-const char jcc_rcs[] = "$Id: jcc.c,v 1.248 2009/05/10 10:25:19 fabiankeil Exp $";
+const char jcc_rcs[] = "$Id: jcc.c,v 1.274 2009/07/19 11:19:50 fabiankeil Exp $";
/*********************************************************************
*
* File : $Source: /cvsroot/ijbswa/current/jcc.c,v $
* or write to the Free Software Foundation, Inc., 59
* Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
- * Revisions :
- * $Log: jcc.c,v $
- * Revision 1.248 2009/05/10 10:25:19 fabiankeil
- * Change wait_for_alive_connection() prototype to use (void).
- *
- * Revision 1.247 2009/05/10 10:19:23 fabiankeil
- * Reenable server-side-only keep-alive support, but only share
- * outgoing connections if the connection-sharing option is set.
- *
- * Revision 1.246 2009/05/10 10:12:30 fabiankeil
- * Initial keep-alive support for the client socket.
- * Temporarily disable the server-side-only keep-alive code.
- *
- * Revision 1.245 2009/04/24 15:29:43 fabiankeil
- * Allow to limit the number of of client connections.
- *
- * Revision 1.244 2009/04/17 11:34:34 fabiankeil
- * Style cosmetics for the IPv6 code.
- *
- * Revision 1.243 2009/04/17 11:27:49 fabiankeil
- * Petr Pisar's privoxy-3.0.12-ipv6-3.diff.
- *
- * Revision 1.242 2009/04/11 10:44:47 fabiankeil
- * Update a comment. We're not in Kansas anymore.
- *
- * Revision 1.241 2009/04/11 10:37:23 fabiankeil
- * When dropping connections due to ACL, don't leak csp->ip_addr_str.
- *
- * Revision 1.240 2009/04/09 10:12:54 fabiankeil
- * Fix two cases in which an invalid server response would result
- * in the client connection being closed without sending an error
- * message first.
- *
- * Revision 1.239 2009/04/07 11:43:50 fabiankeil
- * If the server rudely resets the connection directly after sending the
- * headers, pass the mess to the client instead of sending an incorrect
- * connect-failed message. Fixes #2698674 reported by mybugaccount.
- *
- * Revision 1.238 2009/03/27 14:42:30 fabiankeil
- * Correct the status code for CONNECTION_TIMEOUT_RESPONSE.
- *
- * Revision 1.237 2009/03/27 14:32:04 fabiankeil
- * If spawning a child in listen_loop() fails, send a real
- * HTTP response to the client and continue listening for
- * new connections without artificial delay.
- *
- * Revision 1.236 2009/03/25 17:30:24 fabiankeil
- * In serve(), keep the client socket open until we marked the
- * server socket as unused. This should increase the chances
- * that we reuse the connection for the client's next request
- * to the same destination.
- *
- * Revision 1.235 2009/03/18 21:01:20 fabiankeil
- * Comment fix. Spotted by Roland.
- *
- * Revision 1.234 2009/03/18 20:48:42 fabiankeil
- * If the --no-daemon option is used, enable LOG_LEVEL_INFO
- * before the config file has been parsed (as we always did).
- *
- * Revision 1.233 2009/03/13 14:10:07 fabiankeil
- * Fix some more harmless warnings on amd64.
- *
- * Revision 1.232 2009/03/08 19:29:16 fabiankeil
- * Reinitialize the timeout structure every time before passing
- * it to select(). Apparently some implementations mess with it.
- * Probably fixes #2669131 reported by cyberpatrol.
- *
- * Revision 1.231 2009/03/08 14:19:23 fabiankeil
- * Fix justified (but harmless) compiler warnings
- * on platforms where sizeof(int) < sizeof(long).
- *
- * Revision 1.230 2009/03/07 13:09:17 fabiankeil
- * Change csp->expected_content and_csp->expected_content_length from
- * size_t to unsigned long long to reduce the likelihood of integer
- * overflows that would let us close the connection prematurely.
- * Bug found while investigating #2669131, reported by cyberpatrol.
- *
- * Revision 1.229 2009/03/07 11:17:01 fabiankeil
- * Fix compiler warning.
- *
- * Revision 1.228 2009/03/06 20:30:13 fabiankeil
- * Log unsigned values as such.
- *
- * Revision 1.227 2009/03/02 19:18:11 fabiankeil
- * Streamline parse_http_request()'s prototype. As
- * cparser pointed out it doesn't actually use csp.
- *
- * Revision 1.226 2009/03/01 18:28:24 fabiankeil
- * Help clang understand that we aren't dereferencing
- * NULL pointers here.
- *
- * Revision 1.225 2009/02/19 18:09:32 fabiankeil
- * Unbreak build without FEATURE_CONNECTION_KEEP_ALIVE.
- * Noticed by David.
- *
- * Revision 1.224 2009/02/14 15:32:04 fabiankeil
- * Add the request URL to the timeout message in chat().
- * Suggested by Lee.
- *
- * Revision 1.223 2009/02/09 21:21:16 fabiankeil
- * Now that init_log_module() is called earlier, call show_version()
- * later on from main() directly so it doesn't get called for --help
- * or --version.
- *
- * Revision 1.222 2009/02/08 12:56:51 fabiankeil
- * Call initialize_mutexes() before init_log_module() again.
- * Broken since r220, might be the cause of Lee's #2579448.
- *
- * Revision 1.221 2009/02/06 18:02:58 fabiankeil
- * When dropping privileges, also give up membership in supplementary
- * groups. Thanks to Matthias Drochner for reporting the problem,
- * providing the initial patch and testing the final version.
- *
- * Revision 1.220 2009/02/04 18:29:07 fabiankeil
- * Initialize the log module before parsing arguments.
- * Thanks to Matthias Drochner for the report.
- *
- * Revision 1.219 2009/01/31 16:08:21 fabiankeil
- * Remove redundant error check in receive_client_request().
- *
- * Revision 1.218 2009/01/31 12:25:54 fabiankeil
- * Flatten indentation in receive_client_request().
- *
- * Revision 1.217 2009/01/07 19:50:09 fabiankeil
- * - If the socket-timeout has been reached and the client
- * hasn't received any data yet, send an explanation before
- * closing the connection.
- * - In get_request_line(), signal timeouts the right way.
- *
- * Revision 1.216 2008/12/24 22:13:11 ler762
- * fix GCC 3.4.4 warning
- *
- * Revision 1.215 2008/12/24 17:06:19 fabiankeil
- * Keep a thread around to timeout alive connections
- * even if no new requests are coming in.
- *
- * Revision 1.214 2008/12/20 14:53:55 fabiankeil
- * Add config option socket-timeout to control the time
- * Privoxy waits for data to arrive on a socket. Useful
- * in case of stale ssh tunnels or when fuzz-testing.
- *
- * Revision 1.213 2008/12/15 18:45:51 fabiankeil
- * When logging crunches, log the whole URL, so one can easily
- * differentiate between vanilla HTTP and CONNECT requests.
- *
- * Revision 1.212 2008/12/14 15:46:22 fabiankeil
- * Give crunched requests their own log level.
- *
- * Revision 1.211 2008/12/06 10:05:03 fabiankeil
- * Downgrade "Received x bytes while expecting y." message to
- * LOG_LEVEL_CONNECT as it doesn't necessarily indicate an error.
- *
- * Revision 1.210 2008/12/02 22:03:18 fabiankeil
- * Don't miscalculate byte_count if we don't get all the
- * server headers with one read_socket() call. With keep-alive
- * support enabled, this caused delays until the server closed
- * the connection.
- *
- * Revision 1.209 2008/11/27 09:44:04 fabiankeil
- * Cosmetics for the last commit: Don't watch out for
- * the last chunk if the content isn't chunk-encoded or
- * if we already determined the content length previously.
- *
- * Revision 1.208 2008/11/26 18:24:17 fabiankeil
- * Recognize that the server response is complete if the
- * last chunk is read together with the server headers.
- * Reported by Lee.
