-const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.13 2006/12/06 19:50:54 fabiankeil Exp $";
+const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.20 2007/09/02 15:31:20 fabiankeil Exp $";
/*********************************************************************
*
* File : $Source: /cvsroot/ijbswa/current/urlmatch.c,v $
* Purpose : Declares functions to match URLs against URL
* patterns.
*
- * Copyright : Written by and Copyright (C) 2001-2003, 2006 the SourceForge
+ * Copyright : Written by and Copyright (C) 2001-2003, 2006-2007 the SourceForge
* Privoxy team. http://www.privoxy.org/
*
* Based on the Internet Junkbuster originally written
*
* Revisions :
* $Log: urlmatch.c,v $
+ * Revision 1.20 2007/09/02 15:31:20 fabiankeil
+ * Move match_portlist() from filter.c to urlmatch.c.
+ * It's used for url matching, not for filtering.
+ *
+ * Revision 1.19 2007/09/02 13:42:11 fabiankeil
+ * - Allow port lists in url patterns.
+ * - Ditch unused url_spec member pathlen.
+ *
+ * Revision 1.18 2007/07/30 16:42:21 fabiankeil
+ * Move the method check into unknown_method()
+ * and loop through the known methods instead
+ * of using a screen-long OR chain.
+ *
+ * Revision 1.17 2007/04/15 16:39:21 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 1.16 2007/02/13 13:59:24 fabiankeil
+ * Remove redundant log message.
+ *
+ * Revision 1.15 2007/01/28 16:11:23 fabiankeil
+ * Accept WebDAV methods for subversion
+ * in parse_http_request(). Closes FR 1581425.
+ *
+ * Revision 1.14 2007/01/06 14:23:56 fabiankeil
+ * Fix gcc43 warnings. Mark *csp as immutable
+ * for parse_http_url() and url_match().
+ * Replace a sprintf call with snprintf.
+ *
* Revision 1.13 2006/12/06 19:50:54 fabiankeil
* parse_http_url() now handles intercepted
* HTTP request lines as well. Moved parts
}
+/*********************************************************************
+ *
+ * Function : unknown_method
+ *
+ * Description : Checks whether a method is unknown.
+ *
+ * Parameters :
+ * 1 : method = points to a http method
+ *
+ * Returns : TRUE if it's unknown, FALSE otherwise.
+ *
+ *********************************************************************/
+static int unknown_method(const char *method)
+{
+ static const char *known_http_methods[] = {
+ /* Basic HTTP request type */
+ "GET", "HEAD", "POST", "PUT", "DELETE", "OPTIONS", "TRACE", "CONNECT",
+ /* webDAV extensions (RFC2518) */
+ "PROPFIND", "PROPPATCH", "MOVE", "COPY", "MKCOL", "LOCK", "UNLOCK",
+ /*
+ * Microsoft webDAV extension for Exchange 2000. See:
+ * http://lists.w3.org/Archives/Public/w3c-dist-auth/2002JanMar/0001.html
+ * http://msdn.microsoft.com/library/en-us/wss/wss/_webdav_methods.asp
+ */
+ "BCOPY", "BMOVE", "BDELETE", "BPROPFIND", "BPROPPATCH",
+ /*
+ * Another Microsoft webDAV extension for Exchange 2000. See:
+ * http://systems.cs.colorado.edu/grunwald/MobileComputing/Papers/draft-cohen-gena-p-base-00.txt
+ * http://lists.w3.org/Archives/Public/w3c-dist-auth/2002JanMar/0001.html
+ * http://msdn.microsoft.com/library/en-us/wss/wss/_webdav_methods.asp
+ */
+ "SUBSCRIBE", "UNSUBSCRIBE", "NOTIFY", "POLL",
+ /*
+ * Yet another WebDAV extension, this time for
+ * Web Distributed Authoring and Versioning (RFC3253)
+ */
+ "VERSION-CONTROL", "REPORT", "CHECKOUT", "CHECKIN", "UNCHECKOUT",
+ "MKWORKSPACE", "UPDATE", "LABEL", "MERGE", "BASELINE-CONTROL", "MKACTIVITY",
+ NULL
+ };
+ int i;
+
+ for (i = 0; NULL != known_http_methods[i]; i++)
+ {
+ if (0 == strcmpic(method, known_http_methods[i]))
+ {
+ return FALSE;
+ }
+ }
+
+ return TRUE;
+
+}
+
+
/*********************************************************************
*
