# ******************************************************************** # # File : $Source: /cvsroot/ijbswa/current/default.filter,v $ # # $Id: default.filter,v 1.28 2006/10/11 13:31:13 fabiankeil Exp $ # # Purpose : Rules to process the content of web pages # # Copyright : Written by and Copyright (C) 2001 - 2006 the # Privoxy team. http://www.privoxy.org/ # # We value your feedback. However, to provide you with the best support, # please note: # # * Use the support forum to get help: # http://sourceforge.net/tracker/?group_id=11118&atid=211118 # * Submit bugs only thru our bug forum: # http://sourceforge.net/tracker/?group_id=11118&atid=111118 # Make sure that the bug has not already been submitted. Please try # to verify that it is a Privoxy bug, and not a browser or site # bug first. If you are using your own custom configuration, please # try the stock configs to see if the problem is a configuration # related bug. And if not using the latest development snapshot, # please try the latest one. Or even better, CVS sources. # * Submit feature requests only thru our feature request forum: # http://sourceforge.net/tracker/?atid=361118&group_id=11118&func=browse # # For any other issues, feel free to use the mailing lists: # http://sourceforge.net/mail/?group_id=11118 # # Anyone interested in actively participating in development and related # discussions can join the appropriate mailing list here: # http://sourceforge.net/mail/?group_id=11118. Archives are available # here too. # ################################################################################# # # Syntax: # # Filters start with a line "FILTER: name description". They are then referrable # from the actionsfile with +filter{name} # # Inside the filters, write one Perl-Style substitution (job) per line. # Jobs that precede the first FILTER: line are ignored. # # For Details see the pcrs manpage contained in this distribution. # (and the perlre, perlop and pcre manpages) # # Note that you are free to choose the delimter as you see fit. # # Note2: In addidion 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. # 'T' (trivial) prevents parsing for backreferences in the substitute. # Use if you want to include text like '$&' in your substitute without # quoting. # ################################################################################# ################################################################################# # # js-annoyances: Get rid of particularly annoying JavaScript abuse # ################################################################################# FILTER: js-annoyances Get rid of particularly annoying JavaScript abuse # Note: Most of these jobs would be safer if restricted to a # )|$1never|sigU # 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*\([^\)]+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 s/(open\s*\([^\)]+hotkeys=)(["']?)(?:yes|1)\2/$1$2no$2/sigU s/(open\s*\([^\)]+titlebar=)(["']?)(?:no|0)\2/$1$2yes$2/sigU s/(open\s*\([^\)]+always(?:raised|lowered)=)(["']?)(?:yes|1)\2/$1$2no$2/sigU ################################################################################# # # js-events: Kill all JS event bindings and timers (Radically destructive! Only for extra nasty sites) # ################################################################################# FILTER: js-events Kill all JS event bindings and timers (Radically destructive! Only for extra nasty sites) s/(on|event\.)((mouse(over|out|down|up|move))|(un)?load|contextmenu|selectstart)/never/ig # Not events, but abused on the same type of sites: s/(alert|confirm)\s*\(/concat(/ig s/settimeout\(/concat(/ig ################################################################################# # # html-annoyances: Get rid of particularly annoying HTML abuse # ################################################################################# 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/(]+resizable=)(['"]?)(?:no|0)\2/$1$2yes$2/igU s/(]+location=)(['"]?)(?:no|0)\2/$1$2yes$2/igU s/(]+status=)(['"]?)(?:no|0)\2/$1$2yes1$2/igU s/(]+scrolling=)(['"]?)(?:no|0)\2/$1$2auto$2/igU s/(]+menubar=)(['"]?)