-const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.30 2008/04/10 04:24:24 fabiankeil Exp $";
+const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.74 2012/12/07 12:49:20 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, 2006-2008 the SourceForge
- * Privoxy team. http://www.privoxy.org/
+ * Copyright : Written by and Copyright (C) 2001-2011
+ * the Privoxy team. http://www.privoxy.org/
*
* Based on the Internet Junkbuster originally written
* by and Copyright (C) 1997 Anonymous Coders and
* or write to the Free Software Foundation, Inc., 59
* Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
- * Revisions :
- * $Log: urlmatch.c,v $
- * Revision 1.30 2008/04/10 04:24:24 fabiankeil
- * Stop duplicating the plain text representation of the path regex
- * (and keeping the copy around). Once the regex is compiled it's no
- * longer useful.
- *
- * Revision 1.29 2008/04/10 04:17:56 fabiankeil
- * In url_match(), check the right member for NULL when determining
- * whether there's a path regex to execute. Looking for a plain-text
- * representation works as well, but it looks "interesting" and that
- * member will be removed soonish anyway.
- *
- * Revision 1.28 2008/04/08 16:07:39 fabiankeil
- * Make it harder to mistake url_match()'s
- * second parameter for an url_spec.
- *
- * Revision 1.27 2008/04/08 15:44:33 fabiankeil
- * Save a bit of memory (and a few cpu cycles) by not bothering to
- * compile slash-only path regexes that don't affect the result.
- *
- * Revision 1.26 2008/04/07 16:57:18 fabiankeil
- * - Use free_url_spec() more consistently.
- * - Let it reset url->dcount just in case.
- *
- * Revision 1.25 2008/04/06 15:18:38 fabiankeil
- * Oh well, rename the --enable-pcre-host-patterns option to
- * --enable-extended-host-patterns as it's not really PCRE syntax.
- *
- * Revision 1.24 2008/04/06 14:54:26 fabiankeil
- * Use PCRE syntax in host patterns when configured
- * with --enable-pcre-host-patterns.
- *
- * Revision 1.23 2008/04/05 12:19:20 fabiankeil
- * Factor compile_host_pattern() out of create_url_spec().
- *
- * Revision 1.22 2008/03/30 15:02:32 fabiankeil
- * SZitify unknown_method().
- *
- * Revision 1.21 2007/12/24 16:34:23 fabiankeil
- * Band-aid (and micro-optimization) that makes it less likely to run out of
- * stack space with overly-complex path patterns. Probably masks the problem
- * reported by Lee in #1856679. Hohoho.
- *
- * 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
- * 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
- * Closes bug #736344
- * - Added a comment to what might look like a bug in
- * create_url_spec (see !bug #736931)
- * - Comment cosmetics
- *
- * Revision 1.10.2.6 2003/05/07 12:39:48 oes
- * Fix typo: Default port for https URLs is 443, not 143.
- * Thanks to Scott Tregear for spotting this one.
- *
- * Revision 1.10.2.5 2003/02/28 13:09:29 oes
- * Fixed a rare double free condition as per Bug #694713
- *
- * Revision 1.10.2.4 2003/02/28 12:57:44 oes
- * Moved freeing of http request structure to its owner
- * as per Dan Price's observations in Bug #694713
- *
- * Revision 1.10.2.3 2002/11/12 16:50:40 oes
- * Fixed memory leak in parse_http_request() reported by Oliver Stoeneberg. Fixes bug #637073
- *
- * Revision 1.10.2.2 2002/09/25 14:53:15 oes
- * Added basic support for OPTIONS and TRACE HTTP methods:
- * parse_http_url now recognizes the "*" URI as well as
- * the OPTIONS and TRACE method keywords.
- *
- * Revision 1.10.2.1 2002/06/06 19:06:44 jongfoster
- * Adding support for proprietary Microsoft WebDAV extensions
- *
- * Revision 1.10 2002/05/12 21:40:37 jongfoster
- * - Removing some unused code
- *
- * Revision 1.9 2002/04/04 00:36:36 gliptak
- * always use pcre for matching
- *
- * Revision 1.8 2002/04/03 23:32:47 jongfoster
- * Fixing memory leak on error
- *
- * Revision 1.7 2002/03/26 22:29:55 swa
- * we have a new homepage!
