3 #define PROJECT_H_VERSION "$Id: project.h,v 1.11 2001/06/01 10:32:47 oes Exp $"
4 /*********************************************************************
6 * File : $Source: /cvsroot/ijbswa/current/project.h,v $
8 * Purpose : Defines data structures which are widely used in the
9 * project. Does not define any variables or functions
10 * (though it does declare some macros).
12 * Copyright : Written by and Copyright (C) 2001 the SourceForge
13 * IJBSWA team. http://ijbswa.sourceforge.net
15 * Based on the Internet Junkbuster originally written
16 * by and Copyright (C) 1997 Anonymous Coders and
17 * Junkbusters Corporation. http://www.junkbusters.com
19 * This program is free software; you can redistribute it
20 * and/or modify it under the terms of the GNU General
21 * Public License as published by the Free Software
22 * Foundation; either version 2 of the License, or (at
23 * your option) any later version.
25 * This program is distributed in the hope that it will
26 * be useful, but WITHOUT ANY WARRANTY; without even the
27 * implied warranty of MERCHANTABILITY or FITNESS FOR A
28 * PARTICULAR PURPOSE. See the GNU General Public
29 * License for more details.
31 * The GNU General Public License should be included with
32 * this file. If not, you can view it at
33 * http://www.gnu.org/copyleft/gpl.html
34 * or write to the Free Software Foundation, Inc., 59
35 * Temple Place - Suite 330, Boston, MA 02111-1307, USA.
39 * Revision 1.11 2001/06/01 10:32:47 oes
40 * Added constants for anchoring selection bitmap
42 * Revision 1.10 2001/05/31 21:33:53 jongfoster
43 * Changes for new actions file, replacing permissionsfile
44 * and parts of the config file. Also added support for
47 * Revision 1.9 2001/05/31 17:32:31 oes
49 * - Enhanced domain part globbing with infix and prefix asterisk
50 * matching and optional unanchored operation
52 * Revision 1.8 2001/05/29 20:09:15 joergs
53 * HTTP_REDIRECT_TEMPLATE fixed.
55 * Revision 1.7 2001/05/29 09:50:24 jongfoster
56 * Unified blocklist/imagelist/actionslist.
57 * File format is still under discussion, but the internal changes
60 * Also modified interceptor behaviour:
61 * - We now intercept all URLs beginning with one of the following
62 * prefixes (and *only* these prefixes):
64 * * http://ijbswa.sf.net/config/
65 * * http://ijbswa.sourceforge.net/config/
66 * - New interceptors "home page" - go to http://i.j.b/ to see it.
67 * - Internal changes so that intercepted and fast redirect pages
68 * are not replaced with an image.
69 * - Interceptors now have the option to send a binary page direct
70 * to the client. (i.e. ijb-send-banner uses this)
71 * - Implemented show-url-info interceptor. (Which is why I needed
72 * the above interceptors changes - a typical URL is
73 * "http://i.j.b/show-url-info?url=www.somesite.com/banner.gif".
74 * The previous mechanism would not have intercepted that, and
75 * if it had been intercepted then it then it would have replaced
78 * Revision 1.6 2001/05/27 22:17:04 oes
80 * - re_process_buffer no longer writes the modified buffer
81 * to the client, which was very ugly. It now returns the
82 * buffer, which it is then written by chat.
84 * - content_length now adjusts the Content-Length: header
85 * for modified documents rather than crunch()ing it.
86 * (Length info in csp->content_length, which is 0 for
87 * unmodified documents)
89 * - For this to work, sed() is called twice when filtering.
91 * Revision 1.5 2001/05/26 00:28:36 jongfoster
92 * Automatic reloading of config file.
93 * Removed obsolete SIGHUP support (Unix) and Reload menu option (Win32).
94 * Most of the global variables have been moved to a new
95 * struct configuration_spec, accessed through csp->config->globalname
96 * Most of the globals remaining are used by the Win32 GUI.
98 * Revision 1.4 2001/05/22 18:46:04 oes
100 * - Enabled filtering banners by size rather than URL
101 * by adding patterns that replace all standard banner
102 * sizes with the "Junkbuster" gif to the re_filterfile
104 * - Enabled filtering WebBugs by providing a pattern
105 * which kills all 1x1 images
107 * - Added support for PCRE_UNGREEDY behaviour to pcrs,
108 * which is selected by the (nonstandard and therefore
109 * capital) letter 'U' in the option string.
110 * It causes the quantifiers to be ungreedy by default.
111 * Appending a ? turns back to greedy (!).
113 * - Added a new interceptor ijb-send-banner, which
114 * sends back the "Junkbuster" gif. Without imagelist or
115 * MSIE detection support, or if tinygif = 1, or the
116 * URL isn't recognized as an imageurl, a lame HTML
117 * explanation is sent instead.
119 * - Added new feature, which permits blocking remote
120 * script redirects and firing back a local redirect
122 * The feature is conditionally compiled, i.e. it
123 * can be disabled with --disable-fast-redirects,
124 * plus it must be activated by a "fast-redirects"
125 * line in the config file, has its own log level
126 * and of course wants to be displayed by show-proxy-args
127 * Note: Boy, all the #ifdefs in 1001 locations and
128 * all the fumbling with configure.in and acconfig.h
129 * were *way* more work than the feature itself :-(
131 * - Because a generic redirect template was needed for
132 * this, tinygif = 3 now uses the same.
134 * - Moved GIFs, and other static HTTP response templates
139 * - Removed some >400 CRs again (Jon, you really worked
142 * Revision 1.3 2001/05/20 01:21:20 jongfoster
143 * Version 2.9.4 checkin.
144 * - Merged popupfile and cookiefile, and added control over PCRS
145 * filtering, in new "actionsfile".
146 * - Implemented LOG_LEVEL_FATAL, so that if there is a configuration
147 * file error you now get a message box (in the Win32 GUI) rather
148 * than the program exiting with no explanation.
149 * - Made killpopup use the PCRS MIME-type checking and HTTP-header
151 * - Removed tabs from "config"
152 * - Moved duplicated url parsing code in "loaders.c" to a new funcition.
153 * - Bumped up version number.
155 * Revision 1.2 2001/05/17 23:01:01 oes
156 * - Cleaned CRLF's from the sources and related files
158 * Revision 1.1.1.1 2001/05/15 13:59:03 oes
159 * Initial import of version 2.9.3 source tree
162 *********************************************************************/
165 /* Declare struct FILE for vars and funcs. */
168 /* Need time_t for file_list */
172 * Include appropriate regular expression libraries.
174 * PCRS ==> Include pcre
175 * REGEX && PCRE ==> Include pcre and pcreposix
176 * REGEX && !PCRE ==> Include gnu_regex
178 * STATIC ==> Use #include "pcre.h" (compiling at same time)
179 * !STATIC ==> Use #include <pcre.h> (System library)
182 #if (defined(REGEX) && defined(PCRE)) || defined(PCRS)
188 #endif /* (defined(REGEX) && defined(PCRE)) || defined(PCRS) */
190 #if defined(REGEX) && defined(PCRE)
192 # include "pcreposix.h"
194 # include <pcreposix.h>
196 #endif /* defined(REGEX) && defined(PCRE) */
198 #if defined(REGEX) && !defined(PCRE)
199 # include "gnu_regex.h"
204 #endif /* def PCRS */
208 #endif /* def AMIGA */
216 /* Default IP and port to listen on */
217 #define HADDR_DEFAULT "127.0.0.1"
218 #define HADDR_PORT 8000
221 /* Need this for struct gateway */
224 /* Need this for struct client_state */
225 struct configuration_spec;
236 char *hostport; /* "host[:port]" */
242 /* generic attributes */
244 int (*conn)(const struct gateway *, struct http_request *, struct client_state *);
247 /* domain specific attributes */
256 /* Generic linked list of strings */
257 struct list /* FIXME: Why not separate entries and header? */
259 char * str; /* valid in an entry */
260 struct list *last; /* valid in header */
269 char *domain; /* fqdn */
271 char **dvec; /* Domain ptr vector */
