X-Git-Url: http://www.privoxy.org/gitweb/?p=privoxy.git;a=blobdiff_plain;f=default.filter;h=4a2857a471e839c841eb936eebf484937780e9de;hp=4017e3085c67ffc7c91d67719ef8c57a8c4e0917;hb=42e4fa330b6bae707912dc050900fda316a5224f;hpb=51152962760824003f1da2c85ac744f35b9248ac

diff --git a/default.filter b/default.filter
index 4017e308..4a2857a4 100644
--- a/default.filter
+++ b/default.filter
@@ -2,11 +2,11 @@
 # 
 #  File        :  $Source: /cvsroot/ijbswa/current/default.filter,v $
 # 
-#  $Id: default.filter,v 1.31 2006/10/21 13:12:28 fabiankeil Exp $
+#  $Id: default.filter,v 1.39 2007/02/21 14:10:23 fabiankeil Exp $
 #
 #  Purpose     :  Rules to process the content of web pages
 # 
-#  Copyright   :  Written by and Copyright (C) 2001 - 2006 the
+#  Copyright   :  Written by and Copyright (C) 2001 - 2007 the
 #                 Privoxy team. http://www.privoxy.org/
 #
 # We value your feedback. However, to provide you with the best support,
@@ -81,8 +81,11 @@ FILTER: js-annoyances Get rid of particularly annoying JavaScript abuse
 s|(?:\w+\.)+referrer|"Not Your Business!"|gisU
 
 # The status bar is for displaying link targets, not pointless blahblah
-# 
-s/(\W\s*)((this|window)\.(default)?status)\s*=\s*((['"]).*?(?<!\\)\6)/$1if(typeof(this.href) != 'undefined') $2 = $5 + ' URL: ' + this.href;else return false/ig
+#
+s@([\W]\s*)((?:this|window)\.(?:default)?status)\s*=\s*((['"]).*?\4)@$1$2 =\
+ (typeof(this.href) != 'undefined')?($3 + ' URL: ' + this.href):($2)@ig
+
+s/(?:(?:this|window)\.(?:default)?status)\s*=\s*\w*\s*;//ig
 
 # Kill OnUnload popups. Yummy.
 # Test: http://www.zdnet.com/zdsubs/yahoo/tree/yfs.html
@@ -178,7 +181,7 @@ s/<meta\s+http-equiv\s*=\s*(['"]?)refresh\1\s+content\s*=\s*(['"]?)\d{2,}\s*(;\s
 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*)((window|this|parent)\.)?open\s*\(+$1PrivoxyWindowOpen(+ig
+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     
 
 
@@ -189,7 +192,7 @@ s+([^'"]\s*</html>)(?!\s*(\\n|'|"))+$1<script>function PrivoxyWindowOpen(a, b, c
 #################################################################################
 FILTER: all-popups Kill all popups in JavaScript and HTML
 
-s/((\W\s*)(window|this|parent)\.)open\s*\\?\(/$1concat(/ig  # JavaScript
+s@((\W\s*)(?:map)?(window|this|parent)\.?)open\s*\\?\(@$1concat(@ig  # JavaScript
 s/\starget\s*=\s*(['"]?)_?(blank|new)\1?/ notarget/ig        # HTML
 
 
@@ -198,7 +201,7 @@ s/\starget\s*=\s*(['"]?)_?(blank|new)\1?/ notarget/ig        # HTML
 # img-reorder: Reorder attributes in <img> tags to make the banners-by-* filters more effective
 #
 #################################################################################
-FILTER: img-reorder Reorder attributes in &lt;img> tags to make the banners-by-* filters more effective
+FILTER: img-reorder Reorder attributes in <img> tags to make the banners-by-* filters more effective
 
 # In the first step src is moved to the start, then width is moved to the second
 # place to guarantee an order of src, width, height.
@@ -229,27 +232,37 @@ s|<img (src=(?:(['"])[^>\\\\2]*\2\|[^'">\\\s]+?))([^>]*)\s+width\s*=\s*(["']?)(\
 FILTER: banners-by-size Kill banners by size
 
