2 /* Washtml, a HTML sanityzer.
4 * Copyright (c) 2007 Frederic Motte <fmotte@ubixis.com>
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
17 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
18 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
19 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
20 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
21 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
22 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
23 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
25 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
28 /* Please send me your comments about this code if you have some, thanks, Fred. */
32 * Wahstml take an untrusted HTML and return a safe html string.
36 * washtml::wash($html, $config, $full);
37 * It return a sanityzed string of the $html parameter without html and head tags.
38 * $html is a string containing the html code to wash.
39 * $config is an array containing options:
40 * $config['allow_remote'] is a boolean to allow link to remote images.
41 * $config['blocked_src'] string with image-src to be used for blocked remote images
42 * $config['show_washed'] is a boolean to include washed out attributes as x-washed
43 * $config['cid_map'] is an array where cid urls index urls to replace them.
44 * $config['charset'] is a string containing the charset of the HTML document if it is not defined in it.
45 * $full is a reference to a boolean that is set to true if no remote images are removed. (FE: show remote images link)
49 * Only tags and attributes in the globals $html_elements and $html_attributes
50 * are kept, inline styles are also filtered: all style identifiers matching
51 * /[a-z\-]/i are allowed. Values matching colors, sizes, /[a-z\-]/i and safe
52 * urls if allowed and cid urls if mapped are kept.
54 * BUGS: It MUST be safe !
56 * - urlencode URLs instead of htmlspecials
57 * - Check is a 3 bytes utf8 first char can eat '">'
58 * - Update PCRE: CVE-2007-1659 - CVE-2007-1660 - CVE-2007-1661 - CVE-2007-1662
59 * CVE-2007-4766 - CVE-2007-4767 - CVE-2007-4768
60 * http://lists.debian.org/debian-security-announce/debian-security-announce-2007/msg00177.html
64 * - relative links, can be implemented by prefixing an absolute path, ask me
69 * - Dont alter data on a GET: '<img src="http://yourhost/mail?action=delete&uid=3267" />'
76 /* Allowed HTML elements */
77 static $html_elements = array('a', 'abbr', 'acronym', 'address', 'area', 'b', 'basefont', 'bdo', 'big', 'blockquote', 'body', 'br', 'caption', 'center', 'cite', 'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', 'em', 'fieldset', 'font', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'ins', 'label', 'legend', 'li', 'map', 'menu', 'ol', 'p', 'pre', 'q', 's', 'samp', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table', 'tbody', 'td', 'tfoot', 'th', 'thead', 'title', 'tr', 'tt', 'u', 'ul', 'var', 'img');
79 /* Allowed HTML attributes */
80 static $html_attribs = array('name', 'class', 'title', 'alt', 'width', 'height', 'align', 'nowrap', 'col', 'row', 'id', 'rowspan', 'colspan', 'cellspacing', 'cellpadding', 'valign', 'bgcolor', 'color', 'border', 'bordercolorlight', 'bordercolordark', 'face', 'marginwidth', 'marginheight', 'axis', 'border', 'abbr', 'char', 'charoff', 'clear', 'compact', 'coords', 'vspace', 'hspace', 'cellborder', 'size', 'lang', 'dir');
83 static function wash_style($style, $config, &$full) {
86 foreach(explode(';', $style) as $declaration) {
87 if(preg_match('/^\s*([a-z\-]+)\s*:\s*(.*)\s*$/i', $declaration, $match)) {
91 while(sizeof($str) > 0 &&
92 preg_match('/^(url\(\s*[\'"]?([^\'"\)]*)[\'"]?\s*\)'./*1,2*/
93 '|rgb\(\s*[0-9]+\s*,\s*[0-9]+\s*,\s*[0-9]+\s*\)'.
94 '|-?[0-9.]+\s*(em|ex|px|cm|mm|in|pt|pc|deg|rad|grad|ms|s|hz|khz|%)?'.
