Ref: Tutorial of React.js

A quick start of React.js. Learn React.js by implementing a tic tac toe game.

Passing Data Through Props

In Board’s renderSquare method, change the code to pass a prop called value to the Square.

class Board extends React.Component {
  renderSquare(i) {
    return <Square value={i} />;
  }
}

Change Square’s render method to show that value by replacing {/* TODO */} with {this.props.value}:

class Square extends React.Component {
  render() {
    return (
      <button className="square">
        {this.props.value}
      </button>
    );
  }
}

Making an Interactive Component

Let’s fill the Square component with an “X” when we click it. First, change the button tag that is returned from the Square component’s render() function to this:

class Square extends React.Component {
 render() {
   return (
     <button className="square" onClick={() => alert('click')}>
       {this.props.value}
     </button>
   );
 }
}

If you click on a Square now, you should see an alert in your browser.

As a next step, we want the Square component to “remember” that it got clicked, and fill it with an “X” mark. To “remember” things, components use state.

React components can have state by setting this.state in their constructors. this.state should be considered as private to a React component that it’s defined in. Let’s store the current value of the Square in this.state, and change it when the Square is clicked.

First, we’ll add a constructor to the class to initialize the state:

class Square extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      value: null,
    };
  }

  render() {
    return (
      <button className="square" onClick={() => alert('click')}>
        {this.props.value}
      </button>
    );
  }
}

In JavaScript classes, you need to always call super when defining the constructor of a subclass. All React component classes that have a constructor should start with a super(props) call.

Now we’ll change the Square’s render method to display the current state’s value when clicked:

  • Replace this.props.value with this.state.value inside the <button> tag.
  • Replace the onClick={...} event handler with onClick={() => this.setState({value: 'X'})}.
  • Put the className and onClick props on separate lines for better readability.

After these changes, the <button> tag that is returned by the Square’s render method looks like this:

class Square extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      value: null,
    };
  }

  render() {
    return (
      <button
        className="square"
        onClick={() => this.setState({value: 'X'})}
      >
        {this.state.value}
      </button>
    );
  }
}

By calling this.setState from an onClick handler in the Square’s render method, we tell React to re-render that Square whenever its <button> is clicked. After the update, the Square’s this.state.value will be ‘X’, so we’ll see the X on the game board. If you click on any Square, an X should show up.

When you call setState in a component, React automatically updates the child components inside of it too.

Lifting State Up

To collect data from multiple children, or to have two child components communicate with each other, you need to declare the shared state in their parent component instead. The parent component can pass the state back down to the children by using props; this keeps the child components in sync with each other and with the parent component.

Add a constructor to the Board and set the Board’s initial state to contain an array of 9 nulls corresponding to the 9 squares:

class Board extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      squares: Array(9).fill(null),
    };
  }

  renderSquare(i) {
    return <Square value={i} />;
  }

Next, we need to change what happens when a Square is clicked. The Board component now maintains which squares are filled. We need to create a way for the Square to update the Board’s state. Since state is considered to be private to a component that defines it, we cannot update the Board’s state directly from Square.

Instead, we’ll pass down a function from the Board to the Square, and we’ll have Square call that function when a square is clicked. We’ll change the renderSquare method in Board to:

 renderSquare(i) {
    return (
      <Square
        value={this.state.squares[i]}
        onClick={() => this.handleClick(i)}
      />
    );
  }

Now we’re passing down two props from Board to Square: value and onClick. The onClick prop is a function that Square can call when clicked. We’ll make the following changes to Square:

  • Replace this.state.value with this.props.value in Square’s render method
  • Replace this.setState() with this.props.onClick() in Square’s render method
  • Delete the constructor from Square because Square no longer keeps track of the game’s state

After these changes, the Square component looks like this:

class Square extends React.Component {
  render() {
    return (
      <button
        className="square"
        onClick={() => this.props.onClick()}
      >
        {this.props.value}
      </button>
    );
  }
}

We’ll now add handleClick to the Board class:

class Board extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      squares: Array(9).fill(null),
    };
  }

  handleClick(i) {
    const squares = this.state.squares.slice();
    squares[i] = 'X';
    this.setState({squares: squares});
  }

  renderSquare(i) {
    return (
      <Square
        value={this.state.squares[i]}
        onClick={() => this.handleClick(i)}
      />
    );
  }

  render() {
    const status = 'Next player: X';

    return (
      <div>
        <div className="status">{status}</div>
        <div className="board-row">
          {this.renderSquare(0)}
          {this.renderSquare(1)}
          {this.renderSquare(2)}
        </div>
        <div className="board-row">
          {this.renderSquare(3)}
          {this.renderSquare(4)}
          {this.renderSquare(5)}
        </div>
        <div className="board-row">
          {this.renderSquare(6)}
          {this.renderSquare(7)}
          {this.renderSquare(8)}
        </div>
      </div>
    );
  }
}

Since the Square components no longer maintain state, the Square components receive values from the Board component and inform the Board component when they’re clicked. In React terms, the Square components are now controlled components. The Board has full control over them.