- *
- * Revision 1.6 2002/03/24 13:25:43 swa
- * name change related issues
- *
- * Revision 1.5 2002/03/13 00:27:05 jongfoster
- * Killing warnings
- *
- * Revision 1.4 2002/03/07 03:46:17 oes
- * Fixed compiler warnings
- *
- * Revision 1.3 2002/03/03 14:51:11 oes
- * Fixed CLF logging: Added ocmd member for client's request to struct http_request
- *
- * Revision 1.2 2002/01/21 00:14:09 jongfoster
- * Correcting comment style
- * Fixing an uninitialized memory bug in create_url_spec()
- *
- * Revision 1.1 2002/01/17 20:53:46 jongfoster
- * Moving all our URL and URL pattern parsing code to the same file - it
- * was scattered around in filters.c, loaders.c and parsers.c.
- *
- * Providing a single, simple url_match(pattern,url) function - rather than
- * the 3-line match routine which was repeated all over the place.
- *
- * Renaming free_url to free_url_spec, since it frees a struct url_spec.
- *
- * Providing parse_http_url() so that URLs can be parsed without faking a
- * HTTP request line for parse_http_request() or repeating the parsing
- * code (both of which were techniques that were actually in use).
- *
- * 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).
- *
- *
*********************************************************************/
-\f
+
#include "config.h"
const char urlmatch_h_rcs[] = URLMATCH_H_VERSION;
+enum regex_anchoring
+{
+ NO_ANCHORING,
+ LEFT_ANCHORED,
+ RIGHT_ANCHORED,
+ RIGHT_ANCHORED_HOST
+};
+static jb_err compile_host_pattern(struct url_spec *url, const char *host_pattern);
/*********************************************************************
*
freez(http->path);
freez(http->ver);
freez(http->host_ip_addr_str);
+#ifndef FEATURE_EXTENDED_HOST_PATTERNS
freez(http->dbuffer);
freez(http->dvec);
http->dcount = 0;
+#endif
}
+#ifndef FEATURE_EXTENDED_HOST_PATTERNS
/*********************************************************************
*
* Function : init_domain_components
* 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.
*
size_t size;
char *p;
- http->dbuffer = strdup(http->host);
- if (NULL == http->dbuffer)
- {
- return JB_ERR_MEMORY;
- }
+ http->dbuffer = strdup_or_die(http->host);
/* map to lower case */
for (p = http->dbuffer; *p ; p++)
{
- *p = (char)tolower((int)(unsigned char)*p);
+ *p = (char)privoxy_tolower(*p);
}
/* split the domain name into components */
- http->dcount = ssplit(http->dbuffer, ".", vec, SZ(vec), 1, 1);
+ http->dcount = ssplit(http->dbuffer, ".", vec, SZ(vec));
if (http->dcount <= 0)
{
/* 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;
- }
+ http->dvec = malloc_or_die(size);
memcpy(http->dvec, vec, size);
return JB_ERR_OK;
}
+#endif /* ndef FEATURE_EXTENDED_HOST_PATTERNS */
+
+
+/*********************************************************************
+ *
+ * Function : url_requires_percent_encoding
+ *
+ * Description : Checks if an URL contains invalid characters
+ * according to RFC 3986 that should be percent-encoded.
+ * Does not verify whether or not the passed string
+ * actually is a valid URL.
+ *
+ * Parameters :
+ * 1 : url = URL to check
+ *
+ * Returns : True in case of valid URLs, false otherwise
+ *
+ *********************************************************************/
+int url_requires_percent_encoding(const char *url)
+{
+ static const char allowed_characters[128] = {
+ '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0',
+ '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0',
+ '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0',
+ '\0', '\0', '\0', '!', '\0', '#', '$', '%', '&', '\'',
+ '(', ')', '*', '+', ',', '-', '.', '/', '0', '1',
+ '2', '3', '4', '5', '6', '7', '8', '9', ':', ';',
+ '\0', '=', '\0', '?', '@', 'A', 'B', 'C', 'D', 'E',
+ 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O',
+ 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y',
+ 'Z', '[', '\0', ']', '\0', '_', '\0', 'a', 'b', 'c',
+ 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
+ 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w',
+ 'x', 'y', 'z', '\0', '\0', '\0', '~', '\0'
+ };
+
+ while (*url != '\0')
+ {
+ const unsigned int i = (unsigned char)*url++;
+ if (i >= sizeof(allowed_characters) || '\0' == allowed_characters[i])
+ {
+ return TRUE;
+ }
+ }
+
+ return FALSE;
+
+}
/*********************************************************************
* Parameters :
* 1 : url = URL (or is it URI?) to break down
* 2 : http = pointer to the http structure to hold elements.
- * Will be zeroed before use. Note that this
- * function sets the http->gpc and http->ver
- * members to NULL.
- * 3 : csp = Current client state (buffers, headers, etc...)
+ * Must be initialized with valid values (like NULLs).
+ * 3 : require_protocol = Whether or not URLs without
+ * protocol are acceptable.
