JavaScript Error Handling

ES6 JavaScript: What’s New? – Part 4

Today we continue our journey of exploring all the cool new features that ES6 JavaScript has to offer. This isn’t the first part in this series by any means and thus far we’ve covered quite a bit of ground:

  • In Part 1 we took a look at default parameters, classes, and block-scoping with the let keyword.
  • In Part 2 we explored constants, destructuring, and constant literals syntax.
  • In Part 3 we dove deep into just one major feature known as template literals.

For Part 4 today we’ll be looking at iterators and generators, so let’s get to it!

Iterators

The concept of iterators is not new to programming by any means. Most modern languages offer an interface for defining an iterator, which is a pattern that specifies that an object can be traversed in order to pull a single, consecutive item from a larger list.

In previous versions of JavaScript developers would be forced to create their own ad-hoc iterators. Here’s a basic iterator example from ES5 where we have a doubler object with a next() property that retrieves the next iteration (in this case, the doubled value of our number):

As expected, we get the doubled output from this every time until we reach our maximum of 1024:

Unfortunately, this old iterator pattern isn’t the most user-friendly implementation. For one thing we’re using an infinite loop, which always spells trouble if we fail to hit our break statement. We could get around that by performing our check inside the next() method call and if the iteration is complete return some value indicating as much (like undefined).

Regardless, the new iterator syntax introduced in ES6 aims to standardize these patterns. Here we have the same doubleriterator as before:

ES6 iterators require a next() method which is used to retrieve the next IteratorResult. The IteratorResult should be an object with two properties: value and done. value can be any value (or omitted if done is true). done is a boolean indicating whether the iterator has completed its sequence.

Therefore, you can see that our iterator example returns { value: value, done: false } for most calls of next(), but when we reach the end of our sequence — when value is greater than 1024 — we return { done: true }. This results in the exact same output as our previous example since we’re iterating through our doubler via a for (var...of) loop:

Generators and Yield

While iterators require that execution be completed all at once, and therefore looping through the full set of iterations before any additional action can be taken, ES6 also introduces a new tool to get around this limitation called generators. A generator is a special kind of function that allows execution to be paused at any time, only to be resumed later on. Generators take a lot of functionality from iterators (in fact, underneath each generator is an iterator created by the JavaScript engine).

The syntax for a generator is the same as a normal function except an asterisk (*) must precede the function name: function *myGenerator() { }

The special sauce of the generator largely comes from the new yield keyword that it also introduces. You may be familiar with yield from other programming languages, but the basic idea is that the yield statement tells the executing code that it should temporarily pause at that location, stepping out of the current function scope for the time being, while remembering that location for later resumption.

For example, here we have a simple *myGenerator() function that outputs a before message, then issues a yield call, before finally outputting an after message:

As discussed, since a generator is merely a wrapper for an iterator, we assign it to a variable and call it via the next()method. In our example above, our first next() call outputs the before message, then we’re free to execute any other code we want since the inner-execution of our *myGenerator() function was halted when it encountered the first yield statement. Finally we make another call to next() to produce our post-yield after message:

In addition to acting as a pause point within the generator function, the yield statement can also be used to return a value, which is merely substituted into the value property of our IteratorResult object of our wrapped iterator. Let’s modify our above example to yield the value 50 and see how output changes:

Notice that, in addition to adding our output value of 50 after our yield statement, we also want to retrieve (and output) the value that we get from our first next() iteration call, which will contain the yielded value in our output:

Things really start to get cool when we stick yield inside a loop. This allows us to easily generate a pseudo-infinite iterator that can be called anywhere and at anytime to grab the next value that our generator spits out. Let’s modify our previous doubler iterator to be a generator instead that can be called over and over to give us the next doubled output:

As you can see, we can call doubler.next() as many times we want, wherever we want, and we’ll generate and return the next iteration. Calling it six times, as seen above, produces the expected output result:

Just as an asterisk (*) prior to our function names converts a normal function into an iterable generator, we can also precede a yield value with an asterisk, so long as the value is itself an iterable. This will cause the parent generator which houses the yield to use the underlying iterator associated with the yielded value to retrieve (and return) each iteration.

As an example, let’s take our infinite *doubler() generator from before, which you’ll recall is an iterator unto itself, and call make a generator call to it via yield inside a new *useDoubler() generator:

Now we can assign our *useDoubler() generator just as we did before and make however many next() calls we wish. This will cause the yield to call the next() method of the underlying iterator it is associated with, which is *doubler() in this case:

The returned value propagates back up through the chain so we get the same output result as before:

To help you and your team with JavaScript development, particularly when dealing with unexpected errors, check out the revolutionary Airbrake JavaScript error tracking tool for real-time alerts and instantaneous insight into what went wrong with your JavaScript code.