1 # ********************************************************************
3 # File : $Source: /cvsroot/ijbswa/current/re_filterfile,v $
5 # Purpose : Rules to process the content of web pages
7 # Copyright : Written by and Copyright (C) 2001 the SourceForge
8 # IJBSWA team. http://ijbswa.sourceforge.net
10 # This program is free software; you can redistribute it
11 # and/or modify it under the terms of the GNU General
12 # Public License as published by the Free Software
13 # Foundation; either version 2 of the License, or (at
14 # your option) any later version.
16 # This program is distributed in the hope that it will
17 # be useful, but WITHOUT ANY WARRANTY; without even the
18 # implied warranty of MERCHANTABILITY or FITNESS FOR A
19 # PARTICULAR PURPOSE. See the GNU General Public
20 # License for more details.
22 # The GNU General Public License should be included with
23 # this file. If not, you can view it at
24 # http://www.gnu.org/copyleft/gpl.html
25 # or write to the Free Software Foundation, Inc., 59
26 # Temple Place - Suite 330, Boston, MA 02111-1307, USA.
29 # $Log: re_filterfile,v $
30 # Revision 1.21 2002/03/12 11:59:20 oes
31 # Beefed up Buzzword Bingo
33 # Revision 1.20 2002/03/12 01:42:50 oes
34 # Introduced modular filters
36 # Revision 1.19 2002/03/10 19:49:24 oes
37 # Added expression to kill referer tracking in JavaScripts
39 # Revision 1.18 2002/03/08 17:14:12 oes
40 # PNG -> image in comments
42 # Revision 1.17 2002/03/07 03:50:54 oes
43 # Adapted comments to new built-in images
45 # Revision 1.16 2002/02/21 00:12:19 jongfoster
46 # Modifying the banner regexps to use long URLS and to autodetect
47 # whether to show a logo or a transparent GIF, based on actionsfile
50 # Revision 1.15 2001/12/28 23:54:20 steudten
51 # Fix for feature Req #495374: http-equiv problem
53 # Revision 1.14 2001/12/09 18:55:11 david__schmidt
54 # Updated CODE_STATUS to beta, commented out microsuck line in re_filterfile
57 # Revision 1.13 2001/10/13 13:11:20 joergs
58 # Fixed WebBug filter.
60 # Revision 1.12 2001/10/07 15:46:42 oes
61 # Followed Guy's proposal to change the document.cookie job
63 # Revision 1.11 2001/09/21 12:34:00 joergs
64 # Added filter to replace "Nimda" code by a warning.
66 # Revision 1.10 2001/07/20 11:04:26 oes
67 # Added Rodneys javascript cookie filter
69 # Revision 1.9 2001/07/13 14:03:48 oes
70 # Elimiated yet another bug in the banner-by-size jobs. Shame on me!
72 # Revision 1.8 2001/06/29 13:34:00 oes
73 # - Added explanation for U and T options
74 # - Added hint on image replacement by CGI call
75 # - Fixed bug in banner-by-size jobs
77 # Revision 1.7 2001/06/19 14:21:56 oes
78 # Fixed microsuck line
80 # Revision 1.6 2001/06/09 14:01:57 swa
81 # header. cosmetics. default: no messing ala microsuck.
86 #################################################################################
90 #################################################################################
92 # Filters start with a line "FILTER: name". They are then referrable
93 # from the actionsfile with +filter{name}
95 # Inside the filters, write one Perl-Style substitution per line.
97 # For Details see the pcrs manpage contained in this distribution.
98 # (and the perlre, perlop and pcre manpages)
100 # Note that you are free to choose the delimter as you see fit.
102 # Note2: In addidion to the Perl options gimsx, the following nonstandard
103 # options are supported:
105 # 'U' turns the default to ungreedy matching. Add ? to quantifiers to
106 # switch back to greedy.
107 # 'T' (trivial) prevents parsing for backreferences in the substitute.
108 # Use if you want to include text like '$&' in your substitute without
111 #################################################################################
114 #################################################################################
116 # html-annoyances: Get rid of particularly annoying HTML abuse
118 #################################################################################
119 FILTER: html-annoyances
121 # New browser windows (if allowed -- see no-popups filter below) should be
122 # resizeable and have a location and status bar
124 s/(<a\s+href[^>]+)resizable="?(no|0)"?(.*>)/$1resizable=1$3/igU
125 s/(<a\s+href[^>]+)location="?(no|0)"?(.*>)/$1location=1$3/igU
126 s/(<a\s+href[^>]+)status="?(no|0)"?(.*>)/$1status=1$3/igU
127 s/(<a\s+href[^>]+)scrolling="?(no|0|auto)"?(.*>)/$1scrolling=no$3/igU
128 s/(<a\s+href[^>]+)menubar="?(no|0)"?(.*>)/$1menubar=1$3/igU
130 # The <BLINK> tag was a crime!
