-const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.10.2.7 2003/05/17 15:57:24 oes Exp $";
+const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.16 2007/02/13 13:59:24 fabiankeil Exp $";
/*********************************************************************
*
- * File : $Source: /cvsroot/ijbswa/current/Attic/urlmatch.c,v $
+ * File : $Source: /cvsroot/ijbswa/current/urlmatch.c,v $
*
* Purpose : Declares functions to match URLs against URL
* patterns.
*
- * Copyright : Written by and Copyright (C) 2001 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 1.16 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
+ * of parse_http_url()'s code into
+ * init_domain_components() so that it can
+ * be reused in chat().
+ *
+ * Revision 1.12 2006/07/18 14:48:47 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 1.10.2.7 2003/05/17 15:57:24 oes
* - parse_http_url now checks memory allocation failure for
* duplication of "*" URL and rejects "*something" URLs
http->dcount = 0;
}
+/*********************************************************************
+ *
+ * Function : init_domain_components
+ *
+ * Description : Splits the domain name so we can compare it
+ * against wildcards. It used to be part of
+ * parse_http_url, but was separated because the
+ * same code is required in chat in case of
+ * intercepted requests.
+ *
+ * Parameters :
+ * 1 : http = pointer to the http structure to hold elements.
+ *
+ * Returns : JB_ERR_OK on success
+ * JB_ERR_MEMORY on out of memory
+ * JB_ERR_PARSE on malformed command/URL
+ * or >100 domains deep.
+ *
+ *********************************************************************/
+jb_err init_domain_components(struct http_request *http)
+{
+ char *vec[BUFFER_SIZE];
+ size_t size;
+ char *p;
+
+ http->dbuffer = strdup(http->host);
+ if (NULL == http->dbuffer)
+ {
+ return JB_ERR_MEMORY;
+ }
+
+ /* map to lower case */
+ for (p = http->dbuffer; *p ; p++)
+ {
+ *p = (char)tolower((int)(unsigned char)*p);
+ }
+
+ /* split the domain name into components */
+ http->dcount = ssplit(http->dbuffer, ".", vec, SZ(vec), 1, 1);
+
+ if (http->dcount <= 0)
+ {
+ /*
+ * Error: More than SZ(vec) components in domain
+ * or: no components in domain
+ */
+ log_error(LOG_LEVEL_ERROR, "More than SZ(vec) components in domain or none at all.");
+ return JB_ERR_PARSE;
+ }
+
+ /* save a copy of the pointers in dvec */
+ size = (size_t)http->dcount * sizeof(*http->dvec);
+
+ http->dvec = (char **)malloc(size);
+ if (NULL == http->dvec)
+ {
+ return JB_ERR_MEMORY;
+ }
+
+ memcpy(http->dvec, vec, size);
+
+ return JB_ERR_OK;
+}
/*********************************************************************
*
*********************************************************************/
jb_err parse_http_url(const char * url,
struct http_request *http,
- struct client_state *csp)
+ const struct client_state *csp)
{
+ int host_available = 1; /* A proxy can dream. */
+
/*
* Zero out the results structure
*/
url_noproto += 8;
http->ssl = 1;
}
+ else if (*url_noproto == '/')
+ {
+ /*
+ * Short request line without protocol and host.
+ * Most likely because the client's request
+ * was intercepted and redirected into Privoxy.
+ */
+ http->ssl = 0;
+ http->host = NULL;
+ host_available = 0;
+ }
else
{
http->ssl = 0;
}
}
+ if (!host_available)
+ {
+ /* Without host, there is nothing left to do here */
+ return JB_ERR_OK;
+ }
/*
* Split hostport into user/password (ignored), host, port.
/*
* Split domain name so we can compare it against wildcards
*/
-
- {
- char *vec[BUFFER_SIZE];
- size_t size;
- char *p;
-
- http->dbuffer = strdup(http->host);
- if (NULL == http->dbuffer)
- {
- return JB_ERR_MEMORY;
- }
-
- /* map to lower case */
- for (p = http->dbuffer; *p ; p++)
- {
- *p = tolower((int)(unsigned char)*p);
- }
-
- /* split the domain name into components */
- http->dcount = ssplit(http->dbuffer, ".", vec, SZ(vec), 1, 1);
-
- if (http->dcount <= 0)
- {
- /*
- * Error: More than SZ(vec) components in domain
- * or: no components in domain
- */
- return JB_ERR_PARSE;
- }
-
- /* save a copy of the pointers in dvec */
- size = http->dcount * sizeof(*http->dvec);
-
- http->dvec = (char **)malloc(size);
- if (NULL == http->dvec)
- {
- return JB_ERR_MEMORY;
- }
-
- memcpy(http->dvec, vec, size);
- }
-
- return JB_ERR_OK;
+ return init_domain_components(http);
}
*********************************************************************/
jb_err parse_http_request(const char *req,
struct http_request *http,
- struct client_state *csp)
+ const struct client_state *csp)
{
char *buf;
char *v[10];
|| (0 == strcmpic(v[0], "unsubscribe"))
|| (0 == strcmpic(v[0], "notify"))
|| (0 == strcmpic(v[0], "poll"))
+
+ /*
+ * Or yet another WebDAV extension, this time for
+ * Web Distributed Authoring and Versioning (RFC3253)
+ */
+ || (0 == strcmpic(v[0], "version-control"))
+ || (0 == strcmpic(v[0], "report"))
+ || (0 == strcmpic(v[0], "checkout"))
+ || (0 == strcmpic(v[0], "checkin"))
+ || (0 == strcmpic(v[0], "uncheckout"))
+ || (0 == strcmpic(v[0], "mkworkspace"))
+ || (0 == strcmpic(v[0], "update"))
+ || (0 == strcmpic(v[0], "label"))
+ || (0 == strcmpic(v[0], "merge"))
+ || (0 == strcmpic(v[0], "baseline-control"))
+ || (0 == strcmpic(v[0], "mkactivity"))
)
{
/* Normal */
else
{
/* Unknown HTTP method */
+ log_error(LOG_LEVEL_ERROR, "Unknown HTTP method detected: %s", v[0]);
free(buf);
return JB_ERR_PARSE;
}
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;
}
+ /* 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 */
if ((p = strchr(buf, '/')) != NULL)
{
if (NULL == (url->path = strdup(p)))
}
if (url->path)
{
- int errcode;
- char rebuf[BUFFER_SIZE];
-
if (NULL == (url->preg = zalloc(sizeof(*url->preg))))
{
freez(url->spec);
return JB_ERR_MEMORY;
}
- sprintf(rebuf, "^(%s)", url->path);
+ snprintf(rebuf, sizeof(rebuf), "^(%s)", url->path);
errcode = regcomp(url->preg, rebuf,
(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))
{
*/
for (p = url->dbuffer; *p ; p++)
{
- *p = tolower((int)(unsigned char)*p);
+ *p = (char)tolower((int)(unsigned char)*p);
}
/*
/*
* Save a copy of the pointers in dvec
*/
- size = url->dcount * sizeof(*url->dvec);
+ size = (size_t)url->dcount * sizeof(*url->dvec);
url->dvec = (char **)malloc(size);
if (NULL == url->dvec)
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)
{
+ if (pattern->tag_regex != NULL)
+ {
+ /* It's a tag pattern and shouldn't be matched against URLs */
+ return 0;
+ }
+
return ((pattern->port == 0) || (pattern->port == url->port))
&& ((pattern->dbuffer == NULL) || (domain_match(pattern, url) == 0))
&& ((pattern->path == NULL) ||