Mariusz Nosiński
Written By
Mariusz Nosiński
Fullstack Developer @ Scalac.io

Experienced Scala Developer who enjoys learning and using many other techs. Years ago fell in love in Graph Databases now sharing this emotion with GraphQL.

Getting Started

In this chapter you will learn how to:

  • Initialize the SBT project from giter8 template.,
  • Setup Database schema and connection.

Initialize new project

For the purpose of this tutorial I’ve prepared a giter8 template. You can use to easily bootstrap a project. All you need is the latest version of SBT.

You will be asked about project’s name and port number to use by the HTTP server. You can hit ENTER to keep default values.

After this process you will see a simple project with the structure like this:

howtographql-sangria
├── README.md
├── build.sbt
├── project
│   ├── build.properties
│   └── plugins.sbt
└── src
    └── main
        ├── resources
        │   └── application.conf
        |   └── graphiql.html
        └── scala
            └── com
                └── howtographql
                    └── scala
                        └── sangria
                            ├── DAO.scala
                            ├── DBSchema.scala
                            └── Server.scala

I will explain shortly the most important files here.

  • build.sbt
  • project/plugins.sbt
  • project/build.properties

Files above are related to SBT. There you can find all dependencies to external libraries and plugins we will be using in the project.
I assume you’re at least a beginner in the scala and you understand what is going on in those files. One thing you could be unfamiliar with is Revolver plugin.
This plugin is responsible for restarting server every time you save the files, so akka-http will always serve the updated version. It’s very helpful during development.

HTTP Server

Our server extends an App trait so SBT can find it and run when you’ll use sbt run command. All the App does is implementing a main function which is a default entry point when it’s executed. In case there are more files like this in your project, SBT will ask you which one you want to run.

At the 2nd point, there is defined port number we want to use, you could choose it during project initialization.

What is worth pointing out here: In our example I use Spray JSON library for marshalling and unmarshalling JSON objects, but it isn’t obligatory for you. You can use whatever JSON library you want. On this page you can find which JSON libraries Sangria can play with.

Database configuration

In our project I chose to use H2 database. It’s easy to configure and is able to run in memory - you don’t need to install any additional packages in your OS. H2 works perfectly in cases like this tutorial. If you want to use another DB, it’s up to you, Slick supports many of them.

It’s all we need to configure a database. Now we’re ready to use it. For the future purposes we will create two additional files.

DAO.scala is almost empty for now. It will be responsible for managing database connection.

In the second class: DBSchema, we will put database schema configuration and some helper functions related with managing data.

The object above will be useful in the future. We will use it to setup and configure the database. For the sake of simplicity we won’t worry too much about blocking.

To recap, in this chapter we have learnt how to:

  • Initialize the SBT project,
  • Setup Database schema and connection.
Unlock the next chapter
Does GraphQL need HTTP Server?
Yes. It needs HTTP server.
Yes, it needs HTTP server but some of features can be used without that
No, but it strictly recommended to use. Without HTTP layer, GraphQL is losing some of its features.
No, GraphQL is a specification and it's far from transport protocol. You can use HTTP, Websockets, sockets or even use it internally in you application.