Note how in handleClick, we call .slice() to create a copy of the squares array to modify instead of modifying the existing array. We will explain why we create a copy of the squares array in the next section.

Why Immutability Is Important

There are generally two approaches to changing data. The first approach is to mutate the data by directly changing the data’s values. The second approach is to replace the data with a new copy which has the desired changes.

Data Change with Mutation

var player = {score: 1, name: 'Jeff'};
player.score = 2;
// Now player is {score: 2, name: 'Jeff'}

Data Change without Mutation

var player = {score: 1, name: 'Jeff'};

var newPlayer = Object.assign({}, player, {score: 2});
// Now player is unchanged, but newPlayer is {score: 2, name: 'Jeff'}

// Or if you are using object spread syntax proposal, you can write:
// var newPlayer = {...player, score: 2};

Complex Features Become Simple

Immutability makes complex features much easier to implement. Later in this tutorial, we will implement a “time travel” feature that allows us to review the tic-tac-toe game’s history and “jump back” to previous moves. This functionality isn’t specific to games — an ability to undo and redo certain actions is a common requirement in applications. Avoiding direct data mutation lets us keep previous versions of the game’s history intact, and reuse them later.

Detecting Changes

Detecting changes in mutable objects is difficult because they are modified directly. This detection requires the mutable object to be compared to previous copies of itself and the entire object tree to be traversed.

Detecting changes in immutable objects is considerably easier. If the immutable object that is being referenced is different than the previous one, then the object has changed.

Determining When to Re-Render in React

The main benefit of immutability is that it helps you build pure components in React. Immutable data can easily determine if changes have been made which helps to determine when a component requires re-rendering.

Function Components

We’ll now change the Square to be a function component.

In React, function component are a simpler way to write components that only contain a render method and don’t have their own state. Instead of defining a class which extends React.Component, we can write a function that takes props as input and returns what should be rendered. Function components are less tedious to write than classes, and many components can be expressed this way.

Replace the Square class with this function:

function Square(props) {
  return (
    <button className="square" onClick={props.onClick}>
      {props.value}
    </button>
  );
}

We have changed this.props to props both times it appears. When we modified the Square to be a function component, we also changed onClick={() => this.props.onClick()} to a shorter onClick={props.onClick} (note the lack of parentheses on both sides).

Taking Turns

We now need to fix an obvious defect in our tic-tac-toe game: the “O”s cannot be marked on the board.

We’ll set the first move to be “X” by default. We can set this default by modifying the initial state in our Board constructor:

class Board extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      squares: Array(9).fill(null),
      xIsNext: true,
    };
  }

Each time a player moves, xIsNext (a boolean) will be flipped to determine which player goes next and the game’s state will be saved. We’ll update the Board’s handleClick function to flip the value of xIsNext:

  handleClick(i) {
    const squares = this.state.squares.slice();
    squares[i] = this.state.xIsNext ? 'X' : 'O';
    this.setState({
      squares: squares,
      xIsNext: !this.state.xIsNext,
    });
  }

Let’s also change the “status” text in Board’s render so that it displays which player has the next turn:

 render() {
    const status = 'Next player: ' + (this.state.xIsNext ? 'X' : 'O');

    return (
      // the rest has not changed

After applying these changes, you should have this Board component:

class Board extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      squares: Array(9).fill(null),
      xIsNext: true,
    };
  }

  handleClick(i) {
    const squares = this.state.squares.slice();
    squares[i] = this.state.xIsNext ? 'X' : 'O';
    this.setState({
      squares: squares,
      xIsNext: !this.state.xIsNext,
    });
  }

  renderSquare(i) {
    return (
      <Square
        value={this.state.squares[i]}
        onClick={() => this.handleClick(i)}
      />
    );
  }

  render() {
    const status = 'Next player: ' + (this.state.xIsNext ? 'X' : 'O');

    return (
      <div>
        <div className="status">{status}</div>
        <div className="board-row">
          {this.renderSquare(0)}
          {this.renderSquare(1)}
          {this.renderSquare(2)}
        </div>
        <div className="board-row">
          {this.renderSquare(3)}
          {this.renderSquare(4)}
          {this.renderSquare(5)}
        </div>
        <div className="board-row">
          {this.renderSquare(6)}
          {this.renderSquare(7)}
          {this.renderSquare(8)}
        </div>
      </div>
    );
  }
}

