X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=urlmatch.c;h=86d211be5081b55068717f99cd3afcdb255bf4de;hp=f93b6bec770c396e1546709ec0b8e76e1253de0f;hb=473cfd051580edfa1e2a3f6beeb9a0d09a8253fd;hpb=6f7198353a4203c0016c821d5ca7d5c17839d403 diff --git a/urlmatch.c b/urlmatch.c index f93b6bec..86d211be 100644 --- a/urlmatch.c +++ b/urlmatch.c @@ -1,4 +1,4 @@ -const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.31 2008/04/10 14:41:04 fabiankeil Exp $"; +const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.44 2008/05/04 16:18:32 fabiankeil Exp $"; /********************************************************************* * * File : $Source: /cvsroot/ijbswa/current/urlmatch.c,v $ @@ -33,6 +33,52 @@ const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.31 2008/04/10 14:41:04 fabianke * * Revisions : * $Log: urlmatch.c,v $ + * Revision 1.44 2008/05/04 16:18:32 fabiankeil + * Provide parse_http_url() with a third parameter to specify + * whether or not URLs without protocol are acceptable. + * + * Revision 1.43 2008/05/04 13:30:55 fabiankeil + * Streamline parse_http_url()'s prototype. + * + * Revision 1.42 2008/05/04 13:24:16 fabiankeil + * If the method isn't CONNECT, reject URLs without protocol. + * + * Revision 1.41 2008/05/02 09:51:34 fabiankeil + * In parse_http_url(), don't muck around with values + * that are none of its business: require an initialized + * http structure and never unset http->ssl. + * + * Revision 1.40 2008/04/23 16:12:28 fabiankeil + * Free with freez(). + * + * Revision 1.39 2008/04/22 16:27:42 fabiankeil + * In parse_http_request(), remove a pointless + * temporary variable and free the buffer earlier. + * + * Revision 1.38 2008/04/18 05:17:18 fabiankeil + * Mark simplematch()'s parameters as immutable. + * + * Revision 1.37 2008/04/17 14:53:29 fabiankeil + * Move simplematch() into urlmatch.c as it's only + * used to match (old-school) domain patterns. + * + * Revision 1.36 2008/04/14 18:19:48 fabiankeil + * Remove now-pointless cast in create_url_spec(). + * + * Revision 1.35 2008/04/14 18:11:21 fabiankeil + * The compiler might not notice it, but the buffer passed to + * create_url_spec() is modified later on and thus shouldn't + * be declared immutable. + * + * Revision 1.34 2008/04/13 13:32:07 fabiankeil + * Factor URL pattern compilation out of create_url_spec(). + * + * Revision 1.33 2008/04/12 14:03:13 fabiankeil + * Remove an obvious comment and improve another one. + * + * Revision 1.32 2008/04/12 12:38:06 fabiankeil + * Factor out duplicated code to compile host, path and tag patterns. + * * Revision 1.31 2008/04/10 14:41:04 fabiankeil * Ditch url_spec's path member now that it's no longer used. * @@ -223,6 +269,7 @@ const char urlmatch_rcs[] = "$Id: urlmatch.c,v 1.31 2008/04/10 14:41:04 fabianke const char urlmatch_h_rcs[] = URLMATCH_H_VERSION; enum regex_anchoring {NO_ANCHORING, LEFT_ANCHORED, RIGHT_ANCHORED}; +static jb_err compile_host_pattern(struct url_spec *url, const char *host_pattern); /********************************************************************* * @@ -330,10 +377,9 @@ jb_err init_domain_components(struct http_request *http) * 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 @@ -341,18 +387,10 @@ jb_err init_domain_components(struct http_request *http) * 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 */ @@ -400,10 +438,12 @@ jb_err parse_http_url(const char * url, 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; } @@ -414,13 +454,13 @@ jb_err parse_http_url(const char * url, * 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, '/'); @@ -507,7 +547,7 @@ jb_err parse_http_url(const char * url, http->host = strdup(host); - free(buf); + freez(buf); if (http->host == NULL) { @@ -603,7 +643,6 @@ jb_err parse_http_request(const char *req, 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)); @@ -616,7 +655,7 @@ jb_err parse_http_request(const char *req, n = ssplit(buf, " \r\n", v, SZ(v), 1, 1); if (n != 3) { - free(buf); + freez(buf); return JB_ERR_PARSE; } @@ -632,39 +671,35 @@ jb_err parse_http_request(const char *req, 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) - { - is_connect = 1; - } + http->ssl = !strcmpic(v[0], "CONNECT"); - err = parse_http_url(v[1], http, csp); + 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]); + freez(buf); + if ( (http->cmd == NULL) || (http->gpc == NULL) || (http->ver == NULL) ) { - free(buf); return JB_ERR_MEMORY; } - free(buf); return JB_ERR_OK; } @@ -695,7 +730,7 @@ static jb_err compile_pattern(const char *pattern, enum regex_anchoring anchorin { int errcode; char rebuf[BUFFER_SIZE]; - const char *fmt; + const char *fmt = NULL; assert(pattern); assert(strlen(pattern) < sizeof(rebuf) - 2); @@ -752,6 +787,73 @@ static jb_err compile_pattern(const char *pattern, enum regex_anchoring anchorin } + +/********************************************************************* + * + * 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'; + } + + 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; + +} + + #ifdef FEATURE_EXTENDED_HOST_PATTERNS /********************************************************************* * @@ -861,6 +963,144 @@ static jb_err compile_host_pattern(struct url_spec *url, const char *host_patter } +/********************************************************************* + * + * 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)))) ) + { + /* + * Sucess: 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 @@ -1004,22 +1244,16 @@ static int domain_match(const struct url_spec *pattern, const struct http_reques * 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; - assert(url); assert(buf); - /* - * Zero memory - */ memset(url, '\0', sizeof(*url)); - /* - * Save a copy of the orignal specification - */ - if ((url->spec = strdup(buf)) == NULL) + /* Remember the original specification for the CGI pages. */ + url->spec = strdup(buf); + if (NULL == url->spec) { return JB_ERR_MEMORY; } @@ -1032,51 +1266,8 @@ jb_err create_url_spec(struct url_spec * url, const char * buf) return compile_pattern(tag_pattern, NO_ANCHORING, url, &url->tag_regex); } - /* Only reached for URL patterns. XXX: should be factored out. */ - 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'; - } - - 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 a URL pattern. */ + return compile_url_pattern(url, buf); } @@ -1200,7 +1391,7 @@ int match_portlist(const char *portlist, int port) */ if (port == atoi(min)) { - free(portlist_copy); + freez(portlist_copy); return(1); } } @@ -1213,7 +1404,7 @@ int match_portlist(const char *portlist, int port) *max++ = '\0'; if(port >= atoi(min) && port <= (atoi(max) ? atoi(max) : 65535)) { - free(portlist_copy); + freez(portlist_copy); return(1); } @@ -1233,7 +1424,7 @@ int match_portlist(const char *portlist, int port) } } - free(portlist_copy); + freez(portlist_copy); return 0; }