 # 88*31
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)88\4)[^>]*?(height=(['"]?)31\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)88\4)[^>]*?(height=(['"]?)31\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 120*60, 120*90, 120*240, 120*600
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)120\4)[^>]*?(height=(['"]?)(?:600?|90|240)\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)120\4)[^>]*?(height=(['"]?)(?:600?|90|240)\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 125*125
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)125\4)[^>]*?(height=(['"]?)125\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)125\4)[^>]*?(height=(['"]?)125\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 160*600
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)160\4)[^>]*?(height=(['"]?)600\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)160\4)[^>]*?(height=(['"]?)600\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 180*150
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)180\4)[^>]*?(height=(['"]?)150\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)180\4)[^>]*?(height=(['"]?)150\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 234*60, 468*60 (Most Banners!)
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:234|468)\4)[^>]*?(height=(['"]?)60\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:234|468)\4)[^>]*?(height=(['"]?)60\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 240*400
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)240\4)[^>]*?(height=(['"]?)400\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)240\4)[^>]*?(height=(['"]?)400\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 250*250, 300*250
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:250|300)\4)[^>]*?(height=(['"]?)250\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:250|300)\4)[^>]*?(height=(['"]?)250\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 # 336*280
-s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)336\4)[^>]*?(height=(['"]?)280\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)336\4)[^>]*?(height=(['"]?)280\6)[^>]*?(?=/?>)@\
+  <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 
 # Note: 200*50 was also proposed, but it probably causes too much collateral damage:
 #
-#s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)200\4)[^>]*?(height=(['"]?)50\6)[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed-$2-by-size$1 $3 $5>@sig
+#s@<img\s+(?:src\s*=\s*(['"]?)([^>\\\1\s]+)\1)?[^>]*?(width=(['"]?)200\4)[^>]*?(height=(['"]?)50\6)[^>]*?(?=/?>)@\
+#   <img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed-$2-by-size" $3 $5@sig
 
 
 #################################################################################
@@ -263,16 +276,17 @@ FILTER: banners-by-link Kill banners by their links to known clicktrackers
 #
 s@<a\s+href\s*=\s*(['"]?)([^>\1\s]*?(?:\
   adclick                             # See www.dn.se \
+| advert                              # see dict.leo.org \
 | atwola\.com/(?:link|redir)          # see www.cnn.com \
 | /jump/                              # redirs for doublecklick.net ads \
 | tracker | counter                   # common \
 | adlog\.pl                           # see sf.net \
-)[^>\1\s]*)\1[^>]*>\s*<img\s+(?:src\s*=\s*(['"]?)([^>\\\3\s]+)\3)?[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\6)[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\8)[^>]*>\
-@<img $5 $7 src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed $4 by link to $2$1>@sigx
+)[^>\1\s]*)\1[^>]*>\s*<img\s+(?:src\s*=\s*(['"]?)([^>\\\3\s]+)\3)?[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\6)[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\8)[^>]*?(?=/?>)\
+@<img $5 $7 src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed $4 by link to $2"@sigx
 
 # Rare case w/o explicit dimensions:
 #
-s@<a\s+href\s*=\s*(['"]?)([^>\1\s]*?(?:adclick|atwola\.com/(?:link|redir)|doubleclick\.net/jump/|tracker|counter|adlog\.pl)[^>\1\s]*)\1[^>]*>\s*<img\s+(?:src\s*=\s*(['"]?)([^>\\\3\s]+)\3)?[^>]*>@<img src=$1http://config.privoxy.org/send-banner?type=auto$1 border="0" title=$1Killed $4 by link to $2$1>@sig
+s@<a\s+href\s*=\s*(['"]?)([^>\1\s]*?(?:ad(?:click|vert)|atwola\.com/(?:link|redir)|doubleclick\.net/jump/|tracker|counter|adlog\.pl)[^>\1\s]*)\1[^>]*>\s*<img\s+(?:src\s*=\s*(['"]?)([^>\\\3\s]+)\3)?[^>]*?(?=/?>)@<img src="http://config.privoxy.org/send-banner?type=auto" border="0" title="Killed $4 by link to $2"@sig
 
 
 ################################################################################
@@ -282,8 +296,7 @@ s@<a\s+href\s*=\s*(['"]?)([^>\1\s]*?(?:adclick|atwola\.com/(?:link|redir)|double
 #################################################################################
 FILTER: webbugs Squish WebBugs (1x1 invisible GIFs used for user tracking)
 
-s@<img[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*?>@<!--\
- Filtered by Privoxy's webbugs filter: $0 -->@siUg
+s@<img[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*?>@@siUg
 
 
 #################################################################################
@@ -413,18 +426,47 @@ s/(M|m)edien(?![^<]*>)/$1&auml;dchen/Ug
 