132 s*<blink>|</blink>**ig
136 #s/margin(height|width)=[0-9]*//gi
137 #s/noresize/yesresize/igU
139 #################################################################################
141 # js-annoyances: Get rid of particularly annoying JavaScript abuse
143 #################################################################################
144 FILTER: js-annoyances
146 # JS cookies, like found on privacy.net:
148 s|(document\.cookie)([ \t\r\n]*=)|documenZapCooky$2|g
150 # Get rid of Javascript referrer tracking. Test page: http://www.randomoddness.com/untitled.htm
152 s|(<script.*)document\.referrer(.*</script>)|$1"Not Your Business!"$2|Usg
154 # The status bar is for displaying link targets, not pointless blahblah
158 # Kill OnUnload popups. Yummy. Test: http://www.zdnet.com/zdsubs/yahoo/tree/yfs.html
160 s/(<body .*)onunload(.*>)/$1never$2/iU
163 ##################################################################################
165 # no-popups: Kill all popups in JS and HTML
167 #################################################################################
170 s/window\.open\(/1;''\.concat\(/ig # JavaScript
171 s/target=['"]?_blank['"]?/target_crunched/ig # HTML
172 s/target=['"]?_new['"]?/target_crunched/ig # HTML
174 #################################################################################
176 # frameset-borders: Give frames a border and make them resizable
178 #################################################################################
179 FILTER: frameset-borders
181 s/(<frameset[^>]+)framespacing=['"]?(no|0['"]?(.*>)/$1$3/igU
182 s/(<frameset[^>]+)border=['"]?(no|0)['"]?(.*>)/$1$3/igU
183 s/(<frameset[^>]+)frameborder=['"]?(no|0)['"]?(.*>)/$1$3/igU
186 #################################################################################
188 # webbugs: Squish WebBugs (1x1 invisible GIFs used for user tracking)
190 #################################################################################
193 s/<img\s+[^>]*?(width|height)\s*=\s*['"]?1\D[^>]*?(width|height)\s*=\s*['"]?1(\D[^>]*?)?>/<!-- Squished WebBug -->/sig
196 #################################################################################
198 # no-refresh: Automatic refresh sucks on auto-dialup lines
200 #################################################################################
203 # FIXME: second line like first line for content value
205 s/<meta[^>]*http-equiv[^>]*refresh.*([0-9]+[0-9]|"[2-9]);URL=([^>]*?)"?>/<link rev="x-refresh" href=$2>/i
206 s/<meta[^>]*http-equiv="?page-enter"?[^>]*content=[^>]*>/<!--no page enter for me-->/i
209 #################################################################################
211 # fun: Text replacements for subversive browsing fun!
213 #################################################################################
216 s/microsoft(?!.com)/MicroSuck/ig
218 # Buzzword Bingo (example for extended syntax)
220 s* industry[ -]leading \
222 | award[ -]winning # Comments are OK, too! \
223 | high[ -]performance \
224 | solutions[ -]based \
228 *<font color=red><b>BINGO!</b></font> \
232 #################################################################################
234 # nimda: Remove Nimda (virus) code
236 #################################################################################
239 s%<script language="JavaScript">(window\.open|1;''\.concat)\("readme\.eml", null, "resizable=no,top=6000,left=6000"\)</script>%<br><hr><font size=7><b>Internet J</b></font><b><font size=6>UNK</font><font size=5 color="red"><i>BUSTER</i></font></b><font size=7> WARNING: This Server is infected with <a href="http://www.cert.org/advisories/CA-2001-26.html">Nimda</a>!</font>%g
242 #################################################################################
244 # banners-by-size: Kill banners by size
246 #################################################################################
248 # Standard banner sizes taken from http://www.iab.net/iab_banner_standards/bannersizes.html
250 # Note: Use http://ijbswa.sourceforge.net/config/send-banner?type=trans for a transparent 1x1 image
251 # Use http://ijbswa.sourceforge.net/config/send-banner?type=logo for the logo image
252 # Use http://ijbswa.sourceforge.net/config/send-banner?type=logo for a grey/white pattern image
253 # Use http://ijbswa.sourceforge.net/config/send-banner?type=auto to auto-select.
255 #################################################################################
256 FILTER: banners-by-size
258 s|<img\s+[^>]*?(width=['"]?468\D)[^>]*(height=['"]?60[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
259 s|<img\s+[^>]*?(width=['"]?234\D)[^>]*(height=['"]?60[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
260 s|<img\s+[^>]*?(width=['"]?88\D)[^>]*(height=['"]?31[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
261 s|<img\s+[^>]*?(width=['"]?120\D)[^>]*(height=['"]?90[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
262 s|<img\s+[^>]*?(width=['"]?120\D)[^>]*(height=['"]?600[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
263 s|<img\s+[^>]*?(width=['"]?120\D)[^>]*(height=['"]?60[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
264 s|<img\s+[^>]*?(width=['"]?160\D)[^>]*(height=['"]?600[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
265 s|<img\s+[^>]*?(width=['"]?125\D)[^>]*(height=['"]?125[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
266 s|<img\s+[^>]*?(width=['"]?120\D)[^>]*(height=['"]?240[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
267 s|<img\s+[^>]*?(width=['"]?180\D)[^>]*(height=['"]?150[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
268 s|<img\s+[^>]*?(width=['"]?300\D)[^>]*(height=['"]?250[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
269 s|<img\s+[^>]*?(width=['"]?250\D)[^>]*(height=['"]?250[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
270 s|<img\s+[^>]*?(width=['"]?240\D)[^>]*(height=['"]?400[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
271 s|<img\s+[^>]*?(width=['"]?336\D)[^>]*(height=['"]?280[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
273 # One more. (Where is 200x50 from?)
275 s|<img\s+[^>]*?(width=['"]?200\D)[^>]*(height=['"]?50[^>]*?)>|<img src=http://ijbswa.sourceforge.net/config/send-banner?type=auto $1 $2>|sig
278 #################################################################################
280 # crude-parental: Crude parental filtering? (Use along with a suitable blocklist).
281 # Shows how to deny access to whole page based on a keyword.
283 #################################################################################
285 # (Note: Middlesex, Sussex and Essex are counties in the UK, not rude words)
286 # (Note #2: Is 'sex' a rude word?!)
288 #################################################################################
289 FILTER: crude-parental
291 s%^.*(?<!middle)(?<!sus)(?<!es)sex.*$%<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
292 s+^.*warez.*$+<html><head><title>No Warez</title></head><body><h3>You're not searching for illegal stuff, are you?</h3></body></html>+is