X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=filters.c;h=aa798bcd89a69fd71b7da4feec3cc5a7cf8054df;hp=e5497c2e353ffc02e21ce60b23138a3018d14581;hb=2afa3221231342abdf0298bf1c463042843e05ab;hpb=9412ec32b9c306ace8d22bbd90b872296792e8eb diff --git a/filters.c b/filters.c index e5497c2e..aa798bcd 100644 --- a/filters.c +++ b/filters.c @@ -1,4 +1,4 @@ -const char filters_rcs[] = "$Id: filters.c,v 1.10 2001/05/29 09:50:24 jongfoster Exp $"; +const char filters_rcs[] = "$Id: filters.c,v 1.31 2001/09/16 11:38:02 jongfoster Exp $"; /********************************************************************* * * File : $Source: /cvsroot/ijbswa/current/filters.c,v $ @@ -6,9 +6,9 @@ const char filters_rcs[] = "$Id: filters.c,v 1.10 2001/05/29 09:50:24 jongfoster * Purpose : Declares functions to parse/crunch headers and pages. * Functions declared include: * `acl_addr', `add_stats', `block_acl', `block_imageurl', - * `block_url', `url_permissions', `domaincmp', `dsplit', + * `block_url', `url_actions', `domaincmp', `dsplit', * `filter_popups', `forward_url', 'redirect_url', - * `ij_untrusted_url', `intercept_url', `re_process_buffer', + * `ij_untrusted_url', `intercept_url', `pcrs_filter_respose', * `show_proxy_args', 'ijb_send_banner', and `trust_url' * * Copyright : Written by and Copyright (C) 2001 the SourceForge @@ -38,6 +38,121 @@ const char filters_rcs[] = "$Id: filters.c,v 1.10 2001/05/29 09:50:24 jongfoster * * Revisions : * $Log: filters.c,v $ + * 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 @@ -185,6 +300,10 @@ const char filters_rcs[] = "$Id: filters.c,v 1.10 2001/05/29 09:50:24 jongfoster #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" @@ -202,61 +321,7 @@ 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.
See why" -#ifdef FORCE_LOAD - " or " - "go there anyway." -#endif /* def FORCE_LOAD */ - "

