Mercurial > dnsbl

comparison src/dnsbl.cpp @ 119:d9d2f8699621

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
uribl patch from Jeff Evans <jeffe@tricab.com>
author carl
date Sun, 12 Mar 2006 12:38:43 -0800
parents aa07452e641b
children 1d9e6c1b8872
comparison
equal deleted inserted replaced
118:13fcb0c66763 119:d9d2f8699621
230 only_whites = true; 230 only_whites = true;
231 memory = NULL; 231 memory = NULL;
232 scanner = NULL; 232 scanner = NULL;
233 content_suffix = NULL; 233 content_suffix = NULL;
234 content_message = NULL; 234 content_message = NULL;
235 uribl_suffix = NULL;
236 uribl_message = NULL;
235 content_host_ignore = NULL; 237 content_host_ignore = NULL;
236 } 238 }
237 239
238 mlfiPriv::~mlfiPriv() { 240 mlfiPriv::~mlfiPriv() {
239 return_fd(); 241 return_fd();
257 only_whites = true; 259 only_whites = true;
258 memory = NULL; 260 memory = NULL;
259 scanner = NULL; 261 scanner = NULL;
260 content_suffix = NULL; 262 content_suffix = NULL;
261 content_message = NULL; 263 content_message = NULL;
264 uribl_suffix = NULL;
265 uribl_message = NULL;
262 content_host_ignore = NULL; 266 content_host_ignore = NULL;
263 } 267 }
264 } 268 }
265 269
266 void mlfiPriv::get_fd() { 270 void mlfiPriv::get_fd() {
367 // the content filtering parameters 371 // the content filtering parameters
368 memory = new recorder(this, con.get_html_tags(), con.get_content_tlds(), con.get_content_cctlds()); 372 memory = new recorder(this, con.get_html_tags(), con.get_content_tlds(), con.get_content_cctlds());
369 scanner = new url_scanner(memory); 373 scanner = new url_scanner(memory);
370 content_suffix = con.get_content_suffix(); 374 content_suffix = con.get_content_suffix();
371 content_message = con.get_content_message(); 375 content_message = con.get_content_message();
376 uribl_suffix = con.get_uribl_suffix();
377 uribl_message = con.get_uribl_message();
372 content_host_ignore = &con.get_content_host_ignore(); 378 content_host_ignore = &con.get_content_host_ignore();
373 } 379 }
374 } 380 }
375 381
376 #define MLFIPRIV ((struct mlfiPriv *) smfi_getpriv(ctx)) 382 #define MLFIPRIV ((struct mlfiPriv *) smfi_getpriv(ctx))
667 *top = '\0'; 673 *top = '\0';
668 char *x = strrchr(hostname, '.'); 674 char *x = strrchr(hostname, '.');
669 if (x) hostname = x+1; 675 if (x) hostname = x+1;
670 *top = '.'; 676 *top = '.';
671 } 677 }
672 for (int i=0; i<2; i++) { 678 snprintf(buf, sizeof(buf), "%s.%s", hostname, priv.uribl_suffix);
673 snprintf(buf, sizeof(buf), "%s.%s", hostname, uriblname[i]); 679 if (dns_interface(priv, buf, false, NULL)) {
674 if (debug_syslog > 2) { 680 if (debug_syslog > 2) {
675 char tmp[maxlen]; 681 char tmp[maxlen];
676 snprintf(tmp, sizeof(tmp), "Looking up %s on %s", hostname, uriblname[i]); 682 snprintf(tmp, sizeof(tmp), "found %s on %s", hostname, priv.uribl_suffix);
677 my_syslog(tmp); 683 my_syslog(tmp);
678 } 684 }
679 if (dns_interface(priv, buf, false, NULL)) return true; 685 return true;
680 } 686 }
681 return false; 687 return false;
682 } 688 }
683 689
684 690
725 return false; 731 return false;
726 } 732 }
727 733
728 734
729 //////////////////////////////////////////////// 735 ////////////////////////////////////////////////
730 // check the hosts from the body against the content dnsbl 736 // check the hosts from the body against the content filter and uribl dnsbls
731 // 737 //
732 bool check_hosts(mlfiPriv &priv, bool random, int limit, char *&host, int &ip); 738 bool check_hosts(mlfiPriv &priv, bool random, int limit, char *&msg, char *&host, int &ip);
733 bool check_hosts(mlfiPriv &priv, bool random, int limit, char *&host, int &ip) { 739 bool check_hosts(mlfiPriv &priv, bool random, int limit, char *&msg, char *&host, int &ip) {
740 if (!priv.content_suffix && !priv.uribl_suffix) return false; // nothing to check
734 CONFIG &dc = *priv.pc; 741 CONFIG &dc = *priv.pc;
