof(*ret->dvec); - while ((pn > 0) && (fn > 0)) + ret->dvec = (char **)malloc(size); + if (NULL == ret->dvec) { - p = pv[--pn]; - f = fv[--fn]; - - while (*p && *f && (*p == tolower(*f))) - { - p++, f++; - } - - if ((*p != tolower(*f)) && (*p != '*')) return(1); + free(ret->dbuf); + memset(ret, '\0', sizeof(ret)); + return *ret; } - if (pn > 0) return(1); + memcpy(ret->dvec, v, size); - return(0); + return *ret; } -/* intercept functions */ - /********************************************************************* * - * Function : show_proxy_args + * Function : simple_domaincmp * - * Description : This "crunch"es "http:/any.thing/show-proxy-args" and - * returns a web page describing the current status of IJB. + * Description : Domain-wise Compare fqdn's. The comparison is + * both left- and right-anchored. The individual + * domain names are compared with simplematch(). + * This is only used by domaincmp. * * Parameters : - * 1 : http = ignored - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : pv = array of patterns to compare + * 2 : fv = array of domain components to compare + * 3 : len = length of the arrays (both arrays are the + * same length - if they weren't, it couldn't + * possibly be a match). * - * Returns : A string that contains the current status of IJB. + * Returns : 0 => domains are equivalent, else no match. * *********************************************************************/ -char *show_proxy_args(struct http_request *http, struct client_state *csp) +static int simple_domaincmp(char **pv, char **fv, int len) { - char *s = NULL; - -#ifdef SPLIT_PROXY_ARGS - FILE * fp; - char buf[BUFSIZ]; - char * p; - const char * filename = NULL; - const char * file_description = NULL; - char * query_string = strrchr(http->path, '?'); - char which_file = '\0'; - + int n; - if (query_string != NULL) - { - /* first char past the last '?' (maybe '\0')*/ - which_file = query_string[1]; - } - switch (which_file) + for (n = 0; n < len; n++) { - case 'b': - if (csp->blist) - { - filename = csp->blist->filename; - file_description = "Block List"; - } - break; - case 'p': - if (csp->permissions_list) + if (simplematch(pv[n], fv[n])) { - filename = csp->permissions_list->filename; - file_description = "Permissions List"; + return 1; } - break; - case 'f': - if (csp->flist) - { - filename = csp->flist->filename; - file_description = "Forward List"; - } - break; - -#ifdef ACL_FILES - case 'a': - if (csp->alist) - { - filename = csp->alist->filename; - file_description = "Access Control List"; - } - break; -#endif /* def ACL_FILES */ - -#ifdef USE_IMAGE_LIST - case 'i': - if (csp->ilist) - { - filename = csp->ilist->filename; - file_description = "Image List"; - } - break; -#endif /* def USE_IMAGE_LIST */ - -#ifdef PCRS - case 'r': - if (csp->rlist) - { - filename = csp->rlist->filename; - file_description = "RE Filter List"; - } - break; -#endif /* def PCRS */ - -#ifdef TRUST_FILES - case 't': - if (csp->tlist) - { - filename = csp->tlist->filename; - file_description = "Trust List"; - } - break; -#endif /* def TRUST_FILES */ - } - - if (filename) - { - /* Display specified file */ - /* FIXME: Add HTTP headers so this isn't cached */ - s = strsav(s, - "HTTP/1.0 200 OK\n" - "Server: IJ/" VERSION "\n" - "Content-type: text/html\n" - "Pragma: no-cache\n" - "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "\n" - - "" - "" - "Internet Junkbuster Proxy Status - "); - s = strsav(s, file_description); - s = strsav(s, - "" - "\n" - "\n" - "
    \n" - "

    " BANNER "\n"); - s = strsav(s, file_description); - s = strsav(s, - "

    \n" - "

    Back to proxy status

    \n" - "

    "); - s = strsav(s, file_description); - s = strsav(s, - "

    \n" - "Contents of file ""); - p = html_encode(filename); - s = strsav(s, p); - freez(p); - s = strsav(s, - "":
    \n" - "

    \n" - "
    ");
-      
-      if ((fp = fopen(filename, "r")) == NULL)
-      {
-         s = strsav(s, "

    ERROR OPENING FILE!

