ts-html-sanitizer v1.0.5
ts-html-sanitizer
Extendable HTML sanitizer written in typescript on top of htmlparser2 suitable for both JS and TS projects
Usage
import * as s from 'ts-html-sanitizer'; // for TS users
// let s = require('ts-html-sanitizer'); // for JS users
const INPUT = `
<html>
<head>
<title>Something</title>
</head>
<body class="content" style="color: red">
<form action="/message" method="post">
<input name="text" />
<button type="submit">Send</button>
</form>
<main>
<h1 class="heading">Awesome page</h1>
<section style="width: 600px">
<img src="/1.jpg" alt="some image">
<P>
Some <big>BIG</big> text <a href="/boom">here</a>
With custom <font>font</font>
</P>
</section>
</main>
<script>
alert('malicious script');
</script>
</body>
</html>
`;
let sanitizer = s.sanitizer();
console.log(sanitizer.sanitize(INPUT));
// outputs
// <section><h1>Awesome page</h1><section><img src="/1.jpg" alt="some image" /><p> Some <strong>BIG</strong> text <a href="/boom" target="_blank" rel="nofollow">here</a> With custom font</p></section></section>sanitizer()
sanitizer function creates an instance of Sanitize.Sanitizer and takes an optional parameter:
tagMap: Object— transforms mapped to tag names
Sanitize.Sanitizer
Interface declares a single sanitize() function which takes following arguments:
input: string— html input to sanitizeoptions: Sanitize.Options— an object with following keys:sourceHost: string— hostname of a document source used by link transformer, it will be appended to links starting with"/"sosrc="/1.png"will becomesrc="${sourceHost}/1.png". Defaults to empty stringsourcePath: string— path of a source document, it will be append to realtive links sohref="next.html"will becomehref="${sourceHost}/${sourcePath}/next.html". Defaults to empty string
Customization
Parsing logic can be greatly altered by creating a new or modifying the default Sanitize.TagMap.
All you need to do is register your own function implementing Transforms.Transform interface within the tag map.
Transfomations can be used as middleware functions by utilizing Transfomrs.chain() which takes an array of tranformations and runs is top down.
One may also consider writing custom Tags.Tag implmentation.
More documentation to come, consult the source code for any details it's really simple