272 int dcnt; /* How many domains in fqdn? */
273 int unanchored; /* bitmap 0: left, 1: right */
283 #define ANCHOR_LEFT 1
284 #define ANCHOR_RIGHT 2
296 #define IOB_PEEK(CSP) ((CSP->iob->cur > CSP->iob->eod) ? (CSP->iob->eod - CSP->iob->cur) : 0)
297 #define IOB_RESET(CSP) if(CSP->iob->buf) free(CSP->iob->buf); memset(CSP->iob, '\0', sizeof(CSP->iob));
301 #define ACTION_MASK_ALL (~0U)
303 #define ACTION_MOST_COMPATIBLE 0U
305 #define ACTION_BLOCK 0x0001U
306 #define ACTION_FAST_REDIRECTS 0x0002U
307 #define ACTION_FILTER 0x0004U
308 #define ACTION_HIDE_FORWARDED 0x0008U
309 #define ACTION_HIDE_FROM 0x0010U
310 #define ACTION_HIDE_REFERER 0x0020U /* sic - follow HTTP, not English */
311 #define ACTION_HIDE_USER_AGENT 0x0040U
312 #define ACTION_IMAGE 0x0080U
313 #define ACTION_NO_COOKIE_READ 0x0100U
314 #define ACTION_NO_COOKIE_SET 0x0200U
315 #define ACTION_NO_POPUPS 0x0400U
316 #define ACTION_VANILLA_WAFER 0x0800U
318 #define ACTION_STRING_FROM 0
319 #define ACTION_STRING_REFERER 1
320 #define ACTION_STRING_USER_AGENT 2
321 #define ACTION_STRING_COUNT 3
323 #define ACTION_MULTI_ADD_HEADER 0
324 #define ACTION_MULTI_WAFER 1
325 #define ACTION_MULTI_COUNT 2
328 struct current_action_spec
330 unsigned flags; /* a bit set to "1" = add action */
332 /* For those actions that require parameters: */
334 /* each entry is valid if & only if corresponding entry in "add" set. */
335 char * string[ACTION_STRING_COUNT];
338 struct list multi[ACTION_MULTI_COUNT][1];
343 unsigned mask; /* a bit set to "0" = remove action */
344 unsigned add; /* a bit set to "1" = add action */
346 /* For those actions that require parameters: */
348 /* each entry is valid if & only if corresponding entry in "add" set. */
349 char * string[ACTION_STRING_COUNT];
351 /* Strings to remove. */
352 struct list multi_remove[ACTION_MULTI_COUNT][1];
354 /* If nonzero, remove *all* strings. */
355 int multi_remove_all[ACTION_MULTI_COUNT];
358 struct list multi_add[ACTION_MULTI_COUNT][1];
363 struct url_spec url[1];
365 struct action_spec action[1];
367 struct url_actions * next;
371 /* Constants defining bitmask for csp->accept_types */
373 #ifdef DETECT_MSIE_IMAGES
375 /* MSIE detected by user-agent string */
376 #define ACCEPT_TYPE_IS_MSIE 0x0001
379 * *If* this is MSIE, it wants an image. (Or this is a shift-reload, or
380 * it's got an image from this URL before... yuck!)
381 * Only meaningful if ACCEPT_TYPE_IS_MSIE set
383 #define ACCEPT_TYPE_MSIE_IMAGE 0x0002
386 * *If* this is MSIE, it wants a HTML document.
387 * Only meaningful if ACCEPT_TYPE_IS_MSIE set
389 #define ACCEPT_TYPE_MSIE_HTML 0x0004
391 #endif /* def DETECT_MSIE_IMAGES */
396 /* The proxy's configuration */
397 struct configuration_spec * config;
400 /* The actions to perform on the current request */
401 struct current_action_spec action[1];
404 /* socket to talk to client (web browser) */
407 /* socket to talk to server (web server or proxy) */
412 /* 1 if this URL was rejected, 0 otherwise. Allows actual stats inc to
413 * occur in main thread only for thread-safety.
416 #endif /* def STATISTICS */
420 #endif /* def FORCE_LOAD */
424 #endif /* def TOGGLE */
427 * Client PC's IP address, as reported by the accept()_ function.
428 * Both as string and number
434 /* The referer in this request, if one was specified. */
436 #endif /* def TRUST_FILES */
438 #if defined(DETECT_MSIE_IMAGES)
439 /* Types the client will accept.
440 * Bitmask - see ACCEPT_TYPE_XXX constants.
