X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=filters.c;h=b31cb434d155568e9240d53b121467c2be10faba;hp=42f46aff89d41c957bd51a6743dde13f60f260f3;hb=299a43a9012f4c361ff92b302e57990b3016bb64;hpb=1952a87fe55ceeff77c376d61e281df7aee11db7 diff --git a/filters.c b/filters.c index 42f46aff..b31cb434 100644 --- a/filters.c +++ b/filters.c @@ -1,4 +1,4 @@ -const char filters_rcs[] = "$Id: filters.c,v 1.3 2001/05/20 16:44:47 jongfoster Exp $"; +const char filters_rcs[] = "$Id: filters.c,v 1.15 2001/06/03 19:12:00 oes Exp $"; /********************************************************************* * * File : $Source: /cvsroot/ijbswa/current/filters.c,v $ @@ -6,7 +6,7 @@ const char filters_rcs[] = "$Id: filters.c,v 1.3 2001/05/20 16:44:47 jongfoster * Purpose : Declares functions to parse/crunch headers and pages. * Functions declared include: * `acl_addr', `add_stats', `block_acl', `block_imageurl', - * `block_url', `url_permissions', `domaincmp', `dsplit', + * `block_url', `url_actions', `domaincmp', `dsplit', * `filter_popups', `forward_url', 'redirect_url', * `ij_untrusted_url', `intercept_url', `re_process_buffer', * `show_proxy_args', 'ijb_send_banner', and `trust_url' @@ -38,6 +38,177 @@ const char filters_rcs[] = "$Id: filters.c,v 1.3 2001/05/20 16:44:47 jongfoster * * Revisions : * $Log: filters.c,v $ + * Revision 1.15 2001/06/03 19:12:00 oes + * extracted-CGI relevant stuff + * + * Revision 1.15 2001/06/03 11:03:48 oes + * Makefile/in + * + * introduced cgi.c + * + * actions.c: + * + * adapted to new enlist_unique arg format + * + * conf loadcfg.c + * + * introduced confdir option + * + * filters.c filtrers.h + * + * extracted-CGI relevant stuff + * + * jbsockets.c + * + * filled comment + * + * jcc.c + * + * support for new cgi mechansim + * + * list.c list.h + * + * functions for new list type: "map" + * extended enlist_unique + * + * miscutil.c .h + * introduced bindup() + * + * parsers.c parsers.h + * + * deleted const struct interceptors + * + * pcrs.c + * added FIXME + * + * project.h + * + * added struct map + * added struct http_response + * changes struct interceptors to struct cgi_dispatcher + * moved HTML stuff to cgi.h + * + * re_filterfile: + * + * changed + * + * showargs.c + * NO TIME LEFT + * + * 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 + * + * Revision 1.4 2001/05/22 18:46:04 oes + * + * - Enabled filtering banners by size rather than URL + * by adding patterns that replace all standard banner + * sizes with the "Junkbuster" gif to the re_filterfile + * + * - Enabled filtering WebBugs by providing a pattern + * which kills all 1x1 images + * + * - Added support for PCRE_UNGREEDY behaviour to pcrs, + * which is selected by the (nonstandard and therefore + * capital) letter 'U' in the option string. + * It causes the quantifiers to be ungreedy by default. + * Appending a ? turns back to greedy (!). + * + * - Added a new interceptor ijb-send-banner, which + * sends back the "Junkbuster" gif. Without imagelist or + * MSIE detection support, or if tinygif = 1, or the + * URL isn't recognized as an imageurl, a lame HTML + * explanation is sent instead. + * + * - Added new feature, which permits blocking remote + * script redirects and firing back a local redirect + * to the browser. + * The feature is conditionally compiled, i.e. it + * can be disabled with --disable-fast-redirects, + * plus it must be activated by a "fast-redirects" + * line in the config file, has its own log level + * and of course wants to be displayed by show-proxy-args + * Note: Boy, all the #ifdefs in 1001 locations and + * all the fumbling with configure.in and acconfig.h + * were *way* more work than the feature itself :-( + * + * - Because a generic redirect template was needed for + * this, tinygif = 3 now uses the same. + * + * - Moved GIFs, and other static HTTP response templates + * to project.h + * + * - Some minor fixes + * + * - Removed some >400 CRs again (Jon, you really worked + * a lot! ;-) + * * Revision 1.3 2001/05/20 16:44:47 jongfoster * Removing