Chris@0: <?php
Chris@0: 
Chris@0: namespace Drupal\Component\Utility;
Chris@0: 
Chris@0: /**
Chris@0:  * Provides helper to filter for cross-site scripting.
Chris@0:  *
Chris@0:  * @ingroup utility
Chris@0:  */
Chris@0: class Xss {
Chris@0: 
Chris@0:   /**
Chris@0:    * The list of HTML tags allowed by filterAdmin().
Chris@0:    *
Chris@0:    * @var array
Chris@0:    *
Chris@0:    * @see \Drupal\Component\Utility\Xss::filterAdmin()
Chris@0:    */
Chris@0:   protected static $adminTags = ['a', 'abbr', 'acronym', 'address', 'article', 'aside', 'b', 'bdi', 'bdo', 'big', 'blockquote', 'br', 'caption', 'cite', 'code', 'col', 'colgroup', 'command', 'dd', 'del', 'details', 'dfn', 'div', 'dl', 'dt', 'em', 'figcaption', 'figure', 'footer', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'header', 'hgroup', 'hr', 'i', 'img', 'ins', 'kbd', 'li', 'mark', 'menu', 'meter', 'nav', 'ol', 'output', 'p', 'pre', 'progress', 'q', 'rp', 'rt', 'ruby', 's', 'samp', 'section', 'small', 'span', 'strong', 'sub', 'summary', 'sup', 'table', 'tbody', 'td', 'tfoot', 'th', 'thead', 'time', 'tr', 'tt', 'u', 'ul', 'var', 'wbr'];
Chris@0: 
Chris@0:   /**
Chris@0:    * The default list of HTML tags allowed by filter().
Chris@0:    *
Chris@0:    * @var array
Chris@0:    *
Chris@0:    * @see \Drupal\Component\Utility\Xss::filter()
Chris@0:    */
Chris@0:   protected static $htmlTags = ['a', 'em', 'strong', 'cite', 'blockquote', 'code', 'ul', 'ol', 'li', 'dl', 'dt', 'dd'];
Chris@0: 
Chris@0:   /**
Chris@0:    * Filters HTML to prevent cross-site-scripting (XSS) vulnerabilities.
Chris@0:    *
Chris@0:    * Based on kses by Ulf Harnhammar, see http://sourceforge.net/projects/kses.
Chris@0:    * For examples of various XSS attacks, see: http://ha.ckers.org/xss.html.
Chris@0:    *
Chris@0:    * This code does four things:
Chris@0:    * - Removes characters and constructs that can trick browsers.
Chris@0:    * - Makes sure all HTML entities are well-formed.
Chris@0:    * - Makes sure all HTML tags and attributes are well-formed.
Chris@0:    * - Makes sure no HTML tags contain URLs with a disallowed protocol (e.g.
Chris@0:    *   javascript:).
Chris@0:    *
Chris@0:    * @param $string
Chris@0:    *   The string with raw HTML in it. It will be stripped of everything that
Chris@0:    *   can cause an XSS attack.
Chris@0:    * @param array $html_tags
Chris@0:    *   An array of HTML tags.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   An XSS safe version of $string, or an empty string if $string is not
Chris@0:    *   valid UTF-8.
Chris@0:    *
Chris@0:    * @see \Drupal\Component\Utility\Unicode::validateUtf8()
Chris@0:    *
Chris@0:    * @ingroup sanitization
Chris@0:    */
Chris@0:   public static function filter($string, array $html_tags = NULL) {
Chris@0:     if (is_null($html_tags)) {
Chris@0:       $html_tags = static::$htmlTags;
Chris@0:     }
Chris@0:     // Only operate on valid UTF-8 strings. This is necessary to prevent cross
Chris@0:     // site scripting issues on Internet Explorer 6.
Chris@0:     if (!Unicode::validateUtf8($string)) {
Chris@0:       return '';
Chris@0:     }
Chris@0:     // Remove NULL characters (ignored by some browsers).
Chris@0:     $string = str_replace(chr(0), '', $string);
Chris@0:     // Remove Netscape 4 JS entities.
Chris@0:     $string = preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
Chris@0: 
Chris@0:     // Defuse all HTML entities.
