Arrow functions

An arrow function expression (also known as fat arrow function) has a shorter syntax compared to function expressions and lexically binds the this value. Arrow functions are always anonymous.

Syntax

// Basic syntax:
(param1, param2, paramN) => { statements }
(param1, param2, paramN) => expression
   // equivalent to:  => { return expression; }

// Parentheses are optional when there's only one argument:
singleParam => { statements }
singleParam => expression

// A function with no arguments requires parentheses:
() => { statements }

// Advanced:
// Parenthesize the body to return an object literal expression:
params => ({foo: bar})

// Rest parameters are supported
(param1, param2, ...rest) => { statements }

Detailed syntax examples can be seen here.

Description

See also "ES6 In Depth: Arrow functions" on hacks.mozilla.org.

Two factors influenced the introduction of arrow functions: shorter functions and lexical this.

Shorter functions

In some functional patterns, shorter functions are welcome. Compare:

var a = [
  "Hydrogen",
  "Helium",
  "Lithium",
  "Beryl­lium"
];

var a2 = a.map(function(s){ return s.length });

var a3 = a.map( s => s.length );

Lexical this

Until arrow functions, every new function defined its own this value (a new object in case of a constructor, undefined in strict mode function calls, the context object if the function is called as an "object method", etc.). This proved to be annoying with an object-oriented style of programming.

function Person() {
  // The Person() constructor defines `this` as an instance of itself.
  this.age = 0;

  setInterval(function growUp() {
    // In nonstrict mode, the growUp() function defines `this` 
    // as the global object, which is different from the `this`
    // defined by the Person() constructor.
    this.age++;
  }, 1000);
}

var p = new Person();

In ECMAScript 3/5, this issue was fixed by assigning the value in this to a variable that could be closed over.

function Person() {
  var self = this; // Some choose `that` instead of `self`. 
                   // Choose one and be consistent.
  self.age = 0;

  setInterval(function growUp() {
    // The callback refers to the `self` variable of which
    // the value is the expected object.
    self.age++;
  }, 1000);
}

Alternatively, a bound function could be created so that the proper this value would be passed to the growUp() function.

Arrow functions capture the this value of the enclosing context, so the following code works as expected.

function Person(){
  this.age = 0;

  setInterval(() => {
    this.age++; // |this| properly refers to the person object
  }, 1000);
}

var p = new Person();

Relation with strict mode

Given that this is lexical, strict mode rules with regard to this are just ignored.

var f = () => {'use strict'; return this};
f() === window; // or the global object

The rest of strict mode rules apply normally.

Invoked through call or apply

Since this is already bound lexically, invoking an arrow function through the call() or apply() methods can only pass in arguments, but has no effect on this:

var adder = {
  base : 1,
    
  add : function(a) {
    var f = v => v + this.base;
    return f(a);
  },

  addThruCall: function(a) {
    var f = v => v + this.base;
    var b = {
      base : 2
    };
            
    return f.call(b, a);
  }
};

console.log(adder.add(1));         // This would log to 2
console.log(adder.addThruCall(1)); // This would log to 2 still

Use of arguments object or yield keyword

The arguments object may not be used inside the body of an arrow function. Likewise, the yield keyword may not be used inside an arrow function's body and an arrow function may not be used as a generator.

Returning object literals

Keep in mind that returning object literals using the consise syntax params => {object:literal} will not work as expected:

var func = () => {  foo: 1  };               // Calling func() returns undefined!
var func = () => {  foo: function() {}  };   // SyntaxError: function statement requires a name

This is because the code inside braces ({}) is parsed as a sequence of statements (i.e. foo is treated like a label, not a key in an object literal).

Remember to wrap the object literal in parentheses:

var func = () => ({ foo: 1 });

Examples

// An empty arrow function returns undefined
let empty = () => {};

(() => "foobar")() // returns "foobar" 

var simple = a => a > 15 ? 15 : a; 
simple(16); // 15
simple(10); // 10

let max = (a, b) => a > b ? a : b;

// Easy array filtering, mapping, ...

var arr = [5, 6, 13, 0, 1, 18, 23];
var sum = arr.reduce((a, b) => a + b);  // 66
var even = arr.filter(v => v % 2 == 0); // [6, 0, 18]
var double = arr.map(v => v * 2);       // [10, 12, 26, 0, 2, 36, 46]

Specifications

Browser compatibility

Firefox-specific notes

• The initial implementation of arrow functions in Firefox made them automatically strict. This has been changed as of Firefox 24. The use of "use strict"; is now required.
• Arrow functions are semantically different from the non-standard Expression Closures added in Firefox 3 (details: Javascript 1.8), for Expression Closures do not bind this lexically.
• Prior to Firefox 39, a line terminator (\n) was incorrectly allowed after arrow function arguments. This has been fixed to conform to the ES6 specification and code like () \n => {} will now throw a SyntaxError in this and later versions.

See also

• "ES6 In Depth: Arrow functions" on hacks.mozilla.org