*
* 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 parse_http_url(const char * url,
- struct http_request *http,
- const struct client_state *csp)
+jb_err parse_http_url(const char *url, struct http_request *http, int require_protocol)
{
int host_available = 1; /* A proxy can dream. */
- /*
- * Zero out the results structure
- */
- memset(http, '\0', sizeof(*http));
-
-
/*
* Save our initial URL
*/
- http->url = strdup(url);
- if (http->url == NULL)
- {
- return JB_ERR_MEMORY;
- }
-
+ http->url = strdup_or_die(url);
/*
* Check for * URI. If found, we're done.
- */
+ */
if (*http->url == '*')
{
- if ( NULL == (http->path = strdup("*"))
- || NULL == (http->hostport = strdup("")) )
- {
- return JB_ERR_MEMORY;
- }
+ http->path = strdup_or_die("*");
+ http->hostport = strdup_or_die("");
if (http->url[1] != '\0')
{
return JB_ERR_PARSE;
char *url_noproto;
char *url_path;
- buf = strdup(url);
- if (buf == NULL)
- {
- return JB_ERR_MEMORY;
- }
+ buf = strdup_or_die(url);
/* Find the start of the URL in our scratch space */
url_noproto = buf;
if (strncmpic(url_noproto, "http://", 7) == 0)
{
url_noproto += 7;
- http->ssl = 0;
}
else if (strncmpic(url_noproto, "https://", 8) == 0)
{
+ /*
+ * Should only happen when called from cgi_show_url_info().
+ */
url_noproto += 8;
http->ssl = 1;
}
* Most likely because the client's request
* was intercepted and redirected into Privoxy.
*/
- http->ssl = 0;
http->host = NULL;
host_available = 0;
}
- else
+ else if (require_protocol)
{
- http->ssl = 0;
+ freez(buf);
+ return JB_ERR_PARSE;
}
url_path = strchr(url_noproto, '/');
* https URL in and it's parsed by the function. (When the
* URL is actually retrieved, SSL hides the path part).
*/
- http->path = strdup(http->ssl ? "/" : url_path);
+ http->path = strdup_or_die(http->ssl ? "/" : url_path);
*url_path = '\0';
- http->hostport = strdup(url_noproto);
+ http->hostport = strdup_or_die(url_noproto);
}
else
{
* Repair broken HTTP requests that don't contain a path,
* or CONNECT requests
*/
- http->path = strdup("/");
- http->hostport = strdup(url_noproto);
+ http->path = strdup_or_die("/");
+ http->hostport = strdup_or_die(url_noproto);
}
freez(buf);
-
- if ( (http->path == NULL)
- || (http->hostport == NULL))
- {
- return JB_ERR_MEMORY;
- }
}
if (!host_available)
char *host;
char *port;
- buf = strdup(http->hostport);
- if (buf == NULL)
- {
- return JB_ERR_MEMORY;
- }
+ buf = strdup_or_die(http->hostport);
/* check if url contains username and/or password */
host = strchr(buf, '@');
host = buf;
}
+ /* Move after hostname before port number */
+ if (*host == '[')
+ {
+ /* Numeric IPv6 address delimited by brackets */
+ host++;
+ port = strchr(host, ']');
+
+ if (port == NULL)
+ {
+ /* Missing closing bracket */
+ freez(buf);
+ return JB_ERR_PARSE;
+ }
+
+ *port++ = '\0';
+
+ if (*port == '\0')
+ {
+ port = NULL;
+ }
+ else if (*port != ':')
+ {
+ /* Garbage after closing bracket */
+ freez(buf);
+ return JB_ERR_PARSE;
+ }
+ }
+ else
+ {
+ /* Plain non-escaped hostname */
+ port = strchr(host, ':');
+ }
+
/* check if url contains port */
- port = strchr(host, ':');
if (port != NULL)
{
/* Contains port */
+ char *endptr;
+ long parsed_port;
/* Terminate hostname and point to start of port string */
*port++ = '\0';
- http->port = atoi(port);
+ parsed_port = strtol(port, &endptr, 10);
+ if ((parsed_port <= 0) || (parsed_port > 65535) || (*endptr != '\0'))
+ {
+ log_error(LOG_LEVEL_ERROR, "Invalid port in URL: %s.", url);
+ freez(buf);
+ return JB_ERR_PARSE;
+ }
+ http->port = (int)parsed_port;
}
else
{
http->port = (http->ssl ? 443 : 80);
}
- http->host = strdup(host);
-
- free(buf);
+ http->host = strdup_or_die(host);
- if (http->host == NULL)
- {
- return JB_ERR_MEMORY;
- }
+ freez(buf);
}
- /*
- * Split domain name so we can compare it against wildcards
- */
+#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+ return JB_ERR_OK;
+#else
+ /* Split domain name so we can compare it against wildcards */
return init_domain_components(http);
+#endif /* def FEATURE_EXTENDED_HOST_PATTERNS */
}
*********************************************************************/
static int unknown_method(const char *method)
{
- static const char *known_http_methods[] = {
+ static const char * const known_http_methods[] = {
/* Basic HTTP request type */
"GET", "HEAD", "POST", "PUT", "DELETE", "OPTIONS", "TRACE", "CONNECT",
/* webDAV extensions (RFC2518) */
* 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
* Parameters :
* 1 : req = HTTP request line to break down
* 2 : http = pointer to the http structure to hold elements
- * 3 : csp = Current client state (buffers, headers, etc...)
