2 rcs_id('$Id: RawHtml.php,v 1.9 2004-07-05 13:04:47 rurban Exp $');
4 Copyright 1999,2000,2001,2002,2004 $ThePhpWikiProgrammingTeam
6 This file is part of PhpWiki.
8 PhpWiki is free software; you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation; either version 2 of the License, or
11 (at your option) any later version.
13 PhpWiki is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with PhpWiki; if not, write to the Free Software
20 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
23 //FIXME: Move this to IniConfig and config-default.ini
24 // Define ENABLE_RAW_HTML to false (in config.ini) to disable the RawHtml
26 if (!defined('ENABLE_RAW_HTML'))
27 define('ENABLE_RAW_HTML', true);
29 if (!defined('ENABLE_RAW_HTML_LOCKEDONLY'))
30 define('ENABLE_RAW_HTML_LOCKEDONLY', true);
31 // sanitize to safe html code
32 if (!defined('ENABLE_RAW_HTML_SAFE'))
33 define('ENABLE_RAW_HTML_SAFE', true);
35 /** We'll define a better policy when to allow RawHtml:
36 * ENABLE_RAW_HTML_LOCKEDONLY:
37 * - Allowed if page is locked by ADMIN_USER.
38 * ENABLE_RAW_HTML_SAFE:
39 * - Allow some sort of "safe" html tags and attributes.
40 * See http://phpwiki.sourceforge.net/phpwiki/allowing%20safe%20HTML
44 * A plugin to provide for raw HTML within wiki pages.
46 class WikiPlugin_RawHtml
53 function getDescription () {
54 return _("A plugin to provide for raw HTML within wiki pages.");
57 function getVersion() {
58 return preg_replace("/[Revision: $]/", '',
62 function managesValidators() {
63 // The plugin output will only change if the plugin
64 // invocation (page text) changes --- so the necessary
65 // validators have already been handled by displayPage.
69 function run($dbi, $argstr, &$request, $basepage) {
70 if (!defined('ENABLE_RAW_HTML') || ! ENABLE_RAW_HTML) {
71 return $this->disabled(_("Raw HTML is disabled in this wiki."));
74 return $this->error("$basepage unset?");
77 $page = $request->getPage($basepage);
78 if (ENABLE_RAW_HTML_LOCKEDONLY) {
79 if (! $page->get('locked')) {
80 return $this->disabled(fmt("%s is only allowed in locked pages.",
84 if (ENABLE_RAW_HTML_SAFE) {
85 // check for javascript handlers (on*) and style tags with external urls. no javascript urls.
86 // See also http://simon.incutio.com/archive/2003/02/23/safeHtmlChecker
87 // But we should allow not only code semantic meaning, presentational markup also.
89 // http://chxo.com/scripts/safe_html-test.php looks better
90 $argstr = $this->safe_html($argstr);
91 /*return $this->disabled(HTML(fmt("This %s plugin on %s is disabled because of unsafe HTML code. ",$this->getName(), $basepage),
92 fmt("See PhpWiki:allowing%20safe%20HTML")
97 return HTML::raw($argstr);
101 // From http://chxo.com/scripts/safe_html-test.php
102 // safe_html by Chris Snyder (csnyder@chxo.com) for http://pcoms.net
103 // - Huge thanks to James Wetterau for testing and feedback!
106 Copyright 2003 Chris Snyder. All rights reserved.
108 Redistribution and use in source and binary forms, with or without modification,
109 are permitted provided that the following conditions are met:
111 1. Redistributions of source code must retain the above copyright
112 notice, this list of conditions and the following disclaimer.
114 2. Redistributions in binary form must reproduce the above
115 copyright notice, this list of conditions and the following
116 disclaimer in the documentation and/or other materials provided
117 with the distribution.
119 THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
120 INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
121 FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
122 AUTHOR OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
123 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
124 PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
125 OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
126 WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
127 OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
128 ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
132 set of functions for sanitizing user input:
133 keeps "friendly" tags but strips javascript events and style attributes
134 closes any open comment tags
135 closes any open HTML tags - results may not be valid HTML, but
136 at least they will keep the rest of the page from breaking
138 treats the following as malicious conditions and returns text stripped
140 any instances of ='javascript:
141 event or style attributes remaining after initial replacement
144 function strip_attributes ($html, $attrs) {
145 if (!is_array($attrs)) {
146 $array= array( "$attrs" );
151 foreach ($attrs AS $attribute) {
152 // once for ", once for ', s makes the dot match linebreaks, too.
