#ifndef PROJECT_H_INCLUDED
#define PROJECT_H_INCLUDED
-#define PROJECT_H_VERSION "$Id: project.h,v 1.38 2001/10/23 21:19:04 jongfoster Exp $"
+#define PROJECT_H_VERSION "$Id: project.h,v 1.63 2002/03/31 17:19:00 jongfoster Exp $"
/*********************************************************************
*
* File : $Source: /cvsroot/ijbswa/current/project.h,v $
* (though it does declare some macros).
*
* Copyright : Written by and Copyright (C) 2001 the SourceForge
- * IJBSWA team. http://ijbswa.sourceforge.net
+ * Privoxy team. http://www.privoxy.org/
*
* Based on the Internet Junkbuster originally written
* by and Copyright (C) 1997 Anonymous Coders and
*
* Revisions :
* $Log: project.h,v $
+ * Revision 1.63 2002/03/31 17:19:00 jongfoster
+ * Win32 only: Enabling STRICT to fix a VC++ compile warning.
+ *
+ * Revision 1.62 2002/03/26 22:48:49 swa
+ * new homepage url
+ *
+ * Revision 1.61 2002/03/26 22:29:55 swa
+ * we have a new homepage!
+ *
+ * Revision 1.60 2002/03/24 15:52:17 jongfoster
+ * Changing CGI URL prefixes for new name
+ *
+ * Revision 1.59 2002/03/24 15:23:33 jongfoster
+ * Name changes
+ *
+ * Revision 1.58 2002/03/24 13:25:43 swa
+ * name change related issues
+ *
+ * Revision 1.57 2002/03/16 20:28:34 oes
+ * Added descriptions to the filters so users will know what they select in the cgi editor
+ *
+ * Revision 1.56 2002/03/13 20:27:30 oes
+ * Fixing bug with CT_TABOO
+ *
+ * Revision 1.55 2002/03/12 01:42:50 oes
+ * Introduced modular filters
+ *
+ * Revision 1.54 2002/03/09 20:03:52 jongfoster
+ * - Making various functions return int rather than size_t.
+ * (Undoing a recent change). Since size_t is unsigned on
+ * Windows, functions like read_socket that return -1 on
+ * error cannot return a size_t.
+ *
+ * THIS WAS A MAJOR BUG - it caused frequent, unpredictable
+ * crashes, and also frequently caused JB to jump to 100%
+ * CPU and stay there. (Because it thought it had just
+ * read ((unsigned)-1) == 4Gb of data...)
+ *
+ * - The signature of write_socket has changed, it now simply
+ * returns success=0/failure=nonzero.
+ *
+ * - Trying to get rid of a few warnings --with-debug on
+ * Windows, I've introduced a new type "jb_socket". This is
+ * used for the socket file descriptors. On Windows, this
+ * is SOCKET (a typedef for unsigned). Everywhere else, it's
+ * an int. The error value can't be -1 any more, so it's
+ * now JB_INVALID_SOCKET (which is -1 on UNIX, and in
+ * Windows it maps to the #define INVALID_SOCKET.)
+ *
+ * - The signature of bind_port has changed.
+ *
+ * Revision 1.53 2002/03/08 16:48:55 oes
+ * Added FEATURE_NO_GIFS and BUILTIN_IMAGE_MIMETYPE
+ *
+ * Revision 1.52 2002/03/07 03:46:17 oes
+ * Fixed compiler warnings
+ *
+ * Revision 1.51 2002/03/05 04:52:42 oes
+ * Deleted non-errlog debugging code
+ *
+ * Revision 1.50 2002/03/04 19:32:07 oes
+ * Changed default port to 8118
+ *
+ * Revision 1.49 2002/03/04 18:28:55 oes
+ * Deleted PID_FILE_NAME
+ *
+ * Revision 1.48 2002/03/03 14:50:40 oes
+ * Fixed CLF logging: Added ocmd member for client's request to struct http_request
+ *
+ * Revision 1.47 2002/02/20 23:15:13 jongfoster
+ * Parsing functions now handle out-of-memory gracefully by returning
+ * an error code.
+ *
+ * Revision 1.46 2002/01/17 21:06:09 jongfoster
+ * Now #defining the URLs of the config interface
+ *
+ * Minor changes to struct http_request and struct url_spec due to
+ * standardizing that struct http_request is used to represent a URL, and
+ * struct url_spec is used to represent a URL pattern. (Before, URLs were
+ * represented as seperate variables and a partially-filled-in url_spec).