- *
- * Revision 1.207 2008/11/25 17:25:16 fabiankeil
- * Don't convert the client-header list to text until we need to.
- *
- * Revision 1.206 2008/11/23 17:00:11 fabiankeil
- * Some more chat() cosmetics.
- *
- * Revision 1.205 2008/11/16 12:43:49 fabiankeil
- * Turn keep-alive support into a runtime feature
- * that is disabled by setting keep-alive-timeout
- * to a negative value.
- *
- * Revision 1.204 2008/11/06 19:42:17 fabiankeil
- * Fix last-chunk detection hack to also apply
- * if buf[] contains nothing but the last-chunk.
- *
- * Revision 1.203 2008/11/06 18:34:35 fabiankeil
- * Factor receive_client_request() and
- * parse_client_request() out of chat().
- *
- * Revision 1.202 2008/11/02 18:40:34 fabiankeil
- * If we received a different amount of data than we expected,
- * log a warning and make sure the server socket isn't reused.
- *
- * Revision 1.201 2008/11/02 16:48:20 fabiankeil
- * Revert revision 1.195 and try again.
- *
- * Revision 1.200 2008/10/26 16:53:18 fabiankeil
- * Fix gcc44 warning.
- *
- * Revision 1.199 2008/10/26 15:36:10 fabiankeil
- * Remove two debug messages with LOG_LEVEL_INFO.
- *
- * Revision 1.198 2008/10/22 15:19:55 fabiankeil
- * Once More, With Feeling: if there is no logfile
- * because the user didn't specify one, we shouldn't
- * call init_error_log() after receiving SIGHUP either.
- *
- * Revision 1.197 2008/10/20 17:02:40 fabiankeil
- * If SIGHUP is received while we aren't running in daemon
- * mode, calling init_error_log() would be a mistake.
- *
- * Revision 1.196 2008/10/16 09:16:41 fabiankeil
- * - Fix two gcc44 conversion warnings.
- * - Don't bother logging the last five bytes
- * of the 0-chunk.
- *
- * Revision 1.195 2008/10/13 16:04:37 fabiankeil
- * Make sure we don't try to reuse tainted server sockets.
- *
- * Revision 1.194 2008/10/12 18:35:18 fabiankeil
- * The last commit was a bit too ambitious, apparently the content
- * length adjustment is only necessary if we aren't buffering.
- *
- * Revision 1.193 2008/10/12 15:57:35 fabiankeil
- * Fix content length calculation if we read headers
- * and the start of the body at once. Now that we have
- * FEATURE_CONNECTION_KEEP_ALIVE, it actually matters.
- *
- * Revision 1.192 2008/10/11 18:19:14 fabiankeil
- * Even more chat() cosmetics.
- *
- * Revision 1.191 2008/10/11 18:00:14 fabiankeil
- * Reformat some comments in chat().
- *
- * Revision 1.190 2008/10/11 14:58:00 fabiankeil
- * In case of chunk-encoded content, stop reading if
- * the buffer looks like it ends with the last chunk.
- *
- * Revision 1.189 2008/10/11 09:53:00 fabiankeil
- * Let server_response_is_complete() deal properly with
- * content that is neither buffered nor read all at once.
- *
- * Revision 1.188 2008/10/09 18:21:41 fabiankeil
- * Flush work-in-progress changes to keep outgoing connections
- * alive where possible. Incomplete and mostly #ifdef'd out.
- *
- * Revision 1.187 2008/09/07 12:35:05 fabiankeil
- * Add mutex lock support for _WIN32.
- *
- * Revision 1.186 2008/09/04 08:13:58 fabiankeil
- * Prepare for critical sections on Windows by adding a
- * layer of indirection before the pthread mutex functions.
- *
- * Revision 1.185 2008/08/30 12:03:07 fabiankeil
- * Remove FEATURE_COOKIE_JAR.
- *
- * Revision 1.184 2008/08/22 15:34:45 fabiankeil
- * - Silence LLVM/Clang complaint.
- * - Make received_hup_signal static.
- * - Hide definitions for basedir, pidfile and received_hup_signal
- * from __EMX__ as they only seem to be used in case of #ifdef unix.
- *
- * Revision 1.183 2008/08/21 07:09:35 fabiankeil
- * Accept Shoutcast responses again. Problem reported
- * and fix suggested by Stefan in #2062860.
- *
- * Revision 1.182 2008/06/27 11:13:56 fabiankeil
- * Fix possible NULL-pointer dereference reported
- * by din_a4 in #2003937. Pointy hat to me.
- *
- * Revision 1.181 2008/05/21 15:47:15 fabiankeil
- * Streamline sed()'s prototype and declare
- * the header parse and add structures static.
- *
- * Revision 1.180 2008/05/21 15:26:32 fabiankeil
- * - Mark csp as immutable for send_crunch_response().
- * - Fix comment spelling.
- *
- * Revision 1.179 2008/05/20 20:13:32 fabiankeil
- * Factor update_server_headers() out of sed(), ditch the
- * first_run hack and make server_patterns_light static.
- *
- * Revision 1.178 2008/05/10 13:23:38 fabiankeil
- * Don't provide get_header() with the whole client state
- * structure when it only needs access to csp->iob.
- *
- * Revision 1.177 2008/05/10 11:51:12 fabiankeil
- * Make the "read the rest of the headers" loop a bit more readable.
- *
- * Revision 1.176 2008/05/10 11:37:57 fabiankeil
- * - Instead of logging when the IIS5 hack is enabled, log when it fails.
- * - Remove useless comment.
- *
- * Revision 1.175 2008/05/09 18:53:59 fabiankeil
- * Fix comment grammar.
- *
- * Revision 1.174 2008/05/07 18:05:53 fabiankeil
- * Remove the pointless buffer in client_protocol_is_unsupported().
- *
- * Revision 1.173 2008/05/06 15:09:00 fabiankeil
- * Least-effort fix for bug #1821930 (reported by Lee):
- * If the response doesn't look like HTTP,
- * tell the client and log the problem.
- *
- * Revision 1.172 2008/04/16 16:38:21 fabiankeil
- * Don't pass the whole csp structure to flush_socket()
- * when it only needs a file descriptor and a buffer.
- *
- * Revision 1.171 2008/03/27 18:27:25 fabiankeil
- * Remove kill-popups action.
- *
- * Revision 1.170 2008/03/06 16:33:46 fabiankeil
- * If limit-connect isn't used, don't limit CONNECT requests to port 443.
- *
- * Revision 1.169 2008/03/04 18:30:39 fabiankeil
- * Remove the treat-forbidden-connects-like-blocks action. We now
- * use the "blocked" page for forbidden CONNECT requests by default.
- *
- * Revision 1.168 2008/03/02 12:25:25 fabiankeil
- * Also use shiny new connect_port_is_forbidden() in jcc.c.
- *
- * Revision 1.167 2008/02/23 16:57:12 fabiankeil
- * Rename url_actions() to get_url_actions() and let it
- * use the standard parameter ordering.
- *
- * Revision 1.166 2008/02/23 16:33:43 fabiankeil
- * Let forward_url() use the standard parameter ordering
- * and mark its second parameter immutable.
- *
- * Revision 1.165 2008/02/02 19:36:56 fabiankeil
- * Remove the "Listening ... for local connections only" log message.
- * Whether or not remote connections are able to reach Privoxy is up
- * to the operating system.
- *
- * Revision 1.164 2007/12/16 18:32:46 fabiankeil
- * Prevent the log messages for CONNECT requests to unacceptable
- * ports from printing the limit-connect argument as [null] if
- * limit-connect hasn't been explicitly enabled.
- *
- * Revision 1.163 2007/12/13 01:47:11 david__schmidt
- * Make sure all console-mode apps get a usage() instance
- *
- * Revision 1.162 2007/12/06 17:54:57 fabiankeil
- * Reword NO_SERVER_DATA_RESPONSE to make it harder
- * to misunderstand what the message is all about.
