In this post, I will cover how async-await can help us write better and clean tests. async-await makes working with asynchronous programming a bit easier for us, let’s see how it works –
Async-Await
async functions and await keywords were introduced in ECMAScript 2017 edition. When you pass an async keyword to a function, it returns a promise. And, the await keyword waits for the promise to be fulfilled before moving on to the next step.
So how does it relate to our API tests? Well, let’s take a look at an example. We’ll write a test to create a user post, this test will first create a user and then use the userId to create a post.
it('/posts', () => { // create user request .post('users') .set('Authorization', `Bearer ${TOKEN}`) .send(data) .then((res) => { expect(res.body.data).to.deep.include(data); userId = res.body.data.id; // create post using the above userId request .post('posts') .set('Authorization', `Bearer ${TOKEN}`) .send(data) .then((res) => { expect(res.body.data).to.deep.include(data); }); }); });
Instead of creating that massive chain and callbacks like the way we did above, we can instead use async-await to make it look a bit cleaner –
// create async function with 'it' block it('/posts', async () => { // use await to fulfill the promise and get response const userRes = await request .post('users') .set('Authorization', `Bearer ${TOKEN}`) .send(data); expect(userRes.body.data).to.deep.include(data); userId = res.body.data.id; // do the same for post request too const postRes = await request .post('posts') .set('Authorization', `Bearer ${TOKEN}`) .send(data); expect(postRes.body.data).to.deep.include(data); });
With the help of async-await, we are making the code look synchronous where it’ll do one thing, complete that and then move to another task. To make it even cleaner, we can create an async function for the user generation and call it like this –
userId = await createRandomUser();
To see a detailed explanation of the code above along with other optimization tips, check out the video below:
You can also clone the GitHub repo to access this code