*
* SYNOPSIS:
*
- * washtml::wash($html, $config, $full);
+ * $washer = new washtml($config);
+ * $washer->wash($html);
* It return a sanityzed string of the $html parameter without html and head tags.
* $html is a string containing the html code to wash.
* $config is an array containing options:
* $config['show_washed'] is a boolean to include washed out attributes as x-washed
* $config['cid_map'] is an array where cid urls index urls to replace them.
* $config['charset'] is a string containing the charset of the HTML document if it is not defined in it.
- * $full is a reference to a boolean that is set to true if no remote images are removed. (FE: show remote images link)
+ * $washer->extlinks is a reference to a boolean that is set to true if remote images were removed. (FE: show remote images link)
*
* INTERNALS:
*
- * Only tags and attributes in the globals $html_elements and $html_attributes
+ * Only tags and attributes in the static lists $html_elements and $html_attributes
* are kept, inline styles are also filtered: all style identifiers matching
* /[a-z\-]/i are allowed. Values matching colors, sizes, /[a-z\-]/i and safe
* urls if allowed and cid urls if mapped are kept.
* Dont be a fool:
* - Dont alter data on a GET: '<img src="http://yourhost/mail?action=delete&uid=3267" />'
* - ...
+ *
+ * Roundcube Changes:
+ * - added $block_elements
+ * - changed $ignore_elements behaviour
+ * - added RFC2397 support
+ * - base URL support
+ * - invalid HTML comments removal before parsing
*/
class washtml
{
+ /* Allowed HTML elements (default) */
+ static $html_elements = array('a', 'abbr', 'acronym', 'address', 'area', 'b',
+ 'basefont', 'bdo', 'big', 'blockquote', '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', 'nobr', 'ol', 'p', 'pre', 'q',
+ 's', 'samp', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table',
+ 'tbody', 'td', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u', 'ul', 'var', 'wbr', 'img',
+ // form elements
+ 'button', 'input', 'textarea', 'select', 'option', 'optgroup'
+ );
+
+ /* Ignore these HTML tags and their content */
+ static $ignore_elements = array('script', 'applet', 'embed', 'object', 'style');
+
+ /* Allowed HTML attributes */
+ 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',
+ // attributes of form elements
+ 'type', 'rows', 'cols', 'disabled', 'readonly', 'checked', 'multiple', 'value'
+ );
+
+ /* Block elements which could be empty but cannot be returned in short form (<tag />) */
+ static $block_elements = array('div', 'p', 'pre', 'blockquote', 'a', 'font', 'center',
+ 'table', 'ul', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'ol', 'dl', 'strong', 'i', 'b', 'u');
+
+ /* State for linked objects in HTML */
+ public $extlinks = false;
+
+ /* Current settings */
+ private $config = array();
+
+ /* Registered callback functions for tags */
+ private $handlers = array();
/* Allowed HTML elements */
- 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');
+ private $_html_elements = array();
+
+ /* Ignore these HTML tags but process their content */
+ private $_ignore_elements = array();
+
+ /* Block elements which could be empty but cannot be returned in short form (<tag />) */
+ private $_block_elements = array();
/* Allowed HTML attributes */
- 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');
+ private $_html_attribs = array();
+
+
+ /* Constructor */
+ public function __construct($p = array()) {
+ $this->_html_elements = array_flip((array)$p['html_elements']) + array_flip(self::$html_elements) ;
+ $this->_html_attribs = array_flip((array)$p['html_attribs']) + array_flip(self::$html_attribs);
+ $this->_ignore_elements = array_flip((array)$p['ignore_elements']) + array_flip(self::$ignore_elements);
+ $this->_block_elements = array_flip((array)$p['block_elements']) + array_flip(self::$block_elements);
+ unset($p['html_elements'], $p['html_attribs'], $p['ignore_elements'], $p['block_elements']);
+ $this->config = $p + array('show_washed'=>true, 'allow_remote'=>false, 'cid_map'=>array());
+ }
+
+ /* Register a callback function for a certain tag */
+ public function add_callback($tagName, $callback)
+ {
+ $this->handlers[$tagName] = $callback;
+ }
/* Check CSS style */
- static function wash_style($style, $config, &$full) {
+ private function wash_style($style) {
$s = '';
- foreach(explode(';', $style) as $declaration) {
- if(preg_match('/^\s*([a-z\-]+)\s*:\s*(.*)\s*$/i', $declaration, $match)) {
+ foreach (explode(';', $style) as $declaration) {
+ if (preg_match('/^\s*([a-z\-]+)\s*:\s*(.*)\s*$/i', $declaration, $match)) {
$cssid = $match[1];
$str = $match[2];
$value = '';
- while(sizeof($str) > 0 &&
+ while (sizeof($str) > 0 &&
preg_match('/^(url\(\s*[\'"]?([^\'"\)]*)[\'"]?\s*\)'./*1,2*/
'|rgb\(\s*[0-9]+\s*,\s*[0-9]+\s*,\s*[0-9]+\s*\)'.
