> A source is a JavaScript property that accepts data that is potentially attacker-controlled. An example of a source is the `location.search` property because it reads input from the query string, which is relatively simple for an attacker to control. Ultimately, any property that can be controlled by the attacker is a potential source. This includes the referring URL (exposed by the `document.referrer` string), the user's cookies (exposed by the `document.cookie` string), and web messages.
> A sink is a potentially dangerous JavaScript function or DOM object that can cause undesirable effects if attacker-controlled data is passed to it. For example, the `eval()` function is a sink because it processes the argument that is passed to it as JavaScript. An example of an HTML sink is `document.body.innerHTML` because it potentially allows an attacker to inject malicious HTML and execute arbitrary JavaScript.
Fundamentally, DOM-based vulnerabilities arise when a website **passes data from a source to a sink**, which then handles the data in an unsafe way in the context of the client's session.
The **`innerHTML`** sink doesn't accept `script` elements on any modern browser, nor will `svg onload` events fire. This means you will need to use alternative elements like `img` or `iframe`.
This kind of XSS is probably the **hardest to find**, as you need to look inside the JS code, see if it's **using** any object whose **value you control**, and in that case, see if there is **any way to abuse** it to execute arbitrary JS.
DOM-based open-redirection vulnerabilities arise when a script writes **attacker-controllable data** into a **sink** that can trigger **cross-domain navigation**.
Remember that **if you can start the URL** were the victim is going to be **redirected**, you could execute **arbitrary code** like: **`javascript:alert(1)`**
This could be abuse to make the page behaves on unexpected manner (if the cookie is used in the web) or to perform a [session fixation](../hacking-with-cookies/#session-fixation) attack (if the cookie is used to track the user's session).
The `document.domain` property is used by browsers in their **enforcement** of the **same origin policy**. If **two pages** from **different** origins explicitly set the **same `document.domain`** value, then those two pages can **interact in unrestricted ways**.\
Browsers **generally enforce some restrictions** on the values that can be assigned to `document.domain`, and may prevent the use of completely different values than the actual origin of the page. **But this doesn't occur always** and they usually **allow to use child** or **parent** domains.
DOM-based link-manipulation vulnerabilities arise when a script writes **attacker-controllable data to a navigation target** within the current page, such as a clickable link or the submission URL of a form.
Ajax request manipulation vulnerabilities arise when a script writes **attacker-controllable data into the an Ajax request** that is issued using an `XmlHttpRequest` object.
Local file-path manipulation vulnerabilities arise when a script passes **attacker-controllable data to a file-handling API** as the `filename` parameter. An attacker may be able to use this vulnerability to construct a URL that, if visited by another user, will cause the **user's browser to open/write an arbitrary local file**.
Client-side SQL-injection vulnerabilities arise when a script incorporates **attacker-controllable data into a client-side SQL query in an unsafe way**.
HTML5-storage manipulation vulnerabilities arise when a script **stores attacker-controllable data in the HTML5 storage** of the web browser (either `localStorage` or `sessionStorage`).\
This **behavior does not in itself constitute a security vulnerability**. However, if the application later **reads data back from storage and processes it in an unsafe way**, an attacker may be able to leverage the storage mechanism to deliver other DOM-based attacks, such as cross-site scripting and JavaScript injection.
DOM-based JSON-injection vulnerabilities arise when a script incorporates **attacker-controllable data into a string that is parsed as a JSON data structure and then processed by the application**.
Web-message vulnerabilities arise when a script sends **attacker-controllable data as a web message to another document** within the browser.\
**Example** of vulnerable Web-message manipulation in [https://portswigger.net/web-security/dom-based/controlling-the-web-message-source](https://portswigger.net/web-security/dom-based/controlling-the-web-message-source)
The `postMessage()` method for sending web messages can lead to vulnerabilities if the event listener for receiving messages handles the incoming data in an unsafe way.
DOM-data manipulation vulnerabilities arise when a script writes **attacker-controllable data to a field within the DOM** that is used within the visible UI or client-side logic. An attacker may be able to use this vulnerability to construct a URL that, if visited by another user, will modify the appearance or behaviour of the client-side UI.
DOM-based denial-of-service vulnerabilities arise when a script passes **attacker-controllable data in an unsafe way to a problematic platform API**, such as an API whose invocation can cause the user's computer to consume **excessive amounts of CPU or disk space**. This may result in side effects if the browser restricts the functionality of the website, for example, by rejecting attempts to store data in `localStorage` or killing busy scripts.
A common pattern used by JavaScript developers is:
`var someObject = window.someObject || {};`
If you can control some of the HTML on the page, you can clobber the `someObject` reference with a DOM node, such as an anchor. Consider the following code:
```python
<script>
window.onload = function(){
let someObject = window.someObject || {};
let script = document.createElement('script');
script.src = someObject.url;
document.body.appendChild(script);
};
</script>
```
To exploit this vulnerable code, you could inject the following HTML to clobber the `someObject` reference with an anchor element:
**Trick**: `DOMPurify` allows you to use the **`cid:`** protocol, which **does not URL-encode double-quotes**. This means you can **inject an encoded double-quote that will be decoded at runtime**. Therefore, injecting something like `<a id=defaultAvatar><a id=defaultAvatar name=avatar href="cid:"onerror=alert(1)//">` will make the HTML encoded `"` to be **decoded on runtime** and **escape** from the attribute value to **create** the **`onerror`** event.
Another common technique consists on using **`form`** element. Some client-side libraries will go through the attributes of the created form element to sanitised it. But, if you **create an `input`**inside the form with `id=attributes` , you will **clobber the attributes property** and the sanitizer **won't** be able to go through the **real attributes**.