- *
- * Revision 1.161 2007/12/04 19:44:22 fabiankeil
- * Unbreak trustfile which previously didn't work without
- * FEATURE_TOGGLE. Fixes BR#1843585, reported by Lee.
- *
- * Revision 1.160 2007/11/29 18:00:29 fabiankeil
- * Plug memory leak. Spotted by Valgrind, triggered by
- * Privoxy-Regression-Test feeding proxyfuzz.py.
- *
- * Revision 1.159 2007/11/24 14:34:09 fabiankeil
- * In the HTTP snipplets, refer to the client as client.
- *
- * Revision 1.158 2007/11/11 16:44:17 fabiankeil
- * Emit a log message when activating the MS IIS5 hack.
- *
- * Revision 1.157 2007/11/03 17:34:49 fabiankeil
- * Log the "weak randomization factor" warning only
- * once for mingw32 and provide some more details.
- *
- * Revision 1.156 2007/11/01 18:20:58 fabiankeil
- * Initialize log module after initializing mutexes, future
- * deadlocks in that code should now work cross-platform.
- *
- * Revision 1.155 2007/10/23 20:12:45 fabiankeil
- * Fix first CSUCCEED line to end in \r\n as required by RFC1945.
- * Reported by Bert van Leeuwen in BR#1818808.
- *
- * Revision 1.154 2007/10/19 17:00:08 fabiankeil
- * Downgrade "Flushing header and buffers" message to LOG_LEVEL_INFO.
- *
- * Revision 1.153 2007/10/14 14:12:41 fabiankeil
- * When in daemon mode, close stderr after the configuration file has been
- * parsed the first time. If logfile isn't set, stop logging. Fixes BR#897436.
- *
- * Revision 1.152 2007/10/04 18:03:34 fabiankeil
- * - Fix a crash when parsing invalid requests whose first header
- * is rejected by get_header(). Regression (re?)introduced
- * in r1.143 by yours truly.
- * - Move ACTION_VANILLA_WAFER handling into parsers.c's
- * client_cookie_adder() to make sure send-vanilla-wafer can be
- * controlled through tags (and thus regression-tested).
- *
- * Revision 1.151 2007/09/29 10:21:16 fabiankeil
- * - Move get_filter_function() from jcc.c to filters.c
- * so the filter functions can be static.
- * - Don't bother filtering body-less responses.
- *
- * Revision 1.150 2007/09/28 16:39:29 fabiankeil
- * Execute content filters through execute_content_filter().
- *
- * Revision 1.149 2007/09/04 15:08:48 fabiankeil
- * Initialize req to NULL to make sure it's defined if the
- * first read_socket() call fails. Reported by icmp30.
- *
- * Revision 1.148 2007/08/26 16:47:13 fabiankeil
- * Add Stephen Gildea's --pre-chroot-nslookup patch [#1276666],
- * extensive comments moved to user manual.
- *
- * Revision 1.147 2007/08/25 14:42:40 fabiankeil
- * Don't crash if a broken header filter wiped out the request line.
- *
- * Revision 1.146 2007/08/20 17:09:32 fabiankeil
- * Fix byte_count calculation in case of flushes
- * and don't parse the server headers a second time.
- *
- * Revision 1.145 2007/08/19 13:13:31 fabiankeil
- * - If there's a connection problem after we already forwarded
- * parts of the original content, just hang up. Fixes BR#1776724.
- * - Fix warnings about unused code on mingw32.
- * - In case of flushes, calculate the byte count
- * less incorrectly (I think).
- *
- * Revision 1.144 2007/08/11 14:43:22 fabiankeil
- * Add some more prototypes for static functions.
- *
- * Revision 1.143 2007/08/05 13:58:19 fabiankeil
- * Comment out request_contains_null_bytes() until it's used again.
- *
- * Revision 1.142 2007/08/05 13:50:26 fabiankeil
- * #1763173 from Stefan Huehner: s@const static@static const@
- * and declare some more functions static.
- *
- * Revision 1.141 2007/08/04 09:56:23 fabiankeil
- * - Log rejected CONNECT requests with LOG_LEVEL_INFO
- * and explain why they were rejected in the first place.
- * - Fix the LOG_LEVEL_CLF message for crunches of unallowed
- * CONNECT requests. The request line was missing.
- * - Add two more XXX reminders as we don't have enough already.
- *
- * Revision 1.140 2007/07/21 11:51:36 fabiankeil
- * As Hal noticed, checking dispatch_cgi() as the last cruncher
- * looks like a bug if CGI requests are blocked unintentionally,
- * so don't do it unless the user enabled the new config option
- * "allow-cgi-request-crunching".
- *
- * Revision 1.139 2007/07/14 07:46:41 fabiankeil
- * - Allow to rewrite the request destination behind the client's back.
- * - Turn the weird-looking unconditional for loop that
- * reads the client request into a conditional while loop.
- * Move the stuff that only runs once out of the loop.
- * - Move parts of chat(), server_content_type() and the
- * necessary stuff to fix BR#1750917 into get_filter_function().
- *
- * Revision 1.138 2007/06/03 18:45:18 fabiankeil
- * Temporary workaround for BR#1730105.
- *
- * Revision 1.137 2007/06/01 18:16:36 fabiankeil
- * Use the same mutex for gethostbyname() and gethostbyaddr() to prevent
- * deadlocks and crashes on OpenBSD and possibly other OS with neither
- * gethostbyname_r() nor gethostaddr_r(). Closes BR#1729174.
- * Thanks to Ralf Horstmann for report and solution.
- *
- * Revision 1.136 2007/06/01 16:41:11 fabiankeil
- * Add forward-override{} to change the forwarding settings through
- * action sections. This is mainly interesting to forward different
- * clients differently (for example based on User-Agent or request
- * origin).
- *
- * Revision 1.135 2007/05/24 17:03:50 fabiankeil
- * - Let usage() mention the --chroot parameter.
- * - Use read_socket() consistently and always leave
- * the last buffer byte alone, even in cases where
- * null termination (currently) doesn't matter.
- *
- * Revision 1.134 2007/05/16 14:59:46 fabiankeil
- * - Fix config file loading on Unix if no config file is specified.
- * Since r1.97 Privoxy would always interpret the last argument as
- * config file, even if it's a valid command line option.
- * - Abort in case of unrecognized command line options. Closes #1719696.
- * - Remove a bunch of unnecessary strcpy() calls (yay for c&p without thinking).
- * - Replace the remaining strcpy() and strcat() calls with strlcpy() and strcat().
- *
- * Revision 1.133 2007/05/04 11:23:19 fabiankeil
- * - Don't rerun crunchers that only depend on the request URL.
- * - Don't count redirects and CGI requests as "blocked requests".
- *
- * Revision 1.132 2007/04/25 15:15:17 fabiankeil
- * Support crunching based on tags created by server-header taggers.
- *
- * Revision 1.131 2007/04/22 13:24:50 fabiankeil
- * Make HTTP snippets static (again). Add a Content-Type for those
- * with content so the browser doesn't guess it based on the URL.
- *
- * Revision 1.130 2007/04/19 13:47:34 fabiankeil
- * Move crunching and request line rebuilding out of chat().
- *
- * Revision 1.129 2007/04/15 16:39:20 fabiankeil
- * Introduce tags as alternative way to specify which
- * actions apply to a request. At the moment tags can be
- * created based on client and server headers.
- *
- * Revision 1.128 2007/03/25 16:55:54 fabiankeil
- * Don't CLF-log CONNECT requests twice.
- *
- * Revision 1.127 2007/03/20 13:53:17 fabiankeil
- * Log the source address for ACL-related connection drops.
- *
- * Revision 1.126 2007/03/17 15:20:05 fabiankeil
- * New config option: enforce-blocks.
- *
- * Revision 1.125 2007/03/09 14:12:00 fabiankeil
- * - Move null byte check into separate function.
- * - Don't confuse the client with error pages
- * if a CONNECT request was already confirmed.
- *
- * Revision 1.124 2007/02/23 14:59:54 fabiankeil
- * Speed up NULL byte escaping and only log the complete
- * NULL byte requests with header debugging enabled.
