Merge branch 'master' of ssh://git.privoxy.org:23/git/privoxy

diff --git a/TODO b/TODO
index 11b58ca..4dfebb5 100644 (file)
--- a/TODO
+++ b/TODO
@@ -179,25 +179,6 @@ https://www.privoxy.org/faq/general.html#DONATE
 
 79) Evaluate pcre alternatives.
 
-80) Change FEATURE_EXTENDED_HOST_PATTERNS to support both
-    extended and vanilla host patterns at the same time.
-
-    Note that the requirement is to allow the user to decide
-    if the domain pattern should be interpreted as regex or
-    traditional host pattern and if it's not obvious that the
-    user made any decision, default to the latter.
-
-    Possible solutions would be:
-
-    1. An always-use-regex-domain-patterns config option
-    2. An enable-regex-domain-patterns-for-this-action-file option
-    3. An enable-regex-domain-patterns-for-this-action-file-until-the-user-says-otherwise option
-    4. A treat-the-domain-pattern-in-this-line-as-regex(-or-not) option
-    5. Combinations of the options above
-
-    With 2+4, 3+4 or 2+3+4 being the preferences until
-    further discussion.
-
 82) Detect if the system time goes back in time let the user
     know if it caused any connections to get closed.
 

diff --git a/acconfig.h b/acconfig.h
index abbe2e2..a55d20a 100644 (file)
--- a/acconfig.h
+++ b/acconfig.h
@@ -125,7 +125,7 @@
 /*
  * Allow PCRE syntax in host patterns.
  */
-#undef FEATURE_EXTENDED_HOST_PATTERNS
+#undef FEATURE_PCRE_HOST_PATTERNS
 
 /*
  * Allow filtering with scripts and programs.

diff --git a/configure.in b/configure.in
index 2ca7e24..ee3539f 100644 (file)
--- a/configure.in
+++ b/configure.in
@@ -978,16 +978,12 @@ AC_ARG_ENABLE(graceful-termination,
   AC_DEFINE(FEATURE_GRACEFUL_TERMINATION)
 fi])
 
-AC_ARG_ENABLE(extended-host-patterns,
-[  --enable-extended-host-patterns Enable and require PCRE syntax in host patterns. This feature hasn't
-                                  been announced yet and it's not clear if it's a good idea. It's expected
-                                  to work, but undocumented. You should only enable it if you know what
-                                  PCRE is and are sure that you need it for your host patterns. You can
-                                  use tools/url-pattern-translator.pl to convert existing action files to
-                                  use PCRE host patterns. Please don't enable this option when creating
-                                  packages for others that may not be expecting it.],
+AC_ARG_ENABLE(pcre-host-patterns,
+[  --enable-pcre-host-patterns Allow to use PCRE syntax in host patterns by prefixing the pattern with
+                                  "PCRE-HOST-PATTERN:". You can use tools/url-pattern-translator.pl to
+                                  convert existing action files to use PCRE host patterns.],
 [if test $enableval = yes; then
-  AC_DEFINE(FEATURE_EXTENDED_HOST_PATTERNS)
+  AC_DEFINE(FEATURE_PCRE_HOST_PATTERNS)
 fi])
 
 AC_ARG_ENABLE(external-filters,

diff --git a/doc/source/user-manual.sgml b/doc/source/user-manual.sgml
index 1c15cc5..0b085aa 100644 (file)
--- a/doc/source/user-manual.sgml
+++ b/doc/source/user-manual.sgml
@@ -2294,6 +2294,12 @@ for details.
  While flexible, this is not the sophistication of full regular expression based syntax.
 </para>
 
+<para>
+ When compiled with FEATURE_PCRE_HOST_PATTERNS patterns can be prefixed with
+ <quote>PCRE-HOST-PATTERN:</quote> in which case full regular expression
+ (PCRE) can be used for the host pattern as well.
+</para>
+
 </sect3>
 