\n" - "\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 @@ -273,36 +338,33 @@ static const char CTRUST[] = * Returns : 0 = FALSE (don't block) and 1 = TRUE (do block) * *********************************************************************/ -int block_acl(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_list *acl = csp->config->acl; /* if not using an access control list, then permit the connection */ - if (((fl = csp->alist) == NULL) || - ((acl = (struct access_control_list *) fl->f) == NULL)) + if (acl == NULL) { return(0); } /* search the list */ - for (a = acl->next ; a ; a = a->next) + while (acl != NULL) { - if ((csp->ip_addr_long & a->src->mask) == a->src->addr) + if ((csp->ip_addr_long & acl->src->mask) == acl->src->addr) { if (dst == NULL) { /* Just want to check if they have any access */ - if (a->action == ACL_PERMIT) + if (acl->action == ACL_PERMIT) { return(0); } } - else if ( ((dst->addr & a->dst->mask) == a->dst->addr) - && ((dst->port == a->dst->port) || (a->dst->port == 0))) + else if ( ((dst->addr & acl->dst->mask) == acl->dst->addr) + && ((dst->port == acl->dst->port) || (acl->dst->port == 0))) { - if (a->action == ACL_PERMIT) + if (acl->action == ACL_PERMIT) { return(0); } @@ -312,6 +374,7 @@ int block_acl(struct access_control_addr *dst, } } } + acl = acl->next; } return(1); @@ -392,7 +455,7 @@ int acl_addr(char *aspec, struct access_control_addr *aca) return(0); } -#endif /* def ACL_FILES */ +#endif /* def FEATURE_ACL */ /********************************************************************* @@ -402,229 +465,385 @@ int acl_addr(char *aspec, struct access_control_addr *aca) * Description : Called from `chat'. Check to see if we need to block this. * * Parameters : - * 1 : http = http_request request to "check" for blocked - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : NULL => unblocked, else string to HTML block description. + * Returns : NULL => unblocked, else HTTP block response * *********************************************************************/ -char *block_url(struct http_request *http, struct client_state *csp) +struct http_response *block_url(struct client_state *csp) { +#ifdef FEATURE_IMAGE_BLOCKING char *p; - int n; - int factor = 2; +#endif /* def FEATURE_IMAGE_BLOCKING */ + struct http_response *rsp; - if ((csp->permissions & PERMIT_BLOCK) == 0) + /* + * If it's not blocked, don't block it ;-) + */ + if ((csp->action->flags & ACTION_BLOCK) == 0) { return(NULL); } - else + + /* + * Else, prepare a response + */ + if (NULL == (rsp = alloc_http_response())) { -#ifdef FORCE_LOAD - factor++; -#endif /* def FORCE_LOAD */ + return NULL; + } - n = strlen(CBLOCK); - n += factor * strlen(http->hostport); - n += factor * strlen(http->path); + /* + * 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]; - p = (char *)malloc(n); + /* and handle accordingly: */ + if ((p == NULL) || (0 == strcmpic(p, "logo"))) + { + 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"); + } -#ifdef FORCE_LOAD - sprintf(p, CBLOCK, http->hostport, http->path, http->hostport, http->path, - http->hostport, http->path); -#else - sprintf(p, CBLOCK, http->hostport, http->path, http->hostport, http->path); -#endif /* def FORCE_LOAD */ + else if (0 == strcmpic(p, "blank")) + { + 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"); + } + + else + { + rsp->status = strdup("302 Local Redirect from Junkbuster"); + enlist_unique_header(rsp->headers, "Location", p); + } + } + else +#endif /* def FEATURE_IMAGE_BLOCKING */ + + /* + * Else, generate an HTML "blocked" message: + */ + { + 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); + + /* + * 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(p); } + + return(finish_http_response(rsp)); + } -#ifdef IMAGE_BLOCKING +#ifdef FEATURE_TRUST /********************************************************************* * - * Function : block_imageurl + * Function : trust_url FIXME: I should be called distrust_url * - * Description : Given a URL which is blocked, decide whether to - * send the "blocked" image or HTML. + * 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 : 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 => trusted, else http_response. * *********************************************************************/ -int block_imageurl(struct http_request *http, struct client_state *csp) +struct http_response *trust_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)) + struct http_response *rsp; + struct map * exports; + char buf[BUFFER_SIZE], *p = NULL; + struct url_spec **tl, *t; + + /* + * Don't bother to work on trusted URLs + */ + if (!is_untrusted_url(csp)) { - 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 NULL; } -#endif - return ((csp->permissions & PERMIT_IMAGE) != 0); + exports = default_exports(csp, NULL); + + /* + * Export the host, port, and referrer information + */ + 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) + { + map(exports, "referrer", 1, csp->referrer + 9, 1); + map(exports, "referrer-html", 1, html_encode(csp->referrer + 9), 0); + } + else + { + map(exports, "referrer", 1, "unknown", 1); + map(exports, "referrer-html", 1, "unknown", 1); + } + + /* + * Export the trust list + */ + for (tl = csp->config->trust_list; (t = *tl) ; tl++) + { + sprintf(buf, "
