xss.js

//This module is adapted from the CodeIgniter framework
//The license is available at http://codeigniter.com/

var html_entity_decode = require('./entities').decode;

var never_allowed_str = {
    'document.cookie':              '',
    'document.write':               '',
    '.parentNode':                  '',
    '.innerHTML':                   '',
    'window.location':              '',
    '-moz-binding':                 '',
    '<!--':                         '&lt;!--',
    '-->':                          '--&gt;',
    '(<!\\[CDATA\\[)':              '&lt;![CDATA['
};

var never_allowed_regex = {
    'javascript\\s*:':              '',
    'expression\\s*(\\(|&\\#40;)':  '',
    'vbscript\\s*:':                '',
    'Redirect\\s+302':              ''
};

var non_displayables = [
    /%0[0-8bcef]/g,           // url encoded 00-08, 11, 12, 14, 15
    /%1[0-9a-f]/g,            // url encoded 16-31
    /[\x00-\x08]/g,           // 00-08
    /\x0b/g, /\x0c/g,         // 11,12
    /[\x0e-\x1f]/g            // 14-31
];

var compact_words = [
    'javascript', 'expression', 'vbscript',
    'script', 'applet', 'alert', 'document',
    'write', 'cookie', 'window'
];

exports.clean = function(str, is_image) {

    //Recursively clean objects and arrays
    if (typeof str === 'object') {
        for (var i in str) {
            str[i] = exports.clean(str[i]);
        }
        return str;
    }

    //Remove invisible characters
    str = remove_invisible_characters(str);

    //Protect query string variables in URLs => 901119URL5918AMP18930PROTECT8198
    var hash;
    do {
      // ensure str does not contain hash before inserting it
      hash = xss_hash();
    } while(str.indexOf(hash) >= 0)
    str = str.replace(/\&([a-z\_0-9]+)\=([a-z\_0-9]+)/ig, hash + '$1=$2');

    //Validate UTF16 two byte encoding (x00) - just as above, adds a semicolon if missing.
    str = str.replace(/(&\#x?)([0-9A-F]+);?/ig, '$1$2;');

    //Un-protect query string variables
    str = str.replace(new RegExp(hash, 'g'), '&');

    //Decode just in case stuff like this is submitted:
    //<a href="http://%77%77%77%2E%67%6F%6F%67%6C%65%2E%63%6F%6D">Google</a>
    try{  
      str = decodeURIComponent(str);
    }
    catch(error){
      // str was not actually URI-encoded
    }

    //Convert character entities to ASCII - this permits our tests below to work reliably.
    //We only convert entities that are within tags since these are the ones that will pose security problems.
    str = str.replace(/[a-z]+=([\'\"]).*?\1/gi, function(m, match) {
        return m.replace(match, convert_attribute(match));
    });

    //Remove invisible characters again
    str = remove_invisible_characters(str);

    //Convert tabs to spaces
    str = str.replace('\t', ' ');

    //Captured the converted string for later comparison
    var converted_string = str;

    //Remove strings that are never allowed
    for (var i in never_allowed_str) {
        str = str.replace(new RegExp(i, "gi"), never_allowed_str[i]);
    }

    //Remove regex patterns that are never allowed
    for (var i in never_allowed_regex) {
        str = str.replace(new RegExp(i, 'gi'), never_allowed_regex[i]);
    }

    //Compact any exploded words like:  j a v a s c r i p t
    // We only want to do this when it is followed by a non-word character
    for (var i = 0, l = compact_words.length; i < l; i++) {
        var spacified = compact_words[i].split('').join('\\s*')+'\\s*';

        str = str.replace(new RegExp('('+spacified+')(\\W)', 'ig'), function(m, compat, after) {
            return compat.replace(/\s+/g, '') + after;
        });
    }