(?:no|0)\2/$1$2yes$2/igU # The and tags were crimes! # s---sigU ################################################################################# # # content-cookies: Kill cookies that come in the HTML or JS content # ################################################################################# FILTER: content-cookies Kill cookies that come in the HTML or JS content # JS cookies, except those used by antiadbuster.com to detect us: # s|(\w+\.)+cookie(?=[ \t\r\n]*=)(?!='aab)|ZappedCookie|ig # HTML cookies: # s|||igU ################################################################################# # # refresh-tags: Kill automatic refresh tags (for dial-on-demand setups) # ################################################################################# FILTER: refresh-tags Kill automatic refresh tags (for dial-on-demand setups) # Note: Only deactivates refreshes with more than 9 seconds delay to # preserve monster-stupid but common redirections via meta tags. # s/\2]*))?\2/)(?=\s*[^'"])+$1+isU s+([^\w\s.]\s*)((window|this|parent)\.)?open\s*\(+$1PrivoxyWindowOpen(+ig s+([^'"]\s*)(?!\s*(\\n|'|"))+$1+iU ################################################################################## # # all-popups: Kill all popups in JavaScript and HTML # ################################################################################# FILTER: all-popups Kill all popups in JavaScript and HTML s/((\W\s*)(window|this|parent)\.)open\s*\\?\(/$1concat(/ig # JavaScript s/\starget\s*=\s*(['"]?)_?(blank|new)\1?/ notarget/ig # HTML ################################################################################## # # img-reorder: Reorder attributes in 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. # 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|]*) src\s*=\s*(['"])([^>\\\2]+)\2|]*) src\s*=\s*([^'">\\\s]+)|\\\\2]*\2\|[^'">\\\s]+?))([^>]*)\s+width\s*=\s*(["']?)(\d+?)\4|\\\1\s]+)\1)?[^>]*?(width=(['"]?)88\4)[^>]*?(height=(['"]?)31\6)[^>]*>@@sig # 120*60, 120*90, 120*240, 120*600 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)120\4)[^>]*?(height=(['"]?)(?:600?|90|240)\6)[^>]*>@@sig # 125*125 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)125\4)[^>]*?(height=(['"]?)125\6)[^>]*>@@sig # 160*600 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)160\4)[^>]*?(height=(['"]?)600\6)[^>]*>@@sig # 180*150 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)180\4)[^>]*?(height=(['"]?)150\6)[^>]*>@@sig # 234*60, 468*60 (Most Banners!) s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:234|468)\4)[^>]*?(height=(['"]?)60\6)[^>]*>@@sig # 240*400 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)240\4)[^>]*?(height=(['"]?)400\6)[^>]*>@@sig # 250*250, 300*250 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)(?:250|300)\4)[^>]*?(height=(['"]?)250\6)[^>]*>@@sig # 336*280 s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)336\4)[^>]*?(height=(['"]?)280\6)[^>]*>@@sig # Note: 200*50 was also proposed, but it probably causes too much collateral damage: # #s@\\\1\s]+)\1)?[^>]*?(width=(['"]?)200\4)[^>]*?(height=(['"]?)50\6)[^>]*>@@sig ################################################################################# # # banners-by-link: Kill banners by their links to known clicktrackers (Experimental) # ################################################################################# FILTER: banners-by-link Kill banners by their links to known clicktrackers # Common case with width and height attributes: # s@\1\s]*?(?:\ adclick # See www.dn.se \ | 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*\\\3\s]+)\3)?[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\6)[^>]*((?:width|height)\s*=\s*(['"]?)\d+?\8)[^>]*>\ @@sigx # Rare case w/o explicit dimensions: # s@\1\s]*?(?:adclick|atwola\.com/(?:link|redir)|doubleclick\.net/jump/|tracker|counter|adlog\.pl)[^>\1\s]*)\1[^>]*>\s*\\\3\s]+)\3)?[^>]*>@@sig ################################################################################ # # webbugs: Squish WebBugs (1x1 invisible GIFs used for user tracking) # ################################################################################# FILTER: webbugs Squish WebBugs (1x1 invisible GIFs used for user tracking) s@]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*\s(?:width|height)\s*=\s*['"]?