  • %s
    • \n", t->spec); + p = strsav(p, buf); + } + map(exports, "trusted-referrers", 1, p, 0); + p = NULL; + + /* + * Export the trust info, if available + */ + if (csp->config->trust_info->first) + { + struct list_entry *l; + + for (l = csp->config->trust_info->first; l ; l = l->next) + { + sprintf(buf, "
  • %s
    \n",l->str, l->str); + p = strsav(p, buf); + } + map(exports, "trust-info", 1, p, 0); + } + else + { + map_block_killer(exports, "have-trust-info"); + } + + /* + * Export the force prefix or the force conditional block killer + */ +#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 = template_load(csp, "untrusted"); + template_fill(&rsp->body, exports); + free_map(exports); + + return(finish_http_response(rsp)); + } -#endif /* def IMAGE_BLOCKING */ +#endif /* def FEATURE_TRUST */ -#ifdef PCRS +#ifdef FEATURE_FAST_REDIRECTS /********************************************************************* * - * Function : re_process_buffer + * Function : redirect_url * - * Description : Apply all jobs from the joblist (aka. Perl regexp's) to - * the text buffer that's been accumulated in csp->iob->buf - * and set csp->content_length to the modified size. + * Description : Checks for redirection URLs and returns a HTTP redirect + * to the destination URL, if necessary * * Parameters : * 1 : csp = Current client state (buffers, headers, etc...) * - * Returns : a pointer to the (newly allocated) modified buffer. - * + * Returns : NULL if URL was clean, HTTP redirect otherwise. * *********************************************************************/ -char *re_process_buffer(struct client_state *csp) +struct http_response *redirect_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; + char *p, *q; + struct http_response *rsp; - struct file_list *fl; - struct re_filterfile_spec *b; + p = q = csp->http->path; + log_error(LOG_LEVEL_REDIRECTS, "checking path for redirects: %s", p); - /* Sanity first ;-) */ - if (size <= 0) + /* + * find the last URL encoded in the request + */ + while ((p = strstr(p, "http://"))) { - return(strdup("")); + q = p++; } - if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) ) + /* + * if there was any, generate and return a HTTP redirect + */ + if (q != csp->http->path) { - log_error(LOG_LEVEL_ERROR, "Unable to get current state of regexp filtering."); - return(strdup("")); - } - - joblist = b->joblist; + log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q); + if (NULL == (rsp = alloc_http_response())) + { + return NULL; + } - log_error(LOG_LEVEL_RE_FILTER, "re_filtering %s%s (size %d) ...", - csp->http->hostport, csp->http->path, size); + rsp->status = strdup("302 Local Redirect from Junkbuster"); + enlist_unique_header(rsp->headers, "Location", q); - /* Apply all jobs from the joblist */ - for (job = joblist; NULL != job; job = job->next) + return(finish_http_response(rsp)); + } + else { - hits += pcrs_exec_substitution(job, old, size, &new, &size); - if (old != csp->iob->cur) free(old); - old=new; + return(NULL); } - log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size); +} +#endif /* def FEATURE_FAST_REDIRECTS */ - csp->content_length = size; - /* fwiw, reset the iob */ - IOB_RESET(csp); - return(new); +#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 PCRS */ +#endif /* def FEATURE_IMAGE_BLOCKING */ -#ifdef TRUST_FILES +#ifdef FEATURE_COOKIE_JAR /********************************************************************* * - * Function : trust_url + * Function : is_untrusted_url * - * Description : Should we "trust" this URL? See "trustfile" line in config. + * 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 : 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 : 0 => trusted, 1 => untrusted * *********************************************************************/ -char *trust_url(struct http_request *http, struct client_state *csp) +int is_untrusted_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, *h; + /* + * If we don't have a trustlist, we trust everybody + */ if (((fl = csp->tlist) == NULL) || ((b = fl->f) == NULL)) { - return(NULL); + return(0); } - *url = dsplit(http->host); + + /* + * Do we trust the request URL itself? + */ + *url = dsplit(csp->http->host); /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(NULL); + 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); + if (b->reject == 0) return(0); - p = (char *)malloc(n); - - sprintf(p, CTRUST, hostport, path, refer); - - freez(hostport); - freez(path); - freez(refer); - - return(p); + return(1); } } } @@ -636,7 +855,7 @@ 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 @@ -649,19 +868,21 @@ 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++) { @@ -692,9 +913,9 @@ 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++ == '~')) { @@ -704,227 +925,198 @@ 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); } } } } - -trust_url_not_trusted: - free_http_request(rhttp); - - 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); - - return(p); - + return(1); } -#endif /* def TRUST_FILES */ - - -static const char C_HOME_PAGE[] = - "HTTP/1.0 200 OK\n" - "Pragma: no-cache\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: Information\n" - "\n" - BODY - "