- *
- * Revision 1.123 2007/02/21 18:42:10 fabiankeil
- * Answer requests that contain NULL bytes with
- * a custom response instead of waiting for more
- * data until the client eventually hangs up.
- *
- * Revision 1.122 2007/02/07 11:12:02 fabiankeil
- * - Move delivery and logging of crunched responses
- * from chat() into send_crunch_response().
- * - Display the reason for generating http_responses.
- * - Log the content length for LOG_LEVEL_CLF correctly
- * (still incorrect for some fixed responses).
- * - Reword an incorrect comment about
- * treat-forbidden-connects-like-blocks violating
- * the specs.
- * - Add some log messages.
- *
- * Revision 1.121 2007/01/27 10:52:56 fabiankeil
- * Move mutex initialization into separate
- * function and exit in case of errors.
- *
- * Revision 1.120 2007/01/26 14:18:42 fabiankeil
- * - Start to reduce chat()'s line count and move
- * parts of it into separate functions.
- * - Add "HTTP/1.1 100 Continue" hack for BR 756734.
- *
- * Revision 1.119 2007/01/25 14:02:30 fabiankeil
- * - Add Proxy-Agent header to HTTP snippets that are
- * supposed to reach HTTP clients only.
- * - Made a few CONNECT log messages more descriptive.
- * - Catch completely empty server responses (as seen
- * with Tor's fake ".noconnect" top level domain).
- * - Use shiny new "forwarding-failed" template for socks errors.
- *
- * Revision 1.118 2007/01/07 07:43:43 joergs
- * AmigaOS4 support added.
- *
- * Revision 1.117 2006/12/31 17:56:37 fabiankeil
- * Added config option accept-intercepted-requests
- * and disabled it by default.
- *
- * Revision 1.116 2006/12/29 19:08:22 fabiankeil
- * Reverted parts of my last commit
- * to keep error handling working.
- *
- * Revision 1.115 2006/12/29 17:38:57 fabiankeil
- * Fixed gcc43 conversion warnings.
- *
- * Revision 1.114 2006/12/27 18:52:02 fabiankeil
- * Fix -pedantic ISO C warning about converting
- * from function pointer to object pointer.
- *
- * Revision 1.113 2006/12/26 17:38:50 fabiankeil
- * Silence compiler warning I introduced with my last commit.
- *
- * Revision 1.112 2006/12/26 17:31:41 fabiankeil
- * Mutex protect rand() if POSIX threading
- * is used, warn the user if that's not possible
- * and stop using it on _WIN32 where it could
- * cause crashes.
- *
- * Revision 1.111 2006/12/23 16:15:06 fabiankeil
- * Don't prevent core dumps by catching SIGABRT.
- * It's rude and makes debugging unreasonable painful.
- *
- * Revision 1.110 2006/12/13 14:52:53 etresoft
- * Fix build failure on MacOS X. Global symbols can be either static or extern, but not both.
- *
- * Revision 1.109 2006/12/06 19:41:40 fabiankeil
- * Privoxy is now able to run as intercepting
- * proxy in combination with any packet filter
- * that does the port redirection. The destination
- * is extracted from the "Host:" header which
- * should be available for nearly all requests.
- *
- * Moved HTTP snipplets into jcc.c.
- * Added error message for gopher proxy requests.
- *
- * Revision 1.108 2006/11/28 15:38:51 fabiankeil
- * Only unlink the pidfile if it's actually used.
- *
- * Change order of interception checks to make
- * it possible to block or redirect requests for
- * the cgi pages.
- *
- * Revision 1.107 2006/11/13 19:05:51 fabiankeil
- * Make pthread mutex locking more generic. Instead of
- * checking for OSX and OpenBSD, check for FEATURE_PTHREAD
- * and use mutex locking unless there is an _r function
- * available. Better safe than sorry.
- *
- * Fixes "./configure --disable-pthread" and should result
- * in less threading-related problems on pthread-using platforms,
- * but it still doesn't fix BR#1122404.
- *
- * Revision 1.106 2006/11/06 19:58:23 fabiankeil
- * Move pthread.h inclusion from jcc.c to jcc.h.
- * Fixes build on x86-freebsd1 (FreeBSD 5.4-RELEASE).
- *
- * Revision 1.105 2006/11/06 14:26:02 fabiankeil
- * Don't exit after receiving the second SIGHUP on Solaris.
- *
- * Fixes BR 1052235, but the same problem may exist on other
- * systems. Once 3.0.6 is out we should use sigset()
- * where available and see if it breaks anything.
- *
- * Revision 1.104 2006/09/23 13:26:38 roro
- * Replace TABs by spaces in source code.
- *
- * Revision 1.103 2006/09/21 12:54:43 fabiankeil
- * Fix +redirect{}. Didn't work with -fast-redirects.
- *
- * Revision 1.102 2006/09/06 13:03:04 fabiankeil
- * Respond with 400 and a short text message
- * if the client tries to use Privoxy as FTP proxy.
- *
- * Revision 1.101 2006/09/06 09:23:37 fabiankeil
- * Make number of retries in case of forwarded-connect problems
- * a config file option (forwarded-connect-retries) and use 0 as
- * default.
- *
- * Revision 1.100 2006/09/03 19:42:59 fabiankeil
- * Set random(3) seed.
- *
- * Revision 1.99 2006/09/02 15:36:42 fabiankeil
- * Follow the OpenBSD port's lead and protect the resolve
- * functions on OpenBSD as well.
- *
- * Revision 1.98 2006/08/24 11:01:34 fabiankeil
- * --user fix. Only use the user as group if no group is specified.
- * Solves BR 1492612. Thanks to Spinor S. and David Laight.
- *
- * Revision 1.97 2006/08/18 15:23:17 david__schmidt
- * Windows service (re-)integration
- *
- * The new args are:
- *
- * --install[:service_name]
- * --uninstall[:service_name]
- * --service
- *
- * They work as follows:
- * --install will create a service for you and then terminate.
- * By default the service name will be "privoxy" (without the quotes).
- * However you can run multiple services if you wish, just by adding
- * a colon and then a name (no spaces).
- *
- * --uninstall follows the exact same rules a --install.
- *
- * --service is used when the program is executed by the service
- * control manager, and in normal circumstances would never be
- * used as a command line argument.
- *
- * Revision 1.96 2006/08/15 20:12:36 david__schmidt
- * Windows service integration
- *
- * Revision 1.95 2006/08/03 02:46:41 david__schmidt
- * Incorporate Fabian Keil's patch work:
-http://www.fabiankeil.de/sourcecode/privoxy/
- *
- * Revision 1.94 2006/07/18 14:48:46 david__schmidt
- * Reorganizing the repository: swapping out what was HEAD (the old 3.1 branch)
- * with what was really the latest development (the v_3_0_branch branch)
- *
- * Revision 1.92.2.16 2005/04/03 20:10:50 david__schmidt
- * Thanks to Jindrich Makovicka for a race condition fix for the log
- * file. The race condition remains for non-pthread implementations.
- * Reference patch #1175720.
- *
- * Revision 1.92.2.15 2004/10/03 12:53:32 david__schmidt
- * Add the ability to check jpeg images for invalid
- * lengths of comment blocks. Defensive strategy
- * against the exploit:
- * Microsoft Security Bulletin MS04-028
- * Buffer Overrun in JPEG Processing (GDI+) Could
- * Allow Code Execution (833987)
- * Enabled with +inspect-jpegs in actions files.
- *
- * Revision 1.92.2.14 2003/12/12 12:52:53 oes
- * - Fixed usage info for non-unix platforms
- * - Fixed small cmdline parsing bug
- *
- * Revision 1.92.2.13 2003/11/27 19:20:27 oes
- * Diagnostics: Now preserve the returncode of pthread_create
- * in errno. Closes BR #775721. Thanks to Geoffrey Hausheer.
