3 #define PROJECT_H_VERSION "$Id: project.h,v 1.9 2001/05/31 17:32:31 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 1.9 2001/05/31 17:32:31 oes
41 * - Enhanced domain part globbing with infix and prefix asterisk
42 * matching and optional unanchored operation
44 * Revision 1.8 2001/05/29 20:09:15 joergs
45 * HTTP_REDIRECT_TEMPLATE fixed.
47 * Revision 1.7 2001/05/29 09:50:24 jongfoster
48 * Unified blocklist/imagelist/actionslist.
49 * File format is still under discussion, but the internal changes
52 * Also modified interceptor behaviour:
53 * - We now intercept all URLs beginning with one of the following
54 * prefixes (and *only* these prefixes):
56 * * http://ijbswa.sf.net/config/
57 * * http://ijbswa.sourceforge.net/config/
58 * - New interceptors "home page" - go to http://i.j.b/ to see it.
59 * - Internal changes so that intercepted and fast redirect pages
60 * are not replaced with an image.
61 * - Interceptors now have the option to send a binary page direct
62 * to the client. (i.e. ijb-send-banner uses this)
63 * - Implemented show-url-info interceptor. (Which is why I needed
64 * the above interceptors changes - a typical URL is
65 * "http://i.j.b/show-url-info?url=www.somesite.com/banner.gif".
66 * The previous mechanism would not have intercepted that, and
67 * if it had been intercepted then it then it would have replaced
70 * Revision 1.6 2001/05/27 22:17:04 oes
72 * - re_process_buffer no longer writes the modified buffer
73 * to the client, which was very ugly. It now returns the
74 * buffer, which it is then written by chat.
76 * - content_length now adjusts the Content-Length: header
77 * for modified documents rather than crunch()ing it.
78 * (Length info in csp->content_length, which is 0 for
79 * unmodified documents)
81 * - For this to work, sed() is called twice when filtering.
83 * Revision 1.5 2001/05/26 00:28:36 jongfoster
84 * Automatic reloading of config file.
85 * Removed obsolete SIGHUP support (Unix) and Reload menu option (Win32).
86 * Most of the global variables have been moved to a new
87 * struct configuration_spec, accessed through csp->config->globalname
88 * Most of the globals remaining are used by the Win32 GUI.
90 * Revision 1.4 2001/05/22 18:46:04 oes
92 * - Enabled filtering banners by size rather than URL
93 * by adding patterns that replace all standard banner
94 * sizes with the "Junkbuster" gif to the re_filterfile
96 * - Enabled filtering WebBugs by providing a pattern
97 * which kills all 1x1 images
99 * - Added support for PCRE_UNGREEDY behaviour to pcrs,
100 * which is selected by the (nonstandard and therefore
101 * capital) letter 'U' in the option string.
102 * It causes the quantifiers to be ungreedy by default.
103 * Appending a ? turns back to greedy (!).
105 * - Added a new interceptor ijb-send-banner, which
106 * sends back the "Junkbuster" gif. Without imagelist or
107 * MSIE detection support, or if tinygif = 1, or the
108 * URL isn't recognized as an imageurl, a lame HTML
109 * explanation is sent instead.
111 * - Added new feature, which permits blocking remote
112 * script redirects and firing back a local redirect
114 * The feature is conditionally compiled, i.e. it
115 * can be disabled with --disable-fast-redirects,
116 * plus it must be activated by a "fast-redirects"
117 * line in the config file, has its own log level
118 * and of course wants to be displayed by show-proxy-args
119 * Note: Boy, all the #ifdefs in 1001 locations and
120 * all the fumbling with configure.in and acconfig.h
121 * were *way* more work than the feature itself :-(
123 * - Because a generic redirect template was needed for
124 * this, tinygif = 3 now uses the same.
126 * - Moved GIFs, and other static HTTP response templates
131 * - Removed some >400 CRs again (Jon, you really worked
134 * Revision 1.3 2001/05/20 01:21:20 jongfoster
135 * Version 2.9.4 checkin.