'|-?[0-9.]+\s*(em|ex|px|cm|mm|in|pt|pc|deg|rad|grad|ms|s|hz|khz|%)?'.
- '|#[0-9a-f]{3,6}|[a-z0-9\-]+'.
+ '|#[0-9a-f]{3,6}|[a-z0-9", -]+'.
')\s*/i', $str, $match)) {
- if($match[2]) {
- if(preg_match('/^(http|https|ftp):.*$/i', $match[2], $url)) {
- if($config['allow_remote'])
- $value .= ' url(\''.htmlspecialchars($url[0], ENT_QUOTES).'\')';
+ if ($match[2]) {
+ if (($src = $this->config['cid_map'][$match[2]])
+ || ($src = $this->config['cid_map'][$this->config['base_url'].$match[2]])) {
+ $value .= ' url('.htmlspecialchars($src, ENT_QUOTES) . ')';
+ }
+ else if (preg_match('!^(https?:)?//[a-z0-9/._+-]+$!i', $match[2], $url)) {
+ if ($this->config['allow_remote'])
+ $value .= ' url('.htmlspecialchars($url[0], ENT_QUOTES).')';
else
- $full = false;
- } else if(preg_match('/^cid:(.*)$/i', $match[2], $cid))
- $value .= ' url(\''.htmlspecialchars($config['cid_map']['cid:'.$cid[1]], ENT_QUOTES) . '\')';
- } else if($match[0] != 'url' && $match[0] != 'rbg')//whitelist ?
+ $this->extlinks = true;
+ }
+ else if (preg_match('/^data:.+/i', $match[2])) { // RFC2397
+ $value .= ' url('.htmlspecialchars($match[2], ENT_QUOTES).')';
+ }
+ }
+ else if ($match[0] != 'url' && $match[0] != 'rgb') //whitelist ?
$value .= ' ' . $match[0];
+
$str = substr($str, strlen($match[0]));
}
- if($value)
+ if ($value)
$s .= ($s?' ':'') . $cssid . ':' . $value . ';';
}
}
}
/* Take a node and return allowed attributes and check values */
- static function wash_attribs($node, $config, &$full) {
+ private function wash_attribs($node) {
$t = '';
$washed;
- foreach($node->attributes as $key => $plop) {
+ foreach ($node->attributes as $key => $plop) {
$key = strtolower($key);
$value = $node->getAttribute($key);
- if((in_array($key, self::$html_attribs)) ||
- ($key == 'href' && preg_match('/^(http|https|ftp|mailto):.*/i', $value)))
+ if (isset($this->_html_attribs[$key]) ||
+ ($key == 'href' && preg_match('/^(http:|https:|ftp:|mailto:|#).+/i', $value)))
$t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
- else if($key == 'style' && ($style = self::wash_style($value, $config, $full)))
- $t .= ' style="' . $style . '"';
- else if($key == 'src' && strtolower($node->tagName) == 'img') { //check tagName anyway
- if(preg_match('/^(http|https|ftp):.*/i', $value)) {
- if($config['allow_remote'])
+ else if ($key == 'style' && ($style = $this->wash_style($value))) {
+ $quot = strpos($style, '"') !== false ? "'" : '"';
+ $t .= ' style=' . $quot . $style . $quot;
+ }
+ else if ($key == 'background' || ($key == 'src' && strtolower($node->tagName) == 'img')) { //check tagName anyway
+ if (($src = $this->config['cid_map'][$value])
+ || ($src = $this->config['cid_map'][$this->config['base_url'].$value])) {
+ $t .= ' ' . $key . '="' . htmlspecialchars($src, ENT_QUOTES) . '"';
+ }
+ else if (preg_match('/^(http|https|ftp):.+/i', $value)) {
+ if ($this->config['allow_remote'])
$t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
else {
- $full = false;
- if ($config['blocked_src'])
- $t .= ' src="' . htmlspecialchars($config['blocked_src'], ENT_QUOTES) . '"';
+ $this->extlinks = true;
+ if ($this->config['blocked_src'])
+ $t .= ' ' . $key . '="' . htmlspecialchars($this->config['blocked_src'], ENT_QUOTES) . '"';
}
- } else if(preg_match('/^cid:(.*)$/i', $value, $cid))
- $t .= ' ' . $key . '="' . htmlspecialchars($config['cid_map']['cid:'.$cid[1]], ENT_QUOTES) . '"';
+ }
+ else if (preg_match('/^data:.+/i', $value)) { // RFC2397
+ $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
+ }
} else
$washed .= ($washed?' ':'') . $key;
}
- return $t . ($washed && $config['show_washed']?' x-washed="'.$washed.'"':'');
+ return $t . ($washed && $this->config['show_washed']?' x-washed="'.$washed.'"':'');
}
/* The main loop that recurse on a node tree.
