React's main goals are to render data in a fast way for users and a maintainable way for developers.
So far you've learned about React components. However, while you could make multiple components, they all displayed the same data. With props (properties), you can customize each component to render different data.
- Understand what props are and what problem they solve
- Build an application that uses props
- Articulate the distinction between container and display components
In previous lessons, we've seen how to build React components and display them on a website. However, we had no way to make two different versions of the same component. For example, we should be able to create a UserProfile component with the name "Robert Kahn", then create a separate UserProfile component with the name "Vinton Cerf".
Another example of this is Amazon's product search. Take a look at, say, a search for t-shirts:
Each of these t-shirt listings represents a separate instance of the exact same component. This component is reused in many different places in the application.
In order to have the same component display different information, we can use props.
Props allow components to pass information to child components that they're rendering. Components can use props to consume data from their parents and render it to the user. This permits us to reuse the same component with relative ease. For example, we don't need distinct UserProfile components if each user's profile consumes the same shape of information and renders that information in the same way. We can use the same component and pass down unique user information as props.
Props also allow us to intelligently divide responsibilities between components. We can create components whose only job is to fetch, store, and update information, and we can have components whose only job is to receive props and render information. More on that separation of concerns in a future lesson.
Returning to our Amazon search page example, let's say our components (ProductItem) are the child components of a component for the entire page (ProductPage). ProductPage stores the result of a large network request on the first render of the page, getting an array with fifty products. We then map through this array and send each individual product to a ProductItem component by way of props, which processes the product's information into something user-facing.
Let's take a look at a small-scale example of this.
Props are passed from a parent component to a child component. This is similar to how we call a function and pass arguments to it! Under the hood, this is actually what react is doing.
The syntax here looks very similar to writing HTML attributes, and that is by design.
Parent component:
// App.jsx
const App = () => {
let userDetails = {
name: "Jimmy B",
userId: 7348,
email: "jimmy@pursuit.org",
};
return <User name={userDetails.name} />;
};Child component:
const User = (props) => {
const { name } = props;
return (
<div className="user-details">
<h1>{name}</h1>
</div>
);
};Note that we are only passing in one prop. If we want to add more, we can just pass them to the child component with more key/value pairs.
// App.jsx
const App = () => {
let userDetails = {
name: "Jimmy B",
userId: 7348,
email: "jimmy@pursuit.org",
};
return <User name={userDetails.name} userId={userDetails.userId} />;
};When the code gets too long, we can put each k/v pair on its own line.
// App.jsx
const App = () => {
let userDetails = {
name: "Jimmy B",
userId: 7348,
email: "jimmy@pursuit.org",
};
return (
<User
name={userDetails.name}
userId={userDetails.userId}
email={userDetails.email}
/>
);
};Bonus: What if we used the spread operator here?
Here, we will build a simple application that uses props to display a list of shoes.
import ProductItem from './ProductItem';
const ProductPage = () => {
// You can see here that we start our functional `ProductPage` component with an array of objects, each with certain consistent traits: `name`, `manufacturer`, and `price`.
const products = [
{ name: "Ultra Boost", manufacturer: "Adidas", price: 160 },
{ name: "Air Force One Low", manufacturer: "Nike", price: 100 },
{ name: "Classic Leather", manufacturer: "Reebok", price: 120 },
{ name: "Sk8-Hi", manufacturer: "Vans", price: 60 }
];
// We then then use this array to render our products in JSX:
const listItems = products.map(product => {
return (
<ProductItem
name={product.name}
price={product.price}
/>
);
});
return (
<div>
<ul>{listItems}</ul>
</div>
);
}
}First, we're creating a variable called listItems, which maps through our products and returns React components named ProductItem. These ProductItem components have three attributes that correspond to keys of the objects in our products array. At first glance, however, these attributes might look confusing. They definitely aren't anything we've seen before in HTML. That's because, while these items share the same syntax, they don't translate directly to HTML attributes. Instead, they represent props that we're passing down to our ProductItem components.
Think of props as arguments for a component. They compile into an attribute on the component which we can reference using this.props (in the case of a class component) or as our function's argument (for a functional component). In this case, for each item in our products array, we're inputting the name, manufacturer, and price into our ProductItem component. You'll notice, because this component is only responsible for receiving props and rendering data, we're making this one functional, rather than using a class:
const ProductItem = (props) => {
const { name, price } = props;
return (
<li>
{name} - ${price}
</li>
);
};
export default ProductItem;This is the entire ProductItem component. Pretty clean, right? Let's break this down.
We define ProductItem as an anonymous function that takes its props as an argument. ProductItem then uses object destructuring to break out two of those same props that we passed into ProductItem in the ProductPage component. Finally, we place our three items in an li tag, which our component returns.
Importantly, what you're seeing rendered is four separate instances of this component, each with different props. We create those instances in our map method in our parent component.
Exercise: Notice that the objects in our products array have a third key: manufacturer. Pass this key down to each ProductItem and render it alongside name and price.
For React function components, we saw that you can add props as the argument to the function:
const UserProfile = (props) => {
const { name, age, location } = props;
return (
<ul>
<li>Name: {name}</li>
<li>Age: {age}</li>
<li>Location: {location}</li>
</ul>
);
};
export default ProductItem;In a React class component, props are accessed differently. Instead of using arguments to a function, you can access props with the expression this.props. This is because class components extend React.Component, which has a props attribute.
Note that we aren't even including a constructor, because we don't have to!
import React from "react";
class UserProfile extends React.Component {
render() {
const { name, age, location } = this.props;
return (
<ul>
<li>Name: {name}</li>
<li>Age: {age}</li>
<li>Location: {location}</li>
</ul>
);
}
}
export default ProductItem;Exercise 1: For UserProfile, the function and class component access props differently. What differences do you notice?
Exercise 2: In the UserProfile class component, this.props is accessed inside the render() method. What do you think will happen if we move that line of code outside the render method? Where else could you put it?
Exercise 3: Refactor the ProductItem component above to be a class component. Make sure it still handles props.
In our previous example, you might have noticed that our ProductPage component doesn't really render much of anything independently. Its main purpose is to handle its data and send props down to child components. It's only when we get to the ProductItem components that we actually render something that looks even close to HTML.
This is a design philosophy called separation of concerns, which means that we should break out different functionalities into different parts of our app.
Why is this important? Well, let's go back to our Amazon example. Product items can show up on the homepage, on search index pages, or as a recommended/related item on a product page. If we simply have Home, Search and Product components that handle all rendering, we'll have to re-write the same JSX to render the same product previews. Much easier and more efficient to make a ProductItem or ProductThumbnail component and use it across all of these pages. So, the top-level pages handle their data, but they all feed into the same child components.
Additionally, if we have a parent component that does a single API request and passes the result down as props to several different components, as we saw above, we can avoid a situation where each separate component fires an AJAX request, which would be taxing to our backend server or remote API.
Finally, it keeps us organized and it keeps our app consistent. The higher up in our component structure we can store state, and the more components share the same state, the more consistently we present information to the user, and the less we have to update the same information in different places.
Separating our components into these two categories is a pattern known as Smart Components and Dumb Components. Sometimes also called Container Components and Pure Components.
At the end of the day, a full-fledged app should have this general kind of component structure:
Clearly, there's a lot of syntax we're introducing to you here, but there's also a lot of creative work that goes into deciding exactly what information should be stored in which component, which component should display what, and so on. We highly recommend thinking about this stuff as early as possible when you get around to building your own full-scale React projects!