443 #endif /* defined(DETECT_MSIE_IMAGES) */
445 /* The URL that was requested */
446 struct http_request http[1];
448 /* An I/O buffer used for buffering data read from the client */
451 /* List of all headers for this request */
452 struct list headers[1];
454 /* List of all cookies for this request */
455 struct list cookie_list[1];
457 #if defined(PCRS) || defined(KILLPOPUPS)
458 /* Nonzero if this has a text MIME type */
460 #endif /* defined(PCRS) || defined(KILLPOPUPS) */
462 /* The "X-Forwarded-For:" header sent by the client */
466 * Nonzero if this client is processing data.
467 * Set to zero when the thread associated with this structure dies.
471 /* files associated with this client */
472 struct file_list *flist; /* forwardfile */
473 struct file_list *actions_list;
477 struct file_list *alist; /* aclfile */
478 #endif /* def ACL_FILES */
481 struct file_list *rlist; /* Perl re_filterfile */
482 size_t content_length; /* Length after processing */
483 #endif /* def PCRS */
486 struct file_list *tlist; /* trustfile */
487 #endif /* def TRUST_FILES */
489 struct client_state *next;
497 char *(*parser)(const struct parsers *, char *, struct client_state *);
505 char *(*interceptor)(struct http_request *http, struct client_state *csp);
512 * this is a pointer to the data structures associated with the file.
513 * Read-only once the structure has been created.
517 /* Normally NULL. When we are finished with file (i.e. when we have
518 * loaded a new one), set to a pointer to an unloader function.
519 * Unloader will be called by sweep() (called from main loop) when
520 * all clients using this file are done. This prevents threading
523 void (*unloader)(void *);
525 /* Used internally by sweep(). Do not access from elsewhere. */
528 #ifndef SPLIT_PROXY_ARGS
529 /* String to be displayed as part of show-proxy-args display.
530 * Read-only once the structure has been created.
533 #endif /* ndef SPLIT_PROXY_ARGS */
535 /* Following variables allow us to check if file has been changed.
536 * Read-only once the structure has been created.
541 /* Pointer to next entry in the linked list of all "file_list"s.
542 * This linked list is so that sweep() can navigate it.
543 * Since sweep() can remove items from the list, we must be careful
544 * to only access this value from main thread (when we know sweep
547 struct file_list *next;
554 struct url_spec url[1];
556 struct block_spec *next;
558 #endif /* def TRUST_FILES */
563 struct url_spec url[1];
564 struct gateway gw[1];
565 struct forward_spec *next;
570 struct re_filterfile_spec
572 struct list patterns[1];
573 /* See README.re_filter */
576 #endif /* def PCRS */
580 #define ACL_PERMIT 1 /* accept connection request */
581 #define ACL_DENY 2 /* reject connection request */
583 struct access_control_addr
590 struct access_control_list
592 struct access_control_addr src[1];
593 struct access_control_addr dst[1];
596 struct access_control_list *next;
598 #endif /* def ACL_FILES */
601 /* Maximum number of loaders (actions, block, forward, acl...) */
605 * Data loaded from the configuration file.
607 * (Anomaly: toggle is still handled through a global, not this structure)
609 struct configuration_spec
614 #ifdef IMAGE_BLOCKING
615 int tinygif; /* FIXME Should be an action */
616 const char *tinygifurl; /* FIXME Should be an action */
617 #endif /* def IMAGE_BLOCKING */
621 const char *actions_file;
622 const char *forwardfile;
626 #endif /* def ACL_FILES */
629 const char *re_filterfile;
630 #endif /* def PCRS */
633 const char * jarfile;
635 #endif /* def JAR_FILES */
638 * Port and IP to bind to.