153 $search[]= "/".$attribute.'\s*=\s*".+"/Uis';
154 $search[]= "/".$attribute."\s*=\s*'.+'/Uis";
155 // and once more for unquoted attributes
156 $search[]= "/".$attribute."\s*=\s*\S+/i";
158 $html= preg_replace($search, "", $html);
160 // check for additional matches and strip all tags if found
161 foreach ($search AS $pattern) {
162 if (preg_match($pattern, $html)) {
163 $html= strip_tags($html);
171 function safe_html ($html, $allowedtags="") {
172 $version= "safe_html.php/0.4";
174 // anything with ="javascript: is right out -- strip all tags and return if found
175 $pattern= "/=\s*\S+script:\S+/Ui";
176 if (preg_match($pattern, $html)) {
177 $html= strip_tags($html);
181 // setup -- $allowedtags is an array of $tag=>$closeit pairs, where $tag is an HTML tag to allow and $closeit is 1 if the tag requires a matching, closing tag
182 if ($allowedtags=="") {
183 $allowedtags= array ( "p"=>1, "br"=>0, "a"=>1, "img"=>0, "li"=>1, "ol"=>1, "ul"=>1, "b"=>1, "i"=>1, "em"=>1, "strong"=>1, "del"=>1, "ins"=>1, "u"=>1, "blockquote"=>1, "pre"=>1, "hr"=>0);
185 elseif (!is_array($allowedtags)) {
186 $array= array( "$allowedtags" );
188 $allowedtags= $array;
191 // there's some debate about this.. is strip_tags() better than rolling your own regex?
192 // note: a bug in PHP 4.3.1 caused improper handling of ! in tag attributes when using strip_tags()
194 foreach ($allowedtags AS $tag=>$closeit) {
195 $stripallowed.= "<$tag>";
198 //print "Stripallowed: $stripallowed -- ".print_r($allowedtags,1);
199 $html= strip_tags($html, $stripallowed);
201 // also, lets get rid of some pesky attributes that may be set on the remaining tags...
202 $badattrs= array("on\w+", "style");
203 $html= $this->strip_attributes($html, $badattrs);
205 // close html tags if necessary -- note that this WON'T be graceful formatting-wise, it just has to fix any maliciousness
206 foreach ($allowedtags AS $tag=>$closeit) {
207 if (!$closeit) continue;
208 $patternopen= "/<$tag\b[^>]*>/Ui";
209 $patternclose= "/<\/$tag\b[^>]*>/Ui";
210 $totalopen= preg_match_all ( $patternopen, $html, $matches );
211 $totalclose= preg_match_all ( $patternclose, $html, $matches2 );
212 if ($totalopen>$totalclose) {
213 $html.= str_repeat("</$tag>", ($totalopen - $totalclose));
217 // close any open <!--'s and identify version just in case
218 $html.= "<!-- $version -->";
223 // $Log: not supported by cvs2svn $
224 // Revision 1.8 2003/11/22 17:50:32 carstenklapp
225 // Minor internal change: Removed redundant call to gettext within
226 // fmt(). (locale make: RawHtml.php:65: warning: keyword nested in
229 // Revision 1.7 2003/03/17 22:32:26 dairiki
230 // Minor HTTP caching fix.
232 // Revision 1.6 2003/03/17 21:24:53 dairiki
233 // Fix security bugs in the RawHtml plugin.
235 // Change the default configuration to allow use of plugin, since
236 // I believe the plugin is now safe for general use. (Raw HTML will only
237 // work on locked pages.)
239 // Revision 1.5 2003/01/18 22:01:43 carstenklapp
241 // Reformatting & tabs to spaces;
242 // Added copyleft, getVersion, getDescription, rcs_id.
250 // c-hanging-comment-ender-p: nil
251 // indent-tabs-mode: nil