    " - BANNER - "

    \n" - "

    JunkBuster web site

    \n" - "

    Proxy configuration

    \n" - "

    Look up a URL

    \n" - "\n" - "\n"; +#endif /* def FEATURE_COOKIE_JAR */ /********************************************************************* * - * Function : intercept_url + * Function : pcrs_filter_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 : 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. * * Parameters : - * 1 : http = http_request request, check `basename's of blocklist - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : 1 if it intercepts & handles the request. + * 1 : csp = Current client state (buffers, headers, etc...) * + * Returns : a pointer to the (newly allocated) modified buffer. + * or NULL in case something went wrong + * *********************************************************************/ -int intercept_url(struct http_request *http, struct client_state *csp) +char *pcrs_filter_response(struct client_state *csp) { - char *basename = NULL; - const struct interceptors *v; + int hits=0; + int size = csp->iob->eod - csp->iob->cur; - if (0 == strcmpic(http->host,"i.j.b")) - { - /* - * Catch http://i.j.b/... - */ - basename = http->path; - } - else if ( ( (0 == strcmpic(http->host,"ijbswa.sourceforge.net")) - || (0 == strcmpic(http->host,"ijbswa.sf.net")) ) - && (0 == strncmpic(http->path,"/config", 7)) - && ((http->path[7] == '/') || (http->path[7] == '\0'))) + char *old = csp->iob->cur, *new = NULL; + pcrs_job *job; + + struct file_list *fl; + struct re_filterfile_spec *b; + + /* Sanity first ;-) */ + if (size <= 0) { - /* - * Catch http://ijbswa.sourceforge.net/config/... - * and http://ijbswa.sf.net/config/... - */ - basename = http->path + 7; + return(NULL); } - if (!basename) + if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) ) { - /* Don't want to intercept */ - return(0); + log_error(LOG_LEVEL_ERROR, "Unable to get current state of regexp filtering."); + return(NULL); } - /* We have intercepted it. */ - - /* remove any leading slash */ - if (*basename == '/') + if ( NULL == b->joblist ) { - basename++; + log_error(LOG_LEVEL_RE_FILTER, "Empty joblist. Nothing to do."); + return(NULL); } - log_error(LOG_LEVEL_GPC, "%s%s intercepted!", http->hostport, http->path); - log_error(LOG_LEVEL_CLF, "%s - - [%T] \"%s\" 200 3", - csp->ip_addr_str, http->cmd); + log_error(LOG_LEVEL_RE_FILTER, "re_filtering %s%s (size %d) ...", + csp->http->hostport, csp->http->path, size); - for (v = intercept_patterns; v->str; v++) + /* Apply all jobs from the joblist */ + for (job = b->joblist; NULL != job; job = job->next) { - if (strncmp(basename, v->str, v->len) == 0) - { - char * p = ((v->interceptor)(http, csp)); + hits += pcrs_execute(job, old, size, &new, &size); + if (old != csp->iob->cur) free(old); + old=new; + } - if (p != NULL) - { - /* Send HTML redirection result */ - write_socket(csp->cfd, p, strlen(p)); + log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size); - freez(p); - } - return(1); - } - } + csp->content_length = size; - write_socket(csp->cfd, C_HOME_PAGE, strlen(C_HOME_PAGE)); + /* fwiw, reset the iob */ + IOB_RESET(csp); + return(new); - return(1); } -#ifdef FAST_REDIRECTS + /********************************************************************* * - * Function : redirect_url + * Function : gif_deanimate_response * - * Description : Checks for redirection URLs and returns a HTTP redirect - * to the destination URL. + * Description : Deanimate the GIF image that has been accumulated in + * csp->iob->buf and set csp->content_length to the modified + * size. * * Parameters : - * 1 : http = http_request request, check `basename's of blocklist - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : NULL if URL was clean, HTTP redirect otherwise. + * 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 *redirect_url(struct http_request *http, struct client_state *csp) +char *gif_deanimate_response(struct client_state *csp) { - char *p, *q; + struct binbuffer *in, *out; + char *p; + int size = csp->iob->eod - csp->iob->cur; - p = q = csp->http->path; - log_error(LOG_LEVEL_REDIRECTS, "checking path: %s", p); - - /* find the last URL encoded in the request */ - while (p = strstr(p, "http://")) + if ( (NULL == (in = (struct binbuffer *)zalloc(sizeof *in ))) + || (NULL == (out = (struct binbuffer *)zalloc(sizeof *out))) ) { - q = p++; + log_error(LOG_LEVEL_DEANIMATE, "failed! (no