    ");
-      }
-      else
-      {
-         while (fgets(buf, sizeof(buf), fp))
-         {
-            p = html_encode(buf);
-            if (p)
-            {
-               s = strsav(s, p);
-               freez(p);
-               s = strsav(s, "
    ");
-            }
-         }
-         fclose(fp);
-      }
-
-      s = strsav(s,
-         "
    \n" - "
    \n" - "

    Back to proxy status

    \n" - "
    \n" - "

    \n" - "The " BANNER " Proxy - \n" - "" HOME_PAGE_URL "\n" - "" - "\n"); - return(s); - } -#endif /* def SPLIT_PROXY_ARGS */ - - s = strsav(s, csp->config->proxy_args->header); - s = strsav(s, csp->config->proxy_args->invocation); -#ifdef STATISTICS - s = add_stats(s); -#endif /* def STATISTICS */ - s = strsav(s, csp->config->proxy_args->gateways); - -#ifdef SPLIT_PROXY_ARGS - s = strsav(s, - "

    The following files are in use:

    \n" - "

    (Click a filename to view it)

    \n" - ""); - -#else /* ifndef SPLIT_PROXY_ARGS */ - if (csp->blist) - { - s = strsav(s, csp->blist->proxy_args); - } - - if (csp->clist) - { - s = strsav(s, csp->clist->proxy_args); - } - - if (csp->flist) - { - s = strsav(s, csp->flist->proxy_args); - } - -#ifdef ACL_FILES - if (csp->alist) - { - s = strsav(s, csp->alist->proxy_args); - } -#endif /* def ACL_FILES */ - -#ifdef USE_IMAGE_LIST - if (csp->ilist) - { - s = strsav(s, csp->ilist->proxy_args); - } -#endif /* def USE_IMAGE_LIST */ - -#ifdef PCRS - if (csp->rlist) - { - s = strsav(s, csp->rlist->proxy_args); - } -#endif /* def PCRS */ - -#ifdef TRUST_FILES - if (csp->tlist) - { - s = strsav(s, csp->tlist->proxy_args); - } -#endif /* def TRUST_FILES */ - -#endif /* ndef SPLIT_PROXY_ARGS */ - - s = strsav(s, csp->config->proxy_args->trailer); - - return(s); + return 0; } /********************************************************************* * - * Function : ijb_send_banner - * - * Description : This "crunch"es "http:/any.thing/ijb-send-banner and - * thus triggers sending the image in jcc.c:chat. - * For the unlikely case, that the imagefile/MSIE - * mechanism is not used, or tinygif = 0, a page - * describing the reson of the interception is generated. - * - * Parameters : - * 1 : http = http_request request for crunched URL - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : A string that contains why this was intercepted. - * - *********************************************************************/ -char *ijb_send_banner(struct http_request *http, struct client_state *csp) -{ - return(strdup(CNOBANNER)); -} - -#ifdef TRUST_FILES -/********************************************************************* - * - * Function : ij_untrusted_url + * Function : domaincmp * - * Description : This "crunch"es "http:/any.thing/ij-untrusted-url" and - * returns a web page describing why it was untrusted. + * Description : Domain-wise Compare fqdn's. Governed by the bimap in + * pattern->unachored, the comparison is un-, left-, + * right-anchored, or both. + * The individual domain names are compared with + * simplematch(). * * Parameters : - * 1 : http = http_request request for crunched URL - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : pattern = a domain that may contain a '*' as a wildcard. + * 2 : fqdn = domain name against which the patterns are compared. * - * Returns : A string that contains why this was untrusted. + * Returns : 0 => domains are equivalent, else no match. * *********************************************************************/ -char *ij_untrusted_url(struct http_request *http, struct client_state *csp) +int domaincmp(struct url_spec *pattern, struct url_spec *fqdn) { - int n; - char *hostport, *path, *refer, *p, *v[9]; - char buf[BUFSIZ]; - struct url_spec **tl, *t; - - - static const char format[] = - "HTTP/1.0 200 OK\r\n" - "Pragma: no-cache\n" - "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: Request for untrusted URL\n" - "\n" - BODY - "

    " - BANNER - "

    " - "The " BANNER " Proxy " - "" - "(" HOME_PAGE_URL ") " - "intercepted the request for %s%s\n" - "because the URL is not trusted.\n" - "