 <!--  ~  End section  ~  -->

diff --git a/doc/webserver/user-manual/actions-file.html b/doc/webserver/user-manual/actions-file.html
index a130aee..a333825 100644 (file)
--- a/doc/webserver/user-manual/actions-file.html
+++ b/doc/webserver/user-manual/actions-file.html
@@ -394,6 +394,9 @@
           </dl>
         </div>
         <p>While flexible, this is not the sophistication of full regular expression based syntax.</p>
+        <p>When compiled with FEATURE_PCRE_HOST_PATTERNS patterns can be prefixed with <span class=
+        "QUOTE">"PCRE-HOST-PATTERN:"</span> in which case full regular expression (PCRE) can be used for the host
+        pattern as well.</p>
       </div>
       <div class="SECT3">
         <h3 class="SECT3"><a name="PATH-PATTERN" id="PATH-PATTERN">8.4.2. The Path Pattern</a></h3>

diff --git a/doc/webserver/user-manual/appendix.html b/doc/webserver/user-manual/appendix.html
index 723290d..06b472f 100644 (file)
--- a/doc/webserver/user-manual/appendix.html
+++ b/doc/webserver/user-manual/appendix.html
@@ -202,7 +202,7 @@
       these. If not, you will get a friendly error message. Internet access is not necessary either.</p>
       <ul>
         <li>
-          <p>Privoxy main page:</p><a name="AEN6301" id="AEN6301"></a>
+          <p>Privoxy main page:</p><a name="AEN6303" id="AEN6303"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/" target="_top">http://config.privoxy.org/</a></p>
           </blockquote>
@@ -211,7 +211,7 @@
           "APPLICATION">Privoxy</span>)</p>
         </li>
         <li>
-          <p>View and toggle client tags:</p><a name="AEN6309" id="AEN6309"></a>
+          <p>View and toggle client tags:</p><a name="AEN6311" id="AEN6311"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/client-tags" target=
             "_top">http://config.privoxy.org/client-tags</a></p>
@@ -219,21 +219,21 @@
         </li>
         <li>
           <p>Show information about the current configuration, including viewing and editing of actions
-          files:</p><a name="AEN6314" id="AEN6314"></a>
+          files:</p><a name="AEN6316" id="AEN6316"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/show-status" target=
             "_top">http://config.privoxy.org/show-status</a></p>
           </blockquote>
         </li>
         <li>
-          <p>Show the browser's request headers:</p><a name="AEN6319" id="AEN6319"></a>
+          <p>Show the browser's request headers:</p><a name="AEN6321" id="AEN6321"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/show-request" target=
             "_top">http://config.privoxy.org/show-request</a></p>
           </blockquote>
         </li>
         <li>
-          <p>Show which actions apply to a URL and why:</p><a name="AEN6324" id="AEN6324"></a>
+          <p>Show which actions apply to a URL and why:</p><a name="AEN6326" id="AEN6326"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/show-url-info" target=
             "_top">http://config.privoxy.org/show-url-info</a></p>
@@ -242,15 +242,15 @@
         <li>
           <p>Toggle Privoxy on or off. This feature can be turned off/on in the main <tt class="FILENAME">config</tt>
           file. When toggled <span class="QUOTE">"off"</span>, <span class="QUOTE">"Privoxy"</span> continues to run,
-          but only as a pass-through proxy, with no actions taking place:</p><a name="AEN6332" id="AEN6332"></a>
+          but only as a pass-through proxy, with no actions taking place:</p><a name="AEN6334" id="AEN6334"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/toggle" target="_top">http://config.privoxy.org/toggle</a></p>
           </blockquote>
-          <p>Short cuts. Turn off, then on:</p><a name="AEN6336" id="AEN6336"></a>
+          <p>Short cuts. Turn off, then on:</p><a name="AEN6338" id="AEN6338"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/toggle?set=disable" target=
             "_top">http://config.privoxy.org/toggle?set=disable</a></p>
-          </blockquote><a name="AEN6339" id="AEN6339"></a>
+          </blockquote><a name="AEN6341" id="AEN6341"></a>
           <blockquote class="BLOCKQUOTE">
             <p><a href="http://config.privoxy.org/toggle?set=enable" target=
             "_top">http://config.privoxy.org/toggle?set=enable</a></p>

diff --git a/jcc.c b/jcc.c
index 69c0180..d29b511 100644 (file)
--- a/jcc.c
+++ b/jcc.c
@@ -587,10 +587,8 @@ static jb_err get_request_destination_elsewhere(struct client_state *csp, struct
    }
    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;
    }
@@ -2357,10 +2355,8 @@ static jb_err process_encrypted_request(struct client_state *csp)
       return JB_ERR_PARSE;
    }
 
