2 rcs_id('$Id: RawHtml.php,v 1.11 2004-07-05 13:12:58 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 // Moved to IniConfig and config-default.ini
24 // Define ENABLE_RAW_HTML to false (in config.ini) to disable the RawHtml
27 if (!defined('ENABLE_RAW_HTML'))
28 define('ENABLE_RAW_HTML', true);
30 if (!defined('ENABLE_RAW_HTML_LOCKEDONLY'))
31 define('ENABLE_RAW_HTML_LOCKEDONLY', true);
32 // sanitize to safe html code
33 if (!defined('ENABLE_RAW_HTML_SAFE'))
34 define('ENABLE_RAW_HTML_SAFE', true);
37 /** We defined a better policy when to allow RawHtml:
38 * ENABLE_RAW_HTML_LOCKEDONLY:
39 * - Allowed if page is locked by ADMIN_USER.
40 * ENABLE_RAW_HTML_SAFE:
41 * - Allow some sort of "safe" html tags and attributes.
42 * Unsafe attributes are automatically stripped. (Experimental!)
43 * See http://phpwiki.sourceforge.net/phpwiki/allowing%20safe%20HTML
47 * A plugin to provide for raw HTML within wiki pages.
49 class WikiPlugin_RawHtml
56 function getDescription () {
57 return _("A plugin to provide for raw HTML within wiki pages.");
60 function getVersion() {
61 return preg_replace("/[Revision: $]/", '',
62 "\$Revision: 1.11 $");
65 function managesValidators() {
66 // The plugin output will only change if the plugin
67 // invocation (page text) changes --- so the necessary
68 // validators have already been handled by displayPage.
72 function run($dbi, $argstr, &$request, $basepage) {
73 if (!defined('ENABLE_RAW_HTML') || ! ENABLE_RAW_HTML) {
74 return $this->disabled(_("Raw HTML is disabled in this wiki."));
77 return $this->error("$basepage unset?");
80 $page = $request->getPage($basepage);
81 if (ENABLE_RAW_HTML_LOCKEDONLY) {
82 if (! $page->get('locked')) {
83 return $this->disabled(fmt("%s is only allowed in locked pages.",
87 if (ENABLE_RAW_HTML_SAFE) {
88 // check for javascript handlers (on*) and style tags with external urls. no javascript urls.
89 // See also http://simon.incutio.com/archive/2003/02/23/safeHtmlChecker
90 // But we should allow not only code semantic meaning, presentational markup also.
92 // http://chxo.com/scripts/safe_html-test.php looks better
93 $argstr = $this->safe_html($argstr);
94 /*return $this->disabled(HTML(fmt("This %s plugin on %s is disabled because of unsafe HTML code. ",$this->getName(), $basepage),
95 fmt("See PhpWiki:allowing%20safe%20HTML")
100 return HTML::raw($argstr);
104 // From http://chxo.com/scripts/safe_html-test.php
105 // safe_html by Chris Snyder (csnyder@chxo.com) for http://pcoms.net
106 // - Huge thanks to James Wetterau for testing and feedback!
109 Copyright 2003 Chris Snyder. All rights reserved.
111 Redistribution and use in source and binary forms, with or without modification,
112 are permitted provided that the following conditions are met:
114 1. Redistributions of source code must retain the above copyright
115 notice, this list of conditions and the following disclaimer.
117 2. Redistributions in binary form must reproduce the above
118 copyright notice, this list of conditions and the following
119 disclaimer in the documentation and/or other materials provided
120 with the distribution.
122 THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
123 INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
124 FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
125 AUTHOR OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
126 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
127 PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
128 OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
129 WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
130 OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
131 ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
135 set of functions for sanitizing user input:
136 keeps "friendly" tags but strips javascript events and style attributes
137 closes any open comment tags
138 closes any open HTML tags - results may not be valid HTML, but
139 at least they will keep the rest of the page from breaking
141 treats the following as malicious conditions and returns text stripped
143 any instances of ='javascript:
144 event or style attributes remaining after initial replacement
147 function strip_attributes ($html, $attrs) {
148 if (!is_array($attrs)) {
149 $array= array( "$attrs" );
154 foreach ($attrs AS $attribute) {
155 // once for ", once for ', s makes the dot match linebreaks, too.