mem)"); + return NULL; } - /* if there was any, generate and return a HTTP redirect */ - if (q != csp->http->path) - { - log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q); + in->buffer = csp->iob->cur; + in->size = size; - p = (char *)malloc(strlen(HTTP_REDIRECT_TEMPLATE) + strlen(q)); - sprintf(p, HTTP_REDIRECT_TEMPLATE, q); - return(p); - } - else + 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; + p = out->buffer; + free(in); + free(out); + return(p); + } } -#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]; - int permissions = csp->config->default_permissions; - if (((fl = csp->permissions_list) == NULL) || ((b = fl->f) == NULL)) + if (b == NULL) { - return(permissions); + /* Should never happen */ + return; } *url = dsplit(http->host); @@ -932,7 +1124,7 @@ int url_permissions(struct http_request *http, struct client_state *csp) /* if splitting the domain fails, punt */ if (url->dbuf == NULL) { - return(permissions); + return; } for (b = b->next; NULL != b; b = b->next) @@ -949,8 +1141,7 @@ int url_permissions(struct http_request *http, struct client_state *csp) #endif ) { - permissions &= b->mask; - permissions |= b->add; + merge_current_action(action, b->action); } } } @@ -958,8 +1149,6 @@ int url_permissions(struct http_request *http, struct client_state *csp) freez(url->dbuf); freez(url->dvec); - return(permissions); - } @@ -973,51 +1162,55 @@ 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); } @@ -1039,18 +1232,20 @@ const struct gateway *forward_url(struct http_request *http, struct client_state 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); @@ -1082,722 +1277,121 @@ struct url_spec dsplit(char *domain) /********************************************************************* * - * Function : domaincmp + * Function : simple_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) + * 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 : pattern = a domain that may contain a '*' as a wildcard. - * 2 : fqdn = domain name against which the patterns are compared. + * 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 : 0 => domains are equivalent, else no match. * *********************************************************************/ -int domaincmp(struct url_spec *pattern, struct url_spec *fqdn) +static int simple_domaincmp(char **pv, char **fv, int len) { - char **pv, **fv; /* vectors */ - int pn, fn; /* counters */ - char *p, *f; /* chars */ - - pv = pattern->dvec; - pn = pattern->dcnt; - - fv = fqdn->dvec; - fn = fqdn->dcnt; + int n; - while ((pn > 0) && (fn > 0)) + for (n = 0; n < len; n++) { - p = pv[--pn]; - f = fv[--fn]; - - while (*p && *f && (*p == tolower(*f))) + if (simplematch(pv[n], fv[n])) { - p++, f++; + return 1; } - - if ((*p != tolower(*f)) && (*p != '*')) return(1); } - if (pn > 0) return(1); - - return(0); + return 0; } -/* intercept functions */ - /********************************************************************* * - * Function : show_proxy_args + * Function : 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. 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 = ignored - * 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 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) +int domaincmp(struct url_spec *pattern, struct url_spec *fqdn) { - 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'; - - - if (query_string != NULL) - { - /* first char past the last '?' (maybe '\0')*/ - which_file = query_string[1]; - } - switch (which_file) - { - case 'p': - if (csp->permissions_list) - { - filename = csp->permissions_list->filename; - file_description = "Permissions List"; - } - 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 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->clist) - { - s = strsav(s, csp->clist->proxy_args); - } - - if (csp->flist) - { - s = strsav(s, csp->flist->proxy_args); - } + char **pv, **fv; /* vectors */ + int plen, flen; + int unanchored = pattern->unanchored & (ANCHOR_RIGHT | ANCHOR_LEFT); -#ifdef ACL_FILES - if (csp->alist) - { - s = strsav(s, csp->alist->proxy_args); - } -#endif /* def ACL_FILES */ + plen = pattern->dcnt; + flen = fqdn->dcnt; -#ifdef PCRS - if (csp->rlist) + if (flen < plen) { - 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); - -} - - -static const char C_URL_INFO_HEADER[] = - "HTTP/1.0 200 OK\n" - "Pragma: no-cache\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: URL Info\n" - "\n" - BODY - "