-#ifndef FEATURE_EXTENDED_HOST_PATTERNS
    /* Split the domain we just got for pattern matching */
    init_domain_components(csp->http);
-#endif
 
 #ifdef FEATURE_TOGGLE
    if ((csp->flags & CSP_FLAG_TOGGLED_ON) != 0)

diff --git a/project.h b/project.h
index e241360..4c594ff 100644 (file)
--- a/project.h
+++ b/project.h
@@ -331,11 +331,9 @@ struct http_request
    char *host_ip_addr_str; /**< String with dotted decimal representation
                                 of host's IP. NULL before connect_to() */
 
-#ifndef FEATURE_EXTENDED_HOST_PATTERNS
    char  *dbuffer; /**< Buffer with '\0'-delimited domain name.           */
    char **dvec;    /**< List of pointers to the strings in dbuffer.       */
    int    dcount;  /**< How many parts to this domain? (length of dvec)   */
-#endif /* ndef FEATURE_EXTENDED_HOST_PATTERNS */
 
 #ifdef FEATURE_HTTPS_INSPECTION
    int client_ssl;                                                  /**< Flag if we should communicate with client over ssl   */
@@ -396,14 +394,14 @@ struct http_response
 
 struct url_spec
 {
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+#ifdef FEATURE_PCRE_HOST_PATTERNS
    regex_t *host_regex;/**< Regex for host matching                          */
-#else
+   enum host_regex_type { VANILLA_HOST_PATTERN, PCRE_HOST_PATTERN } host_regex_type;
+#endif /* defined FEATURE_PCRE_HOST_PATTERNS */
    char  *dbuffer;     /**< Buffer with '\0'-delimited domain name, or NULL to match all hosts. */
    char **dvec;        /**< List of pointers to the strings in dbuffer.       */
    int    dcount;      /**< How many parts to this domain? (length of dvec)   */
    int    unanchored;  /**< Bitmap - flags are ANCHOR_LEFT and ANCHOR_RIGHT.  */
-#endif /* defined FEATURE_EXTENDED_HOST_PATTERNS */
 
    char  *port_list;   /**< List of acceptable ports, or NULL to match all ports */
 

diff --git a/ssl.c b/ssl.c
index e2389f0..85701ce 100644 (file)
--- a/ssl.c
+++ b/ssl.c
@@ -1548,13 +1548,10 @@ static int host_is_ip_address(const char *host)
 
    for (p = host; *p; p++)
    {
-      if (*p != '.')
+      if ((*p != '.') && !privoxy_isdigit(*p))
       {
-         if (!privoxy_isdigit(*p))
-         {
-            /* Not a dot or digit so it can't be an IPv4 address. */
-            return 0;
-         }
+         /* Not a dot or digit so it can't be an IPv4 address. */
+         return 0;
       }
    }
 

diff --git a/tools/url-pattern-translator.pl b/tools/url-pattern-translator.pl
index 0653e76..5477a87 100755 (executable)
--- a/tools/url-pattern-translator.pl
+++ b/tools/url-pattern-translator.pl
@@ -15,9 +15,6 @@
 #
 # url-pattern-translator.pl old.action > new.action 
 #
-# Only convert your files once, or, as RoboCop used to say,
-# there will be... trouble.
-#
 # Copyright (c) 2008 Fabian Keil <fk@fabiankeil.de>
 #
 # Permission to use, copy, modify, and distribute this software for any
@@ -72,6 +69,9 @@ sub convert_host_pattern ($) {
     # Match single character with a dot.
     $hp =~ s@(?<!\))\?@.@g;
 
+    # Add the prefix
+    $hp = "PCRE-HOST-PATTERN:" . $hp;
+
     return $hp;
 }
 
@@ -91,7 +91,11 @@ sub looks_interesting($) {
 
         $type_to_skip = "whitespace";
 