+ *
+ * Revision 1.45 2002/01/09 14:33:27 oes
+ * Added HOSTENT_BUFFER_SIZE
+ *
+ * Revision 1.44 2001/12/30 14:07:32 steudten
+ * - Add signal handling (unix)
+ * - Add SIGHUP handler (unix)
+ * - Add creation of pidfile (unix)
+ * - Add action 'top' in rc file (RH)
+ * - Add entry 'SIGNALS' to manpage
+ * - Add exit message to logfile (unix)
+ *
+ * Revision 1.43 2001/11/22 21:57:51 jongfoster
+ * Making action_spec->flags into an unsigned long rather than just an
+ * unsigned int.
+ * Adding ACTION_NO_COOKIE_KEEP
+ *
+ * Revision 1.42 2001/11/05 21:42:41 steudten
+ * Include DBG() macro.
+ *
+ * Revision 1.41 2001/10/28 19:12:06 jongfoster
+ * Adding ijb_toupper()
+ *
+ * Revision 1.40 2001/10/26 17:40:47 oes
+ * Moved ijb_isspace and ijb_tolower to project.h
+ * Removed http->user_agent, csp->referrer and csp->accept_types
+ *
+ * Revision 1.39 2001/10/25 03:45:02 david__schmidt
+ * Adding a (void*) cast to freez() because Visual Age C++ won't expand the
+ * macro when called with a cast; so moving the cast to the macro def'n
+ * seems to both eliminate compiler warnings (on darwin and OS/2, anyway) and
+ * doesn't make macro expansion complain. Hope this works for everyone else
+ * too...
+ *
* Revision 1.38 2001/10/23 21:19:04 jongfoster
* New error-handling support: jb_err type and JB_ERR_xxx constants
* CGI functions now return a jb_err, and their parameters map is const.
* and are included anyway.
*/
-#if defined(REGEX_PCRE) || defined (REGEX_GNU)
+#if defined(REGEX_PCRE)
# define REGEX
-#endif /* defined(REGEX_PCRE) || defined (REGEX_GNU) */
+#endif /* defined(REGEX_PCRE) */
#ifdef STATIC_PCRE
# include "pcre.h"
# endif
#endif /* defined(REGEX_PCRE) */
-#if defined(REGEX_GNU)
-# include "gnu_regex.h"
-#endif
-
#ifdef AMIGA
#include "amiga.h"
#endif /* def AMIGA */
+#ifdef _WIN32
+/*
+ * I don't want to have to #include all this just for the declaration
+ * of SOCKET. However, it looks like we have to...
+ */
+#ifndef STRICT
+#define STRICT
+#endif
+#include <windows.h>
+#endif
+
+
#ifdef __cplusplus
extern "C" {
#endif
+/*
+ * The type used by sockets. On UNIX it's an int. Microsoft decided to
+ * make it an unsigned.
+ */
+#ifdef _WIN32
+typedef SOCKET jb_socket;
+#define JB_INVALID_SOCKET INVALID_SOCKET
+#else /* ndef _WIN32 */
+typedef int jb_socket;
+#define JB_INVALID_SOCKET (-1)
+#endif /* ndef _WIN32 */
+
/*
* Error codes. Functions returning these should return a jb_err
*/
#define freez(X) { if(X) { free((void*)X); X = NULL ; } }
+
+/* Fix a problem with Solaris. There should be no effect on other
+ * platforms.
+ * Solaris's isspace() is a macro which uses it's argument directly
+ * as an array index. Therefore we need to make sure that high-bit
+ * characters generate +ve values, and ideally we also want to make
+ * the argument match the declared parameter type of "int".
+ *
+ * Note: Remember to #include <ctype.h> if you use these macros.
+ */
+#define ijb_toupper(__X) toupper((int)(unsigned char)(__X))
+#define ijb_tolower(__X) tolower((int)(unsigned char)(__X))
+#define ijb_isspace(__X) isspace((int)(unsigned char)(__X))
+
/*
* Use for statically allocated buffers if you have no other choice.
* Remember to check the length of what you write into the buffer
*/
#define BUFFER_SIZE 5000
+/*
+ * Buffer size for capturing struct hostent data in the
+ * gethostby(name|addr)_r library calls. Since we don't
+ * loop over gethostbyname_r, the buffer must be sufficient
+ * to accomodate multiple IN A RRs, as used in DNS round robin
+ * load balancing. W3C's wwwlib uses 1K, so that should be
+ * good enough for us, too.