*
* Returns : JB_ERR_OK on success
- * JB_ERR_MEMORY on out of memory
* JB_ERR_CGI_PARAMS on malformed command/URL
* or >100 domains deep.
*
*********************************************************************/
-jb_err parse_http_request(const char *req,
- struct http_request *http,
- const struct client_state *csp)
+jb_err parse_http_request(const char *req, struct http_request *http)
{
char *buf;
char *v[10]; /* XXX: Why 10? We should only need three. */
int n;
jb_err err;
- int is_connect = 0;
memset(http, '\0', sizeof(*http));
- buf = strdup(req);
- if (buf == NULL)
- {
- return JB_ERR_MEMORY;
- }
+ buf = strdup_or_die(req);
- n = ssplit(buf, " \r\n", v, SZ(v), 1, 1);
+ n = ssplit(buf, " \r\n", v, SZ(v));
if (n != 3)
{
- free(buf);
+ freez(buf);
return JB_ERR_PARSE;
}
if (unknown_method(v[0]))
{
log_error(LOG_LEVEL_ERROR, "Unknown HTTP method detected: %s", v[0]);
- free(buf);
+ freez(buf);
return JB_ERR_PARSE;
}
- if (strcmpic(v[0], "CONNECT") == 0)
+ if (strcmpic(v[2], "HTTP/1.1") && strcmpic(v[2], "HTTP/1.0"))
{
- is_connect = 1;
+ log_error(LOG_LEVEL_ERROR, "The only supported HTTP "
+ "versions are 1.0 and 1.1. This rules out: %s", v[2]);
+ freez(buf);
+ return JB_ERR_PARSE;
}
- err = parse_http_url(v[1], http, csp);
+ http->ssl = !strcmpic(v[0], "CONNECT");
+
+ err = parse_http_url(v[1], http, !http->ssl);
if (err)
{
- free(buf);
+ freez(buf);
return err;
}
/*
* Copy the details into the structure
*/
- http->ssl = is_connect;
- http->cmd = strdup(req);
- http->gpc = strdup(v[0]);
- http->ver = strdup(v[2]);
-
- if ( (http->cmd == NULL)
- || (http->gpc == NULL)
- || (http->ver == NULL) )
- {
- free(buf);
- return JB_ERR_MEMORY;
- }
+ http->cmd = strdup_or_die(req);
+ http->gpc = strdup_or_die(v[0]);
+ http->ver = strdup_or_die(v[2]);
+
+ freez(buf);
- free(buf);
return JB_ERR_OK;
}
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
/*********************************************************************
*
- * Function : compile_host_pattern
+ * Function : compile_pattern
*
- * Description : Parses and compiles a PCRE host pattern..
+ * Description : Compiles a host, domain or TAG pattern.
*
* Parameters :
- * 1 : url = Target url_spec to be filled in.
- * 2 : host_pattern = Host pattern to compile.
+ * 1 : pattern = The pattern to compile.
+ * 2 : anchoring = How the regex should be modified
+ * before compilation. Can be either
+ * one of NO_ANCHORING, LEFT_ANCHORED,
+ * RIGHT_ANCHORED or RIGHT_ANCHORED_HOST.
+ * 3 : url = In case of failures, the spec member is
+ * logged and the structure freed.
+ * 4 : regex = Where the compiled regex should be stored.