-    } elsif (m@^\s*TAG:@) {
+    } elsif (m@^\s*CLIENT-TAG:@i) {
+
+        $type_to_skip = "client tag patttern";
+
+    } elsif (m@^\s*TAG:@i) {
 
         $type_to_skip = "tag patttern";
 
@@ -103,6 +107,10 @@ sub looks_interesting($) {
 
         $type_to_skip = "predefined settings";
 
+    } elsif (m@^\s*PCRE-HOST-PATTERN:@i) {
+
+        $type_to_skip = "already converted pcre host patttern";
+
     }
 
     #p("Skipping " . $type_to_skip . ": " . $_) if defined $type_to_skip;

diff --git a/urlmatch.c b/urlmatch.c
index 6949eed..868a16b 100644 (file)
--- a/urlmatch.c
+++ b/urlmatch.c
@@ -5,7 +5,7 @@
  * Purpose     :  Declares functions to match URLs against URL
  *                patterns.
  *
- * Copyright   :  Written by and Copyright (C) 2001-2014
+ * Copyright   :  Written by and Copyright (C) 2001-2020
  *                the Privoxy team. https://www.privoxy.org/
  *
  *                Based on the Internet Junkbuster originally written
@@ -62,7 +62,10 @@ enum regex_anchoring
    RIGHT_ANCHORED,
    RIGHT_ANCHORED_HOST
 };
-static jb_err compile_host_pattern(struct pattern_spec *url, const char *host_pattern);
+static jb_err compile_vanilla_host_pattern(struct pattern_spec *url, const char *host_pattern);
+#ifdef FEATURE_PCRE_HOST_PATTERNS
+static jb_err compile_pcre_host_pattern(struct pattern_spec *url, const char *host_pattern);
+#endif
 
 /*********************************************************************
  *
@@ -89,15 +92,12 @@ void free_http_request(struct http_request *http)
    freez(http->path);
    freez(http->version);
    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
@@ -152,7 +152,6 @@ jb_err init_domain_components(struct http_request *http)
 
    return JB_ERR_OK;
 }
-#endif /* ndef FEATURE_EXTENDED_HOST_PATTERNS */
 
 
 /*********************************************************************
@@ -408,12 +407,8 @@ jb_err parse_http_url(const char *url, struct http_request *http, int require_pr
       freez(buf);
    }
 
-#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 */
 
 }
 
@@ -705,6 +700,20 @@ static jb_err compile_url_pattern(struct pattern_spec *url, char *buf)
 {
    char *p;
 
+#ifdef FEATURE_PCRE_HOST_PATTERNS
+   const size_t prefix_length = 18;
+   if (strncmpic(buf, "PCRE-HOST-PATTERN:", prefix_length) == 0)
+   {
+      url->pattern.url_spec.host_regex_type = PCRE_HOST_PATTERN;
+      /* Overwrite the "PCRE-HOST-PATTERN:" prefix */
+      memmove(buf, buf+prefix_length, strlen(buf+prefix_length)+1);
+   }
+   else
+   {
+      url->pattern.url_spec.host_regex_type = VANILLA_HOST_PATTERN;
+   }
+#endif
+
    p = strchr(buf, '/');
    if (NULL != p)
    {
@@ -766,7 +775,16 @@ static jb_err compile_url_pattern(struct pattern_spec *url, char *buf)
 
    if (buf[0] != '\0')
    {
-      return compile_host_pattern(url, buf);
+#ifdef FEATURE_PCRE_HOST_PATTERNS
+      if (url->pattern.url_spec.host_regex_type == PCRE_HOST_PATTERN)
+      {
+         return compile_pcre_host_pattern(url, buf);
+      }
+      else
+#endif
+      {
+         return compile_vanilla_host_pattern(url, buf);
+      }
    }
 
    return JB_ERR_OK;
@@ -774,12 +792,12 @@ static jb_err compile_url_pattern(struct pattern_spec *url, char *buf)
 }
 
