There's been a lot written about API design, it's probably not surprising there are several books written about the subject. It also seems like nearly every cloud-based service provides an API to allow access to your data. In this tutorial, we'll attempt to condense this information and answer the following questions:

• Are there different types of API paradigms?
• What kinds of considerations do we need to make when building an API for our decoupled site?
• And, what's this REST thing everyone is talking about?

Let's dig into those questions one at a time.

If you're interested in decoupling Drupal, there's a good chance you've heard at least some of the buzz in the Drupalverse about "headless" or decoupled Drupal. Or perhaps you watched Dries' keynote from DrupalCon Barcelona or read Dries' blog post about the future of decoupled Drupal. Whatever the case may be, this tutorial and the ones that follow will walk you through building a simple decoupled blog. In Dries' terminology the demo site we'll be building is "fully decoupled." While it would be trivial to adopt similar techniques to build a progressively decoupled site, let's dig a bit deeper into what it means to build a decoupled Drupal site.

Andrew Berry, from Lullabot, has written a great article asking Should you Decouple? Like most architectural decisions there are trade-offs to consider with a decoupled approach. Let's take a look at some of the pros and cons of a decoupled approach. Is it the right choice for your project?

At this point, whether we've decided to use a third-party pre-rendering service or we've written our own isomorphic JavaScript application to serve as the front-end of our website, our hosting requirements have definitely gotten more complex. Let's take a look at the continuum of decoupled site architectures and start to come up with a list of things to keep in mind when considering their hosting requirements.

A multisite Drupal installation allows you to host multiple, separate websites while relying on the same set of code. Large organizations often rely on a multisite installation to cut down on the operational upkeep of multiple sites. Hosting a multisite in Docker poses several additional challenges. Fortunately, the process is not dissimilar from configuring a bare-metal server to run a multisite.

In this tutorial, we'll:

• Outline a multisite's additional requirements for Docker containers.
• Configure alternate, local domain names to resolve each site.
• Learn how to configure a multisite to use alternate domain names.

Every Drupal module needs a *.info.yml file; the basic structure of a form controller class and related routing is the same for every form; and much of the code required to create a custom content entity type is boilerplate annotations and extending base classes. Wouldn't it be nice if there was a way to automate some of that repetitive work? Drush can be used to speed up module development by generating scaffolding code for event subscribers, forms, services, module files, routing, and much more. These generators are provided by the Drupal Code Generator project. They're neatly bundled up in Drush under the drush generate command.

Before Drush 9, there were no code generators in Drush, but the Drupal Console project provided them. That project and its code generators, have languished since the release of Drupal 9. While it can still be used, and is often referenced in tutorials about Drupal, we much prefer the code generated by Drush at this point.

In this tutorial we'll:

• Learn about the Drupal Code Generator project
• Learn how Drush integrates with this project
• Demonstrate the drush generate command and its options

By the end of this tutorial, you'll know how to use the drush generate command to speed up development for your Drupal modules.