Skip to content

Instantly share code, notes, and snippets.

@ufologist
Forked from kaznum/gist:3810258
Last active March 26, 2023 00:07
Show Gist options
  • Save ufologist/5a0da51b2b9ef1b861c30254172ac3c9 to your computer and use it in GitHub Desktop.
Save ufologist/5a0da51b2b9ef1b861c30254172ac3c9 to your computer and use it in GitHub Desktop.
sanitize HTML with jQuery prevent Application from XSS attacks
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, user-scalable=no, initial-scale=1.0, maximum-scale=1.0, minimum-scale=1.0">
<title>Sanitize HTML with jQuery</title>
</head>
<body>
<h1>Sanitize HTML with jQuery</h1>
<h2>HTML Code</h2>
<div>
<textarea class="js-html-input" cols="30" rows="10" placeholder="input html code here"></textarea>
<button class="js-do-sanitize" style="display: block">sanitize it</button>
</div>
<h3>sanitize HTML result</h3>
<div class="js-result"></div>
<!-- must use jQuery 3.0+ -->
<script src="http://cdn.bootcss.com/jquery/3.0.0/jquery.js"></script>
<script>
/**
* sanitize HTML with jQuery
* example:
* sanitizer.sanitize('<a href="foo" class="bar">aaa</a><script>alert("...")<\/script>');
*/
var sanitizer = {};
(function($) {
function trimAttributes(node) {
$.each(node.attributes, function() {
var attrName = this.name;
var attrValue = this.value;
// remove attribute name start with "on", possible unsafe,
// for example: onload, onerror...
//
// remvoe attribute value start with "javascript:" pseudo protocol, possible unsafe,
// for example href="javascript:alert(1)"
if (attrName.indexOf('on') == 0 || attrValue.indexOf('javascript:') == 0) {
$(node).removeAttr(attrName);
}
});
}
function sanitize(html) {
// [jQuery.parseHTML(data [, context ] [, keepScripts ])](http://api.jquery.com/jQuery.parseHTML/) added: 1.8
// Parses a string into an array of DOM nodes.
//
// By default, the context is the current document if not specified or given as null or undefined. If the HTML was to be used in another document such as an iframe, that frame's document could be used.
//
// As of 3.0 the default behavior is changed.
//
// If the context is not specified or given as null or undefined, a new document is used.
// This can potentially improve security because inline events will not execute when the HTML is parsed. Once the parsed HTML is injected into a document it does execute, but this gives tools a chance to traverse the created DOM and remove anything deemed unsafe. This improvement does not apply to internal uses of jQuery.parseHTML as they usually pass in the current document. Therefore, a statement like $( "#log" ).append( $( htmlString ) ) is still subject to the injection of malicious code.
//
// without context do not execute script
// $.parseHTML('<div><img src=1 onerror=alert(1)></div>');
// $.parseHTML('<div><img src=1 onerror=alert(2)></div>', null);
//
// with context document execute script!
// $.parseHTML('<div><img src=1 onerror=alert(3)></div>', document);
//
// Most jQuery APIs that accept HTML strings will run scripts that are included in the HTML. jQuery.parseHTML does not run scripts in the parsed HTML unless keepScripts is explicitly true. However, it is still possible in most environments to execute scripts indirectly, for example via the <img onerror> attribute.
//
// will return []
// $.parseHTML('<script>alert(1)<\/script>', null, false);
// will return [script DOM element]
// $.parseHTML('<script>alert(1)<\/script>', null, true);
var output = $($.parseHTML('<div>' + html + '</div>', null, false));
output.find('*').each(function() {
trimAttributes(this);
});
return output.html();
}
sanitizer.sanitize = sanitize;
})(jQuery);
</script>
<script>
var html = '';
html = sanitizer.sanitize('<img src=1 onerror=alert(1)>');
// <img src=1>
console.log(html);
$('.js-result').append(html);
html = sanitizer.sanitize('<a href="javascript:alert(1)">123</a>');
// <a>123</a>
console.log(html);
$('.js-result').append(html);
$('.js-do-sanitize').on('click', function() {
html = sanitizer.sanitize($('.js-html-input').val());
console.log(html);
$('.js-result').append(html);
});
</script>
</body>
</html>
@ufologist
Copy link
Author

ufologist commented Dec 26, 2016

sanitize HTML with jQuery prevent Application from XSS attacks

must use jQuery 3.0+, see demo jquery-sanitize-html.html

jQuery.parseHTML(data [, context ] [, keepScripts ]) added: 1.8

Parses a string into an array of DOM nodes.

By default, the context is the current document if not specified or given as null or undefined. If the HTML was to be used in another document such as an iframe, that frame's document could be used.

As of 3.0 the default behavior is changed.

If the context is not specified or given as null or undefined, a new document is used.
This can potentially improve security because inline events will not execute when the HTML is parsed. Once the parsed HTML is injected into a document it does execute, but this gives tools a chance to traverse the created DOM and remove anything deemed unsafe. This improvement does not apply to internal uses of jQuery.parseHTML as they usually pass in the current document. Therefore, a statement like $( "#log" ).append( $( htmlString ) ) is still subject to the injection of malicious code.

// without context do not execute script
$.parseHTML('<div><img src=1 onerror=alert(1)></div>');
$.parseHTML('<div><img src=1 onerror=alert(2)></div>', null);
// with context document execute script!
$.parseHTML('<div><img src=1 onerror=alert(3)></div>', document);

Most jQuery APIs that accept HTML strings will run scripts that are included in the HTML. jQuery.parseHTML does not run scripts in the parsed HTML unless keepScripts is explicitly true. However, it is still possible in most environments to execute scripts indirectly, for example via the attribute.

// will return []
$.parseHTML('<script>alert(1)<\/script>', null, false);
// will return [script DOM element]
$.parseHTML('<script>alert(1)<\/script>', null, true);

Reference

  • Avoid XSS and allow some html tags with JavaScript

    In order to prevent Application from XSS attacks I usually use following rules:

    1. Determine the level of security for your application.
      • In case you need prevent XSS on client side you can use following tools:
    2. Understand whether you need to perform sanitization on client, server or both sides. In most cases it's enough to do this on server side.
    3. Understand whether you need to preserve html tags or not
  • angular-sanitize.js

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment