React Hooks for Beginners - En hjerne-venlig vejledning om brug Stat og useEffect

"Hvad i helvede er kroge?"

Jeg befandt mig i at spørge dette, ligesom jeg troede, at jeg havde dækket hele grundlaget for React. Sådan er en frontend-udviklers liv, spillet ændrer sig altid. Indtast kroge.

Det er altid rart at lære noget nyt, ikke? Selvfølgelig! Men nogle gange er vi nødt til at spørge os selv: "Hvorfor? Hvad er meningen med denne nye ting? Skal jeg lære det"?

Med kroge er svaret "ikke med det samme". Hvis du har lært at reagere og har brugt klassebaserede komponenter til dato, er der ikke travlt med at flytte til kroge. Kroge er valgfri og kan arbejde sammen med dine eksisterende komponenter. Hader du ikke det, når du skal omskrive hele din codebase for at få nogle nye ting til at fungere?

Under alle omstændigheder er her nogle grunde til, at kroge i første omgang blev introduceret, og hvorfor jeg anbefaler, at begyndere bør lære dem.

Brug af tilstand i funktionelle komponenter

Før kroge kunne vi ikke bruge tilstand i funktionelle komponenter. Det betyder, at hvis du har en pænt udformet og testet funktionel komponent, der pludselig skal gemme tilstand, sidder du fast med den smertefulde opgave at omlægge din funktionelle komponent til en klassekomponent.

Hurra! Tilladelse af tilstand inden for funktionelle komponenter betyder, at vi ikke behøver at omlægge vores præsentationskomponenter. Se denne artikel for mere.

Klassekomponenter er klodset

Lad os indse det, klassekomponenter kommer med en masse kedelplade. Konstruktører, bindende, bruger "dette" overalt. Brug af funktionelle komponenter fjerner meget af dette, så vores kode bliver lettere at følge og vedligeholde.

Du kan læse mere om dette i React docs:

Mere læselig kode

Da kroge lader os bruge funktionelle komponenter, betyder det, at der er mindre kode sammenlignet med klassekomponenter. Dette gør vores kode mere læselig. Nå, det er alligevel ideen.

Vi behøver ikke bekymre os om at binde vores funktioner eller huske, hvad "dette" også vedrører osv. Vi kan bekymre os om at skrive vores kode i stedet.

Hvis du lige er begyndt med React, har jeg en masse indledende indlæg på min blog, der måske kan hjælpe dig! Tjek det her:

Reager statskrog

Ah, stat. En hjørnesten i React-økosystemet. Lad os få våde fødder med kroge ved at introducere den mest almindelige krog, som du vil arbejde med - useState().

Lad os se på en klassekomponent, der har tilstand.

 import React, { Component } from 'react'; import './styles.css'; class Counter extends Component { state = { count: this.props.initialValue, }; setCount = () => { this.setState({ count: this.state.count + 1 }); }; render() { return ( 

This is a counter using a class

{this.state.count}

Click to Increment ); } } export default Counter;

Med React Hooks kan vi omskrive denne komponent og fjerne en masse ting, hvilket gør det lettere at forstå:

 import React, { useState } from 'react'; function CounterWithHooks(props) { const [count, setCount] = useState(props.initialValue); return ( 

This is a counter using hooks

{count}

setCount(count + 1)}>Click to Increment ); } export default CounterWithHooks;

På forsiden er der mindre kode, men hvad sker der?

Reager tilstandssyntaks

Så vi har set vores første krog! Hurra!

 const [count, setCount] = useState(); 

Dybest set bruger dette destruktureringsopgave til arrays. Den useState()funktion giver os 2 ting:

  • en variabel til at holde tilstandsværdien , i dette tilfælde kaldes den count- en funktion til at ændre værdien , i dette tilfælde kaldes den setCount.

Du kan navngive disse, hvad du vil:

 const [myCount, setCount] = useState(0); 

Og du kan bruge dem i hele koden som normale variabler / funktioner:

 function CounterWithHooks() { const [count, setCount] = useState(); return ( 

This is a counter using hooks

{count}

setCount(count + 1)}>Click to Increment ); }

Bemærk useStatekrogen øverst. Vi erklærer / destrukturer to ting:

  • counter: en værdi, der holder vores statsværdi
  • setCounter: en funktion, der vil ændre vores countervariabel

As we continue through the code, you'll see this line:

{count}

This is an example of how we can use a state hook variable. Within our JSX, we place our count variable within {} to execute it as JavaScript, and in turn the count value gets rendered on the page.

Comparing this to the old "class-based" way of using a state variable:

{this.state.count}

You'll notice we no longer need to worry about using this, which makes our life a lot easier - for example, the VS Code editor will give us a warning if {count} is not defined, allowing us to catch errors early. Whereas it won't know if {this.state.count} is undefined until the code is run.

On to the next line!

  setCount(count + 1)}>Click to Increment 

Here, we're using the setCount function (remember we destructured/declared this from the useState() hook) to change the count variable.

When the button is clicked, we update the count variable by 1. Since this is a change of state this triggers a rerender, and React updates the view with the new count value for us. Sweet!

How can I set the initial state?