 #################################################################################
 #
-# crude-parental: Crude parental filtering?  (Use along with a suitable blocklist).
-#                 Shows how to deny access to whole page based on a keyword.
+# crude-parental: Crude parental filtering. Use with a suitable blocklist.
+#                 Pages are "blocked" based on keyword matching.
 #
 #################################################################################
-FILTER: crude-parental Crude parental filtering (demo only)
+FILTER: crude-parental Crude parental filtering 
 
 # (Note: Middlesex, Sussex and Essex are counties in the UK, not rude words)
 # (Note #2: Is 'sex' a rude word?!)
 
 s%^.*(?<!middle)(?<!sus)(?<!es)sex.*$%<html><head><title>Blocked</title></head><body><h3>Blocked due to possible adult content. Please see <a href="http://dmoz.org/Kids_and_Teens/">this site</a>.</h3></body></html>%is
+
 s+^.*warez.*$+<html><head><title>No Warez</title></head><body><h3>You're not searching for illegal stuff, are you?</h3></body></html>+is
 
+# Remove by description
+s/^.*\
+(suck |lick |tounge |rub |fuck |fingering |finger |chicks? )?\
+(her |your |my |hard |with |big |wet |tight |pink |hot |moist |young |teen )+\
+(dicks?|penis|cocks?|balls?|tits?|pussy|cunt|clit|ass|mouth).*$\
+/This page has been blocked by Privoxy's crude-parental content filter\
+/is 
+
+#Remove by link text
+s/^.*\
+(download|broadband|view|watch|free|get|extreem)? \
+(sex|xxx|porn|cumshot|fuck(ing|s)?|anal|ass|asian|adult|Amateur|org(y|ies)|close ups?|hand ?      job|nail(ed)?)+ \
+(movies?|pics?|videos?|dvds?|dvd's|links?).*$\
+/This page has been blocked by Privoxy's crude-parental content filter\
+/is 
+
+#Remove by age disclaimer
+s/^.*\
+(models?|chicks?|girls?|women|persons) \
+(who|are|were)+ (over|at least) (16|18|21) years (old|of age).*$\
+/This page has been blocked by Privoxy's crude-parental content filter\
+/is 
+
+#Remove by regulations
+s/^.*(Section 2257|18 U.?S.?C.? 2257).*$\
+/This page has been blocked by Privoxy's crude-parental content filter\
+/is 
+
 
 #################################################################################
 #
@@ -501,7 +543,7 @@ 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 &lt;a> and &lt;area> tags.
+FILTER: no-ping Removes non-standard ping attributes in <a> and <area> tags.
 s@(<a(?:rea)?[^>]*?)\sping=(['"]?)([^"'>]+)\2([>\s]?)@\
 <strong style="color:white; background-color:red;" title="Privoxy removed ping target '$3'">PING!</strong>\n$1$4@ig
 
@@ -516,7 +558,7 @@ s@</head>@<style type="text/css">\n\
  /* Style sheet inserted by Privoxy's google filter. */\n\
  \#fbc, \#fbl, \#ra, .rhh {visibility: hidden !important;}\n\
  \#tpa1,\#tpa2,\#tpa3,\#tpa4,\#tpa5,\#tpa5, \#spl, .ch, \#ads,\
- \#toolbar, \#google_ads_frame{display: none !important;}\n\
+ \#toolbar, \#google_ads_frame, \#mbEnd {display: none !important;}\n\
  .main_body, .j {width: 100%}\n</style>\n$0@
 s@<div style=\"padding-top:11px;min-width:500px\">@<div id="main_body">@
 s@(<table cellspacing=0 cellpadding=0 width=25% align=right bgcolor=\#ffffff border=0\
@@ -596,36 +638,35 @@ s@(background:\#[a-f\d]{3})( url\(\"http://www.blogblog.com/rounders[^\"]*\"\).*
 
 #################################################################################
 #
-# x-httpd-php-to-html: Header filter to change the Content-Type from
+# x-httpd-php-to-html: Changes the Content-Type header from
 #                      x-httpd-php to html. "Content-Type: x-httpd-php"
 #                      is set by clueless PHP users and causes many
 #                      browsers do open a download menu instead of
 #                      rendering the page.
 #
 #################################################################################
-FILTER: x-httpd-php-to-html Header filter to change the Content-Type from x-httpd-php to html.
+SERVER-HEADER-FILTER: x-httpd-php-to-html Changes the Content-Type header from x-httpd-php to html.
 s@^(Content-Type:) application/x-httpd-php@$1 text/html@
 