+ */
+#define HOSTENT_BUFFER_SIZE 1024
+
/*
* So you can say "while (FOREVER) { ...do something... }"
*/
/* Default IP and port to listen on */
#define HADDR_DEFAULT "127.0.0.1"
-#define HADDR_PORT 8000
-
+#define HADDR_PORT 8118
/* Forward defs for various structures */
struct http_request
{
char *cmd; /* Whole command line: method, URL, Version */
+ char *ocmd; /* Backup of original cmd for CLF logging */
char *gpc; /* HTTP method: GET, POST, .. */
char *url; /* The URL */
char *ver; /* Protocol version */
char *host_ip_addr_str; /* String with dotted decimal representation
* of host's IP. NULL before connect_to() */
- char *user_agent; /* Client's User-Agent: header value */
+
+ char *dbuffer; /* Buffer with '\0'-delimited domain name. */
+ char **dvec; /* List of pointers to the strings in dbuffer. */
+ int dcount; /* How many parts to this domain? (length of dvec) */
};
-/* Response generated by CGI, blocker, or error handler */
+/*
+ * Response generated by CGI, blocker, or error handler
+ */
struct http_response
{
- char *status; /* HTTP status (string)*/
+ char *status; /* HTTP status (string) */
struct list headers[1]; /* List of header lines */
- char *head; /* Formatted http response head */
- int head_length; /* Length of http response head */
- char *body; /* HTTP document body */
- int content_length; /* Length of body, REQUIRED if binary body */
- int is_static; /* Nonzero if the content will never change and
- * should be cached by the broser (e.g. images) */
+ char *head; /* Formatted http response head */
+ size_t head_length; /* Length of http response head */
+ char *body; /* HTTP document body */
+ size_t content_length; /* Length of body, REQUIRED if binary body */
+ int is_static; /* Nonzero if the content will never change and
+ * should be cached by the brwoser (e.g. images) */
};
/* A URL pattern */
char *spec; /* The string which was parsed to produce this */
/* url_spec. Used for debugging or display only. */
- /* Hostname matching: */
- char *domain; /* Fully qalified domain name (FQDN) pattern. */
- /* May contain "*". */
- char *dbuf; /* Buffer with '\0'-delimited fqdn */
- char **dvec; /* Domain ptr vector into dbuf */
- int dcnt; /* How many domains in fqdn? */
- int unanchored; /* Bitmap - flags are ANCHOR_LEFT and ANCHOR_RIGHT */
+ /* Hostname matching, or dbuffer == NULL to match all hosts */
+ char *dbuffer; /* Buffer with '\0'-delimited domain name. */
+ char **dvec; /* List of pointers to the strings in dbuffer. */
+ int dcount; /* How many parts to this domain? (length of dvec) */
+ int unanchored; /* Bitmap - flags are ANCHOR_LEFT and ANCHOR_RIGHT. */
/* Port matching: */
int port; /* The port number, or 0 to match all ports. */
#endif
};
#ifdef REGEX
-#define URL_SPEC_INITIALIZER { NULL, NULL, NULL, NULL, 0, 0, 0, NULL, 0, NULL }
+#define URL_SPEC_INITIALIZER { NULL, NULL, NULL, 0, 0, 0, NULL, 0, NULL }
#else /* ifndef REGEX */
-#define URL_SPEC_INITIALIZER { NULL, NULL, NULL, NULL, 0, 0, 0, NULL, 0 }
+#define URL_SPEC_INITIALIZER { NULL, NULL, NULL, 0, 0, 0, NULL, 0 }
#endif /* ndef REGEX */
/* Constants for host part matching in URLs */
#define IOB_PEEK(CSP) ((CSP->iob->cur > CSP->iob->eod) ? (CSP->iob->eod - CSP->iob->cur) : 0)
#define IOB_RESET(CSP) if(CSP->iob->buf) free(CSP->iob->buf); memset(CSP->iob, '\0', sizeof(CSP->iob));
-/* Keys for csp->content_type */
+/* Bits for csp->content_type */
#define CT_TEXT 1 /* Suitable for pcrs filtering */
#define CT_GIF 2 /* Suitable for GIF filtering */
-#define CT_TABOO 3 /* DONT filter */
+#define CT_TABOO 4 /* DONT filter */
#define ACTION_MASK_ALL (~0U)