Declaring a Winner

Now that we show which player’s turn is next, we should also show when the game is won and there are no more turns to make. Copy this helper function and paste it at the end of the file:

Given an array of 9 squares, this function will check for a winner and return ‘X’, ‘O’, or null as appropriate.

We will call calculateWinner(squares) in the Board’s render function to check if a player has won. If a player has won, we can display text such as “Winner: X” or “Winner: O”. We’ll replace the status declaration in Board’s render function with this code:

  render() {
    const winner = calculateWinner(this.state.squares);
    let status;
    if (winner) {
      status = 'Winner: ' + winner;
    } else {
      status = 'Next player: ' + (this.state.xIsNext ? 'X' : 'O');
    }

    return (
      // the rest has not changed

We can now change the Board’s handleClick function to return early by ignoring a click if someone has won the game or if a Square is already filled:

handleClick(i) {
    const squares = this.state.squares.slice();
    if (calculateWinner(squares) || squares[i]) {
      return;
    }
    squares[i] = this.state.xIsNext ? 'X' : 'O';
    this.setState({
      squares: squares,
      xIsNext: !this.state.xIsNext,
    });
  }

Storing a History of Moves

We used slice() to create a new copy of the squares array after every move, and treated it as immutable. This will allow us to store every past version of the squares array, and navigate between the turns that have already happened.

We’ll store the past squares arrays in another array called history. The history array represents all board states, from the first to the last move, and has a shape like this:

history = [
  // Before first move
  {
    squares: [
      null, null, null,
      null, null, null,
      null, null, null,
    ]
  },
  // After first move
  {
    squares: [
      null, null, null,
      null, 'X', null,
      null, null, null,
    ]
  },
  // After second move
  {
    squares: [
      null, null, null,
      null, 'X', null,
      null, null, 'O',
    ]
  },
  // ...
]

Lifting State Up, Again

We’ll want the top-level Game component to display a list of past moves. It will need access to the history to do that, so we will place the history state in the top-level Game component.

Placing the history state into the Game component lets us remove the squares state from its child Board component. Just like we “lifted state up” from the Square component into the Board component, we are now lifting it up from the Board into the top-level Game component. This gives the Game component full control over the Board’s data, and lets it instruct the Board to render previous turns from the history.

First, we’ll set up the initial state for the Game component within its constructor:

class Game extends React.Component {
  constructor(props) {
    super(props);
    this.state = {
      history: [{
        squares: Array(9).fill(null),
      }],
      xIsNext: true,
    };
  }

  render() {
    return (
      <div className="game">
        <div className="game-board">
          <Board />
        </div>
        <div className="game-info">
          <div>{/* status */}</div>
          <ol>{/* TODO */}</ol>
        </div>
      </div>
    );
  }
}

Next, we’ll have the Board component receive squares and onClick props from the Game component. Since we now have a single click handler in Board for many Squares, we’ll need to pass the location of each Square into the onClick handler to indicate which Square was clicked. Here are the required steps to transform the Board component:

  • Delete the constructor in Board.
  • Replace this.state.squares[i] with this.props.squares[i] in Board’s renderSquare.
  • Replace this.handleClick(i) with this.props.onClick(i) in Board’s renderSquare.

