Strip trailing whitespace

[privoxy.git] / default.filter

diff --git a/default.filter b/default.filter
index 3f63e99..548befe 100644 (file)
--- a/default.filter
+++ b/default.filter
@@ -1,13 +1,11 @@
 #################################################################################
-# 
-#  File        :  $Source: /cvsroot/ijbswa/current/default.filter,v $
-# 
-#  $Id: default.filter,v 1.90 2014/05/05 09:59:51 fabiankeil Exp $
+#
+#  File        :  default.filter
 #
 #  Purpose     :  Rules to process the content of web pages
-# 
-#  Copyright   :  Written by and Copyright (C) 2001-2014 the
-#                 Privoxy team. http://www.privoxy.org/
+#
+#  Copyright   :  Written by and Copyright (C) 2001-2018 the
+#                 Privoxy team. https://www.privoxy.org/
 #
 #                 This program is free software; you can redistribute it
 #                 and/or modify it under the terms of the GNU General
@@ -48,7 +46,7 @@
 #
 # Note2: In addition to the Perl options gimsx, the following nonstandard
 # options are supported:
-# 
+#
 # 'U' turns the default to ungreedy matching.  Add ? to quantifiers to
 #     switch back to greedy.
 #
@@ -57,7 +55,8 @@
 #     quoting.
 #
 # 'D' (Dynamic) allows the use of variables. Supported variables are:
-#     $host, $origin (the IP address the request came from), $path and $url.
+#     $host, $listen-address, $origin (the IP address the request came
+#     from), $path and $url.
 #
 #     Note that '$' is a bad choice as delimiter for dynamic filters as you
 #     might end up with unintended variables if you use a variable name
@@ -65,7 +64,7 @@
 #     escaping anything, therefore you also have to be careful not to chose
 #     delimiters that appear in the replacement text. For example '<' should
 #     be save, while '?' will sooner or later cause conflicts with $url.
-# 
+#
 #################################################################################
 
 
@@ -84,7 +83,7 @@ FILTER: js-annoyances Get rid of particularly annoying JavaScript abuse.
 # but that would make them match only the first occurrence of
 # nasty-item in each <script>. We need nestable jobs!
 
-# Get rid of Javascript referrer tracking. 
+# Get rid of Javascript referrer tracking.
 # Test page: http://www.javascript-page.com/referrer.html
 #
 s|(?:\w+\.)+referrer|false.toString()|gisU
@@ -102,16 +101,16 @@ s/(?:(?:this|window)\.(?:default)?status)\s*=\s*\w*\s*;//ig
 s/(<body\s+[^>]*)onunload/$1never/siU
 s|(<script.*)window\.onunload(?=.*</script>)|$1never|sigU
 
-# If we allow window.open, we want normal window features: 
+# If we allow window.open, we want normal window features:
 # Test: http://www.htmlgoodies.com/beyond/notitle.html
 #
 s/(open\s*\([^\)]+resizable=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
-s/(open\s*\([^\)]+location=)(["']?)(?:no|0)\2/$1$2yes$2/sigU 
-s/(open\s*\([^\)]+status=)(["']?)(?:no|0)\2/$1$2yes$2/sigU 
-s/(open\s*\([^\)]+scroll(?:ing|bars)=)(["']?)(?:no|0)\2/$1$2auto$2/sigU 
-s/(open\s*\([^\)]+menubar=)(["']?)(?:no|0)\2/$1$2yes$2/sigU 
-s/(open\s*\([^\)]+toolbar=)(["']?)(?:no|0)\2/$1$2yes$2/sigU 
-s/(open\s*\([^\)]+directories=)(["']?)(?:no|0)\2/$1$2yes$2/sigU 
+s/(open\s*\([^\)]+location=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
+s/(open\s*\([^\)]+status=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
+s/(open\s*\([^\)]+scroll(?:ing|bars)=)(["']?)(?:no|0)\2/$1$2auto$2/sigU
+s/(open\s*\([^\)]+menubar=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
+s/(open\s*\([^\)]+toolbar=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
+s/(open\s*\([^\)]+directories=)(["']?)(?:no|0)\2/$1$2yes$2/sigU
 s/(open\s*\([^\)]+fullscreen=)(["']?)(?:yes|1)\2/$1$2no$2/sigU
 s/(open\s*\([^\)]+always(?:raised|lowered)=)(["']?)(?:yes|1)\2/$1$2no$2/sigU
 s/(open\s*\([^\)]+z-?lock=)(["']?)(?:yes|1)\2/$1$2no$2/sigU
@@ -142,8 +141,8 @@ FILTER: html-annoyances Get rid of particularly annoying HTML abuse.
 # New browser windows (if allowed -- see no-popups filter below) should be
 # resizeable and have a location and status bar
 #
-s/(<a\s+href[^>]+resizable=)(['"]?)(?:no|0)\2/$1$2yes$2/igU 
-s/(<a\s+href[^>]+location=)(['"]?)(?:no|0)\2/$1$2yes$2/igU 
+s/(<a\s+href[^>]+resizable=)(['"]?)(?:no|0)\2/$1$2yes$2/igU
+s/(<a\s+href[^>]+location=)(['"]?)(?:no|0)\2/$1$2yes$2/igU
 s/(<a\s+href[^>]+status=)(['"]?)(?:no|0)\2/$1$2yes$2/igU
 s/(<a\s+href[^>]+scrolling=)(['"]?)(?:no|0)\2/$1$2auto$2/igU
 s/(<a\s+href[^>]+menubar=)(['"]?)(?:no|0)\2/$1$2yes$2/igU
@@ -191,7 +190,7 @@ FILTER: unsolicited-popups Disable only unsolicited pop-up windows.
 