639 * Defaults to HADDR_DEFAULT:HADDR_PORT == 127.0.0.1:8000
644 #ifndef SPLIT_PROXY_ARGS
645 const char *suppress_message;
646 #endif /* ndef SPLIT_PROXY_ARGS */
648 #ifndef SPLIT_PROXY_ARGS
649 /* suppress listing config files */
650 int suppress_blocklists;
651 #endif /* ndef SPLIT_PROXY_ARGS */
654 const char * trustfile;
656 struct list trust_info[1];
657 struct url_spec *trust_list[64];
658 #endif /* def TRUST_FILES */
660 /* Various strings for show-proxy-args */
661 char *proxy_args_header;
662 char *proxy_args_invocation;
663 char *proxy_args_gateways;
664 char *proxy_args_trailer;
666 /* the configuration file object. */
667 struct file_list *config_file_list;
669 /* List of loaders */
670 int (*loaders[NLOADERS])(struct client_state *);
672 /* bool, nonzero if we need to bind() to the new port */
677 #define SZ(X) (sizeof(X) / sizeof(*X))
679 #define WHITEBG "<body bgcolor=\"#ffffff\" link=\"#000078\" alink=\"#ff0022\" vlink=\"#787878\">\n"
680 #define BODY "<body bgcolor=\"#f8f8f0\" link=\"#000078\" alink=\"#ff0022\" vlink=\"#787878\">\n"
681 #define BANNER "<strong>Internet J<small>UNK<i><font color=\"red\">BUSTER</font></i></small></strong>"
684 #define FORCE_PREFIX "/IJB-FORCE-LOAD"
685 #endif /* def FORCE_LOAD */
687 #define HOME_PAGE_URL "http://ijbswa.sourceforge.net/"
688 #define REDIRECT_URL HOME_PAGE_URL "redirect.php?v=" VERSION "&to="
690 static const char CFAIL[] =
691 "HTTP/1.0 503 Connect failed\n"
692 "Content-Type: text/html\n\n"
695 "<title>Internet Junkbuster: Connect failed</title>\n"
701 "TCP connection to '%s' failed: %s.\n<br>"
705 static const char CNXDOM[] =
706 "HTTP/1.0 404 Non-existent domain\n"
707 "Content-Type: text/html\n\n"
710 "<title>Internet Junkbuster: Non-existent domain</title>\n"
716 "No such domain: %s\n"
720 static const char CNOBANNER[] =
721 "HTTP/1.0 200 No Banner\n"
722 "Content-Type: text/html\n\n"
725 "<title>Internet Junkbuster: No Banner</title>\n"
731 "You asked for a banner that this proxy can't produce because either configuration does not permit.\n<br>"
732 "or the URL didn't end with .gif\n"
736 static const char CSUCCEED[] =
737 "HTTP/1.0 200 Connection established\n"
738 "Proxy-Agent: IJ/" VERSION "\n\n";
740 static const char CHEADER[] =
741 "HTTP/1.0 400 Invalid header received from browser\n\n";
743 static const char SHEADER[] =
744 "HTTP/1.0 502 Invalid header received from server\n\n";
746 #ifdef IMAGE_BLOCKING
749 * Hint: You can encode your own GIFs like that:
750 * perl -e 'while (read STDIN, $c, 1) { printf("\\%.3o,", unpack("C", $c)); }'
753 static const char BLANKGIF[] =
754 "HTTP/1.0 200 OK\r\n"
755 "Pragma: no-cache\r\n"
756 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
757 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
758 "Content-type: image/gif\r\n\r\n"
759 "GIF89a\001\000\001\000\200\000\000\377\377\377\000\000"
760 "\000!\371\004\001\000\000\000\000,\000\000\000\000\001"
761 "\000\001\000\000\002\002D\001\000;";
763 static const char JBGIF[] =
764 "HTTP/1.0 200 OK\r\n"
765 "Pragma: no-cache\r\n"
766 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
767 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
768 "Content-type: image/gif\r\n\r\n"
769 "GIF89aD\000\013\000\360\000\000\000\000\000\377\377\377!"
770 "\371\004\001\000\000\001\000,\000\000\000\000D\000\013\000"
771 "\000\002a\214\217\251\313\355\277\000\200G&K\025\316hC\037"
772 "\200\234\230Y\2309\235S\230\266\206\372J\253<\3131\253\271"
773 "\270\215\342\254\013\203\371\202\264\334P\207\332\020o\266"
774 "N\215I\332=\211\312\3513\266:\026AK)\364\370\365aobr\305"
775 "\372\003S\275\274k2\354\254z\347?\335\274x\306^9\374\276"
778 #endif /* def IMAGE_BLOCKING */
780 #if defined(FAST_REDIRECTS) || defined(IMAGE_BLOCKING)
782 static const char HTTP_REDIRECT_TEMPLATE[] =
783 "HTTP/1.0 302 Local Redirect from Junkbuster\r\n"
784 "Pragma: no-cache\r\n"
785 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
786 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
787 "Location: %s\r\n\r\n";
789 #endif /* defined(FAST_REDIRECTS) || defined(IMAGE_BLOCKING) */
795 #endif /* ndef _PROJECT_H */