JavaScript promises, mastering the asynchronous


Open Source Your Knowledge, Become a Contributor

Technology knowledge has to be shared and made accessible for free. Join the movement.

Create Content

Now we have to deal with the difficult part of promises: chaining. Chaining promises is the very reason we have promises in the first place. It's a proper way to tell JavaScript the next thing to do after an asynchronous task is done, thus avoiding the pyramid of doom we saw in the previous lessons.

Did you ask yourself, "What is the result of the then method?" Here is the answer: It's a promise. But what promise is it? It can be the promise you want. It's easier to explain with an example so take a look at the following code and run it.

Test the code

First of all, when dealing with promises chains, people often put a new line before the first call. It's pretty hard to read and confusing when you don't know the syntax. Read the code carefully. The statement starting on line 3 ends on line 17. Each call of then is part of the same "line of code".

So, what is going on here? Let's take a deep breath and analyze it line by line:

  • Line 1: We call job1 and we store the returned promise in a variable called promise.
  • Line 5: We call then on this promise and we attach a callback for when the promise is resolved
  • Line 6: We print data1 and it is obvioulsy result of job 1 (see line 22)
  • Line 7: On this line, we call job2 and we return the resulting promise. Keep that in mind and go to line 10.
  • Line 10: We call then on the result of the first then. The result of then is always a promise. Always. At worst, it can be a never resvoled promise, but it is a promise. In this case, the promise is the return value of job2 (called at line 7). When you are in a then callback, if you return a promise, it will be the resulting promise of the then call.
  • Line 11: We print data2. According to the resolve call in the promise returned by job2 (called at line 7), data2 is result of job 2 (see line 30). By chaining our 2 promises (job1 then job2), job2 is always executed after job1. Line 6 is executed when the job1 promise is resolved, line 11 is executed when the job2 promise is resolved.
  • Line 12: We return a simple string 'Hello world'.
  • Line 15: We call then on the result of the then call on line 10. The promise here is an auto-resolved promise, and it will pass 'Hello world' in the data. When you are in a then callback, if you return anything but a promise, an auto-resvoled promise is created, and this promise will be the result of the then call.
  • Line 16: We print data3 and this is the 'Hello world' returned at line 12.
  • Line 17: End of line 3 (yes, this is a very long line!). At this point, you can add a then call if you want. then always returns a promise. And if you decide to return nothing (like at line 16, we don't return anything), then returns an auto-resolved promise with no data (if you try to get any data, you'll get undefined).

Chaining promises is very powerfull and simple to use. This is how you master asynchronous code.

Open Source Your Knowledge: become a Contributor and help others learn. Create New Content
var promise = job1();