[01](?=\D)[^>]*?>@@siUg ################################################################################# # # tiny-textforms: Extend those tiny textareas up to 40x80 and kill the hard wrap # ################################################################################# FILTER: tiny-textforms Extend those tiny textareas up to 40x80 and kill the hard wrap s/(]*?)(?:\s*(?:rows|cols)=(['"]?)\d+\2)+/$1 rows=$2\40$2 cols=$2\80$2/ig s/(]*?)wrap=(['"]?)hard\2/$1/ig ################################################################################# # # jumping-windows: Prevent windows from resizing and moving themselves # ################################################################################# FILTER: jumping-windows Prevent windows from resizing and moving themselves s/(?:window|this|self)\.(?:move|resize)(?:to|by)\(/''.concat(/ig ################################################################################# # # frameset-borders: Give frames a border, make them resizable and scrollable # ################################################################################# FILTER: frameset-borders Give frames a border and make them resizable s/(]*)framespacing=(['"]?)(no|0)\2/$1/igU s/(]*)frameborder=(['"]?)(no|0)\2/$1/igU s/(]*)border=(['"]?)(no|0)\2/$1/igU s/(]*)noresize/$1/igU s/(]*)frameborder=(['"]?)(no|0)\2/$1/igU s/(]*)scrolling=(['"]?)(no|0)\2/$1/igU ################################################################################# # # 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: # John Walker -- January 1998, http://www.fourmilab.ch/webtools/demoroniser # ################################################################################# FILTER: demoronizer Fix MS's non-standard use of standard charsets s/(&\#[0-2]\d\d)\s/$1; /g # per Robert Lynch: http://slate.msn.com//?id=2067547, just a guess. # Must come before x94 below. s/\xE2\x80\x94/ -- /g s/\x82/,/g #s-\x83-f-g s/\x84/,,/g s/\x85/.../g #s/\x88/^/g #s-\x89- °/°°-g s/\x8B/~-g #s-\x99-TM-g # per Robert Lynch. s/\x9B/>/g # 155 ################################################################################# # # shockwave-flash: Kill embedded Shockwave Flash objects # Note: Better just block "/.*\.swf$"! # ################################################################################# FILTER: shockwave-flash Kill embedded Shockwave Flash objects s|]*macromedia.*||sigU s|]*(application/x-shockwave-flash\|\.swf).*>(.*)?||sigU ################################################################################# # # quicktime-kioskmode: Make Quicktime movies saveable # ################################################################################# FILTER: quicktime-kioskmode Make Quicktime movies saveable s/(]*)kioskmode\s*=\s*(["']?)true\2/$1/ig ################################################################################# # # fun: Text replacements for subversive browsing fun! # ################################################################################# FILTER: fun Text replacements for subversive browsing fun! # SCNR # s/microsoft(?!.com)/MicroSuck/ig # Buzzword Bingo (example for extended regex syntax) # s* (?:industry|world)[ -]leading \ | cutting[ -]edge \ | customer[ -]focused \ | market[ -]driven \ | award[ -]winning # Comments are OK, too! \ | high[ -]performance \ | solutions[ -]based \ | unmatched \ | unparalleled \ | unrivalled \ *$0Bingo! \ *igx # For Germans only # s/(M|m)edien(?![^<]*>)/$1ädchen/Ug ################################################################################# # # crude-parental: Crude parental filtering? (Use along with a suitable blocklist). # Shows how to deny access to whole page based on a keyword. # ################################################################################# FILTER: crude-parental Crude parental filtering (demo only) # (Note: Middlesex, Sussex and Essex are counties in the UK, not rude words) # (Note #2: Is 'sex' a rude word?!) s%^.*(?Blocked

Blocked due to possible adult content. Please see this site.

%is s+^.*warez.*$+No Warez

You're not searching for illegal stuff, are you?