- *
- * Revision 1.92.2.12 2003/07/11 11:34:19 oes
- * No longer ignore SIGCHLD. Fixes bug #769381
- *
- * Revision 1.92.2.11 2003/05/14 12:32:02 oes
- * Close jarfile on graceful exit, remove stray line
- *
- * Revision 1.92.2.10 2003/05/08 15:13:46 oes
- * Cosmetics: Killed a warning, a typo and an allocation left at exit
- *
- * Revision 1.92.2.9 2003/04/03 15:08:42 oes
- * No longer rely on non-POSIX.1 extensions of getcwd().
- * Fixes bug #711001
- *
- * Revision 1.92.2.8 2003/03/31 13:12:32 oes
- * Replaced setenv() by posix-compliant putenv()
- * Thanks to Neil McCalden (nmcc AT users.sf.net).
- *
- * Revision 1.92.2.7 2003/03/17 16:48:59 oes
- * Added chroot ability, thanks to patch by Sviatoslav Sviridov
- *
- * Revision 1.92.2.6 2003/03/11 11:55:00 oes
- * Clean-up and extension of improvements for forked mode:
- * - Child's return code now consists of flags RC_FLAG_*
- * - Reporting toggle to parent now properly #ifdef'ed
- * - Children now report blocking to parent. This enables
- * statistics in forked mode
- *
- * Revision 1.92.2.5 2003/03/10 23:45:32 oes
- * Fixed bug #700381: Non-Threaded version now capable of being toggled.
- * Children now report having been toggled through _exit(17), parents
- * watch for that code and toggle themselves if found.
- *
- * Revision 1.92.2.4 2003/03/07 03:41:04 david__schmidt
- * Wrapping all *_r functions (the non-_r versions of them) with
- * mutex semaphores for OSX. Hopefully this will take care of all
- * of those pesky crash reports.
- *
- * Revision 1.92.2.3 2003/02/28 12:53:06 oes
- * Fixed two mostly harmless mem leaks
- *
- * Revision 1.92.2.2 2002/11/20 14:37:47 oes
- * Fix: Head of global clients list now initialized to NULL
- *
- * Revision 1.92.2.1 2002/09/25 14:52:24 oes
- * Added basic support for OPTIONS and TRACE HTTP methods:
- * - New interceptor direct_response() added in chat().
- * - sed() moved to earlier in the process, so that the
- * Host: header is evaluated before actions and forwarding
- * are decided on.
- *
- * Revision 1.92 2002/05/08 16:00:46 oes
- * Chat's buffer handling:
- * - Fixed bug with unchecked out-of-mem conditions
- * while reading client request & server headers
- * - No longer predict if the buffer limit will be exceeded
- * in the next read -- check add_to_iob's new
- * return code. If buffer couldn't be extended
- * (policy or out-of-mem) while
- * - reading from client: abort
- * - reading server headers: send error page
- * - buffering server body for filter: flush,
- * and if that fails: send error page
- *
- * Revision 1.91 2002/04/08 20:35:58 swa
- * fixed JB spelling
- *
- * Revision 1.90 2002/04/02 14:57:28 oes
- * Made sending wafers independent of FEATURE_COOKIE_JAR
- *
- * Revision 1.89 2002/03/31 17:18:59 jongfoster
- * Win32 only: Enabling STRICT to fix a VC++ compile warning.
- *
- * Revision 1.88 2002/03/27 14:32:43 david__schmidt
- * More compiler warning message maintenance
- *
- * Revision 1.87 2002/03/26 22:29:54 swa
- * we have a new homepage!
- *
- * Revision 1.86 2002/03/25 17:04:55 david__schmidt
- * Workaround for closing the jarfile before load_config() comes around again
- *
- * Revision 1.85 2002/03/24 15:23:33 jongfoster
- * Name changes
- *
- * Revision 1.84 2002/03/24 13:25:43 swa
- * name change related issues
- *
- * Revision 1.83 2002/03/16 23:54:06 jongfoster
- * Adding graceful termination feature, to help look for memory leaks.
- * If you enable this (which, by design, has to be done by hand
- * editing config.h) and then go to http://i.j.b/die, then the program
- * will exit cleanly after the *next* request. It should free all the
- * memory that was used.
- *
- * Revision 1.82 2002/03/13 00:27:05 jongfoster
- * Killing warnings
- *
- * Revision 1.81 2002/03/12 01:42:50 oes
- * Introduced modular filters
- *
- * Revision 1.80 2002/03/11 22:07:05 david__schmidt
- * OS/2 port maintenance:
- * - Fixed EMX build - it had decayed a little
- * - Fixed inexplicable crash during FD_ZERO - must be due to a bad macro.
- * substituted a memset for now.
- *
- * Revision 1.79 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 1.78 2002/03/08 21:35:04 oes
- * Added optional group supplement to --user option. Will now use default group of user if no group given
- *
- * Revision 1.77 2002/03/07 03:52:06 oes
- * - Fixed compiler warnings etc
- * - Improved handling of failed DNS lookups
- *
- * Revision 1.76 2002/03/06 22:54:35 jongfoster
- * Automated function-comment nitpicking.
- *
- * Revision 1.75 2002/03/06 10:02:19 oes
- * Fixed stupid bug when --user was not given
- *
- * Revision 1.74 2002/03/06 00:49:31 jongfoster
- * Fixing warning on Windows
- * Making #ifdefs that refer to the same variable consistently
- * use #ifdef unix rather than mixing #ifdef unix & #ifndef OS2
- *
- * Revision 1.73 2002/03/05 23:57:30 hal9
- * Stray character 's' on line 1618 was breaking build.
- *
- * Revision 1.72 2002/03/05 21:33:45 david__schmidt
- * - Re-enable OS/2 building after new parms were added
- * - Fix false out of memory report when resolving CGI templates when no IP
- * address is available of failed attempt (a la no such domain)
- *
- * Revision 1.71 2002/03/05 18:13:56 oes
- * Added --user option
- *
- * Revision 1.70 2002/03/05 04:52:42 oes
- * Deleted non-errlog debugging code
- *
- * Revision 1.69 2002/03/04 23:50:00 jongfoster
- * Splitting off bind_port() call into bind_port_helper(), with
- * improved logging.
- *
- * Revision 1.68 2002/03/04 20:17:32 oes
- * Fixed usage info
- *
- * Revision 1.67 2002/03/04 18:18:57 oes
- * - Removed _DEBUG mode
- * - Cleand up cmdline parsing
- * - Introduced --no-daemon, --pidfile options
- * - Cleaned up signal handling:
- * - Terminate cleanly on INT, TERM and ABRT
- * - Schedule logfile for re-opening on HUP
- * - Ignore CHLD and PIPE
- * - Leave the rest with their default handlers
- * - Uniform handler registration
- * - Added usage() function
- * - Played styleguide police
- *
- * Revision 1.66 2002/03/03 15:06:55 oes
- * Re-enabled automatic config reloading
- *
- * Revision 1.65 2002/03/03 14:49:11 oes
- * Fixed CLF logging: Now uses client's original HTTP request
- *
- * Revision 1.64 2002/03/03 09:18:03 joergs
- * Made jumbjuster work on AmigaOS again.
- *
- * Revision 1.63 2002/03/02 04:14:50 david__schmidt
- * Clean up a little CRLF unpleasantness that suddenly appeared
- *
- * Revision 1.62 2002/02/20 23:17:23 jongfoster
- * Detecting some out-of memory conditions and exiting with a log message.
- *
- * Revision 1.61 2002/01/17 21:01:52 jongfoster
- * Moving all our URL and URL pattern parsing code to urlmatch.c.
- *
- * Revision 1.60 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 1.59 2001/12/13 14:07:18 oes
- * Fixed Bug: 503 error page now sent OK
- *
- * Revision 1.58 2001/11/30 23:37:24 jongfoster
- * Renaming the Win32 config file to config.txt - this is almost the
- * same as the corresponding UNIX name "config"
- *
- * Revision 1.57 2001/11/16 00:47:43 jongfoster
- * Changing the tty-disconnection code to use setsid().