 
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+#ifdef FEATURE_PCRE_HOST_PATTERNS
 /*********************************************************************
  *
- * Function    :  compile_host_pattern
+ * Function    :  compile_pcre_host_pattern
  *
- * Description :  Parses and compiles a host pattern.
+ * Description :  Parses and compiles a pcre host pattern.
  *
  * Parameters  :
  *          1  :  url = Target pattern_spec to be filled in.
@@ -790,16 +808,16 @@ static jb_err compile_url_pattern(struct pattern_spec *url, char *buf)
  *                JB_ERR_PARSE - Cannot parse regex
  *
  *********************************************************************/
-static jb_err compile_host_pattern(struct pattern_spec *url, const char *host_pattern)
+static jb_err compile_pcre_host_pattern(struct pattern_spec *url, const char *host_pattern)
 {
    return compile_pattern(host_pattern, RIGHT_ANCHORED_HOST, url, &url->pattern.url_spec.host_regex);
 }
+#endif /* def FEATURE_PCRE_HOST_PATTERNS */
 
-#else
 
 /*********************************************************************
  *
- * Function    :  compile_host_pattern
+ * Function    :  compile_vanilla_host_pattern
  *
  * Description :  Parses and "compiles" an old-school host pattern.
  *
@@ -811,7 +829,7 @@ static jb_err compile_host_pattern(struct pattern_spec *url, const char *host_pa
  *                JB_ERR_PARSE - Cannot parse regex
  *
  *********************************************************************/
-static jb_err compile_host_pattern(struct pattern_spec *url, const char *host_pattern)
+static jb_err compile_vanilla_host_pattern(struct pattern_spec *url, const char *host_pattern)
 {
    char *v[150];
    size_t size;
@@ -1127,7 +1145,6 @@ static int domain_match(const struct pattern_spec *p, const struct http_request
    }
 
 }
-#endif /* def FEATURE_EXTENDED_HOST_PATTERNS */
 
 
 /*********************************************************************
@@ -1223,17 +1240,16 @@ void free_pattern_spec(struct pattern_spec *pattern)
    if (pattern == NULL) return;
 
    freez(pattern->spec);
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
+#ifdef FEATURE_PCRE_HOST_PATTERNS
    if (pattern->pattern.url_spec.host_regex)
    {
       regfree(pattern->pattern.url_spec.host_regex);
       freez(pattern->pattern.url_spec.host_regex);
    }
-#else
+#endif /* def FEATURE_PCRE_HOST_PATTERNS */
    freez(pattern->pattern.url_spec.dbuffer);
    freez(pattern->pattern.url_spec.dvec);
    pattern->pattern.url_spec.dcount = 0;
-#endif /* ndef FEATURE_EXTENDED_HOST_PATTERNS */
    freez(pattern->pattern.url_spec.port_list);
    if (pattern->pattern.url_spec.preg)
    {
@@ -1284,12 +1300,14 @@ static int host_matches(const struct http_request *http,
                         const struct pattern_spec *pattern)
 {
    assert(http->host != NULL);
-#ifdef FEATURE_EXTENDED_HOST_PATTERNS
-   return ((NULL == pattern->pattern.url_spec.host_regex)
-      || (0 == regexec(pattern->pattern.url_spec.host_regex, http->host, 0, NULL, 0)));
-#else
-   return ((NULL == pattern->pattern.url_spec.dbuffer) || (0 == domain_match(pattern, http)));
+#ifdef FEATURE_PCRE_HOST_PATTERNS
+   if (pattern->pattern.url_spec.host_regex_type == PCRE_HOST_PATTERN) {
+      return ((NULL == pattern->pattern.url_spec.host_regex)
+         || (0 == regexec(pattern->pattern.url_spec.host_regex,
+               http->host, 0, NULL, 0)));
+   }
 #endif
+   return ((NULL == pattern->pattern.url_spec.dbuffer) || (0 == domain_match(pattern, http)));
 }
 
 

diff --git a/urlmatch.h b/urlmatch.h
index fcd050a..315e8b2 100644 (file)
--- a/urlmatch.h
+++ b/urlmatch.h
@@ -38,9 +38,7 @@
 #include "project.h"
 
 extern void free_http_request(struct http_request *http);
-#ifndef FEATURE_EXTENDED_HOST_PATTERNS
 extern jb_err init_domain_components(struct http_request *http);
-#endif
 extern jb_err parse_http_request(const char *req, struct http_request *http);
 extern jb_err parse_http_url(const char *url,
                              struct http_request *http,