last hardcoded JunkBusters.com URLs. * @@ -87,6 +258,9 @@ const char filters_rcs[] = "$Id: filters.c,v 1.3 2001/05/20 16:44:47 jongfoster #include "jbsockets.h" #include "errlog.h" #include "jbsockets.h" +#include "miscutil.h" +#include "actions.h" +#include "cgi.h" #ifdef _WIN32 #include "win32.h" @@ -104,60 +278,6 @@ 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 /********************************************************************* * @@ -167,52 +287,52 @@ 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) + while (acl != NULL) { - *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))) + 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); @@ -224,11 +344,11 @@ 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. * @@ -311,71 +431,39 @@ int acl_addr(char *aspec, struct access_control_addr *aca) *********************************************************************/ char *block_url(struct http_request *http, struct client_state *csp) { - struct file_list *fl; - struct block_spec *b; - struct url_spec url[1]; char *p; int n; + int factor = 2; - if (((fl = csp->blist) == NULL) || ((b = fl->f) == NULL)) + if ((csp->action->flags & ACTION_BLOCK) == 0) { return(NULL); } - - *url = dsplit(http->host); - - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(NULL); - - for (b = b->next; b ; b = b->next) + else { - if ((b->url->port == 0) || (b->url->port == 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) -#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); + factor++; #endif /* def FORCE_LOAD */ - p = (char *)malloc(n); + n = strlen(CBLOCK); + n += factor * strlen(http->hostport); + n += factor * strlen(http->path); + + p = (char *)malloc(n); #ifdef FORCE_LOAD - sprintf(p, CBLOCK, http->hostport, http->path, b->url->spec, http->hostport, http->path); + sprintf(p, CBLOCK, http->hostport, http->path, http->hostport, http->path, + http->hostport, http->path); #else - sprintf(p, CBLOCK, http->hostport, http->path, b->url->spec); + sprintf(p, CBLOCK, http->hostport, http->path, http->hostport, http->path); #endif /* def FORCE_LOAD */ - return(p); - } - } - } + return(p); } - freez(url->dbuf); - freez(url->dvec); - return(NULL); - } -#if defined(DETECT_MSIE_IMAGES) || defined(USE_IMAGE_LIST) +#ifdef IMAGE_BLOCKING /********************************************************************* * * Function : block_imageurl @@ -408,150 +496,9 @@ int block_imageurl(struct http_request *http, struct client_state *csp) } #endif -#if defined(USE_IMAGE_LIST) - return block_imageurl_using_imagelist(http, csp); -#else - /* Don't know - assume HTML */ - return 0; -#endif + return ((csp->action->flags & ACTION_IMAGE) != 0); } -#endif /* defined(DETECT_MSIE_IMAGES) || defined(USE_IMAGE_LIST) */ - - -#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 (((fl = csp->ilist) == NULL) || ((b = fl->f) == NULL)) - { - return(0); - } - - *url = dsplit(http->host); - - /* if splitting the domain fails, punt */ - if (url->dbuf == NULL) return(0); - - for (b = b->next; b ; b = b->next) - { - - if ((b->url->port == 0) || (b->url->port == http->port)) - { - /* 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); - - if (b->reject == 0) return(0); - - return(1); - } - } - } - } - freez(url->dbuf); - freez(url->dvec); - return(0); - -} -#endif /* def USE_IMAGE_LIST */ - - -#ifdef PCRS -/********************************************************************* - * - * Function : re_process_buffer - * - * 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). - * - * Parameters : - * 1 : csp = Current client state (buffers, headers, etc...) - * - * Returns : N/A - * - *********************************************************************/ -void re_process_buffer(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 file_list *fl; - struct re_filterfile_spec *b; - - /* Sanity first ;-) */ - if (size <= 0) - { - return; - } - - if ( ( NULL == (fl = csp->rlist) ) || ( NULL == (b = fl->f) ) ) - { - log_error(LOG_LEVEL_ERROR, "Unable to get current state of regexp filtering."); - return; - } - - joblist = b->joblist; - - - 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 = joblist; NULL != job; job = job->next) - { - hits += pcrs_exec_substitution(job, old, size, &new, &size); - if (old != csp->iob->cur) free(old); - old=new; - } - - log_error(LOG_LEVEL_RE_FILTER, " produced %d hits (new size %d).", hits, size); - - if (write_socket(csp->cfd, old, size) != size) - { - log_error(LOG_LEVEL_ERROR, "write to client failed."); - } - - /* fwiw, reset the iob */ - IOB_RESET(csp); - freez(new); - return; - -} -#endif /* def PCRS */ +#endif /* def IMAGE_BLOCKING */ #ifdef TRUST_FILES @@ -673,7 +620,7 @@ char *trust_url(struct http_request *http, struct client_state *csp) /* if splitting the domain fails, punt */ if (url->dbuf == NULL) goto trust_url_not_trusted; - for (tl = trust_list; (t = *tl) ; tl++) + for (tl = csp->config->trust_list; (t = *tl) ; tl++) { if ((t->port == 0) || (t->port == rhttp->port)) { @@ -697,7 +644,7 @@ char *trust_url(struct http_request *http, struct client_state *csp) freez(url->dbuf); freez(url->dvec); - if ((fp = fopen(trustfile, "a"))) + if ((fp = fopen(csp->config->trustfile, "a"))) { h = NULL; @@ -764,118 +711,124 @@ trust_url_not_trusted: #endif /* def TRUST_FILES */ +#ifdef PCRS /********************************************************************* * - * Function : intercept_url + * Function : re_process_buffer * - * 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 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. * * 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. + * * *********************************************************************/ -char *intercept_url(struct http_request *http, struct client_state *csp) +char *re_process_buffer(struct client_state *csp) { - char *basename; - const struct interceptors *v; + int hits=0; + int size = csp->iob->eod - csp->iob->cur; + char *old=csp->iob->cur, *new = NULL; + pcrs_job *job, *joblist; + + struct file_list *fl; + struct re_filterfile_spec *b; + + /* Sanity first ;-) */ + if (size <= 0) + { + return(strdup("")); + } - basename = strrchr(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(strdup("")); + } - if (basename == NULL) return(NULL); + joblist = b->joblist; - basename ++; /* first char past the last slash */ - if (*basename) + 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 = joblist; NULL != job; job = job->next) { - for (v = intercept_patterns; v->str; v++) - { - if (strncmp(basename, v->str, v->len) == 0) - { - return((v->interceptor)(http, csp)); - } - } + 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); + + csp->content_length = size; + + /* fwiw, reset the iob */ + IOB_RESET(csp); + return(new); } +#endif /* def PCRS */ + -#ifdef FAST_REDIRECTS /********************************************************************* * - * Function : redirect_url + * Function : url_actions * - * Description : Checks for redirection URLs and returns a HTTP redirect - * to the destination URL. + * Description : Gets the actions for this URL. * * Parameters : - * 1 : http = http_request request, check `basename's of blocklist + * 1 : http = http_request request for blocked URLs * 2 : csp = Current client state (buffers, headers, etc...) * - * Returns : NULL if URL was clean, HTTP redirect otherwise. + * Returns : N/A * *********************************************************************/ -char *redirect_url(struct http_request *http, struct client_state *csp) +void url_actions(struct http_request *http, + struct client_state *csp) { - char *p, *q; - - 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://")) - { - q = p++; - } + struct file_list *fl; + struct url_actions *b; - /* if there was any, generate and return a HTTP redirect */ - if (q != csp->http->path) - { - log_error(LOG_LEVEL_REDIRECTS, "redirecting to: %s", q); + init_current_action(csp->action); - p = (char *)malloc(strlen(HTTP_REDIRECT_TEMPLATE) + strlen(q)); - sprintf(p, HTTP_REDIRECT_TEMPLATE, q); - return(p); - } - else + if (((fl = csp->actions_list) == NULL) || ((b = fl->f) == NULL)) { - return(NULL); - } + return; + } + apply_url_actions(csp->action, http, b); } -#endif /* def FAST_REDIRECTS */ + /********************************************************************* * - * Function : url_permissions + * Function : apply_url_actions * - * Description : Gets the permissions for this URL. + * Description : Applies a list of URL actions. * * Parameters : - * 1 : http = http_request request for blocked URLs - * 2 : csp = Current client state (buffers, headers, etc...) + * 1 : action = Destination. + * 2 : http = Current URL + * 3 : b = list of URL actions to apply * - * 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 apply_url_actions(struct current_action_spec *action, + struct http_request *http, + struct url_actions *b) { - struct file_list *fl; - struct permissions_spec *b; struct url_spec url[1]; - if (((fl = csp->permissions_list) == NULL) || ((b = fl->f) == NULL)) + if (b == NULL) { - return(default_permissions); + /* Should never happen */ + return; } *url = dsplit(http->host); @@ -883,7 +836,7 @@ int url_permissions(struct http_request *http, struct client_state *csp) /* if splitting the domain fails, punt */ if (url->dbuf == NULL) { - return(default_permissions); + return; } for (b = b->next; NULL != b; b = b->next) @@ -900,9 +853,7 @@ 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); } } } @@ -910,8 +861,6 @@ int url_permissions(struct http_request *http, struct client_state *csp) freez(url->dbuf); freez(url->dvec); - return(default_permissions); - } @@ -925,51 +874,55 @@ 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] = { 0 }; /* All zeroes */ + 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); } @@ -997,13 +950,14 @@ struct url_spec dsplit(char *domain) 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); @@ -1027,6 +981,7 @@ struct url_spec dsplit(char *domain) memcpy(ret->dvec, v, size); } + return(*ret); } @@ -1034,550 +989,120 @@ 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 trivialmatch(). + * 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 + * trivialmatch(). * * 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'; + char **pv, **fv; /* vectors */ + int plen, flen; + int unanchored = pattern->unanchored & (ANCHOR_RIGHT | ANCHOR_LEFT); + plen = pattern->dcnt; + flen = fqdn->dcnt; - if (query_string != NULL) + if (flen < plen) { - /* first char past the last '?' (maybe '\0')*/ - which_file = query_string[1]; + /* fqdn is too short to match this pattern */ + return 1; } - switch (which_file) - { - case 'b': - if (csp->blist) - { - filename = csp->blist->filename; - file_description = "Block List"; - } - break; - 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 */ + pv = pattern->dvec; + fv = fqdn->dvec; -#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 (unanchored == ANCHOR_LEFT) + { + /* + * 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 (filename) + else if (unanchored == 0) { - /* 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)
+      /* Fully anchored, check length */
+      if (flen != plen)
       {
-         s = strsav(s, "

ERROR OPENING FILE!

");
+         return 1;
       }
-      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, proxy_args->header); - s = strsav(s, proxy_args->invocation); -#ifdef STATISTICS - s = add_stats(s); -#endif /* def STATISTICS */ - s = strsav(s, 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, proxy_args->trailer); - - return(s); - -} - - -/********************************************************************* - * - * 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 - * - * 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) - { - 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 = trust_list; (t = *tl) ; tl++) - { - sprintf(buf, "%s
\n", t->spec); - p = strsav(p, buf); - } - - if (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 = 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"); + return 1; } - - 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(s); -} -#endif /* def STATISTICS */ /*