*
* Returns : JB_ERR_OK - Success
* JB_ERR_MEMORY - Out of memory
* JB_ERR_PARSE - Cannot parse regex
*
*********************************************************************/
-static jb_err compile_host_pattern(struct url_spec *url, const char *host_pattern)
+static jb_err compile_pattern(const char *pattern, enum regex_anchoring anchoring,
+ struct url_spec *url, regex_t **regex)
{
int errcode;
char rebuf[BUFFER_SIZE];
+ const char *fmt = NULL;
- assert(host_pattern);
- assert(strlen(host_pattern) < sizeof(rebuf) - 2);
+ assert(pattern);
+ assert(strlen(pattern) < sizeof(rebuf) - 2);
- url->host_regex = zalloc(sizeof(*url->host_regex));
- if (NULL == url->host_regex)
+ if (pattern[0] == '\0')
+ {
+ *regex = NULL;
+ return JB_ERR_OK;
+ }
+
+ switch (anchoring)
+ {
+ case NO_ANCHORING:
+ fmt = "%s";
+ break;
+ case RIGHT_ANCHORED:
+ fmt = "%s$";
+ break;
+ case RIGHT_ANCHORED_HOST:
+ fmt = "%s\\.?$";
+ break;
+ case LEFT_ANCHORED:
+ fmt = "^%s";
+ break;
+ default:
+ log_error(LOG_LEVEL_FATAL,
+ "Invalid anchoring in compile_pattern %d", anchoring);
+ }
+
+ *regex = zalloc(sizeof(**regex));
+ if (NULL == *regex)
{
free_url_spec(url);
return JB_ERR_MEMORY;
}
- snprintf(rebuf, sizeof(rebuf), "%s$", host_pattern);
+ snprintf(rebuf, sizeof(rebuf), fmt, pattern);
- errcode = regcomp(url->host_regex, rebuf,
- (REG_EXTENDED|REG_NOSUB|REG_ICASE));
+ errcode = regcomp(*regex, rebuf, (REG_EXTENDED|REG_NOSUB|REG_ICASE));
if (errcode)
{
- size_t errlen = regerror(errcode, url->host_regex, rebuf, sizeof(rebuf));
+ size_t errlen = regerror(errcode, *regex, rebuf, sizeof(rebuf));
if (errlen > (sizeof(rebuf) - (size_t)1))
{
errlen = sizeof(rebuf) - (size_t)1;
}
rebuf[errlen] = '\0';
- log_error(LOG_LEVEL_ERROR, "error compiling %s: %s", url->spec, rebuf);
+ log_error(LOG_LEVEL_ERROR, "error compiling %s from %s: %s",
+ pattern, url->spec, rebuf);
free_url_spec(url);
return JB_ERR_PARSE;
}
+
+/*********************************************************************
+ *
+ * Function : compile_url_pattern
+ *
+ * Description : Compiles the three parts of an URL pattern.
+ *
+ * Parameters :
+ * 1 : url = Target url_spec to be filled in.
+ * 2 : buf = The url pattern to compile. Will be messed up.
+ *
+ * Returns : JB_ERR_OK - Success
+ * JB_ERR_MEMORY - Out of memory
+ * JB_ERR_PARSE - Cannot parse regex
+ *
+ *********************************************************************/
+static jb_err compile_url_pattern(struct url_spec *url, char *buf)
+{
+ char *p;
+
+ p = strchr(buf, '/');
+ if (NULL != p)
+ {
+ /*
+ * Only compile the regex if it consists of more than
+ * a single slash, otherwise it wouldn't affect the result.
+ */
+ if (p[1] != '\0')
+ {
+ /*
+ * XXX: does it make sense to compile the slash at the beginning?
+ */
+ jb_err err = compile_pattern(p, LEFT_ANCHORED, url, &url->preg);
+
+ if (JB_ERR_OK != err)
+ {
+ return err;
+ }
+ }
+ *p = '\0';
+ }
+
+ /*
+ * IPv6 numeric hostnames can contain colons, thus we need
+ * to delimit the hostname before the real port separator.
+ * As brackets are already used in the hostname pattern,
+ * we use angle brackets ('<', '>') instead.
+ */
+ if ((buf[0] == '<') && (NULL != (p = strchr(buf + 1, '>'))))
+ {
+ *p++ = '\0';
+ buf++;
+
+ if (*p == '\0')
+ {
+ /* IPv6 address without port number */
+ p = NULL;
+ }
+ else if (*p != ':')
+ {
+ /* Garbage after address delimiter */
+ return JB_ERR_PARSE;
+ }
+ }
+ else
+ {
+ p = strchr(buf, ':');
+ }
+
+ if (NULL != p)
+ {
+ *p++ = '\0';
+ url->port_list = strdup_or_die(p);
+ }
+ else
+ {
+ url->port_list = NULL;
+ }
+
+ if (buf[0] != '\0')
+ {
+ return compile_host_pattern(url, buf);
+ }
+
+ return JB_ERR_OK;
+
+}
+
+
+#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+/*********************************************************************
+ *
+ * Function : compile_host_pattern
+ *
+ * Description : Parses and compiles a host pattern.