Chris@0:     $string = str_replace('&', '&amp;', $string);
Chris@0:     // Change back only well-formed entities in our whitelist:
Chris@0:     // Decimal numeric entities.
Chris@0:     $string = preg_replace('/&amp;#([0-9]+;)/', '&#\1', $string);
Chris@0:     // Hexadecimal numeric entities.
Chris@0:     $string = preg_replace('/&amp;#[Xx]0*((?:[0-9A-Fa-f]{2})+;)/', '&#x\1', $string);
Chris@0:     // Named entities.
Chris@0:     $string = preg_replace('/&amp;([A-Za-z][A-Za-z0-9]*;)/', '&\1', $string);
Chris@0:     $html_tags = array_flip($html_tags);
Chris@0:     // Late static binding does not work inside anonymous functions.
Chris@0:     $class = get_called_class();
Chris@0:     $splitter = function ($matches) use ($html_tags, $class) {
Chris@0:       return $class::split($matches[1], $html_tags, $class);
Chris@0:     };
Chris@0:     // Strip any tags that are not in the whitelist.
Chris@0:     return preg_replace_callback('%
Chris@0:       (
Chris@0:       <(?=[^a-zA-Z!/])  # a lone <
Chris@0:       |                 # or
Chris@0:       <!--.*?-->        # a comment
Chris@0:       |                 # or
Chris@0:       <[^>]*(>|$)       # a string that starts with a <, up until the > or the end of the string
Chris@0:       |                 # or
Chris@0:       >                 # just a >
Chris@0:       )%x', $splitter, $string);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Applies a very permissive XSS/HTML filter for admin-only use.
Chris@0:    *
Chris@0:    * Use only for fields where it is impractical to use the
Chris@0:    * whole filter system, but where some (mainly inline) mark-up
Chris@0:    * is desired (so \Drupal\Component\Utility\Html::escape() is
Chris@0:    * not acceptable).
Chris@0:    *
Chris@0:    * Allows all tags that can be used inside an HTML body, save
Chris@0:    * for scripts and styles.
Chris@0:    *
Chris@0:    * @param string $string
Chris@0:    *   The string to apply the filter to.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The filtered string.
Chris@0:    *
Chris@0:    * @ingroup sanitization
Chris@0:    *
Chris@0:    * @see \Drupal\Component\Utility\Xss::getAdminTagList()
Chris@0:    */
Chris@0:   public static function filterAdmin($string) {
Chris@0:     return static::filter($string, static::$adminTags);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Processes an HTML tag.
Chris@0:    *
Chris@0:    * @param string $string
Chris@0:    *   The HTML tag to process.
Chris@0:    * @param array $html_tags
Chris@0:    *   An array where the keys are the allowed tags and the values are not
Chris@0:    *   used.
Chris@0:    * @param string $class
Chris@0:    *   The called class. This method is called from an anonymous function which
Chris@0:    *   breaks late static binding. See https://bugs.php.net/bug.php?id=66622 for
Chris@0:    *   more information.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   If the element isn't allowed, an empty string. Otherwise, the cleaned up
Chris@0:    *   version of the HTML element.
Chris@0:    */
Chris@0:   protected static function split($string, $html_tags, $class) {
Chris@0:     if (substr($string, 0, 1) != '<') {
Chris@0:       // We matched a lone ">" character.
Chris@0:       return '&gt;';
Chris@0:     }
Chris@0:     elseif (strlen($string) == 1) {
Chris@0:       // We matched a lone "<" character.
Chris@0:       return '&lt;';
Chris@0:     }
Chris@0: 
Chris@0:     if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9\-]+)\s*([^>]*)>?|(<!--.*?-->)$%', $string, $matches)) {
Chris@0:       // Seriously malformed.
Chris@0:       return '';
Chris@0:     }
Chris@0:     $slash = trim($matches[1]);
Chris@0:     $elem = &$matches[2];
Chris@0:     $attrlist = &$matches[3];
Chris@0:     $comment = &$matches[4];
Chris@0: 
Chris@0:     if ($comment) {
Chris@0:       $elem = '!--';
Chris@0:     }
Chris@0: 
Chris@0:     // When in whitelist mode, an element is disallowed when not listed.