You can set the initial state by passing an argument to the useState() syntax. This can be a hardcoded value:

 const [count, setCount] = useState(0); 

Or can be taken from the props:

 const [count, setCount] = useState(props.initialValue); 

This would set the count value to whatever the props.initialValue is.

That sums up useState(). The beauty of it is that you can use state variables/functions like any other variable/function you would write yourself.

How do I handle multiple state variables?

This is another cool thing about hooks. We can have as many as we like in a component:

 const [count, setCount] = useState(props.initialValue); const [title, setTitle] = useState("This is my title"); const [age, setAge] = useState(25); 

As you can see, we have 3 seperate state objects. If we wanted to update the age for example, we just call the setAge() function. The same with count and title. We no longer are tied to the old clunky class component way where we have one massive state object stored using setState():

 this.setState({ count: props.initialValue, title: "This is my title", age: 25 }) 

So, what about updating things when props or state changes?

When using hooks and functional components, we no longer have access to React lifecycle methods like componentDidMount, componentDidUpdate, and so on. Oh, dear! Do not panic my friend, React has given us another hook we can use:

  • Drum Roll *

Enter useEffect!

The Effect hook (useEffect()) is where we put "side effects".

Eh, side effects? What? Let's go off-track for a minute and discuss what a side effect actually is. This will help us understand what useEffect() does, and why it's useful.

A boring computer-y explanation would be.

"In programming, a side effect is when a procedure changes a variable from outside its scope"

In React-y terms, this means "when a component's variables or state changes based on some outside thing". For example, this could be:

  • When a component receives new props that change its state
  • When a component makes an API call and does something with the response (e.g, changes the state)

So why is it called a side effect? Well, we cannot be sure what the result of the action will be. We can never be 100% certain what props we are going to receive, or what the response from an API call would be. And, we cannot be sure how this will affect our component.

Sure we can write code to validate, and handle errors, and so on, but ultimately we cannot be sure what the side effects of said things are.

So for example, when we change state, based on some outside thing this is know as a side effect.

With that out of the way, let's get back to React and the useEffect Hook!

When using functional components we no longer have access to life cycle methods like componentDidMount(), componentDidUpdate() etc. So, in effect (pun intended), the useEffect hooks replace the current React Life Cycle hooks.

Let's compare a class-based component with how we use the useEffect hook:

import React, { Component } from 'react'; class App extends Component { componentDidMount() { console.log('I have just mounted!'); } render() { return Insert JSX here ; } } 

And now using useEffect():

function App() { useEffect(() => { console.log('I have just mounted!'); }); return Insert JSX here ; } 

Before we continue, it's important to know that, by default, the useEffect hook runs on every render and re-render. So whenever the state changes in your component or your component receives new props, it will rerender and cause the useEffect hook to run again.

Running an effect once (componentDidMount)

So, if hooks run every time a component renders, how do we ensure a hook only runs once when the component mounts? For example, if a component fetches data from an API, we don't want this happening every time the component re-renders!

The useEffect() hook takes a second parameter, an array, containing the list of things that will cause the useEffect hook to run. When changed, it will trigger the effect hook. The key to running an effect once is to pass in an empty array:

useEffect(() => { console.log('This only runs once'); }, []); 

So this means the useEffect hook will run on the first render as normal. However, when your component rerenders, the useEffect will think "well, I've already run, there's nothing in the array, so I won't have to run again. Back to sleep for me!" and simply does nothing.

In summary, empty array = useEffect hook runs once on mount

Using effects when things change (componentDidUpdate)

We've covered how to make sure a useEffect hook only runs once, but what about when our component receives a new prop? Or we want to run some code when the state changes? Hooks let us do this as well!

 useEffect(() => { console.log("The name props has changed!") }, [props.name]); 

Notice how we are passing stuff to the useEffect array this time, namely props.name.

In this scenario, the useEffect hook will run on the first load as always. Whenever your component receives a new name prop from its parent, the useEffect hook will be triggered, and the code within it will run.

We can do the same thing with state variables:

const [name, setName] = useState("Chris"); useEffect(() => { console.log("The name state variable has changed!"); }, [name]); 

Whenever the name variable changes, the component rerenders and the useEffect hook will run and output the message. Since this is an array, we can add multiple things to it:

const [name, setName] = useState("Chris"); useEffect(() => { console.log("Something has changed!"); }, [name, props.name]); 

This time, when the name state variable changes, or the name prop changes, the useEffect hook will run and display the console message.

Can we use componentWillUnmount()?

To run a hook as the component is about to unmount, we just have to return a function from the useEffect hook:

useEffect(() => { console.log('running effect'); return () => { console.log('unmounting'); }; }); 

Can I use different hooks together?

Yes! You can use as many hooks as you want in a component, and mix and match as you like:

function App = () => { const [name, setName] = useState(); const [age, setAge] = useState(); useEffect(()=>{ console.log("component has changed"); }, [name, age]) return( Some jsx here... ) } 

Conclusion - What Next?

There you have it. Hooks allow us to use good old fashioned JavaScript functions to create simplier React components, and reduce alot of boilerplate code.

Løb nu ud i verdenen af ​​Reac-kroge, og prøv at bygge ting selv! Taler om at bygge ting selv ...