+ *
+ * Parameters :
+ * 1 : url = Target url_spec to be filled in.
+ * 2 : host_pattern = Host pattern to compile.
+ *
+ * Returns : JB_ERR_OK - Success
+ * JB_ERR_MEMORY - Out of memory
+ * JB_ERR_PARSE - Cannot parse regex
+ *
+ *********************************************************************/
+static jb_err compile_host_pattern(struct url_spec *url, const char *host_pattern)
+{
+ return compile_pattern(host_pattern, RIGHT_ANCHORED_HOST, url, &url->host_regex);
+}
+
#else
/*********************************************************************
* 2 : host_pattern = Host pattern to parse.
*
* Returns : JB_ERR_OK - Success
- * JB_ERR_MEMORY - Out of memory
* JB_ERR_PARSE - Cannot parse regex
*
*********************************************************************/
url->unanchored |= ANCHOR_LEFT;
}
- /*
+ /*
* Split domain into components
*/
- url->dbuffer = strdup(host_pattern);
- if (NULL == url->dbuffer)
- {
- free_url_spec(url);
- return JB_ERR_MEMORY;
- }
+ url->dbuffer = strdup_or_die(host_pattern);
- /*
+ /*
* Map to lower case
*/
for (p = url->dbuffer; *p ; p++)
{
- *p = (char)tolower((int)(unsigned char)*p);
+ *p = (char)privoxy_tolower(*p);
}
- /*
+ /*
* Split the domain name into components
*/
- url->dcount = ssplit(url->dbuffer, ".", v, SZ(v), 1, 1);
+ url->dcount = ssplit(url->dbuffer, ".", v, SZ(v));
if (url->dcount < 0)
{
free_url_spec(url);
- return JB_ERR_MEMORY;
+ return JB_ERR_PARSE;
}
else if (url->dcount != 0)
{
- /*
+ /*
* Save a copy of the pointers in dvec
*/
size = (size_t)url->dcount * sizeof(*url->dvec);
-
- url->dvec = (char **)malloc(size);
- if (NULL == url->dvec)
- {
- free_url_spec(url);
- return JB_ERR_MEMORY;
- }
+
+ url->dvec = malloc_or_die(size);
memcpy(url->dvec, v, size);
}
}
+/*********************************************************************
+ *
+ * Function : simplematch
+ *
+ * Description : String matching, with a (greedy) '*' wildcard that
+ * stands for zero or more arbitrary characters and
+ * character classes in [], which take both enumerations
+ * and ranges.
+ *
+ * Parameters :
+ * 1 : pattern = pattern for matching
+ * 2 : text = text to be matched
+ *
+ * Returns : 0 if match, else nonzero
+ *
+ *********************************************************************/
+static int simplematch(const char *pattern, const char *text)
+{
+ const unsigned char *pat = (const unsigned char *)pattern;
+ const unsigned char *txt = (const unsigned char *)text;
+ const unsigned char *fallback = pat;
+ int wildcard = 0;
+
+ unsigned char lastchar = 'a';
+ unsigned i;
+ unsigned char charmap[32];
+
+ while (*txt)
+ {
+
+ /* EOF pattern but !EOF text? */
+ if (*pat == '\0')
+ {
+ if (wildcard)
+ {
+ pat = fallback;
+ }
+ else
+ {
+ return 1;
+ }
+ }
+
+ /* '*' in the pattern? */
+ if (*pat == '*')
+ {
+
+ /* The pattern ends afterwards? Speed up the return. */
+ if (*++pat == '\0')
+ {
+ return 0;
+ }
+
+ /* Else, set wildcard mode and remember position after '*' */
+ wildcard = 1;
+ fallback = pat;
+ }
+
+ /* Character range specification? */
+ if (*pat == '[')
+ {
+ memset(charmap, '\0', sizeof(charmap));
+
+ while (*++pat != ']')
+ {
+ if (!*pat)
+ {
+ return 1;
+ }
+ else if (*pat == '-')
+ {
+ if ((*++pat == ']') || *pat == '\0')
+ {
+ return(1);
+ }
+ for (i = lastchar; i <= *pat; i++)
+ {
+ charmap[i / 8] |= (unsigned char)(1 << (i % 8));
+ }
+ }
+ else
+ {
+ charmap[*pat / 8] |= (unsigned char)(1 << (*pat % 8));
+ lastchar = *pat;
+ }
+ }
+ } /* -END- if Character range specification */
+
+
+ /*
+ * Char match, or char range match?
+ */
+ if ((*pat == *txt)
+ || (*pat == '?')