Chris@0:     if ($class::needsRemoval($html_tags, $elem)) {
Chris@0:       return '';
Chris@0:     }
Chris@0: 
Chris@0:     if ($comment) {
Chris@0:       return $comment;
Chris@0:     }
Chris@0: 
Chris@0:     if ($slash != '') {
Chris@0:       return "</$elem>";
Chris@0:     }
Chris@0: 
Chris@0:     // Is there a closing XHTML slash at the end of the attributes?
Chris@0:     $attrlist = preg_replace('%(\s?)/\s*$%', '\1', $attrlist, -1, $count);
Chris@0:     $xhtml_slash = $count ? ' /' : '';
Chris@0: 
Chris@0:     // Clean up attributes.
Chris@0:     $attr2 = implode(' ', $class::attributes($attrlist));
Chris@0:     $attr2 = preg_replace('/[<>]/', '', $attr2);
Chris@0:     $attr2 = strlen($attr2) ? ' ' . $attr2 : '';
Chris@0: 
Chris@0:     return "<$elem$attr2$xhtml_slash>";
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Processes a string of HTML attributes.
Chris@0:    *
Chris@0:    * @param string $attributes
Chris@0:    *   The html attribute to process.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   Cleaned up version of the HTML attributes.
Chris@0:    */
Chris@0:   protected static function attributes($attributes) {
Chris@0:     $attributes_array = [];
Chris@0:     $mode = 0;
Chris@0:     $attribute_name = '';
Chris@0:     $skip = FALSE;
Chris@0:     $skip_protocol_filtering = FALSE;
Chris@0: 
Chris@0:     while (strlen($attributes) != 0) {
Chris@0:       // Was the last operation successful?
Chris@0:       $working = 0;
Chris@0: 
Chris@0:       switch ($mode) {
Chris@0:         case 0:
Chris@0:           // Attribute name, href for instance.
Chris@0:           if (preg_match('/^([-a-zA-Z][-a-zA-Z0-9]*)/', $attributes, $match)) {
Chris@0:             $attribute_name = strtolower($match[1]);
Chris@0:             $skip = ($attribute_name == 'style' || substr($attribute_name, 0, 2) == 'on');
Chris@0: 
Chris@0:             // Values for attributes of type URI should be filtered for
Chris@0:             // potentially malicious protocols (for example, an href-attribute
Chris@0:             // starting with "javascript:"). However, for some non-URI
Chris@0:             // attributes performing this filtering causes valid and safe data
Chris@0:             // to be mangled. We prevent this by skipping protocol filtering on
Chris@0:             // such attributes.
Chris@0:             // @see \Drupal\Component\Utility\UrlHelper::filterBadProtocol()
Chris@0:             // @see http://www.w3.org/TR/html4/index/attributes.html
Chris@0:             $skip_protocol_filtering = substr($attribute_name, 0, 5) === 'data-' || in_array($attribute_name, [
Chris@0:               'title',
Chris@0:               'alt',
Chris@0:               'rel',
Chris@0:               'property',
Chris@0:             ]);
Chris@0: 
Chris@0:             $working = $mode = 1;
Chris@0:             $attributes = preg_replace('/^[-a-zA-Z][-a-zA-Z0-9]*/', '', $attributes);
Chris@0:           }
Chris@0:           break;
Chris@0: 
Chris@0:         case 1:
Chris@0:           // Equals sign or valueless ("selected").
Chris@0:           if (preg_match('/^\s*=\s*/', $attributes)) {
Chris@0:             $working = 1; $mode = 2;
Chris@0:             $attributes = preg_replace('/^\s*=\s*/', '', $attributes);
Chris@0:             break;
Chris@0:           }
Chris@0: 
Chris@0:           if (preg_match('/^\s+/', $attributes)) {
Chris@0:             $working = 1; $mode = 0;
Chris@0:             if (!$skip) {
Chris@0:               $attributes_array[] = $attribute_name;
Chris@0:             }
Chris@0:             $attributes = preg_replace('/^\s+/', '', $attributes);
Chris@0:           }
Chris@0:           break;
Chris@0: 
Chris@0:         case 2:
Chris@0:           // Attribute value, a URL after href= for instance.