- *
- * Revision 1.56 2001/11/13 20:20:54 jongfoster
- * Tabs->spaces, fixing a bug with missing {} around an if()
- *
- * Revision 1.55 2001/11/13 20:14:53 jongfoster
- * Patch for FreeBSD setpgrp() as suggested by Alexander Lazic
- *
- * Revision 1.54 2001/11/07 00:03:14 steudten
- * Give reliable return value if an error
- * occurs not just 0 with new daemon mode.
- *
- * Revision 1.53 2001/11/05 21:41:43 steudten
- * Add changes to be a real daemon just for unix os.
- * (change cwd to /, detach from controlling tty, set
- * process group and session leader to the own process.
- * Add DBG() Macro.
- * Add some fatal-error log message for failed malloc().
- * Add '-d' if compiled with 'configure --with-debug' to
- * enable debug output.
- *
- * Revision 1.52 2001/10/26 20:11:20 jongfoster
- * Fixing type mismatch
- *
- * Revision 1.51 2001/10/26 17:38:28 oes
- * Cosmetics
- *
- * Revision 1.50 2001/10/25 03:40:48 david__schmidt
- * Change in porting tactics: OS/2's EMX porting layer doesn't allow multiple
- * threads to call select() simultaneously. So, it's time to do a real, live,
- * native OS/2 port. See defines for __EMX__ (the porting layer) vs. __OS2__
- * (native). Both versions will work, but using __OS2__ offers multi-threading.
- *
- * Revision 1.49 2001/10/23 21:41:35 jongfoster
- * Added call to initialize the (statically-allocated of course)
- * "out of memory" CGI response.
- *
- * Revision 1.48 2001/10/10 19:56:46 jongfoster
- * Moving some code that wasn't cookie-related out of an #ifdef
- * FEATURE_COOKIE_JAR
- *
- * Revision 1.47 2001/10/10 16:44:36 oes
- * Added CONNECT destination port limitation check
- *
- * Revision 1.46 2001/10/08 15:17:41 oes
- * Re-enabled SSL forwarding
- *
- * Revision 1.45 2001/10/07 15:42:11 oes
- * Replaced 6 boolean members of csp with one bitmap (csp->flags)
- *
- * Moved downgrading of the HTTP version from parse_http_request to
- * chat(), since we can't decide if it is necessary before we have
- * determined the actions for the URL. The HTTP command is now
- * *always* re-built so the repairs need no longer be special-cased.
- *
- * filter_popups now gets a csp pointer so it can raise the new
- * CSP_FLAG_MODIFIED flag.
- *
- * Bugfix
- *
- * Added configurable size limit for the IOB. If the IOB grows so
- * large that the next read would exceed the limit, the header
- * is generated, and the header & unfiltered buffer are flushed
- * to the client. Chat then continues in non-buffering,
- * non-filtering body mode.
- *
- * Revision 1.44 2001/10/02 18:13:57 oes
- * Ooops
- *
- * Revision 1.43 2001/10/02 15:32:13 oes
- * Moved generation of hdr
- *
- * Revision 1.42 2001/09/21 23:02:02 david__schmidt
- * Cleaning up 2 compiler warnings on OS/2.
- *
- * Revision 1.41 2001/09/16 17:05:14 jongfoster
- * Removing unused #include showarg.h
- *
- * Revision 1.40 2001/09/16 15:41:45 jongfoster
- * Fixing signed/unsigned comparison warning.
- *
- * Revision 1.39 2001/09/16 13:21:27 jongfoster
- * Changes to use new list functions.
- *
- * Revision 1.38 2001/09/16 13:01:46 jongfoster
- * Removing redundant function call that zeroed zalloc()'d memory.
- *
- * Revision 1.37 2001/09/10 11:12:24 oes
- * Deleted unused variable
- *
- * Revision 1.36 2001/09/10 10:56:15 oes
- * Silenced compiler warnings
- *
- * Revision 1.35 2001/07/31 14:44:22 oes
- * Deleted unused size parameter from filter_popups()
- *
- * Revision 1.34 2001/07/30 22:08:36 jongfoster
- * Tidying up #defines:
- * - All feature #defines are now of the form FEATURE_xxx
- * - Permanently turned off WIN_GUI_EDIT
- * - Permanently turned on WEBDAV and SPLIT_PROXY_ARGS
- *
- * Revision 1.33 2001/07/29 19:32:00 jongfoster
- * Renaming _main() [mingw32 only] to real_main(), for ANSI compliance.
- *
- * Revision 1.32 2001/07/29 18:47:05 jongfoster
- * Adding missing #include "loadcfg.h"
- *
- * Revision 1.31 2001/07/29 12:17:48 oes
- * Applied pthread fix by Paul Lieverse
- *
- * Revision 1.30 2001/07/25 22:57:13 jongfoster
- * __BEOS__ no longer overrides FEATURE_PTHREAD.
- * This is because FEATURE_PTHREAD will soon be widely used, so I
- * want to keep it simple.
- *
- * Revision 1.29 2001/07/24 12:47:06 oes
- * Applied BeOS support update by Eugenia
- *
- * Revision 1.28 2001/07/23 13:26:12 oes
- * Fixed bug in popup-killing for the first read that caused binary garbage to be sent between headers and body
- *
- * Revision 1.27 2001/07/19 19:09:47 haroon
- * - Added code to take care of the situation where while processing the first
- * server response (which includes the server header), after finding the end
- * of the headers we were not looking past the end of the headers for
- * content modification. I enabled it for filter_popups.
- * Someone else should look to see if other similar operations should be
- * done to the discarded portion of the buffer.
- *
- * Note 2001/07/20: No, the other content modification mechanisms will process
- * the whole iob later anyway. --oes
- *
- * Revision 1.26 2001/07/18 12:31:36 oes
- * cosmetics
- *
- * Revision 1.25 2001/07/15 19:43:49 jongfoster
- * Supports POSIX threads.
- * Also removed some unused #includes.
- *
- * Revision 1.24 2001/07/13 14:00:40 oes
- * - Generic content modification scheme:
- * Each feature has its own applicability flag that is set
- * from csp->action->flags.
- * Replaced the "filtering" int flag , by a function pointer
- * "content_filter" to the function that will do the content
- * modification. If it is != NULL, the document will be buffered
- * and processed through *content_filter, which must set
- * csp->content_length and return a modified copy of the body
- * or return NULL (on failiure).
- * - Changed csp->is_text to the more generic bitmap csp->content_type
- * which can currently take the valued CT_TEXT or CT_GIF
- * - Reformatting etc
- * - Removed all #ifdef PCRS
- *
- * Revision 1.23 2001/07/02 02:28:25 iwanttokeepanon
- * Added "#ifdef ACL_FILES" conditional compilation to line 1291 to exclude
- * the `block_acl' call. This prevents a compilation error when the user
- * does not wish to use the "ACL" feature.
- *
- * Revision 1.22 2001/06/29 21:45:41 oes
- * Indentation, CRLF->LF, Tab-> Space
- *
- * Revision 1.21 2001/06/29 13:29:36 oes
- * - Cleaned up, improved comments
- * - Unified all possible interceptors (CGI,
- * block, trust, fast_redirect) in one
- * place, with one (CGI) answer generation
- * mechansim. Much clearer now.
- * - Removed the GIF image generation, which
- * is now done in filters.c:block_url()
- * - Made error conditions like domain lookup
- * failiure or (various) problems while talking
- * to the server use cgi.c:error_response()
- * instead of generating HTML/HTTP in chat() (yuck!)
- * - Removed logentry from cancelled commit
- *
- * Revision 1.20 2001/06/09 10:55:28 jongfoster
- * Changing BUFSIZ ==> BUFFER_SIZE
- *
- * Revision 1.19 2001/06/07 23:12:52 jongfoster
- * Replacing function pointer in struct gateway with a directly
- * called function forwarded_connect().