* Function : parse_http_request
const struct client_state *csp)
{
char *buf;
- char *v[10];
+ char *v[10]; /* XXX: Why 10? We should only need three. */
int n;
jb_err err;
int is_connect = 0;
n = ssplit(buf, " \r\n", v, SZ(v), 1, 1);
if (n != 3)
{
- log_error(LOG_LEVEL_ERROR, "Trouble ssplitting: %s", buf);
free(buf);
return JB_ERR_PARSE;
}
- /* this could be a CONNECT request */
- if (strcmpic(v[0], "connect") == 0)
- {
- /* Secure */
- is_connect = 1;
- }
- /* or it could be any other basic HTTP request type */
- else if ((0 == strcmpic(v[0], "get"))
- || (0 == strcmpic(v[0], "head"))
- || (0 == strcmpic(v[0], "post"))
- || (0 == strcmpic(v[0], "put"))
- || (0 == strcmpic(v[0], "delete"))
- || (0 == strcmpic(v[0], "options"))
- || (0 == strcmpic(v[0], "trace"))
-
- /* or a webDAV extension (RFC2518) */
- || (0 == strcmpic(v[0], "propfind"))
- || (0 == strcmpic(v[0], "proppatch"))
- || (0 == strcmpic(v[0], "move"))
- || (0 == strcmpic(v[0], "copy"))
- || (0 == strcmpic(v[0], "mkcol"))
- || (0 == strcmpic(v[0], "lock"))
- || (0 == strcmpic(v[0], "unlock"))
-
- /* Or a Microsoft webDAV extension for Exchange 2000. See: */
- /* http://lists.w3.org/Archives/Public/w3c-dist-auth/2002JanMar/0001.html */
- /* http://msdn.microsoft.com/library/en-us/wss/wss/_webdav_methods.asp */
- || (0 == strcmpic(v[0], "bcopy"))
- || (0 == strcmpic(v[0], "bmove"))
- || (0 == strcmpic(v[0], "bdelete"))
- || (0 == strcmpic(v[0], "bpropfind"))
- || (0 == strcmpic(v[0], "bproppatch"))
-
- /* Or another Microsoft webDAV extension for Exchange 2000. See: */
- /* http://systems.cs.colorado.edu/grunwald/MobileComputing/Papers/draft-cohen-gena-p-base-00.txt */
- /* http://lists.w3.org/Archives/Public/w3c-dist-auth/2002JanMar/0001.html */
- /* http://msdn.microsoft.com/library/en-us/wss/wss/_webdav_methods.asp */
- || (0 == strcmpic(v[0], "subscribe"))
- || (0 == strcmpic(v[0], "unsubscribe"))
- || (0 == strcmpic(v[0], "notify"))
- || (0 == strcmpic(v[0], "poll"))
- )
- {
- /* Normal */
- is_connect = 0;
- }
- else
+ /*
+ * Fail in case of unknown methods
+ * which we might not handle correctly.
+ *
+ * XXX: There should be a config option
+ * to forward requests with unknown methods
+ * anyway. Most of them don't need special
+ * steps.
+ */
+ if (unknown_method(v[0]))
{
- /* Unknown HTTP method */
log_error(LOG_LEVEL_ERROR, "Unknown HTTP method detected: %s", v[0]);
free(buf);
return JB_ERR_PARSE;
}
+ if (strcmpic(v[0], "CONNECT") == 0)
+ {
+ is_connect = 1;
+ }
+
err = parse_http_url(v[1], http, csp);
if (err)
{
jb_err create_url_spec(struct url_spec * url, const char * buf)
{
char *p;
+ int errcode;
+ size_t errlen;
+ char rebuf[BUFFER_SIZE];
assert(url);
assert(buf);
return JB_ERR_MEMORY;
}
- if ((p = strchr(buf, '/')) != NULL)
+ /* Is it tag pattern? */
+ if (0 == strncmpic("TAG:", url->spec, 4))
+ {
+ if (NULL == (url->tag_regex = zalloc(sizeof(*url->tag_regex))))
+ {
+ freez(url->spec);
+ return JB_ERR_MEMORY;
+ }
+
+ /* buf + 4 to skip "TAG:" */
+ errcode = regcomp(url->tag_regex, buf + 4, (REG_EXTENDED|REG_NOSUB|REG_ICASE));
+ if (errcode)
+ {
+ errlen = regerror(errcode, url->preg, rebuf, sizeof(rebuf));
+ if (errlen > (sizeof(rebuf) - 1))
+ {
+ errlen = sizeof(rebuf) - 1;
+ }
+ rebuf[errlen] = '\0';
+
+ log_error(LOG_LEVEL_ERROR, "error compiling %s: %s", url->spec, rebuf);
+
+ freez(url->spec);
+ regfree(url->tag_regex);
+ freez(url->tag_regex);
+
+ return JB_ERR_PARSE;
+ }
+ return JB_ERR_OK;
+ }
+
+ /* Only reached for URL patterns */
+ p = strchr(buf, '/');
+ if (NULL != p)
{
- if (NULL == (url->path = strdup(p)))
+ url->path = strdup(p);
+ if (NULL == url->path)
{
freez(url->spec);
return JB_ERR_MEMORY;