 s+([^'"]\s*<head.*>)(?=\s*[^'"])+$1<script>function PrivoxyWindowOpen(){return(null);}</script>+isU
 s@([^\w\s.]\s*)((?:map)?(window|this|parent)\.?)?open\s*\(@$1PrivoxyWindowOpen(@ig
-s+([^'"]\s*</html>)(?!\s*(\\n|'|"))+$1<script>function PrivoxyWindowOpen(a, b, c){return(window.open(a, b, c));}</script>+iU     
+s+([^'"]\s*</html>)(?!\s*(\\n|'|"))+$1<script>function PrivoxyWindowOpen(a, b, c){return(window.open(a, b, c));}</script>+iU
 
 
 ##################################################################################
@@ -219,11 +218,11 @@ FILTER: img-reorder Reorder attributes in <img> tags to make the banners-by-* fi
 # This makes banners-by-size more effective and allows both banners-by-size
 # and banners-by-link to preserve the original image URL in the title attribute.
 
-s|<img\s+?([^>]*)\ssrc\s*=\s*(['"])([^>\\\2]+)\2|<img src=$2$3$2 $1|siUg
-s|<img\s+?([^>]*)\ssrc\s*=\s*([^'">\\\s]+)|<img src=$2 $1|sig
-s|(<img[^>]+height)\s*=\s*|$1=|sig
+s|<img\s+?([^>]*)\ssrc\s*=\s*(['"])([^>'" ]+)\2|<img src=$2$3$2 $1|siUg
+s|<img\s+?([^>]*)\ssrc\s*=\s*([^'">\\\s]+)|<img src=$2 $1|siUg
+s|(<img[^>]+height)\s*=\s*|$1=|siUg
 
-s|<img (src=(?:(['"])[^>\\\\2]*\2\|[^'">\\\s]+?))([^>]*)\s+width\s*=\s*((["']?)\d+?\5)(?=[\s>])|<img $1 width=$4$3|siUg
+s|<img (src=(?:(['"])[^>'" ]*\2\|[^'">\\\s]+?))([^>]*)\s+width\s*=\s*((["']?)\d+?\5)(?=[\s>])|<img $1 width=$4$3|siUg
 
 
 #################################################################################
@@ -319,7 +318,7 @@ s@<img[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*\s(?:width|height)\s*=\s
 #################################################################################
 FILTER: tiny-textforms Extend those tiny textareas up to 40x80 and kill the hard wrap.
 
-s/(<textarea[^>]*?)(?:\s*(?:rows|cols)=(['"]?)\d+\2)+/$1 rows=$2\40$2 cols=$2\80$2/ig 
+s/(<textarea[^>]*?)(?:\s*(?:rows|cols)=(['"]?)\d+\2)+/$1 rows=$2\40$2 cols=$2\80$2/ig
 s/(<textarea[^>]*?)wrap=(['"]?)hard\2/$1/ig
 
 
@@ -343,7 +342,7 @@ s/(<frameset\s+[^>]*)framespacing=(['"]?)(no|0)\2/$1/igU
 s/(<frameset\s+[^>]*)frameborder=(['"]?)(no|0)\2/$1/igU
 s/(<frameset\s+[^>]*)border=(['"]?)(no|0)\2/$1/igU
 s/(<frame\s+[^>]*)noresize/$1/igU
-s/(<frame\s+[^>]*)frameborder=(['"]?)(no|0)\2/$1/igU 
+s/(<frame\s+[^>]*)frameborder=(['"]?)(no|0)\2/$1/igU
 s/(<frame\s+[^>]*)scrolling=(['"]?)(no|0)\2/$1/igU
 