+is ################################################################################# # # IE-Exploits: Disable some known Internet Explorer bug exploits # ################################################################################# FILTER: ie-exploits Disable some known Internet Explorer bug exploits # Note: This is basically a demo and waits for someone more interested in IE # security (sic!) to take over. # Cross-site-scripting: # s%f\("javascript:location.replace\('mk:@MSITStore:C:'\)"\);%alert\("This page looks like it tries to use a vulnerability described here:\n http://online.securityfocus.com/archive/1/298748/2002-11-02/2002-11-08/2"\);%siU # Address bar spoofing (http://www.secunia.com/advisories/10395/): # s/(]*href[^>]*)(?:\x01|\x02|\x03|%0[012])@/$1MALICIOUS-LINK@/ig # Nimda: # s%%
WARNING: This Server is infected with Nimda!%g ################################################################################# # # # site-specifics: Cure for site-specific problems. Don't apply generally! # # Note: The fixes contained here are so specific to the problems of the # particular web sites they are designed for that they would be a # waste of CPU cycles (or even destructive!) on 99.9% of the web # sites where they don't apply. # ################################################################################# 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 # s/indexOf\("x11"\)/indexOf("x13")/ # www.quelle-bausparkasse.de uses a very stupid redirect mechanism that # relies on a webbug being present. Can we tolerate that? No! # Apply to: www.quelle-bausparkasse.de/$ # s/mylogfunc()//g # groups.yahoo.com has splash pages that one needs to click through in # order to access the actual messages. Let the browser do that. Thanks # to Paul Jobson for this one: # s|(?:Continue to message\|Weiter zu Nachricht)||ig # monster.com has two very similar gimmicks: # s|||i s|||i # nytimes.com triggers popups through the onload handler of dummy images # to fool popup-blockers. # s|(]*)onload|$1never|sig # Pre-check all the "Discard" buttons in GNU Mailman's web interface. # (This saves a lot of mouse aiming practice when flushing spamtraps) # s|(@\n$0@ s@
@
@ s@(
@\n$0@ ################################################################################# # # msn: CSS-based block for MSN text ads. Also removes tracking URLs # and a width limitation. # ################################################################################# FILTER: msn CSS-based block for MSN text ads. Also removes tracking URLs and a width limitation. s@@\n$0@ s@(]*) id=(["']?)ads_[^\2]*\2@$1 class="msn_ads"@Uig s@(]*href=\")http://g.msn.com/.*\?(http://.*)(&&DI=.*)(\")@$1$2$4@Ug s@(]*)gping=\".*\"@$1 title="URL cleaned up by Privoxy's msn filter"@Ug ################################################################################# # # blogspot: Cleans up some Blogspot blogs. Read the fine print before using this. # # This filter also intentionally removes some navigation stuff and # sets the page width to 100%. As a result, some rounded "corners" would # appear to early or not at all and as fixing this would require a browser # that understands background-size (CSS3), they are removed instead. # ################################################################################# FILTER: blogspot Cleans up some Blogspot blogs. Read the fine print before using this. s@@\n$0@ s@|(
tags enclosed in JS strings # # Revision 1.11.2.14 2003/03/19 13:17:50 oes # - Added filter "site-specifics" to address site specific problems # - Fixed a small problem in the img-reorder filter # # Revision 1.11.2.13 2003/03/18 19:28:59 oes # Fixed a minor problem in the img-reorder filter # # Revision 1.11.2.12 2003/03/15 14:06:58 oes # - Assorted refinements, optimizations and fixes in the js-annoyances, # img-reorder, banners-by-size, banners-by-link, webbugs, refresh-tags, # html-annoyances, content-cookies and fun filters # - Replaced filter "popups" by choice between two modes: # - "unsolicited-popups" tries to catch only the unsolicited ones # - "all-popups" tries to kill them all (as before) # - New filter "tiny-textforms" Help those tiny or hard-wrap textareas. # - New filter "jumping-windows" that prevents windows from resizing # and moving themselves # - Replaced "nimda" with more general "ie-exploits" filter