136 * - Merged popupfile and cookiefile, and added control over PCRS
137 * filtering, in new "actionsfile".
138 * - Implemented LOG_LEVEL_FATAL, so that if there is a configuration
139 * file error you now get a message box (in the Win32 GUI) rather
140 * than the program exiting with no explanation.
141 * - Made killpopup use the PCRS MIME-type checking and HTTP-header
143 * - Removed tabs from "config"
144 * - Moved duplicated url parsing code in "loaders.c" to a new funcition.
145 * - Bumped up version number.
147 * Revision 1.2 2001/05/17 23:01:01 oes
148 * - Cleaned CRLF's from the sources and related files
150 * Revision 1.1.1.1 2001/05/15 13:59:03 oes
151 * Initial import of version 2.9.3 source tree
154 *********************************************************************/
157 /* Declare struct FILE for vars and funcs. */
160 /* Need time_t for file_list */
164 * Include appropriate regular expression libraries.
166 * PCRS ==> Include pcre
167 * REGEX && PCRE ==> Include pcre and pcreposix
168 * REGEX && !PCRE ==> Include gnu_regex
170 * STATIC ==> Use #include "pcre.h" (compiling at same time)
171 * !STATIC ==> Use #include <pcre.h> (System library)
174 #if (defined(REGEX) && defined(PCRE)) || defined(PCRS)
180 #endif /* (defined(REGEX) && defined(PCRE)) || defined(PCRS) */
182 #if defined(REGEX) && defined(PCRE)
184 # include "pcreposix.h"
186 # include <pcreposix.h>
188 #endif /* defined(REGEX) && defined(PCRE) */
190 #if defined(REGEX) && !defined(PCRE)
191 # include "gnu_regex.h"
196 #endif /* def PCRS */
200 #endif /* def AMIGA */
208 /* Default IP and port to listen on */
209 #define HADDR_DEFAULT "127.0.0.1"
210 #define HADDR_PORT 8000
213 /* Need this for struct gateway */
216 /* Need this for struct client_state */
217 struct configuration_spec;
228 char *hostport; /* "host[:port]" */
234 /* generic attributes */
236 int (*conn)(const struct gateway *, struct http_request *, struct client_state *);
239 /* domain specific attributes */
248 /* Generic linked list of strings */
249 struct list /* FIXME: Why not separate entries and header? */
251 char * str; /* valid in an entry */
252 struct list *last; /* valid in header */
257 /* Generic linked list of strings */
258 struct list_share /* FIXME: Why not separate entries and header? */
260 const char * str; /* valid in an entry */
261 struct list_share *last; /* valid in header */
262 struct list_share *next;
294 #define IOB_PEEK(CSP) ((CSP->iob->cur > CSP->iob->eod) ? (CSP->iob->eod - CSP->iob->cur) : 0)
295 #define IOB_RESET(CSP) if(CSP->iob->buf) free(CSP->iob->buf); memset(CSP->iob, '\0', sizeof(CSP->iob));
299 #define ACTION_MASK_ALL (~0U)
301 #define ACTION_MOST_COMPATIBLE 0U
303 #define ACTION_BLOCK 0x0001U
304 #define ACTION_FAST_REDIRECTS 0x0002U
305 #define ACTION_FILTER 0x0004U
306 #define ACTION_HIDE_FORWARDED 0x0008U
307 #define ACTION_HIDE_FROM 0x0010U
308 #define ACTION_HIDE_REFERER 0x0020U /* sic - follow HTTP, not English */
309 #define ACTION_HIDE_USER_AGENT 0x0040U
310 #define ACTION_IMAGE 0x0080U
311 #define ACTION_NO_COOKIE_READ 0x0100U
312 #define ACTION_NO_COOKIE_SET 0x0200U
313 #define ACTION_NO_POPUPS 0x0400U
314 #define ACTION_VANILLA_WAFER 0x0800U
316 #define ACTION_STRING_FROM 0
317 #define ACTION_STRING_REFERER 1
318 #define ACTION_STRING_USER_AGENT 2
319 #define ACTION_STRING_COUNT 3
321 #define ACTION_MULTI_ADD_HEADER 0
322 #define ACTION_MULTI_WAFER 1
323 #define ACTION_MULTI_COUNT 2
326 struct current_action_spec
328 unsigned flags; /* a bit set to "1" = add action */