 #################################################################################
 #
-# html-to-xml: Header filter to change the Content-Type from html to xml.
+# html-to-xml: Changes the Content-Type header from html to xml.
 #
 #################################################################################
-FILTER: html-to-xml Header filter to change the Content-Type from html to xml.
+SERVER-HEADER-FILTER: html-to-xml Changes the Content-Type header from html to xml.
 s@^(Content-Type:) text/html(;.*)?$@$1 application/xhtml+xml$2@
 
 #################################################################################
 #
-# xml-to-html: Header filter to change the Content-Type from xml to html.
+# xml-to-html: Changes the Content-Type header from xml to html.
 #
 #################################################################################
-FILTER: xml-to-html Header filter to change the Content-Type from xml to html.
+SERVER-HEADER-FILTER: xml-to-html Changes the Content-Type header from xml to html.
 s@^(Content-Type:) (?:application|text)/(?:xhtml\+)?xml(;.*)?$@$1 text/html$2@
 
 #################################################################################
 #
-# hide-tor-exit-notation: Header filter to remove the Tor exit node notation
-#                         in Host and Referer headers.
+# hide-tor-exit-notation: Remove the Tor exit node notation in Host and Referer headers.
 #
 #   Note: If Privoxy and Tor are chained and Privoxy is configured to
 #         use socks4a, one can use http://www.example.org.foobar.exit/
@@ -647,7 +688,7 @@ s@^(Content-Type:) (?:application|text)/(?:xhtml\+)?xml(;.*)?$@$1 text/html$2@
 #         coming from.
 #
 #################################################################################
-FILTER: hide-tor-exit-notation Header filter to remove the Tor exit node notation in Host and Referer headers.
+CLIENT-HEADER-FILTER: hide-tor-exit-notation Removes the Tor exit node notation in Host and Referer headers.
 s@^((?:Referer|Host):\s*(?:https?://)?[^/]*)\.[^\./]*?\.exit@$1@i
 
 
@@ -655,6 +696,54 @@ s@^((?:Referer|Host):\s*(?:https?://)?[^/]*)\.[^\./]*?\.exit@$1@i
 #
 #  Revisions   :
 #     $Log: default.filter,v $
+#     Revision 1.39  2007/02/21 14:10:23  fabiankeil
+#     - Fix a js-annoyances pcrs command that broke
+#       evaluated code. (BR #1124071, thanks to Bor Gergely)
+#     - Have unsolicited-popups and all-popups catch the
+#       wheather.com popup reported in in AF #1640173.
+#
+#     Revision 1.38  2007/02/19 11:22:48  hal9
+#     Adding back the orginal filter content to offset problems found by Fabian.
+#
+#     Revision 1.37  2007/02/17 13:29:44  hal9
+#     Updates to the crude parental filter per Feature Requests item #1648657.
+#
+#     Revision 1.36  2007/02/05 16:47:31  fabiankeil
+#     - Let banners-by-link look for "advert".
+#     - Fix XML systax problems with banners-by-link
+#       and banners-by-size (AF#1651570).
+#
+#     Revision 1.35  2006/12/21 12:28:12  fabiankeil
+#     Escaping special characters in filter descriptions is no
+#     longer necessary, it's done by Privoxy now.
+#
+#     Revision 1.34  2006/12/12 17:32:23  fabiankeil
+#     Added id mbEnd to google filter, it's now and then
+#     used for the sponsored links.
+#
+#     Have js-annoyances try to prevent status bar
+#     modifications where the status bar text is
+#     inside another variable. Fixes 1605710.
+#
+#     Revision 1.33  2006/11/16 17:10:43  fabiankeil
+#     Removed webbugs debugging comment again.
+#     The apostrophe could break JavaScript and
+#     the comment itself could mess up existing
+#     comments.
+#
+#     Revision 1.32  2006/11/10 18:04:04  fabiankeil
+#     Have no-ping print the ping warning in red.
+#
+#     Modified yahoo to keep in sync with recent
+#     CSS changes and to suppress a useless horizontal
+#     scrollbar.
+#
+#     msn now makes sure that the continue-link boxes
+#     act as links (the original CSS just changes the cursor).
+#
+#     Changed fun filter regex to leave microsoft links alone.
+#     Fixes BR 1019996.
+#
 #     Revision 1.31  2006/10/21 13:12:28  fabiankeil
 #     Added no-ping and hide-tor-exit-notation.
 #