}
- url->pathlen = strlen(url->path);
*p = '\0';
}
else
{
- url->path = NULL;
- url->pathlen = 0;
+ url->path = NULL;
}
if (url->path)
{
- int errcode;
- char rebuf[BUFFER_SIZE];
-
if (NULL == (url->preg = zalloc(sizeof(*url->preg))))
{
freez(url->spec);
(REG_EXTENDED|REG_NOSUB|REG_ICASE));
if (errcode)
{
- size_t errlen = regerror(errcode,
- url->preg, rebuf, sizeof(rebuf));
+ errlen = regerror(errcode, url->preg, rebuf, sizeof(rebuf));
if (errlen > (sizeof(rebuf) - (size_t)1))
{
return JB_ERR_PARSE;
}
}
- if ((p = strchr(buf, ':')) == NULL)
+
+ p = strchr(buf, ':');
+ if (NULL != p)
{
- url->port = 0;
+ *p++ = '\0';
+ url->port_list = strdup(p);
+ if (NULL == url->port_list)
+ {
+ return JB_ERR_MEMORY;
+ }
}
else
{
- *p++ = '\0';
- url->port = atoi(p);
+ url->port_list = NULL;
}
if (buf[0] != '\0')
freez(url->dbuffer);
freez(url->dvec);
freez(url->path);
+ freez(url->port_list);
if (url->preg)
{
regfree(url->preg);
freez(url->preg);
}
+ if (url->tag_regex)
+ {
+ regfree(url->tag_regex);
+ freez(url->tag_regex);
+ }
}
* 1 : pattern = a URL pattern
* 2 : url = URL to match
*
- * Returns : 0 iff the URL matches the pattern, else nonzero.
+ * Returns : Nonzero if the URL matches the pattern, else 0.
*
*********************************************************************/
int url_match(const struct url_spec *pattern,
const struct http_request *url)
{
- return ((pattern->port == 0) || (pattern->port == url->port))
- && ((pattern->dbuffer == NULL) || (domain_match(pattern, url) == 0))
- && ((pattern->path == NULL) ||
- (regexec(pattern->preg, url->path, 0, NULL, 0) == 0)
- );
+ /* XXX: these should probably be functions. */
+#define PORT_MATCHES ((NULL == pattern->port_list) || match_portlist(pattern->port_list, url->port))
+#define DOMAIN_MATCHES ((NULL == pattern->dbuffer) || (0 == domain_match(pattern, url)))
+#define PATH_MATCHES ((NULL == pattern->path) || (0 == regexec(pattern->preg, url->path, 0, NULL, 0)))
+
+ if (pattern->tag_regex != NULL)
+ {
+ /* It's a tag pattern and shouldn't be matched against URLs */
+ return 0;
+ }
+
+ return (PORT_MATCHES && DOMAIN_MATCHES && PATH_MATCHES);
+
+}
+
+
+/*********************************************************************
+ *
+ * Function : match_portlist
+ *
+ * Description : Check if a given number is covered by a comma
+ * separated list of numbers and ranges (a,b-c,d,..)
+ *
+ * Parameters :
+ * 1 : portlist = String with list
+ * 2 : port = port to check
+ *
+ * Returns : 0 => no match
+ * 1 => match
+ *
+ *********************************************************************/
+int match_portlist(const char *portlist, int port)
+{
+ char *min, *max, *next, *portlist_copy;
+
+ min = next = portlist_copy = strdup(portlist);
+
+ /*
+ * Zero-terminate first item and remember offset for next
+ */
+ if (NULL != (next = strchr(portlist_copy, (int) ',')))
+ {
+ *next++ = '\0';
+ }
+
+ /*
+ * Loop through all items, checking for match
+ */
+ while(min)
+ {
+ if (NULL == (max = strchr(min, (int) '-')))
+ {
+ /*
+ * No dash, check for equality
+ */
+ if (port == atoi(min))
+ {
+ free(portlist_copy);
+ return(1);
+ }
+ }
+ else
+ {
+ /*
+ * This is a range, so check if between min and max,
+ * or, if max was omitted, between min and 65K
+ */
+ *max++ = '\0';
+ if(port >= atoi(min) && port <= (atoi(max) ? atoi(max) : 65535))
+ {
+ free(portlist_copy);
+ return(1);
+ }
+
+ }
+
+ /*
+ * Jump to next item
+ */
+ min = next;
+
+ /*
+ * Zero-terminate next item and remember offset for n+1
+ */
+ if ((NULL != next) && (NULL != (next = strchr(next, (int) ','))))
+ {
+ *next++ = '\0';
+ }
+ }
+
+ free(portlist_copy);
+ return 0;
+
}