In this article, we'll cover some best practices to create HTTP requests just using the Angular AsyncPipe. If you ever thought about making HTTP requests without calling .subscribe()
, then you're on the right page.
Why AsyncPipe?
According to Angular AsyncPipe documentation:
The
async
pipe subscribes to anObservable
orPromise
and returns the latest value it has emitted. When a new value is emitted, theasync
pipe marks the component to be checked for changes. When the component gets destroyed, theasync
pipe unsubscribes automatically to avoid potential memory leaks.
The most important part here is "when the component gets destroyed, the
async
pipe unsubscribes automatically to avoid potential memory leaks." These being said, we don't have to worry about unsubscribing the Observables, the AsyncPipe will take care for of it us - good practice (and the most important).
The Observables
Let's create some observables and get started:
As you can see, I defined two Observables, onDestroy$
and user$
. It's a good practice to have an observable which will be called when the component gets destroyed, but I'll talk later more about it.
Based on user$
BehaviorSubject, we'll create a new Observable getUserInfo$
which sends a server request - each time the user$
gets emitted:
If you don't understand what's going on in this code and the .pipe
function is unclear for you, maybe you should first take a look at this article and then continue reading!
There are two Operators to pay attention, exhaustMap
and takeUntil
. Of course, you can add more operators, but these two are the most important.
Using an exhaustMap
operator we can be sure that getUserInfo$
will always wait for the server response, even if the user$
emits a new value, it's the opposite of the switchMap
operator. See the picture below:
You can read more about RxJs Mapping here, to have a better idea of why we used exhaustMap
and not other mapping operators.
takeUntil
subscribes and begins mirroring the user$
Observable. It also monitors a second Observable, this.onDestroy$
that we provided. If the this.onDestroy$
emits a value, the getUserInfo$
Observable stops mirroring the user$
Observable and complete. For more info about takeUntil
, see the official documentation.
Subscribing via Angular AsyncPipe
Now as we created the getUserInfo$
Observable, let's see how to subscribe to it using AsyncPipe :
And then just emit a new value for user$
in order to trigger the server request:
Now, if we click the button, even if we double-click, the getUserInfo$
will send a server request and emit new user info which will be handled by Angular AsyncPipe and in turn, will insert new user info into the page.
As we use takeUntil
operator, we have to emit some value for this.onDestroy
as well as complete it, takeUntil
is optional here, as we subscribed to getUserInfo$
via Angular AsyncPipe which takes care of unsubscribing when element gets destroyed, but anyway, it's a good practice and it's good to know about it.
Conclusion
Now we don't have to worry about memory leaking, as we did the best practices for subscribing and unsubscribing from Observables. We can be sure that our code will work faster as AsyncPipe allows us to change the changeDetection
of our component to ChangeDetectionStrategy.Push
, which (after me) is the best way to optimize a component and as well a good practice.
Haven't you tried yet? Try it and let me know how it works for you π».
Top comments (0)