 
@@ -359,11 +358,11 @@ s@<iframe.*</iframe>@<!-- iframe removed by Privoxy's iframe filter -->@Uisg
 
 #################################################################################
 #
-# demoronizer: Correct Microsoft's abuse of standardized character sets, which 
-#              leave the browser to (mis)-interpret unknown characters, with 
+# demoronizer: Correct Microsoft's abuse of standardized character sets, which
+#              leave the browser to (mis)-interpret unknown characters, with
 #              sometimes bizarre results on non-MS platforms.
 #
-# credit: ripped from the demoroniser.pl script by: 
+# credit: ripped from the demoroniser.pl script by:
 #         John Walker -- January 1998, http://www.fourmilab.ch/webtools/demoroniser
 #
 #################################################################################
@@ -413,7 +412,7 @@ s|<embed [^>]*(application/x-shockwave-flash\|\.swf).*>(.*</embed>)?|<!-- Squish
 #################################################################################
 FILTER: quicktime-kioskmode Make Quicktime movies saveable.
 
-s/(<embed\s+[^>]*)kioskmode\s*=\s*(["']?)true\2/$1/ig 
+s/(<embed\s+[^>]*)kioskmode\s*=\s*(["']?)true\2/$1/ig
 
 
 #################################################################################
@@ -488,7 +487,7 @@ s/^.*\
 #Remove by regulations
 s/^.*(Section 2257|18 U.?S.?C.? 2257).*$\
 /This page has been blocked by Privoxy's crude-parental content filter\
-/is 
+/is
 
 
 #################################################################################
@@ -528,7 +527,7 @@ s%<script language="JavaScript">(window\.open|1;''\.concat)\("readme\.eml", null
 FILTER: site-specifics Cure for site-specific problems. Don't apply generally!
 
 # www.spiegel.de excludes X11 users from viewing Flash5 objects - shame.
-# Apply to: www.spiegel.de/static/js/flash-plugin.js 
+# Apply to: www.spiegel.de/static/js/flash-plugin.js
 #
 s/indexOf\("x11"\)/indexOf("x13")/
 
@@ -547,11 +546,11 @@ s|<a href="(.+?)">(?:Continue to message\|Weiter zu Nachricht)</a>|<meta http-eq
 # monster.com has two very similar gimmicks:
 #
 s|<input type="hidden" name="REDIRECT" value="(.+?)">|<meta http-equiv="refresh" content="0; URL=$1">|i
- 
-s|<IMG SRC="http://media.monster.com/mm/usen/my/no_thanks_211x40.gif".+?>|<meta http-equiv="refresh" content="0; URL=http://my.monster.com/resume.asp">|i 
+
+s|<IMG SRC="http://media.monster.com/mm/usen/my/no_thanks_211x40.gif".+?>|<meta http-equiv="refresh" content="0; URL=http://my.monster.com/resume.asp">|i
 
 # nytimes.com triggers popups through the onload handler of dummy images
-# to fool popup-blockers.  
+# to fool popup-blockers.
 #
 s|(<img [^>]*)onload|$1never|sig
 
@@ -564,7 +563,7 @@ s|<INPUT name="\d{2,4}" type="RADIO" value="3" |$0 checked|g
 #################################################################################
 #
 # no-ping: Removes non-standard ping attributes in <a> and <area> tags.
-#          
+#
 #################################################################################
 FILTER: no-ping Removes non-standard ping attributes in <a> and <area> tags.
 s@(<a(?:rea)?[^>]*?)\sping=(['"]?)([^"'>]+)\2([>\s]?)@\
@@ -812,6 +811,16 @@ CLIENT-HEADER-TAGGER: client-ip-address Tags the request with the client's IP ad
 
 s@^\w*\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D
 
+#################################################################################
+#
+# listen-address: Tags the request with the listen-address on which the request
+#                 came in.
+#
+#################################################################################
+CLIENT-HEADER-TAGGER: listen-address Tags the request with the listen-address on which the request came in.
+
+s@^\w*\s+.*\s+HTTP/\d\.\d\s*@LISTEN-ADDRESS: $listen-address@D
+
 #################################################################################
 #
 # http-method: Tags the request with its HTTP method.