+ || ((*pat == ']') && (charmap[*txt / 8] & (1 << (*txt % 8)))))
+ {
+ /*
+ * Success: Go ahead
+ */
+ pat++;
+ }
+ else if (!wildcard)
+ {
+ /*
+ * No match && no wildcard: No luck
+ */
+ return 1;
+ }
+ else if (pat != fallback)
+ {
+ /*
+ * Increment text pointer if in char range matching
+ */
+ if (*pat == ']')
+ {
+ txt++;
+ }
+ /*
+ * Wildcard mode && nonmatch beyond fallback: Rewind pattern
+ */
+ pat = fallback;
+ /*
+ * Restart matching from current text pointer
+ */
+ continue;
+ }
+ txt++;
+ }
+
+ /* Cut off extra '*'s */
+ if (*pat == '*') pat++;
+
+ /* If this is the pattern's end, fine! */
+ return(*pat);
+
+}
+
+
/*********************************************************************
*
* Function : simple_domaincmp
* are lost forever.
*
* Returns : JB_ERR_OK - Success
- * JB_ERR_MEMORY - Out of memory
* JB_ERR_PARSE - Cannot parse regex (Detailed message
* written to system log)
*
*********************************************************************/
-jb_err create_url_spec(struct url_spec * url, const char * buf)
+jb_err create_url_spec(struct url_spec *url, char *buf)
{
- char *p;
- int errcode;
- size_t errlen;
- char rebuf[BUFFER_SIZE];
-
assert(url);
assert(buf);
- /*
- * Zero memory
- */
memset(url, '\0', sizeof(*url));
- /*
- * Save a copy of the orignal specification
- */
- if ((url->spec = strdup(buf)) == NULL)
- {
- return JB_ERR_MEMORY;
- }
+ /* Remember the original specification for the CGI pages. */
+ url->spec = strdup_or_die(buf);
- /* Is it tag pattern? */
- if (0 == strncmpic("TAG:", url->spec, 4))
+ /* Is it a tag pattern? */
+ if (0 == strncmpic(url->spec, "TAG:", 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);
- free_url_spec(url);
-
- return JB_ERR_PARSE;
- }
- return JB_ERR_OK;
+ /* The pattern starts with the first character after "TAG:" */
+ const char *tag_pattern = buf + 4;
+ return compile_pattern(tag_pattern, NO_ANCHORING, url, &url->tag_regex);
}
- /* Only reached for URL patterns */
- p = strchr(buf, '/');
- if (NULL != p)
- {
- /*
- * Only compile the regex if it consists of more than
- * a single slash, otherwise it wouldn't affect the result.
- */
- if (*(p+1) != '\0')
- {
- /* XXX: mostly duplicated code, should be factored out. */
- url->preg = zalloc(sizeof(*url->preg));
- if (NULL == url->preg)
- {
- free_url_spec(url);
- return JB_ERR_MEMORY;
- }
-
- snprintf(rebuf, sizeof(rebuf), "^(%s)", p);
- errcode = regcomp(url->preg, rebuf,
- (REG_EXTENDED|REG_NOSUB|REG_ICASE));
- if (errcode)
- {
- errlen = regerror(errcode, url->preg, rebuf, sizeof(rebuf));
-
- if (errlen > (sizeof(rebuf) - (size_t)1))
- {
- errlen = sizeof(rebuf) - (size_t)1;
- }
- rebuf[errlen] = '\0';
- log_error(LOG_LEVEL_ERROR, "error compiling %s: %s",
- url->spec, rebuf);
- free_url_spec(url);
-
- return JB_ERR_PARSE;
- }
- }
- *p = '\0';
- }
-
- p = strchr(buf, ':');
- if (NULL != p)
- {
- *p++ = '\0';
- url->port_list = strdup(p);
- if (NULL == url->port_list)
- {
- return JB_ERR_MEMORY;
- }
- }
- else
- {
- url->port_list = NULL;
- }
-
- if (buf[0] != '\0')
- {
- return compile_host_pattern(url, buf);
- }
-
- return JB_ERR_OK;
-
+ /* If it isn't a tag pattern it must be an URL pattern. */
+ return compile_url_pattern(url, buf);
}
}
+/*********************************************************************
+ *
+ * Function : port_matches
+ *
+ * Description : Compares a port against a port list.
+ *
+ * Parameters :
+ * 1 : port = The port to check.
+ * 2 : port_list = The list of port to compare with.
+ *
+ * Returns : TRUE for yes, FALSE otherwise.