    //Remove disallowed Javascript in links or img tags
    do {
        var original = str;

        if (str.match(/<a/i)) {
            str = str.replace(/<a\s+([^>]*?)(>|$)/gi, function(m, attributes, end_tag) {
                attributes = filter_attributes(attributes.replace('<','').replace('>',''));
                if (attributes.match(/href=.*?(alert\(|alert&\#40;|javascript\:|charset\=|window\.|document\.|\.cookie|<script|<xss|base64\s*,)/gi)) {
                    return m.replace(attributes, '');
                }
                return m;
            });
        }

        if (str.match(/<img/i)) {
            str = str.replace(/<img\s+([^>]*?)(\s?\/?>|$)/gi, function(m, attributes, end_tag) {
                attributes = filter_attributes(attributes.replace('<','').replace('>',''));
                if (attributes.match(/src=.*?(alert\(|alert&\#40;|javascript\:|charset\=|window\.|document\.|\.cookie|<script|<xss|base64\s*,)/gi)) {
                    return m.replace(attributes, '');
                }
                return m;
            });
        }

        if (str.match(/script/i) || str.match(/xss/i)) {
            str = str.replace(/<(\/*)(script|xss)(.*?)\>/gi, '');
        }

    } while(original != str);

    //Remove JavaScript Event Handlers - Note: This code is a little blunt.  It removes the event
    //handler and anything up to the closing >, but it's unlikely to be a problem.
    var event_handlers = ['[^a-z_\-]on\\w*'];

    //Adobe Photoshop puts XML metadata into JFIF images, including namespacing,
    //so we have to allow this for images
    if (!is_image) {
        event_handlers.push('xmlns');
    }

    str = str.replace(new RegExp("<([^><]+?)("+event_handlers.join('|')+")(\\s*=\\s*[^><]*)([><]*)", 'i'), '<$1$4');

    //Sanitize naughty HTML elements
    //If a tag containing any of the words in the list
    //below is found, the tag gets converted to entities.
    //So this: <blink>
    //Becomes: &lt;blink&gt;
    var naughty = 'alert|applet|audio|basefont|base|behavior|bgsound|blink|body|embed|expression|form|frameset|frame|head|html|ilayer|iframe|input|isindex|layer|link|meta|object|plaintext|style|script|textarea|title|video|xml|xss';
    str = str.replace(new RegExp('<(/*\\s*)('+naughty+')([^><]*)([><]*)', 'gi'), function(m, a, b, c, d) {
        return '&lt;' + a + b + c + d.replace('>','&gt;').replace('<','&lt;');
    });

    //Sanitize naughty scripting elements Similar to above, only instead of looking for
    //tags it looks for PHP and JavaScript commands that are disallowed.  Rather than removing the
    //code, it simply converts the parenthesis to entities rendering the code un-executable.
    //For example:    eval('some code')
    //Becomes:        eval&#40;'some code'&#41;
    str = str.replace(/(alert|cmd|passthru|eval|exec|expression|system|fopen|fsockopen|file|file_get_contents|readfile|unlink)(\s*)\((.*?)\)/gi, '$1$2&#40;$3&#41;');

    //This adds a bit of extra precaution in case something got through the above filters
    for (var i in never_allowed_str) {
        str = str.replace(new RegExp(i, "gi"), never_allowed_str[i]);
    }
    for (var i in never_allowed_regex) {
        str = str.replace(new RegExp(i, 'gi'), never_allowed_regex[i]);
    }

    //Images are handled in a special way
    if (is_image && str !== converted_string) {
        throw new Error('Image may contain XSS');
    }

    return str;
}

function remove_invisible_characters(str) {
    for (var i = 0, l = non_displayables.length; i < l; i++) {
        str = str.replace(non_displayables[i], '');
    }
    return str;
}

function xss_hash() {
    var str = '', num = 10;
    while (num--) str += String.fromCharCode(Math.random() * 25 | 97);
    return str;
}

function convert_attribute(str) {
    return str.replace('>','&gt;').replace('<','&lt;').replace('\\','\\\\');
}

function filter_attributes(str) {
    var comments = /\/\*.*?\*\//g;
    return str.replace(/\s*[a-z-]+\s*=\s*'[^']*'/gi, function (m) {
        return m.replace(comments, '');
    }).replace(/\s*[a-z-]+\s*=\s*"[^"]*"/gi, function (m) {
        return m.replace(comments, '');
    }).replace(/\s*[a-z-]+\s*=\s*[^\s]+/gi, function (m) {
        return m.replace(comments, '');
    });
}