- * Replacing struct gateway with struct forward_spec
- *
- * Revision 1.18 2001/06/03 19:12:16 oes
- * introduced new cgi handling
- *
- * Revision 1.17 2001/06/01 20:07:23 jongfoster
- * Now uses action +image-blocker{} rather than config->tinygif
- *
- * Revision 1.16 2001/06/01 18:49:17 jongfoster
- * Replaced "list_share" with "list" - the tiny memory gain was not
- * worth the extra complexity.
- *
- * Revision 1.15 2001/05/31 21:24:47 jongfoster
- * Changed "permission" to "action" throughout.
- * Removed DEFAULT_USER_AGENT - it must now be specified manually.
- * Moved vanilla wafer check into chat(), since we must now
- * decide whether or not to add it based on the URL.
- *
- * Revision 1.14 2001/05/29 20:14:01 joergs
- * AmigaOS bugfix: PCRS needs a lot of stack, stacksize for child threads
- * increased.
- *
- * Revision 1.13 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 1.12 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 1.11 2001/05/26 17:27:53 jongfoster
- * Added support for CLF and fixed LOG_LEVEL_LOG.
- * Also did CRLF->LF fix of my previous patch.
- *
- * Revision 1.10 2001/05/26 15:26:15 jongfoster
- * ACL feature now provides more security by immediately dropping
- * connections from untrusted hosts.
- *
- * Revision 1.9 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 1.8 2001/05/25 22:43:18 jongfoster
- * Fixing minor memory leak and buffer overflow.
- *
- * Revision 1.7 2001/05/25 22:34:30 jongfoster
- * Hard tabs->Spaces
- *
- * Revision 1.6 2001/05/23 00:13:58 joergs
- * AmigaOS support fixed.
- *
- * Revision 1.5 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 1.4 2001/05/21 19:34:01 jongfoster
- * Made failure to bind() a fatal error.
- *
- * Revision 1.3 2001/05/20 01:21:20 jongfoster
- * Version 2.9.4 checkin.
- * - Merged popupfile and cookiefile, and added control over PCRS
- * filtering, in new "permissionsfile".
- * - Implemented LOG_LEVEL_FATAL, so that if there is a configuration
- * file error you now get a message box (in the Win32 GUI) rather
- * than the program exiting with no explanation.
- * - Made killpopup use the PCRS MIME-type checking and HTTP-header
- * skipping.
- * - Removed tabs from "config"
- * - Moved duplicated url parsing code in "loaders.c" to a new funcition.
- * - Bumped up version number.
- *
- * Revision 1.2 2001/05/17 22:34:44 oes
- * - Added hint on GIF char array generation to jcc.c
- * - Cleaned CRLF's from the sources and related files
- * - Repaired logging for REF and FRC
- *
- * Revision 1.1.1.1 2001/05/15 13:58:56 oes
- * Initial import of version 2.9.3 source tree
- *
- *
*********************************************************************/
-\f
+
#include "config.h"
"Connection: close\r\n\r\n"
"Bad request. Privoxy was unable to extract the destination.\r\n";
-/* XXX: should be a template */
-static const char NO_SERVER_DATA_RESPONSE[] =
- "HTTP/1.0 502 Server or forwarder response empty\r\n"
- "Proxy-Agent: Privoxy " VERSION "\r\n"
- "Content-Type: text/plain\r\n"
- "Connection: close\r\n\r\n"
- "Empty server or forwarder response.\r\n"
- "The connection has been closed but Privoxy didn't receive any data.\r\n";
-
/* XXX: should be a template */
static const char INVALID_SERVER_HEADERS_RESPONSE[] =
"HTTP/1.0 502 Server or forwarder response invalid\r\n"
"Connection: close\r\n\r\n"
"Maximum number of open connections reached.\r\n";
-/* XXX: should be a template */
-static const char CONNECTION_TIMEOUT_RESPONSE[] =
+static const char CLIENT_CONNECTION_TIMEOUT_RESPONSE[] =
"HTTP/1.0 504 Connection timeout\r\n"
"Proxy-Agent: Privoxy " VERSION "\r\n"
"Content-Type: text/plain\r\n"
"Connection: close\r\n\r\n"
- "The connection timed out.\r\n";
+ "The connection timed out because the client request didn't arrive in time.\r\n";
/* A function to crunch a response */
typedef struct http_response *(*crunch_func_ptr)(struct client_state *);
}
else if (JB_ERR_OK == get_destination_from_headers(headers, csp->http))
{
+#ifndef FEATURE_EXTENDED_HOST_PATTERNS
/* Split the domain we just got for pattern matching */
init_domain_components(csp->http);
+#endif
return JB_ERR_OK;
}
case RSP_REASON_OUT_OF_MEMORY:
reason = "Out of memory (may mask other reasons)";
break;
+ case RSP_REASON_CONNECTION_TIMEOUT:
+ reason = "Connection timeout";
+ break;
+ case RSP_REASON_NO_SERVER_DATA:
+ reason = "No server data received";
+ break;
default:
reason = "No reason recorded";
break;
{
assert(sfd != JB_INVALID_SOCKET);
assert(NULL != http->host);
+
+ server_connection->sfd = sfd;
server_connection->host = strdup(http->host);
if (NULL == server_connection->host)
{
{
if ((csp->flags & CSP_FLAG_SERVER_CONNECTION_KEEP_ALIVE))
{
- log_error(LOG_LEVEL_CONNECT, "Unsetting keep-alive flag.");
- csp->flags &= ~CSP_FLAG_SERVER_CONNECTION_KEEP_ALIVE;
+ log_error(LOG_LEVEL_CONNECT,
+ "Marking the server socket %d tainted.", csp->sfd);
+ csp->flags |= CSP_FLAG_SERVER_SOCKET_TAINTED;
}
}
{
log_error(LOG_LEVEL_ERROR,
"Stopped waiting for the request line.");
- write_socket(csp->cfd, CONNECTION_TIMEOUT_RESPONSE,
- strlen(CONNECTION_TIMEOUT_RESPONSE));
+ write_socket(csp->cfd, CLIENT_CONNECTION_TIMEOUT_RESPONSE,
+ strlen(CLIENT_CONNECTION_TIMEOUT_RESPONSE));
return NULL;
}
{
log_error(LOG_LEVEL_ERROR,
"Stopped grabbing the client headers.");
+ destroy_list(headers);
return JB_ERR_PARSE;
}
struct http_request *http = csp->http;
jb_err err;
+#ifdef FEATURE_CONNECTION_KEEP_ALIVE
+ if ((csp->config->feature_flags & RUNTIME_FEATURE_CONNECTION_KEEP_ALIVE)
+ && (!strcmpic(csp->http->ver, "HTTP/1.1"))
+ && (csp->http->ssl == 0))
+ {
+ /* Assume persistence until further notice */
+ csp->flags |= CSP_FLAG_CLIENT_CONNECTION_KEEP_ALIVE;
+ }
+#endif /* def FEATURE_CONNECTION_KEEP_ALIVE */
+
err = sed(csp, FILTER_CLIENT_HEADERS);
if (JB_ERR_OK != err)
{
return JB_ERR_PARSE;
}
+#ifdef FEATURE_CONNECTION_KEEP_ALIVE
+ if ((csp->flags & CSP_FLAG_CLIENT_CONNECTION_KEEP_ALIVE))
+ {
+ if (csp->iob->cur[0] != '\0')
+ {
+ csp->flags |= CSP_FLAG_SERVER_SOCKET_TAINTED;
+ if (strcmpic(csp->http->gpc, "GET")
+ && strcmpic(csp->http->gpc, "HEAD")
+ && strcmpic(csp->http->gpc, "TRACE")
