links: Phoenix MOC

---

Add Absinthe to Phoenix app

Installation

Add these plugins to the phoenix app

{:absinthe, "~> 1.6"},
{:absinthe_plug, "~> 1.5"}

Schema

Each graphql query or mutation has two parts, schema and resolver.

Schema defines how the data is and resolver finds the data in backend and returns to frontend

Now create a schema directory in the web folder and a schema file so we can start writing schema

defmodule KancheWeb.Schema do
	use Absinthe.Schema
 
	query do
		@desc "Get all users"
		field :users, list_of(:user) do
			resolve &KancheWeb.Resolvers.Accounts.users/3
		end
	end
 
  
 
	object :user do
		field :id, non_null(:id)
		field :email, non_null(:string)
		field :password_hash, non_null(:string)
	end
end

In the above example, we instruct to use Absinthe.Schema and we have a query macro where we define the queries with description, field and resolve.

The object macro can be used to defined user object containing three fields

Resolver

Create a new resolver directory under web and then create a elixir file with context name Context.ex, in this example Accounts.ex

defmodule KancheWeb.Resolvers.Accounts do
	alias Kanche.Accounts
 
	def users(_, _, _) do
		{:ok, Accounts.list_users()}
	end
end

Resolvers are 3-arity functions. Most of the time you care about the second argument which is a map of query arguments. The resolver functions simply delegate the work to the generated Accounts context module and return an :ok tuple

The only thing left is to configure router.ex file

scope "/graphql" do
	pipe_through :api
	
	# Absinthe Graphql Routes
	forward "/graphiql", Absinthe.Plug.GraphiQL,
		schema: KancheWeb.Schema,
		interface: :simple
	forward "/", Absinthe.Plug,
		schema: KancheWeb.Schema
end

That’s it 🥳

---

tags: graphql, phoenix, absinthe, basics sources:

• [https://pragmaticstudio.com/tutorials/how-to-setup-graphql-in-a-phoenix-app](How to setup graphql in a phoenix app)
• [https://www.howtographql.com/graphql-elixir/2-queries/](How to graphql)