330 /* For those actions that require parameters: */
332 /* each entry is valid if & only if corresponding entry in "add" set. */
333 char * string[ACTION_STRING_COUNT];
336 struct list_share multi[ACTION_MULTI_COUNT][1];
341 unsigned mask; /* a bit set to "0" = remove action */
342 unsigned add; /* a bit set to "1" = add action */
344 /* For those actions that require parameters: */
346 /* each entry is valid if & only if corresponding entry in "add" set. */
347 char * string[ACTION_STRING_COUNT];
349 /* Strings to remove. */
350 struct list multi_remove[ACTION_MULTI_COUNT][1];
352 /* If nonzero, remove *all* strings. */
353 int multi_remove_all[ACTION_MULTI_COUNT];
356 struct list multi_add[ACTION_MULTI_COUNT][1];
361 struct url_spec url[1];
363 struct action_spec action[1];
365 struct url_actions * next;
369 /* Constants defining bitmask for csp->accept_types */
371 #ifdef DETECT_MSIE_IMAGES
373 /* MSIE detected by user-agent string */
374 #define ACCEPT_TYPE_IS_MSIE 0x0001
377 * *If* this is MSIE, it wants an image. (Or this is a shift-reload, or
378 * it's got an image from this URL before... yuck!)
379 * Only meaningful if ACCEPT_TYPE_IS_MSIE set
381 #define ACCEPT_TYPE_MSIE_IMAGE 0x0002
384 * *If* this is MSIE, it wants a HTML document.
385 * Only meaningful if ACCEPT_TYPE_IS_MSIE set
387 #define ACCEPT_TYPE_MSIE_HTML 0x0004
389 #endif /* def DETECT_MSIE_IMAGES */
394 /* The proxy's configuration */
395 struct configuration_spec * config;
398 /* The actions to perform on the current request */
399 struct current_action_spec action[1];
402 /* socket to talk to client (web browser) */
405 /* socket to talk to server (web server or proxy) */
410 /* 1 if this URL was rejected, 0 otherwise. Allows actual stats inc to
411 * occur in main thread only for thread-safety.
414 #endif /* def STATISTICS */
418 #endif /* def FORCE_LOAD */
422 #endif /* def TOGGLE */
425 * Client PC's IP address, as reported by the accept()_ function.
426 * Both as string and number
432 /* The referer in this request, if one was specified. */
434 #endif /* def TRUST_FILES */
436 #if defined(DETECT_MSIE_IMAGES)
437 /* Types the client will accept.
438 * Bitmask - see ACCEPT_TYPE_XXX constants.
441 #endif /* defined(DETECT_MSIE_IMAGES) */
443 /* The URL that was requested */
444 struct http_request http[1];
446 /* An I/O buffer used for buffering data read from the client */
449 /* List of all headers for this request */
450 struct list headers[1];
452 /* List of all cookies for this request */
453 struct list cookie_list[1];
455 #if defined(PCRS) || defined(KILLPOPUPS)
456 /* Nonzero if this has a text MIME type */
458 #endif /* defined(PCRS) || defined(KILLPOPUPS) */
460 /* The "X-Forwarded-For:" header sent by the client */
464 * Nonzero if this client is processing data.
465 * Set to zero when the thread associated with this structure dies.
469 /* files associated with this client */
470 struct file_list *flist; /* forwardfile */
471 struct file_list *actions_list;
475 struct file_list *alist; /* aclfile */
476 #endif /* def ACL_FILES */
479 struct file_list *rlist; /* Perl re_filterfile */
480 size_t content_length; /* Length after processing */
481 #endif /* def PCRS */
484 struct file_list *tlist; /* trustfile */
485 #endif /* def TRUST_FILES */
487 struct client_state *next;
495 char *(*parser)(const struct parsers *, char *, struct client_state *);
503 char *(*interceptor)(struct http_request *http, struct client_state *csp);
510 * this is a pointer to the data structures associated with the file.
