We'll set up the example API in the demo. You can view the source code here. If you would like to set this up from scratch, feel free the use the sample data provided for SQLite3 at /data/demo-data.sl3.

Defining SQL Schema

We'll set up a little API for a simple blog site built on SQLite3. It will have Users that can make Posts as well as Comments on people's posts. We will also let them follow other users and like comments. Here is a picture of the SQL schema.

I'll omit the code to set up the SQL tables. Here is an example of how it might be done. Now let's look at how the GraphQL API inteface on top of this will look.

The User is the top-level object type. It has a list of posts and comments the user has authored, as well as a list of users they are following. Post has a list of comments on the post, and the Comment has a list of the users that like it.

Declaring GraphQL Schema

Here is how the graphql-js schema would look, beginning with the top-level User.

import { GraphQLSchema } from 'graphql'
import { GraphQLObjectType, GraphQLList, GraphQLString, GraphQLInt } from 'graphql'

const User = new GraphQLObjectType({
  name: 'User',
  fields: () => ({
    id: {
      type: GraphQLInt
    },
    email: {
      type: GraphQLString
    },
    idEncoded: {
      description: 'The ID base-64 encoded',
      type: GraphQLString,
      resolve: user => toBase64(user.idEncoded)
    },
    fullName: {
      description: 'A user\'s first and last name',
      type: GraphQLString
    }
  })
})

const QueryRoot = new GraphQLObjectType({
  name: 'Query',
  fields: () => ({
    users: {
      type: new GraphQLList(User),
      resolve: () => {} // TODO
    }
  })
})

export default new GraphQLSchema({
  description: 'a test schema',
  query: QueryRoot
})

For now, we'll just provide a list of all the users. We'll fix this later. This won't work yet. There are several things missing. Next, we'll map this object type to our accounts table in SQL.