The Board component now looks like this:

class Board extends React.Component {
  handleClick(i) {
    const squares = this.state.squares.slice();
    if (calculateWinner(squares) || squares[i]) {
      return;
    }
    squares[i] = this.state.xIsNext ? 'X' : 'O';
    this.setState({
      squares: squares,
      xIsNext: !this.state.xIsNext,
    });
  }

  renderSquare(i) {
    return (
      <Square
        value={this.props.squares[i]}
        onClick={() => this.props.onClick(i)}
      />
    );
  }

  render() {
    const winner = calculateWinner(this.state.squares);
    let status;
    if (winner) {
      status = 'Winner: ' + winner;
    } else {
      status = 'Next player: ' + (this.state.xIsNext ? 'X' : 'O');
    }

    return (
      <div>
        <div className="status">{status}</div>
        <div className="board-row">
          {this.renderSquare(0)}
          {this.renderSquare(1)}
          {this.renderSquare(2)}
        </div>
        <div className="board-row">
          {this.renderSquare(3)}
          {this.renderSquare(4)}
          {this.renderSquare(5)}
        </div>
        <div className="board-row">
          {this.renderSquare(6)}
          {this.renderSquare(7)}
          {this.renderSquare(8)}
        </div>
      </div>
    );
  }
}

We’ll update the Game component’s render function to use the most recent history entry to determine and display the game’s status:

 render() {
    const history = this.state.history;
    const current = history[history.length - 1];
    const winner = calculateWinner(current.squares);
    let status;
    if (winner) {
      status = 'Winner: ' + winner;
    } else {
      status = 'Next player: ' + (this.state.xIsNext ? 'X' : 'O');
    }

    return (
      <div className="game">
        <div className="game-board">
          <Board
            squares={current.squares}
            onClick={(i) => this.handleClick(i)}
          />
        </div>
        <div className="game-info">
          <div>{status}</div>
          <ol>{/* TODO */}</ol>
        </div>
      </div>
    );
  }

Since the Game component is now rendering the game’s status, we can remove the corresponding code from the Board’s render method. After refactoring, the Board’s render function looks like this:

  render() {
    return (
      <div>
        <div className="board-row">
          {this.renderSquare(0)}
          {this.renderSquare(1)}
          {this.renderSquare(2)}
        </div>
        <div className="board-row">
          {this.renderSquare(3)}
          {this.renderSquare(4)}
          {this.renderSquare(5)}
        </div>
        <div className="board-row">
          {this.renderSquare(6)}
          {this.renderSquare(7)}
          {this.renderSquare(8)}
        </div>
      </div>
    );
  }

Finally, we need to move the handleClick method from the Board component to the Game component. We also need to modify handleClick because the Game component’s state is structured differently. Within the Game’s handleClick method, we concatenate new history entries onto history.

 handleClick(i) {
    const history = this.state.history;
    const current = history[history.length - 1];
    const squares = current.squares.slice();
    if (calculateWinner(squares) || squares[i]) {
      return;
    }
    squares[i] = this.state.xIsNext ? 'X' : 'O';
    this.setState({
      history: history.concat([{
        squares: squares,
      }]),
      xIsNext: !this.state.xIsNext,
    });
  }

Unlike the array push() method you might be more familiar with, the concat() method doesn’t mutate the original array, so we prefer it.

At this point, the Board component only needs the renderSquare and render methods. The game’s state and the handleClick method should be in the Game component.

Showing the Past Moves

This part doesn’t include some unknown information about React.js. Check this link to finish this step.

Picking a Key

Because React cannot know our intentions, we need to specify a key property for each list item to differentiate each list item from its siblings. One option would be to use the strings alexa, ben, claudia. If we were displaying data from a database, Alexa, Ben, and Claudia’s database IDs could be used as keys.

<li key={user.id}>{user.name}: {user.taskCount} tasks left</li>

When a list is re-rendered, React takes each list item’s key and searches the previous list’s items for a matching key. If the current list has a key that didn’t exist before, React creates a component. If the current list is missing a key that existed in the previous list, React destroys the previous component. If two keys match, the corresponding component is moved. Keys tell React about the identity of each component which allows React to maintain state between re-renders. If a component’s key changes, the component will be destroyed and re-created with a new state.

key is a special and reserved property in React (along with ref, a more advanced feature). When an element is created, React extracts the key property and stores the key directly on the returned element. Even though key may look like it belongs in props, key cannot be referenced using this.props.key. React automatically uses key to decide which components to update. A component cannot inquire about its key.

It’s strongly recommended that you assign proper keys whenever you build dynamic lists. If you don’t have an appropriate key, you may want to consider restructuring your data so that you do.

If no key is specified, React will present a warning and use the array index as a key by default. Using the array index as a key is problematic when trying to re-order a list’s items or inserting/removing list items. Explicitly passing key={i} silences the warning but has the same problems as array indices and is not recommended in most cases.

Keys do not need to be globally unique; they only need to be unique between components and their siblings.

Implementing Time Travel

Check this link for more information.