Become a member
to unlock all features

Level Up!

Access all courses & lessons on egghead today and lock-in your price for life.


    Create an API with Swagger


    Swagger is a project used to describe restful APIs using the OpenAPI Specification. It allows you to document your API so consumers understand the endpoints, parameters, and responses. In this lesson, I'll show you how to install the swagger command line tool, create a new API project using swagger, and introduce you to the swagger API editor.



    Become a Member to view code

    You must be a Pro Member to view code

    Access all courses and lessons, track your progress, gain confidence and expertise.

    Become a Member
    and unlock code for this lesson
    orLog In




    To get started using Swagger, the first thing I need to do is install the package itself. It has a command line interface, I'm going to install it using the -g flag to install it globally.

    With that installed, I can create a new Swagger project using the Swagger command with Swagger project create, and then the name for my project, which is going to be todo-api.

    It supports multiple frameworks, including Express, Hapi, Restify, and Sales. I'm going to use the Express framework, and that's going to kick off the installation from the skeleton project. Switching over to my other pane, I can do a directory listing. There's a new directory called "todo-api."

    I can go into that directory, and then start my project using the Swagger project, start, and the name of my project. It gives us an example that we can use here to start it, we'll just try that over here in our other pane. It returns, "Hello Scott."

    The really cool part about using Swagger is that it has this editor that allows us to define our API. We can get to that using the Swagger project edit command. When I run that command, it opens up a browser for me, points it to localhost, and to a port that it's created to host this Swagger interface.

    Let's start looking on the right-hand side first. You can see that it has the name of our application, the version number, and then, it defines all of the paths that exist in the API. We've already used the /hello endpoint in our curl command that we did earlier, but it also allows us to use that from this editor, as well.

    It gives us a description about the endpoint, and tells us about all the parameters, a description, where it's located at, the type of data that it is. Same thing with the responses, and it gives you the details of the response that you can expect back from the API.

    We also have the ability to try it out from within the editor. It selects the scheme for us. Then, we can change the content type if we're selecting multiple content types. Then parameters, we can supply. You can see, as I supplied that parameter, it updated the URL that will be called with this.

    Then I can send that response, or send that request. It comes back with a success message. I can choose the output, and it shows the output from the API server. It's pretty cool. It allows you to test your API without actually having to have a specific client to do that.

    All of that comes from this definition over here on the left-hand side. Starting off, we've got the Swagger that defines the version of Swagger we're using. You've already seen the version number and the title that were displayed over here on the left, and the host name, as well, where the server's going to be running from.

    The most useful part of this is in this path section, where you'll define each of the endpoints hosted by your API. You can see one, the /hello endpoint that we used a minute ago. You can see how the description gets mapped over here to the UI provided to us.

    Other things to note here are the operation ID, which is the name of the function that's going to be called within your application to respond to this endpoint request. The types of responses that you support is defined.

    Then, there's this concept of definitions down here, where you can define a schema for the response itself. That's going to allow you to do some type validation on the responses that your API server is sending back down to the client so that as a consumer of this API, you can be confident that the data you're getting matches to the schema specification.