156 $search[]= "/".$attribute.'\s*=\s*".+"/Uis';
157 $search[]= "/".$attribute."\s*=\s*'.+'/Uis";
158 // and once more for unquoted attributes
159 $search[]= "/".$attribute."\s*=\s*\S+/i";
161 $html= preg_replace($search, "", $html);
163 // check for additional matches and strip all tags if found
164 foreach ($search AS $pattern) {
165 if (preg_match($pattern, $html)) {
166 $html= strip_tags($html);
174 function safe_html ($html, $allowedtags="") {
175 $version= "safe_html.php/0.4";
177 // anything with ="javascript: is right out -- strip all tags and return if found
178 $pattern= "/=\s*\S+script:\S+/Ui";
179 if (preg_match($pattern, $html)) {
180 $html= strip_tags($html);
184 // setup -- $allowedtags is an array of $tag=>$closeit pairs, where $tag is an HTML tag to allow and $closeit is 1 if the tag requires a matching, closing tag
185 if ($allowedtags=="") {
186 $allowedtags= array ( "p"=>1, "br"=>0, "a"=>1, "img"=>0, "li"=>1,
187 "ol"=>1, "ul"=>1, "b"=>1, "i"=>1, "em"=>1, "strong"=>1, "del"=>1, "ins"=>1,
188 "u"=>1, "blockquote"=>1, "pre"=>1, "hr"=>0,
189 "table"=>1, "tr"=>1, "td"=>1,
192 elseif (!is_array($allowedtags)) {
193 $array= array( "$allowedtags" );
195 $allowedtags= $array;
198 // there's some debate about this.. is strip_tags() better than rolling your own regex?
199 // note: a bug in PHP 4.3.1 caused improper handling of ! in tag attributes when using strip_tags()
201 foreach ($allowedtags AS $tag=>$closeit) {
202 $stripallowed.= "<$tag>";
205 //print "Stripallowed: $stripallowed -- ".print_r($allowedtags,1);
206 $html= strip_tags($html, $stripallowed);
208 // also, lets get rid of some pesky attributes that may be set on the remaining tags...
209 $badattrs= array("on\w+", "style");
210 $html= $this->strip_attributes($html, $badattrs);
212 // close html tags if necessary -- note that this WON'T be graceful formatting-wise, it just has to fix any maliciousness
213 foreach ($allowedtags AS $tag=>$closeit) {
214 if (!$closeit) continue;
215 $patternopen= "/<$tag\b[^>]*>/Ui";
216 $patternclose= "/<\/$tag\b[^>]*>/Ui";
217 $totalopen= preg_match_all ( $patternopen, $html, $matches );
218 $totalclose= preg_match_all ( $patternclose, $html, $matches2 );
219 if ($totalopen>$totalclose) {
220 $html.= str_repeat("</$tag>", ($totalopen - $totalclose));
224 // close any open <!--'s and identify version just in case
225 $html.= "<!-- $version -->";
230 // $Log: not supported by cvs2svn $
231 // Revision 1.10 2004/07/05 13:09:37 rurban
232 // ENABLE_RAW_HTML_LOCKEDONLY, ENABLE_RAW_HTML_SAFE
234 // Revision 1.9 2004/07/05 13:04:47 rurban
235 // new RawHtml policies: ENABLE_RAW_HTML_LOCKEDONLY, ENABLE_RAW_HTML_SAFE
237 // Revision 1.8 2003/11/22 17:50:32 carstenklapp
238 // Minor internal change: Removed redundant call to gettext within
239 // fmt(). (locale make: RawHtml.php:65: warning: keyword nested in
242 // Revision 1.7 2003/03/17 22:32:26 dairiki
243 // Minor HTTP caching fix.
245 // Revision 1.6 2003/03/17 21:24:53 dairiki
246 // Fix security bugs in the RawHtml plugin.
248 // Change the default configuration to allow use of plugin, since
249 // I believe the plugin is now safe for general use. (Raw HTML will only
250 // work on locked pages.)
252 // Revision 1.5 2003/01/18 22:01:43 carstenklapp
254 // Reformatting & tabs to spaces;
255 // Added copyleft, getVersion, getDescription, rcs_id.
263 // c-hanging-comment-ender-p: nil
264 // indent-tabs-mode: nil