735 string_set &hosts = priv.memory->get_hosts(); 742 string_set &hosts = priv.memory->get_hosts();
736 string_set &ignore = *priv.content_host_ignore; 743 string_set &ignore = *priv.content_host_ignore;
737 char *suffix = priv.content_suffix;
738 744
739 int count = 0; 745 int count = 0;
740 int cnt = hosts.size(); // number of hosts we could look at 746 int cnt = hosts.size(); // number of hosts we could look at
741 int_set ips; 747 int_set ips;
742 ns_map nameservers; 748 ns_map nameservers;
777 if (ip) { 783 if (ip) {
778 int_set::iterator i = ips.find(ip); 784 int_set::iterator i = ips.find(ip);
779 if (i == ips.end()) { 785 if (i == ips.end()) {
780 // we haven't looked this up yet 786 // we haven't looked this up yet
781 ips.insert(ip); 787 ips.insert(ip);
782 if (check_single(priv, ip, suffix)) return true; 788 if (check_single(priv, ip, priv.content_suffix)) {
789 msg = priv.content_message;
790 return true;
791 }
783 // Check uribl & surbl 792 // Check uribl & surbl
784 if (check_uribl(priv, host)) return true; 793 if (check_uribl(priv, host)) {
794 msg = priv.uribl_message;
795 return true;
796 }
785 } 797 }
786 } 798 }
787 } 799 }
788 limit *= 4; // allow average of 3 ns per host name 800 limit *= 4; // allow average of 3 ns per host name
789 for (ns_mapper::iterator i=nameservers.ns_ip.begin(); i!=nameservers.ns_ip.end(); i++) { 801 for (ns_mapper::iterator i=nameservers.ns_ip.begin(); i!=nameservers.ns_ip.end(); i++) {
790 count++; 802 count++;
791 if ((count > limit) && (limit > 0)) { 803 if ((count > limit) && (limit > 0)) return false; // too many name servers to check them all
792 if (random) continue; // don't complain
793 return true;
794 }
795 host = (*i).first; // a transient reference that needs to be replaced before we return it 804 host = (*i).first; // a transient reference that needs to be replaced before we return it
796 ip = (*i).second; 805 ip = (*i).second;
797 if (!ip) ip = dns_interface(priv, host, false, NULL); 806 if (!ip) ip = dns_interface(priv, host, false, NULL);
798 if (debug_syslog > 2) { 807 if (debug_syslog > 2) {
799 char buf[maxlen]; 808 char buf[maxlen];
810 } 819 }
811 if (ip) { 820 if (ip) {
812 int_set::iterator i = ips.find(ip); 821 int_set::iterator i = ips.find(ip);
813 if (i == ips.end()) { 822 if (i == ips.end()) {
814 ips.insert(ip); 823 ips.insert(ip);
815 if (check_single(priv, ip, suffix)) { 824 if (check_single(priv, ip, priv.content_suffix)) {
825 msg = priv.content_message;
816 string_map::iterator j = nameservers.ns_host.find(host); 826 string_map::iterator j = nameservers.ns_host.find(host);
817 if (j != nameservers.ns_host.end()) { 827 if (j != nameservers.ns_host.end()) {
818 char *refer = (*j).second; 828 char *refer = (*j).second;
819 char buf[maxlen]; 829 char buf[maxlen];
820 snprintf(buf, sizeof(buf), "%s with nameserver %s", refer, host); 830 snprintf(buf, sizeof(buf), "%s with nameserver %s", refer, host);
983 limit = max(limit, con.get_host_limit()); 993 limit = max(limit, con.get_host_limit());
984 } 994 }
985 } 995 }
986 bool rejecting = alive.empty(); // if alive is empty, we must have set msg above in acceptable_content() 996 bool rejecting = alive.empty(); // if alive is empty, we must have set msg above in acceptable_content()
987 if (!rejecting) { 997 if (!rejecting) {
988 if (check_hosts(priv, random, limit, host, ip)) { 998 char *msg;
999 if (check_hosts(priv, random, limit, msg, host, ip)) {
989 char adr[sizeof "255.255.255.255"]; 1000 char adr[sizeof "255.255.255.255"];
990 adr[0] = '\0'; 1001 adr[0] = '\0';
991 inet_ntop(AF_INET, (const u_char *)&ip, adr, sizeof(adr)); 1002 inet_ntop(AF_INET, (const u_char *)&ip, adr, sizeof(adr));
992 snprintf(buf, sizeof(buf), priv.content_message, host, adr); 1003 snprintf(buf, sizeof(buf), msg, host, adr);
993 msg = buf; 1004 msg = buf;
994 rejecting = true; 1005 rejecting = true;
995 } 1006 }
996 } 1007 }
997 if (!rejecting) { 1008 if (!rejecting) {