511 * Read-only once the structure has been created.
515 /* Normally NULL. When we are finished with file (i.e. when we have
516 * loaded a new one), set to a pointer to an unloader function.
517 * Unloader will be called by sweep() (called from main loop) when
518 * all clients using this file are done. This prevents threading
521 void (*unloader)(void *);
523 /* Used internally by sweep(). Do not access from elsewhere. */
526 #ifndef SPLIT_PROXY_ARGS
527 /* String to be displayed as part of show-proxy-args display.
528 * Read-only once the structure has been created.
531 #endif /* ndef SPLIT_PROXY_ARGS */
533 /* Following variables allow us to check if file has been changed.
534 * Read-only once the structure has been created.
539 /* Pointer to next entry in the linked list of all "file_list"s.
540 * This linked list is so that sweep() can navigate it.
541 * Since sweep() can remove items from the list, we must be careful
542 * to only access this value from main thread (when we know sweep
545 struct file_list *next;
552 struct url_spec url[1];
554 struct block_spec *next;
556 #endif /* def TRUST_FILES */
561 struct url_spec url[1];
562 struct gateway gw[1];
563 struct forward_spec *next;
568 struct re_filterfile_spec
570 struct list patterns[1];
571 /* See README.re_filter */
574 #endif /* def PCRS */
578 #define ACL_PERMIT 1 /* accept connection request */
579 #define ACL_DENY 2 /* reject connection request */
581 struct access_control_addr
588 struct access_control_list
590 struct access_control_addr src[1];
591 struct access_control_addr dst[1];
594 struct access_control_list *next;
596 #endif /* def ACL_FILES */
599 /* Maximum number of loaders (actions, block, forward, acl...) */
603 * Data loaded from the configuration file.
605 * (Anomaly: toggle is still handled through a global, not this structure)
607 struct configuration_spec
612 #ifdef IMAGE_BLOCKING
613 int tinygif; /* FIXME Should be an action */
614 const char *tinygifurl; /* FIXME Should be an action */
615 #endif /* def IMAGE_BLOCKING */
619 const char *actions_file;
620 const char *forwardfile;
624 #endif /* def ACL_FILES */
627 const char *re_filterfile;
628 #endif /* def PCRS */
631 const char * jarfile;
633 #endif /* def JAR_FILES */
636 * Port and IP to bind to.
637 * Defaults to HADDR_DEFAULT:HADDR_PORT == 127.0.0.1:8000
642 #ifndef SPLIT_PROXY_ARGS
643 const char *suppress_message;
644 #endif /* ndef SPLIT_PROXY_ARGS */
646 #ifndef SPLIT_PROXY_ARGS
647 /* suppress listing config files */
648 int suppress_blocklists;
649 #endif /* ndef SPLIT_PROXY_ARGS */
652 const char * trustfile;
654 struct list trust_info[1];
655 struct url_spec *trust_list[64];
656 #endif /* def TRUST_FILES */
658 /* Various strings for show-proxy-args */
659 char *proxy_args_header;
660 char *proxy_args_invocation;
661 char *proxy_args_gateways;
662 char *proxy_args_trailer;
664 /* the configuration file object. */
665 struct file_list *config_file_list;
667 /* List of loaders */
668 int (*loaders[NLOADERS])(struct client_state *);
670 /* bool, nonzero if we need to bind() to the new port */
675 #define SZ(X) (sizeof(X) / sizeof(*X))
677 #define WHITEBG "<body bgcolor=\"#ffffff\" link=\"#000078\" alink=\"#ff0022\" vlink=\"#787878\">\n"