in which # all filters for exploits shall be collected # # Revision 1.11.2.11 2002/11/12 16:14:43 oes # Exchanged js-annoyance filter against status bar rewrites with improved version by Don Libes # # Revision 1.11.2.10 2002/11/11 13:39:47 oes # Make refresh-tags filter work even on incorrect refresh tags like found on usatoday.com # # Revision 1.11.2.9 2002/11/08 16:39:17 oes # Made img-reorder more cautious. Fixes bug #632715 # # Revision 1.11.2.8 2002/10/13 21:56:52 hal9 # Adding demoronizer filter. This should include all the common abuses. I have # left a few of the rare cases commented out (never found these in the wild). # # Revision 1.11.2.7 2002/09/25 15:09:39 oes # Preserve original quoting style in tags wherever possible. Fixes Bug #605956 # # Revision 1.11.2.6 2002/08/23 14:12:26 oes # Proofed frameset-borders against "fremaborder=0 border=0" # # Revision 1.11.2.5 2002/08/22 15:05:20 oes # Added Filter to make Quicktime movies saveable (thanks to aaron@linville.org for the idea) # # Revision 1.11.2.4 2002/08/10 11:32:29 oes # Attribute values in replacement tags of banners-by-size filter now undelimited. (Fixes bug #592493) # # Revision 1.11.2.3 2002/08/05 11:43:56 oes # Fixed a bug in the popups filter that was introduced with the last fix :-( # # Revision 1.11.2.2 2002/08/01 11:20:13 oes # Fixed bugs 587802, 577802 and an unreported one # # Revision 1.11.2.1 2002/07/26 15:18:26 oes # - All filters reviewed and many shorcomings fixed # - New filters: img-reorder, banners-by-link and js-events # - Jobs reorderd because they are now executed in order of # appearance # # Revision 1.11 2002/05/24 00:57:18 oes # Made WeBugs job ungreedy; Fixes bug 559190 # # Revision 1.10 2002/04/18 10:14:19 oes # renamed some filters # # Revision 1.9 2002/04/11 07:36:35 oes # Generalized js-popup filter # # Revision 1.8 2002/04/10 17:07:21 oes # Fixed potentially desctructive jobs, added noflash filter # # Revision 1.7 2002/04/09 18:34:51 oes # Fixed HTML syntax in replacements # # Revision 1.6 2002/04/03 19:49:52 swa # name change # # Revision 1.5 2002/03/27 15:30:26 swa # have a consistent appearance # # Revision 1.4 2002/03/26 22:29:54 swa # we have a new homepage! # # Revision 1.3 2002/03/24 16:08:03 jongfoster # Fixing banners-by-size for new config URLs # # Revision 1.2 2002/03/24 13:02:18 swa # name change related issues. # # Revision 1.1 2002/03/24 11:37:39 jongfoster # Name change # # Revision 1.24 2002/03/16 20:39:54 oes # - Added descriptions to the filters so users will know what they select in the cgi editor # - Added content-cookies filter # - Bugfixed many jobs (Thanks to Al for some hints) # # Revision 1.22 2002/03/12 13:42:50 oes # Fixing & Optimizing REs # # Revision 1.21 2002/03/12 11:59:20 oes # Beefed up Buzzword Bingo # # Revision 1.20 2002/03/12 01:42:50 oes # Introduced modular filters # # Revision 1.19 2002/03/10 19:49:24 oes # Added expression to kill referer tracking in JavaScripts # # Revision 1.18 2002/03/08 17:14:12 oes # PNG -> image in comments # # Revision 1.17 2002/03/07 03:50:54 oes # Adapted comments to new built-in images # # Revision 1.16 2002/02/21 00:12:19 jongfoster # Modifying the banner regexps to use long URLS and to autodetect # whether to show a logo or a transparent GIF, based on actionsfile # setting. # # Revision 1.15 2001/12/28 23:54:20 steudten # Fix for feature Req #495374: http-equiv problem # # Revision 1.14 2001/12/09 18:55:11 david__schmidt # Updated CODE_STATUS to beta, commented out microsuck line in re_filterfile # for 2.9.10 beta # # Revision 1.13 2001/10/13 13:11:20 joergs # Fixed WebBug filter. # # Revision 1.12 2001/10/07 15:46:42 oes # Followed Guy's proposal to change the document.cookie job # # Revision 1.11 2001/09/21 12:34:00 joergs # Added filter to replace "Nimda" code by a warning. # # Revision 1.10 2001/07/20 11:04:26 oes # Added Rodneys javascript cookie filter # # Revision 1.9 2001/07/13 14:03:48 oes # Elimiated yet another bug in the banner-by-size jobs. Shame on me! # # Revision 1.8 2001/06/29 13:34:00 oes # - Added explanation for U and T options # - Added hint on image replacement by CGI call # - Fixed bug in banner-by-size jobs # # Revision 1.7 2001/06/19 14:21:56 oes # Fixed microsuck line # # Revision 1.6 2001/06/09 14:01:57 swa # header. cosmetics. default: no messing ala microsuck. # # #