-#define ACTION_MOST_COMPATIBLE 0x0000U
-
-#define ACTION_BLOCK 0x0001U
-#define ACTION_DEANIMATE 0x0002U
-#define ACTION_DOWNGRADE 0x0004U
-#define ACTION_FAST_REDIRECTS 0x0008U
-#define ACTION_FILTER 0x0010U
-#define ACTION_HIDE_FORWARDED 0x0020U
-#define ACTION_HIDE_FROM 0x0040U
-#define ACTION_HIDE_REFERER 0x0080U /* sic - follow HTTP, not English */
-#define ACTION_HIDE_USER_AGENT 0x0100U
-#define ACTION_IMAGE 0x0200U
-#define ACTION_IMAGE_BLOCKER 0x0400U
-#define ACTION_NO_COMPRESSION 0x0800U
-#define ACTION_NO_COOKIE_READ 0x1000U
-#define ACTION_NO_COOKIE_SET 0x2000U
-#define ACTION_NO_POPUPS 0x4000U
-#define ACTION_VANILLA_WAFER 0x8000U
-#define ACTION_LIMIT_CONNECT 0x010000U
+#define ACTION_MOST_COMPATIBLE 0x00000000UL
+
+#define ACTION_BLOCK 0x00000001UL
+#define ACTION_DEANIMATE 0x00000002UL
+#define ACTION_DOWNGRADE 0x00000004UL
+#define ACTION_FAST_REDIRECTS 0x00000008UL
+#define ACTION_HIDE_FORWARDED 0x00000010UL
+#define ACTION_HIDE_FROM 0x00000020UL
+#define ACTION_HIDE_REFERER 0x00000040UL /* sic - follow HTTP, not English */
+#define ACTION_HIDE_USER_AGENT 0x00000080UL
+#define ACTION_IMAGE 0x00000100UL
+#define ACTION_IMAGE_BLOCKER 0x00000200UL
+#define ACTION_NO_COMPRESSION 0x00000400UL
+#define ACTION_NO_COOKIE_KEEP 0x00000800UL
+#define ACTION_NO_COOKIE_READ 0x00001000UL
+#define ACTION_NO_COOKIE_SET 0x00002000UL
+#define ACTION_NO_POPUPS 0x00004000UL
+#define ACTION_VANILLA_WAFER 0x00008000UL
+#define ACTION_LIMIT_CONNECT 0x00010000UL
#define ACTION_STRING_DEANIMATE 0
#define ACTION_STRING_FROM 1
#define ACTION_MULTI_ADD_HEADER 0
#define ACTION_MULTI_WAFER 1
-#define ACTION_MULTI_COUNT 2
+#define ACTION_MULTI_FILTER 2
+#define ACTION_MULTI_COUNT 3
+
/*
* This structure contains a list of actions to apply to a URL.
*/
struct current_action_spec
{
- unsigned flags; /* a bit set to "1" = add action */
+ unsigned long flags; /* a bit set to "1" = add action */
/* For those actions that require parameters: */
*/
struct action_spec
{
- unsigned mask; /* a bit set to "0" = remove action */
- unsigned add; /* a bit set to "1" = add action */
+ unsigned long mask; /* a bit set to "0" = remove action */
+ unsigned long add; /* a bit set to "1" = add action */
/* For those actions that require parameters: */
};
-/* Constants defining bitmask for csp->accept_types */
-
-#ifdef FEATURE_IMAGE_DETECT_MSIE
-
-/* MSIE detected by user-agent string */
-#define ACCEPT_TYPE_IS_MSIE 0x0001
-
-/*
- * *If* this is MSIE, it wants an image. (Or this is a shift-reload, or
- * it's got an image from this URL before... yuck!)
- * Only meaningful if ACCEPT_TYPE_IS_MSIE set
- */
-#define ACCEPT_TYPE_MSIE_IMAGE 0x0002
-
-/*
- * *If* this is MSIE, it wants a HTML document.
- * Only meaningful if ACCEPT_TYPE_IS_MSIE set
- */
-#define ACCEPT_TYPE_MSIE_HTML 0x0004
-
-#endif /* def FEATURE_IMAGE_DETECT_MSIE */
-
/*
* Flags for use in csp->flags
*/
#define CSP_FLAG_TOGGLED_ON 0x20 /* Set if we are toggled on (FEATURE_TOGGLE) */
/*
- * The state of a JunkBuster processing thread.
+ * The state of a Privoxy processing thread.
*/
struct client_state
{
struct current_action_spec action[1];
/* socket to talk to client (web browser) */
- int cfd;
+ jb_socket cfd;
/* socket to talk to server (web server or proxy) */
- int sfd;
+ jb_socket sfd;
/* Multi-purpose flag container, see CSP_FLAG_* above */
unsigned short int flags;
char *my_ip_addr_str;
char *my_hostname;
-#ifdef FEATURE_TRUST
- /* The referer in this request, if one was specified. */
- char *referrer;
-#endif /* def FEATURE_TRUST */
-
-#if defined(FEATURE_IMAGE_DETECT_MSIE)
- /* Types the client will accept.