Chris@0:           if (preg_match('/^"([^"]*)"(\s+|$)/', $attributes, $match)) {
Chris@0:             $thisval = $skip_protocol_filtering ? $match[1] : UrlHelper::filterBadProtocol($match[1]);
Chris@0: 
Chris@0:             if (!$skip) {
Chris@0:               $attributes_array[] = "$attribute_name=\"$thisval\"";
Chris@0:             }
Chris@0:             $working = 1;
Chris@0:             $mode = 0;
Chris@0:             $attributes = preg_replace('/^"[^"]*"(\s+|$)/', '', $attributes);
Chris@0:             break;
Chris@0:           }
Chris@0: 
Chris@0:           if (preg_match("/^'([^']*)'(\s+|$)/", $attributes, $match)) {
Chris@0:             $thisval = $skip_protocol_filtering ? $match[1] : UrlHelper::filterBadProtocol($match[1]);
Chris@0: 
Chris@0:             if (!$skip) {
Chris@0:               $attributes_array[] = "$attribute_name='$thisval'";
Chris@0:             }
Chris@0:             $working = 1; $mode = 0;
Chris@0:             $attributes = preg_replace("/^'[^']*'(\s+|$)/", '', $attributes);
Chris@0:             break;
Chris@0:           }
Chris@0: 
Chris@0:           if (preg_match("%^([^\s\"']+)(\s+|$)%", $attributes, $match)) {
Chris@0:             $thisval = $skip_protocol_filtering ? $match[1] : UrlHelper::filterBadProtocol($match[1]);
Chris@0: 
Chris@0:             if (!$skip) {
Chris@0:               $attributes_array[] = "$attribute_name=\"$thisval\"";
Chris@0:             }
Chris@0:             $working = 1; $mode = 0;
Chris@0:             $attributes = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attributes);
Chris@0:           }
Chris@0:           break;
Chris@0:       }
Chris@0: 
Chris@0:       if ($working == 0) {
Chris@0:         // Not well formed; remove and try again.
Chris@0:         $attributes = preg_replace('/
Chris@0:           ^
Chris@0:           (
Chris@0:           "[^"]*("|$)     # - a string that starts with a double quote, up until the next double quote or the end of the string
Chris@0:           |               # or
Chris@0:           \'[^\']*(\'|$)| # - a string that starts with a quote, up until the next quote or the end of the string
Chris@0:           |               # or
Chris@0:           \S              # - a non-whitespace character
Chris@0:           )*              # any number of the above three
Chris@0:           \s*             # any number of whitespaces
Chris@0:           /x', '', $attributes);
Chris@0:         $mode = 0;
Chris@0:       }
Chris@0:     }
Chris@0: 
Chris@0:     // The attribute list ends with a valueless attribute like "selected".
Chris@0:     if ($mode == 1 && !$skip) {
Chris@0:       $attributes_array[] = $attribute_name;
Chris@0:     }
Chris@0:     return $attributes_array;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Whether this element needs to be removed altogether.
Chris@0:    *
Chris@0:    * @param $html_tags
Chris@0:    *   The list of HTML tags.
Chris@0:    * @param $elem
Chris@0:    *   The name of the HTML element.
Chris@0:    *
Chris@0:    * @return bool
Chris@0:    *   TRUE if this element needs to be removed.
Chris@0:    */
Chris@0:   protected static function needsRemoval($html_tags, $elem) {
Chris@0:     return !isset($html_tags[strtolower($elem)]);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Gets the list of HTML tags allowed by Xss::filterAdmin().
Chris@0:    *
Chris@0:    * @return array
Chris@0:    *   The list of HTML tags allowed by filterAdmin().
Chris@0:    */
Chris@0:   public static function getAdminTagList() {
Chris@0:     return static::$adminTags;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Gets the standard list of HTML tags allowed by Xss::filter().
Chris@0:    *
Chris@0:    * @return array
Chris@0:    *   The list of HTML tags allowed by Xss::filter().
Chris@0:    */
Chris@0:   public static function getHtmlTagList() {
Chris@0:     return static::$htmlTags;
Chris@0:   }
Chris@0: 
Chris@0: }