+ && strcmpic(csp->http->gpc, "OPTIONS")
+ && strcmpic(csp->http->gpc, "DELETE"))
+ {
+ /* XXX: this is an incomplete hack */
+ csp->flags &= ~CSP_FLAG_CLIENT_REQUEST_COMPLETELY_READ;
+ log_error(LOG_LEVEL_CONNECT,
+ "There might be a request body. The connection will not be kept alive.");
+ }
+ else
+ {
+ /* XXX: and so is this */
+ csp->flags |= CSP_FLAG_CLIENT_REQUEST_COMPLETELY_READ;
+ log_error(LOG_LEVEL_CONNECT,
+ "Possible pipeline attempt detected. The connection will not "
+ "be kept alive and we will only serve the first request.");
+ /* Nuke the pipelined requests from orbit, just to be sure. */
+ csp->iob->buf[0] = '\0';
+ csp->iob->eod = csp->iob->cur = csp->iob->buf;
+ }
+ }
+ else
+ {
+ csp->flags |= CSP_FLAG_CLIENT_REQUEST_COMPLETELY_READ;
+ log_error(LOG_LEVEL_CONNECT, "Complete client request received.");
+ }
+ }
+#endif /* def FEATURE_CONNECTION_KEEP_ALIVE */
+
return JB_ERR_OK;
}
if (fwd->type != SOCKS_NONE)
{
/* Socks error. */
- rsp = error_response(csp, "forwarding-failed", errno);
+ rsp = error_response(csp, "forwarding-failed");
}
else if (errno == EINVAL)
{
- rsp = error_response(csp, "no-such-domain", errno);
+ rsp = error_response(csp, "no-such-domain");
}
else
{
- rsp = error_response(csp, "connect-failed", errno);
+ rsp = error_response(csp, "connect-failed");
log_error(LOG_LEVEL_CONNECT, "connect to: %s failed: %E",
http->hostport);
}
log_error(LOG_LEVEL_CONNECT,
"write header to: %s failed: %E", http->hostport);
- rsp = error_response(csp, "connect-failed", errno);
+ rsp = error_response(csp, "connect-failed");
if (rsp)
{
send_crunch_response(csp, rsp);
log_error(LOG_LEVEL_CONNECT, "to %s successful", http->hostport);
+ csp->server_connection.request_sent = time(NULL);
+
/* we're finished with the client's header */
freez(hdr);
#else
FD_ZERO(&rfds);
#endif
- FD_SET(csp->cfd, &rfds);
+#ifdef FEATURE_CONNECTION_KEEP_ALIVE
+ if ((csp->flags & CSP_FLAG_CLIENT_REQUEST_COMPLETELY_READ))
+ {
+ maxfd = csp->sfd;
+ }
+ else
+#endif /* def FEATURE_CONNECTION_KEEP_ALIVE */
+ {
+ FD_SET(csp->cfd, &rfds);
+ }
+
FD_SET(csp->sfd, &rfds);
#ifdef FEATURE_CONNECTION_KEEP_ALIVE
"Didn't receive data in time: %s", http->url);
if ((byte_count == 0) && (http->ssl == 0))
{
- write_socket(csp->cfd, CONNECTION_TIMEOUT_RESPONSE,
- strlen(CONNECTION_TIMEOUT_RESPONSE));
+ send_crunch_response(csp, error_response(csp, "connection-timeout"));
}
mark_server_socket_tainted(csp);
return;
*/
if (FD_ISSET(csp->sfd, &rfds))
{
- fflush(0);
+#ifdef FEATURE_CONNECTION_KEEP_ALIVE
+ if (!socket_is_still_usable(csp->cfd))
+ {
+#ifdef _WIN32
+ log_error(LOG_LEVEL_CONNECT,
+ "The server still wants to talk, but the client may already have hung up on us.");
+#else
+ log_error(LOG_LEVEL_CONNECT,
+ "The server still wants to talk, but the client hung up on us.");
+ mark_server_socket_tainted(csp);
+ return;
+#endif /* def _WIN32 */
+ }
+#endif /* def FEATURE_CONNECTION_KEEP_ALIVE */
+
+ fflush(NULL);
len = read_socket(csp->sfd, buf, sizeof(buf) - 1);
if (len < 0)
/* Did we actually get anything? */
if (NULL == csp->headers->first)
{
- log_error(LOG_LEVEL_ERROR, "Empty server or forwarder response.");
+ log_error(LOG_LEVEL_ERROR,
+ "Empty server or forwarder response received on socket %d.", csp->sfd);
log_error(LOG_LEVEL_CLF, "%s - - [%T] \"%s\" 502 0", csp->ip_addr_str, http->cmd);
- write_socket(csp->cfd, NO_SERVER_DATA_RESPONSE, strlen(NO_SERVER_DATA_RESPONSE));
+ send_crunch_response(csp, error_response(csp, "no-server-data"));
free_http_request(http);
mark_server_socket_tainted(csp);
return;
log_error(LOG_LEVEL_FATAL, "Out of memory parsing server header");
}
+ csp->server_connection.response_received = time(NULL);
+
if (crunch_response_triggered(csp, crunchers_light))
{
/*
log_error(LOG_LEVEL_CLF, "%s - - [%T] \"%s\" 200 %llu",
csp->ip_addr_str, http->ocmd, csp->content_length);
+
+ csp->server_connection.timestamp = time(NULL);
}
#ifdef FEATURE_CONNECTION_KEEP_ALIVE
static int monitor_thread_running = 0;
int continue_chatting = 0;
+ unsigned int latency = 0;
+
do
{
chat(csp);
+ if ((csp->flags & CSP_FLAG_SERVER_CONNECTION_KEEP_ALIVE)
+ && !(csp->flags & CSP_FLAG_SERVER_KEEP_ALIVE_TIMEOUT_SET))
+ {
+ log_error(LOG_LEVEL_CONNECT, "The server didn't specify how long "
+ "the connection will stay open. Assume it's only a second.");
+ csp->server_connection.keep_alive_timeout = 1;
+ }
+
continue_chatting = (csp->config->feature_flags
& RUNTIME_FEATURE_CONNECTION_KEEP_ALIVE)
&& (csp->flags & CSP_FLAG_SERVER_CONNECTION_KEEP_ALIVE)
+ && !(csp->flags & CSP_FLAG_SERVER_SOCKET_TAINTED)
&& (csp->cfd != JB_INVALID_SOCKET)
&& (csp->sfd != JB_INVALID_SOCKET)
- && socket_is_still_usable(csp->sfd);
+ && socket_is_still_usable(csp->sfd)
+ && (latency < csp->server_connection.keep_alive_timeout);
if (continue_chatting)
{
+ unsigned int client_timeout = (unsigned)csp->server_connection.keep_alive_timeout - latency;
log_error(LOG_LEVEL_CONNECT,
"Waiting for the next client request. "
"Keeping the server socket %d to %s open.",
csp->sfd, csp->server_connection.host);
if ((csp->flags & CSP_FLAG_CLIENT_CONNECTION_KEEP_ALIVE)
- && data_is_available(csp->cfd, (int)csp->server_connection.keep_alive_timeout)
+ && data_is_available(csp->cfd, (int)client_timeout)
&& socket_is_still_usable(csp->cfd))
{
log_error(LOG_LEVEL_CONNECT, "Client request arrived in "
{
log_error(LOG_LEVEL_CONNECT,
"No additional client request received in time.");
- if ((csp->config->feature_flags & RUNTIME_FEATURE_CONNECTION_SHARING))
+ if ((csp->config->feature_flags & RUNTIME_FEATURE_CONNECTION_SHARING)
+ && (socket_is_still_usable(csp->sfd)))
{
- remember_connection(csp->sfd, csp->http,
- forward_url(csp, csp->http),
- csp->server_connection.keep_alive_timeout);
+ remember_connection(csp, forward_url(csp, csp->http));
csp->sfd = JB_INVALID_SOCKET;
close_socket(csp->cfd);
csp->cfd = JB_INVALID_SOCKET;
(NULL != config->haddr) ? config->haddr : "INADDR_ANY", config->hport);
default :
- log_error(LOG_LEVEL_FATAL, "can't bind to %s:%d: because %E",
+ log_error(LOG_LEVEL_FATAL, "can't bind to %s:%d: %E",
(NULL != config->haddr) ? config->haddr : "INADDR_ANY", config->hport);
}