95 '|#[0-9a-f]{3,6}|[a-z0-9\-]+'.
96 ')\s*/i', $str, $match)) {
98 if(preg_match('/^(http|https|ftp):.*$/i', $match[2], $url)) {
99 if($config['allow_remote'])
100 $value .= ' url(\''.htmlspecialchars($url[0], ENT_QUOTES).'\')';
103 } else if(preg_match('/^cid:(.*)$/i', $match[2], $cid))
104 $value .= ' url(\''.htmlspecialchars($config['cid_map']['cid:'.$cid[1]], ENT_QUOTES) . '\')';
105 } else if($match[0] != 'url' && $match[0] != 'rbg')//whitelist ?
106 $value .= ' ' . $match[0];
107 $str = substr($str, strlen($match[0]));
110 $s .= ($s?' ':'') . $cssid . ':' . $value . ';';
116 /* Take a node and return allowed attributes and check values */
117 static function wash_attribs($node, $config, &$full) {
121 foreach($node->attributes as $key => $plop) {
122 $key = strtolower($key);
123 $value = $node->getAttribute($key);
124 if((in_array($key, self::$html_attribs)) ||
125 ($key == 'href' && preg_match('/^(http|https|ftp|mailto):.*/i', $value)))
126 $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
127 else if($key == 'style' && ($style = self::wash_style($value, $config, $full)))
128 $t .= ' style="' . $style . '"';
129 else if($key == 'src' && strtolower($node->tagName) == 'img') { //check tagName anyway
130 if(preg_match('/^(http|https|ftp):.*/i', $value)) {
131 if($config['allow_remote'])
132 $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
135 if ($config['blocked_src'])
136 $t .= ' src="' . htmlspecialchars($config['blocked_src'], ENT_QUOTES) . '"';
138 } else if(preg_match('/^cid:(.*)$/i', $value, $cid))
139 $t .= ' ' . $key . '="' . htmlspecialchars($config['cid_map']['cid:'.$cid[1]], ENT_QUOTES) . '"';
141 $washed .= ($washed?' ':'') . $key;
143 return $t . ($washed && $config['show_washed']?' x-washed="'.$washed.'"':'');
146 /* The main loop that recurse on a node tree.
147 * It output only allowed tags with allowed attributes
148 * and allowed inline styles */
149 static function dumpHtml($node, $config, &$full) {
150 if(!$node->hasChildNodes())
153 $node = $node->firstChild;
157 switch($node->nodeType) {
158 case XML_ELEMENT_NODE: //Check element
159 $tagName = strtolower($node->tagName);
160 if(in_array($tagName, self::$html_elements)) {
161 $content = self::dumpHtml($node, $config, $full);
162 $dump .= '<' . $tagName . self::wash_attribs($node, $config, $full) .
163 ($content?">$content</$tagName>":' />');
164 } else if($tagName == 'html' || $tagName == 'body') {
165 $dump .= self::dumpHtml($node, $config, $full); //Just ignored
167 $dump .= '<!-- ' . htmlspecialchars($tagName, ENT_QUOTES) . ' not allowed -->';
170 $dump .= htmlspecialchars($node->nodeValue);
172 case XML_HTML_DOCUMENT_NODE:
173 $dump .= self::dumpHtml($node, $config, $full);
175 case XML_DOCUMENT_TYPE_NODE: break;
178 } while($node = $node->nextSibling);
183 /* Main function, give it untrusted HTML, tell it if you allow loading
184 * remote images and give it a map to convert "cid:" urls. */
185 static function wash($html, $config=array(), &$full=true) {
186 $config += array('show_washed'=>true, 'allow_remote'=>false, 'cid_map'=>array());
187 //Charset seems to be ignored (probably if defined in the HTML document)
188 $node = new DOMDocument('1.0', $config['charset']);
190 @$node->loadHTML($html);
191 return self::dumpHtml($node, $config, $full);