    \n"; - - if ((n = ssplit(http->path, "?+", v, SZ(v), 0, 0)) == 4) - { - hostport = url_decode(v[1]); - path = url_decode(v[2]); - refer = url_decode(v[3]); - } - else - { - hostport = strdup("undefined_host"); - path = strdup("/undefined_path"); - refer = strdup("undefined"); - } + char **pv, **fv; /* vectors */ + int plen, flen; + int unanchored = pattern->unanchored & (ANCHOR_RIGHT | ANCHOR_LEFT); - n = sizeof(format); - n += strlen(hostport); - n += strlen(path ); + plen = pattern->dcnt; + flen = fqdn->dcnt; - if ((p = (char *)malloc(n))) + if (flen < plen) { - sprintf(p, format, hostport, path); + /* fqdn is too short to match this pattern */ + return 1; } - strsav(p, "The referrer in this request was "); - strsav(p, refer); - strsav(p, "
    \n"); + pv = pattern->dvec; + fv = fqdn->dvec; - freez(hostport); - freez(path ); - freez(refer ); - - p = strsav(p, "

    The following referrers are trusted

    \n"); - - for (tl = csp->config->trust_list; (t = *tl) ; tl++) + if (unanchored == ANCHOR_LEFT) { - sprintf(buf, "%s
    \n", t->spec); - p = strsav(p, buf); + /* + * Right anchored. + * + * Convert this into a fully anchored pattern with + * the fqdn and pattern the same length + */ + fv += (flen - plen); /* flen - plen >= 0 due to check above */ + return simple_domaincmp(pv, fv, plen); } - - if (csp->config->trust_info->next) + else if (unanchored == 0) { - struct list *l; - - strcpy(buf, - "

    " - "You can learn more about what this means " - "and what you may be able to do about it by " - "reading the following documents:
    \n" - "

      \n" - ); - - p = strsav(p, buf); - - for (l = csp->config->trust_info->next; l ; l = l->next) + /* Fully anchored, check length */ + if (flen != plen) { - sprintf(buf, - "
    1. %s
      \n", - l->str, l->str); - p = strsav(p, buf); + return 1; } - - p = strsav(p, "
    \n"); + return simple_domaincmp(pv, fv, plen); } - - p = strsav(p, "\n" "\n"); - - return(p); - -} -#endif /* def TRUST_FILES */ - - -#ifdef STATISTICS -/********************************************************************* - * - * Function : add_stats - * - * Description : Statistics function of JB. Called by `show_proxy_args'. - * - * Parameters : - * 1 : s = string that holds the proxy args description page - * - * Returns : A pointer to the descriptive status web page. - * - *********************************************************************/ -char *add_stats(char *s) -{ - /* - * Output details of the number of requests rejected and - * accepted. This is switchable in the junkbuster config. - * Does nothing if this option is not enabled. - */ - - float perc_rej; /* Percentage of http requests rejected */ - char out_str[81]; - int local_urls_read = urls_read; - int local_urls_rejected = urls_rejected; - - /* - * Need to alter the stats not to include the fetch of this - * page. - * - * Can't do following thread safely! doh! - * - * urls_read--; - * urls_rejected--; * This will be incremented subsequently * - */ - - s = strsav(s,"

    Statistics for this " BANNER ":

    \n"); - - if (local_urls_read == 0) + else if (unanchored == ANCHOR_RIGHT) { - - s = strsav(s,"No activity so far!\n"); - + /* Left anchored, ignore all extra in fqdn */ + return simple_domaincmp(pv, fv, plen); } else { - - perc_rej = (float)local_urls_rejected * 100.0F / - (float)local_urls_read; - - sprintf(out_str, - "%d requests received, %d filtered " - "(%6.2f %%).", - local_urls_read, - local_urls_rejected, perc_rej); - - s = strsav(s,out_str); + /* Unanchored */ + int n; + int maxn = flen - plen; + for (n = 0; n <= maxn; n++) + { + if (!simple_domaincmp(pv, fv, plen)) + { + return 0; + } + /* + * Doesn't match from start of fqdn + * Try skipping first part of fqdn + */ + fv++; + } + return 1; } - return(s); } -#endif /* def STATISTICS */ /*