- * Bitmask - see ACCEPT_TYPE_XXX constants.
- */
- int accept_types;
-#endif /* defined(FEATURE_IMAGE_DETECT_MSIE) */
-
/* The URL that was requested */
struct http_request http[1];
};
+/*
+ * A function to add a header
+ */
+typedef jb_err (*add_header_func_ptr)(struct client_state *);
+
+/*
+ * A function to process a header
+ */
+typedef jb_err (*parser_func_ptr )(struct client_state *, char **);
+
/*
* List of functions to run on a list of headers
*/
struct parsers
{
- char *str;
- char len;
- char *(*parser)(const struct parsers *, const char *, struct client_state *);
+ char *str;
+ size_t len;
+ parser_func_ptr parser;
};
/*
- * A data file used by JunkBuster. Kept in a linked list.
+ * A data file used by Privoxy. Kept in a linked list.
*/
struct file_list
{
#define FORWARD_SPEC_INITIALIZER { { URL_SPEC_INITIALIZER }, 0, NULL, 0, NULL, 0, NULL }
+/*
+ * This struct represents one filter (one block) from
+ * the re_filterfile. If there is more than one filter
+ * in the file, the file will be represented by a
+ * chained list of re_filterfile specs.
+ */
struct re_filterfile_spec
{
- char *username;
- char *filtername;
- struct list patterns[1];
- pcrs_job *joblist;
+ char *name; /* Name from FILTER: statement in re_filterfile */
+ char *description; /* Description from FILTER: statement in re_filterfile */
+ struct list patterns[1]; /* The patterns from the re_filterfile */
+ pcrs_job *joblist; /* The resulting compiled pcrs_jobs */
+ struct re_filterfile_spec *next; /* The pointer for chaining */
};
#ifdef FEATURE_ACL
/*
* Port and IP to bind to.
- * Defaults to HADDR_DEFAULT:HADDR_PORT == 127.0.0.1:8000
+ * Defaults to HADDR_DEFAULT:HADDR_PORT == 127.0.0.1:8118
*/
const char *haddr;
int hport;
#define SZ(X) (sizeof(X) / sizeof(*X))
#ifdef FEATURE_FORCE_LOAD
-#define FORCE_PREFIX "/IJB-FORCE-LOAD"
+#define FORCE_PREFIX "/PRIVOXY-FORCE"
#endif /* def FEATURE_FORCE_LOAD */
+#ifdef FEATURE_NO_GIFS
+#define BUILTIN_IMAGE_MIMETYPE "image/png"
+#else
+#define BUILTIN_IMAGE_MIMETYPE "image/gif"
+#endif /* def FEATURE_NO_GIFS */
+
+
/* Hardwired URLs */
-#define HOME_PAGE_URL "http://ijbswa.sourceforge.net"
-#define REDIRECT_URL HOME_PAGE_URL "/redirect.php?v=" VERSION "&to="
-#define CGI_PREFIX_HOST "i.j.b"
+#define HOME_PAGE_URL "http://www.privoxy.org"
+#define REDIRECT_URL HOME_PAGE_URL "/redirect.php?v=" VERSION "&to="
+
+/*
+ * The "hosts" to intercept and display CGI pages.
+ * First one is a hostname only, second one can specify host and path.
+ *
+ * Notes:
+ * 1) Do not specify the http: prefix
+ * 2) CGI_SITE_2_PATH must not end with /, one will be added automatically.
+ * 3) CGI_SITE_2_PATH must start with /, unless it is the empty string.
+ */
+#define CGI_SITE_1_HOST "p.p"
+#define CGI_SITE_2_HOST "config.privoxy.org"
+#define CGI_SITE_2_PATH ""
+
+/*
+ * The prefix for CGI pages. Written out in generated HTML.
+ * INCLUDES the trailing slash.
+ */
+#define CGI_PREFIX "http://" CGI_SITE_2_HOST CGI_SITE_2_PATH "/"
+
/* HTTP snipplets */
static const char CSUCCEED[] =
"HTTP/1.0 200 Connection established\n"
- "Proxy-Agent: IJ/" VERSION "\r\n\r\n";
+ "Proxy-Agent: Privoxy/" VERSION "\r\n\r\n";
static const char CHEADER[] =
"HTTP/1.0 400 Invalid header received from browser\r\n\r\n";