* It output only allowed tags with allowed attributes
* and allowed inline styles */
- static function dumpHtml($node, $config, &$full) {
+ private function dumpHtml($node) {
if(!$node->hasChildNodes())
return '';
switch($node->nodeType) {
case XML_ELEMENT_NODE: //Check element
$tagName = strtolower($node->tagName);
- if(in_array($tagName, self::$html_elements)) {
- $content = self::dumpHtml($node, $config, $full);
- $dump .= '<' . $tagName . self::wash_attribs($node, $config, $full) .
- ($content?">$content</$tagName>":' />');
- } else if($tagName == 'html' || $tagName == 'body') {
- $dump .= self::dumpHtml($node, $config, $full); //Just ignored
- } else
+ if ($callback = $this->handlers[$tagName]) {
+ $dump .= call_user_func($callback, $tagName, $this->wash_attribs($node), $this->dumpHtml($node), $this);
+ }
+ else if (isset($this->_html_elements[$tagName])) {
+ $content = $this->dumpHtml($node);
+ $dump .= '<' . $tagName . $this->wash_attribs($node) .
+ // create closing tag for block elements, but also for elements
+ // with content or with some attributes (eg. style, class) (#1486812)
+ ($content != '' || $node->hasAttributes() || isset($this->_block_elements[$tagName]) ? ">$content</$tagName>" : ' />');
+ }
+ else if (isset($this->_ignore_elements[$tagName])) {
$dump .= '<!-- ' . htmlspecialchars($tagName, ENT_QUOTES) . ' not allowed -->';
+ }
+ else {
+ $dump .= '<!-- ' . htmlspecialchars($tagName, ENT_QUOTES) . ' ignored -->';
+ $dump .= $this->dumpHtml($node); // ignore tags not its content
+ }
+ break;
+ case XML_CDATA_SECTION_NODE:
+ $dump .= $node->nodeValue;
break;
case XML_TEXT_NODE:
$dump .= htmlspecialchars($node->nodeValue);
break;
case XML_HTML_DOCUMENT_NODE:
- $dump .= self::dumpHtml($node, $config, $full);
+ $dump .= $this->dumpHtml($node);
+ break;
+ case XML_DOCUMENT_TYPE_NODE:
break;
- case XML_DOCUMENT_TYPE_NODE: break;
default:
+ $dump . '<!-- node type ' . $node->nodeType . ' -->';
}
} while($node = $node->nextSibling);
/* Main function, give it untrusted HTML, tell it if you allow loading
* remote images and give it a map to convert "cid:" urls. */
- static function wash($html, $config=array(), &$full=true) {
- $config += array('show_washed'=>true, 'allow_remote'=>false, 'cid_map'=>array());
- //Charset seems to be ignored (probably if defined in the HTML document)
- $node = new DOMDocument('1.0', $config['charset']);
- $full = true;
+ public function wash($html)
+ {
+ // Charset seems to be ignored (probably if defined in the HTML document)
+ $node = new DOMDocument('1.0', $this->config['charset']);
+ $this->extlinks = false;
+
+ // Find base URL for images
+ if (preg_match('/<base\s+href=[\'"]*([^\'"]+)/is', $html, $matches))
+ $this->config['base_url'] = $matches[1];
+ else
+ $this->config['base_url'] = '';
+
+ // Remove invalid HTML comments (#1487759)
+ // Don't remove valid conditional comments
+ $html = preg_replace('/<!--[^->[\n]*>/', '', $html);
+
@$node->loadHTML($html);
- return self::dumpHtml($node, $config, $full);
+ return $this->dumpHtml($node);
+ }
+
+ /**
+ * Getter for config parameters
+ */
+ public function get_config($prop)
+ {
+ return $this->config[$prop];
}
}
-?>
\ No newline at end of file
+?>
projects / roundcube.git / blobdiff