+ *
+ *********************************************************************/
+static int port_matches(const int port, const char *port_list)
+{
+ return ((NULL == port_list) || match_portlist(port_list, port));
+}
+
+
+/*********************************************************************
+ *
+ * Function : host_matches
+ *
+ * Description : Compares a host against a host pattern.
+ *
+ * Parameters :
+ * 1 : url = The URL to match
+ * 2 : pattern = The URL pattern
+ *
+ * Returns : TRUE for yes, FALSE otherwise.
+ *
+ *********************************************************************/
+static int host_matches(const struct http_request *http,
+ const struct url_spec *pattern)
+{
+#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+ return ((NULL == pattern->host_regex)
+ || (0 == regexec(pattern->host_regex, http->host, 0, NULL, 0)));
+#else
+ return ((NULL == pattern->dbuffer) || (0 == domain_match(pattern, http)));
+#endif
+}
+
+
+/*********************************************************************
+ *
+ * Function : path_matches
+ *
+ * Description : Compares a path against a path pattern.
+ *
+ * Parameters :
+ * 1 : path = The path to match
+ * 2 : pattern = The URL pattern
+ *
+ * Returns : TRUE for yes, FALSE otherwise.
+ *
+ *********************************************************************/
+static int path_matches(const char *path, const struct url_spec *pattern)
+{
+ return ((NULL == pattern->preg)
+ || (0 == regexec(pattern->preg, path, 0, NULL, 0)));
+}
+
+
/*********************************************************************
*
* Function : url_match
int url_match(const struct url_spec *pattern,
const struct http_request *http)
{
- /* XXX: these should probably be functions. */
-#define PORT_MATCHES ((NULL == pattern->port_list) || match_portlist(pattern->port_list, http->port))
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
-#define DOMAIN_MATCHES ((NULL == pattern->host_regex) || (0 == regexec(pattern->host_regex, http->host, 0, NULL, 0)))
-#else
-#define DOMAIN_MATCHES ((NULL == pattern->dbuffer) || (0 == domain_match(pattern, http)))
-#endif
-#define PATH_MATCHES ((NULL == pattern->preg) || (0 == regexec(pattern->preg, http->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);
+ return (port_matches(http->port, pattern->port_list)
+ && host_matches(http, pattern) && path_matches(http->path, pattern));
}
{
char *min, *max, *next, *portlist_copy;
- min = next = portlist_copy = strdup(portlist);
+ min = portlist_copy = strdup_or_die(portlist);
/*
* Zero-terminate first item and remember offset for next
/*
* Loop through all items, checking for match
*/
- while(min)
+ while (NULL != min)
{
if (NULL == (max = strchr(min, (int) '-')))
{
*/
if (port == atoi(min))
{
- free(portlist_copy);
+ freez(portlist_copy);
return(1);
}
}
* or, if max was omitted, between min and 65K
*/
*max++ = '\0';
- if(port >= atoi(min) && port <= (atoi(max) ? atoi(max) : 65535))
+ if (port >= atoi(min) && port <= (atoi(max) ? atoi(max) : 65535))
{
- free(portlist_copy);
+ freez(portlist_copy);
return(1);
}
}
}
- free(portlist_copy);
+ freez(portlist_copy);
return 0;
}
+/*********************************************************************
+ *
+ * Function : parse_forwarder_address
+ *
+ * Description : Parse out the host and port from a forwarder address.
+ *
+ * Parameters :
+ * 1 : address = The forwarder address to parse.
+ * 2 : hostname = Used to return the hostname. NULL on error.
+ * 3 : port = Used to return the port. Untouched if no port
+ * is specified.
+ *
+ * Returns : JB_ERR_OK on success
+ * JB_ERR_MEMORY on out of memory
+ * JB_ERR_PARSE on malformed address.
+ *
+ *********************************************************************/
+jb_err parse_forwarder_address(char *address, char **hostname, int *port)
+{
+ char *p = address;
+
+ if ((*address == '[') && (NULL == strchr(address, ']')))
+ {
+ /* XXX: Should do some more validity checks here. */
+ return JB_ERR_PARSE;
+ }
+
+ *hostname = strdup_or_die(address);
+
+ if ((**hostname == '[') && (NULL != (p = strchr(*hostname, ']'))))
+ {
+ *p++ = '\0';
+ memmove(*hostname, (*hostname + 1), (size_t)(p - *hostname));
+ if (*p == ':')
+ {
+ *port = (int)strtol(++p, NULL, 0);
+ }
+ }
+ else if (NULL != (p = strchr(*hostname, ':')))
+ {
+ *p++ = '\0';
+ *port = (int)strtol(p, NULL, 0);
+ }
+
+ return JB_ERR_OK;
+
+}
+
+
/*
Local Variables:
tab-width: 3