    " - BANNER - "

    \n" - "

    Information for: http://%s

    \n"; -static const char C_URL_INFO_FOOTER[] = - "\n

    \n" - "\n" - "\n"; - -static const char C_URL_INFO_FORM[] = - "HTTP/1.0 200 OK\n" - "Pragma: no-cache\n" - "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\n" - "Content-Type: text/html\n\n" - "\n" - "\n" - "Internet Junkbuster: URL Info\n" - "\n" - BODY - "

    " - BANNER - "

    \n" - "
    \n" - "

    Please enter a URL, without the leading "http://":

    " - "

    " - "

    \n" - "
    \n" - "\n" - "\n"; - - -/********************************************************************* - * - * Function : permissions_to_text - * - * Description : Converts a permissionsfil entry from numeric form - * ("mask" and "add") to text. - * - * Parameters : - * 1 : mask = As from struct permissions_spec - * 2 : add = As from struct permissions_spec - * - * Returns : A string. Caller must free it. - * - *********************************************************************/ -char * permissions_to_text(unsigned mask, unsigned add) -{ - char * result = strdup(""); - - /* sanity - prevents "-feature +feature" */ - mask |= add; - -#define PERMISSION_TO_TEXT(__bit, __name) \ - if (!(mask & __bit)) \ - { \ - result = strsav(result, " -" __name); \ - } \ - else if (add & __bit) \ - { \ - result = strsav(result, " +" __name); \ + /* fqdn is too short to match this pattern */ + return 1; } - PERMISSION_TO_TEXT(PERMIT_COOKIE_SET, "cookies-set"); - PERMISSION_TO_TEXT(PERMIT_COOKIE_READ, "cookies-read"); - PERMISSION_TO_TEXT(PERMIT_RE_FILTER, "filter"); - PERMISSION_TO_TEXT(PERMIT_POPUPS, "popup"); - PERMISSION_TO_TEXT(PERMIT_REFERER, "referer"); - PERMISSION_TO_TEXT(PERMIT_FAST_REDIRECTS, "fast-redirects"); - PERMISSION_TO_TEXT(PERMIT_BLOCK, "block"); - PERMISSION_TO_TEXT(PERMIT_IMAGE, "image"); + pv = pattern->dvec; + fv = fqdn->dvec; - return result; -} - - - /********************************************************************* - * - * Function : ijb_show_url_info - * - * Description : (please fill me in) - * - * Parameters : - * 1 : http = http_request request for crunched URL - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : ???FIXME - * - *********************************************************************/ -char *ijb_show_url_info(struct http_request *http, struct client_state *csp) -{ - char * query_string = strchr(http->path, '?'); - char * host = NULL; - - if (query_string != NULL) + if (unanchored == ANCHOR_LEFT) { - query_string = url_decode(query_string + 1); - if (strncmpic(query_string, "url=", 4) == 0) - { - host = strdup(query_string + 4); - } - freez(query_string); + /* + * 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 (host != NULL) + else if (unanchored == 0) { - char * result; - char * path; - char * s; - int port = 80; - struct file_list *fl; - struct permissions_spec *b; - struct url_spec url[1]; - int permissions = csp->config->default_permissions; - - result = (char *)malloc(sizeof(C_URL_INFO_HEADER) + 2 * strlen(host)); - sprintf(result, C_URL_INFO_HEADER, host, host); - - s = permissions_to_text(permissions, permissions); - result = strsav(result, "

    Defaults:

    \n

    {"); - result = strsav(result, s); - result = strsav(result, " }

    \n

    Patterns affecting the URL:

    \n

    \n"); - freez(s); - - s = strchr(host, '/'); - if (s != NULL) - { - path = strdup(s); - *s = '\0'; - } - else - { - path = strdup(""); - } - s = strchr(host, ':'); - if (s != NULL) - { - *s++ = '\0'; - port = atoi(s); - } - - if (((fl = csp->permissions_list) == NULL) || ((b = fl->f) == NULL)) - { - freez(host); - freez(path); - result = strsav(result, C_URL_INFO_FOOTER); - return result; - } - - *url = dsplit(host); - - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) - { - freez(host); - freez(path); - result = strsav(result, C_URL_INFO_FOOTER); - return result; - } - - for (b = b->next; NULL != b; b = b->next) + /* Fully anchored, check length */ + if (flen != plen) { - if ((b->url->port == 0) || (b->url->port == port)) - { - if ((b->url->domain[0] == '\0') || (domaincmp(b->url, url) == 0)) - { - if ((b->url->path == NULL) || -#ifdef REGEX - (regexec(b->url->preg, path, 0, NULL, 0) == 0) -#else - (strncmp(b->url->path, path, b->url->pathlen) == 0) -#endif - ) - { - s = permissions_to_text(b->mask, b->add); - result = strsav(result, "{"); - result = strsav(result, s); - result = strsav(result, " }
    \n"); - result = strsav(result, b->url->spec); - result = strsav(result, "
    \n
    \n"); - freez(s); - permissions &= b->mask; - permissions |= b->add; - } - } - } + return 1; } - - freez(url->dbuf); - freez(url->dvec); - - freez(host); - freez(path); - - s = permissions_to_text(permissions, permissions); - result = strsav(result, "

    \n

    Final Results:

    \n

    {"); - result = strsav(result, s); - result = strsav(result, " }
    \n
    \n"); - freez(s); - - result = strsav(result, C_URL_INFO_FOOTER); - return result; - } - else - { - return strdup(C_URL_INFO_FORM); + return simple_domaincmp(pv, fv, plen); } -} - - -/********************************************************************* - * - * Function : ijb_send_banner - * - * Description : This "crunch"es "http://i.j.b/ijb-send-banner and - * sends the image. - * - * Parameters : - * 1 : http = http_request request for crunched URL - * 2 : csp = Current client state (buffers, headers, etc...) - * - * Returns : NULL, indicating that it has already sent the data. - * - *********************************************************************/ -char *ijb_send_banner(struct http_request *http, struct client_state *csp) -{ - write_socket(csp->cfd, JBGIF, sizeof(JBGIF)-1); - - return(NULL); -} - -#ifdef TRUST_FILES -/********************************************************************* - * - * Function : ij_untrusted_url - * - * Description : This "crunch"es "http:/any.thing/ij-untrusted-url" and - * returns a web page describing why it was untrusted. - * - * 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 untrusted. - * - *********************************************************************/ -char *ij_untrusted_url(struct http_request *http, struct client_state *csp) -{ - 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) + else if (unanchored == ANCHOR_RIGHT) { - hostport = url_decode(v[1]); - path = url_decode(v[2]); - refer = url_decode(v[3]); + /* Left anchored, ignore all extra in fqdn */ + return simple_domaincmp(pv, fv, plen); } else { - hostport = strdup("undefined_host"); - path = strdup("/undefined_path"); - refer = strdup("undefined"); - } - - n = sizeof(format); - n += strlen(hostport); - n += strlen(path ); - - if ((p = (char *)malloc(n))) - { - sprintf(p, format, hostport, path); - } - - strsav(p, "The referrer in this request was "); - strsav(p, refer); - strsav(p, "
    \n"); - - freez(hostport); - freez(path ); - freez(refer ); - - p = strsav(p, "

    The following referrers are trusted

    \n"); - - for (tl = csp->config->trust_list; (t = *tl) ; tl++) - { - sprintf(buf, "%s
    \n", t->spec); - p = strsav(p, buf); - } - - if (csp->config->trust_info->next) - { - 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) + /* Unanchored */ + int n; + int maxn = flen - plen; + for (n = 0; n <= maxn; n++) { - sprintf(buf, - "
    1. %s
      \n", - l->str, l->str); - p = strsav(p, buf); + if (!simple_domaincmp(pv, fv, plen)) + { + return 0; + } + /* + * Doesn't match from start of fqdn + * Try skipping first part of fqdn + */ + fv++; } - - p = strsav(p, "
    \n"); - } - - 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) - { - - s = strsav(s,"No activity so far!\n"); - - } - 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); + return 1; } - return(s); } -#endif /* def STATISTICS */ /*