678 #define BODY "<body bgcolor=\"#f8f8f0\" link=\"#000078\" alink=\"#ff0022\" vlink=\"#787878\">\n"
679 #define BANNER "<strong>Internet J<small>UNK<i><font color=\"red\">BUSTER</font></i></small></strong>"
682 #define FORCE_PREFIX "/IJB-FORCE-LOAD"
683 #endif /* def FORCE_LOAD */
685 #define HOME_PAGE_URL "http://ijbswa.sourceforge.net/"
686 #define REDIRECT_URL HOME_PAGE_URL "redirect.php?v=" VERSION "&to="
688 static const char CFAIL[] =
689 "HTTP/1.0 503 Connect failed\n"
690 "Content-Type: text/html\n\n"
693 "<title>Internet Junkbuster: Connect failed</title>\n"
699 "TCP connection to '%s' failed: %s.\n<br>"
703 static const char CNXDOM[] =
704 "HTTP/1.0 404 Non-existent domain\n"
705 "Content-Type: text/html\n\n"
708 "<title>Internet Junkbuster: Non-existent domain</title>\n"
714 "No such domain: %s\n"
718 static const char CNOBANNER[] =
719 "HTTP/1.0 200 No Banner\n"
720 "Content-Type: text/html\n\n"
723 "<title>Internet Junkbuster: No Banner</title>\n"
729 "You asked for a banner that this proxy can't produce because either configuration does not permit.\n<br>"
730 "or the URL didn't end with .gif\n"
734 static const char CSUCCEED[] =
735 "HTTP/1.0 200 Connection established\n"
736 "Proxy-Agent: IJ/" VERSION "\n\n";
738 static const char CHEADER[] =
739 "HTTP/1.0 400 Invalid header received from browser\n\n";
741 static const char SHEADER[] =
742 "HTTP/1.0 502 Invalid header received from server\n\n";
744 #ifdef IMAGE_BLOCKING
747 * Hint: You can encode your own GIFs like that:
748 * perl -e 'while (read STDIN, $c, 1) { printf("\\%.3o,", unpack("C", $c)); }'
751 static const char BLANKGIF[] =
752 "HTTP/1.0 200 OK\r\n"
753 "Pragma: no-cache\r\n"
754 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
755 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
756 "Content-type: image/gif\r\n\r\n"
757 "GIF89a\001\000\001\000\200\000\000\377\377\377\000\000"
758 "\000!\371\004\001\000\000\000\000,\000\000\000\000\001"
759 "\000\001\000\000\002\002D\001\000;";
761 static const char JBGIF[] =
762 "HTTP/1.0 200 OK\r\n"
763 "Pragma: no-cache\r\n"
764 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
765 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
766 "Content-type: image/gif\r\n\r\n"
767 "GIF89aD\000\013\000\360\000\000\000\000\000\377\377\377!"
768 "\371\004\001\000\000\001\000,\000\000\000\000D\000\013\000"
769 "\000\002a\214\217\251\313\355\277\000\200G&K\025\316hC\037"
770 "\200\234\230Y\2309\235S\230\266\206\372J\253<\3131\253\271"
771 "\270\215\342\254\013\203\371\202\264\334P\207\332\020o\266"
772 "N\215I\332=\211\312\3513\266:\026AK)\364\370\365aobr\305"
773 "\372\003S\275\274k2\354\254z\347?\335\274x\306^9\374\276"
776 #endif /* def IMAGE_BLOCKING */
778 #if defined(FAST_REDIRECTS) || defined(IMAGE_BLOCKING)
780 static const char HTTP_REDIRECT_TEMPLATE[] =
781 "HTTP/1.0 302 Local Redirect from Junkbuster\r\n"
782 "Pragma: no-cache\r\n"
783 "Last-Modified: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
784 "Expires: Thu Jul 31, 1997 07:42:22 pm GMT\r\n"
785 "Location: %s\r\n\r\n";
787 #endif /* defined(FAST_REDIRECTS) || defined(IMAGE_BLOCKING) */
793 #endif /* ndef _PROJECT_H */