Quick Start

In this Quick-Start, we will:

Set up a basic pair of models with a relation between them.
Add them to a graphql schema and serve the graph API.
Query the graph API for model contents.

For a more advanced example of a similar setup including a set of mutations and more queries, please check the example app .

Installation

poetry add strawberry-graphql-django
poetry add django-choices-field  # Not required but recommended

(Not using poetry yet? pip install strawberry-graphql-django works fine too.)

Define your application models

We’ll build an example database of fruit and their colours.

Tip

You’ll notice that for Fruit.category , we use TextChoicesField instead of TextField(choices=...) . This allows strawberry-django to automatically use an enum in the graphQL schema, instead of a string which would be the default behaviour for TextField.

See the choices-field integration for more information.

from django.db import models
from django_choices_field import TextChoicesField

class FruitCategory(models.TextChoices):
    CITRUS = "citrus", "Citrus"
    BERRY = "berry", "Berry"

class Fruit(models.Model):
    """A tasty treat"""

    name = models.CharField(max_length=20, help_text="The name of the fruit variety")
    category = TextChoicesField(choices_enum=FruitCategory, help_text="The category of the fruit")
    color = models.ForeignKey(
        "Color",
        on_delete=models.CASCADE,
        related_name="fruits",
        blank=True,
        null=True,
        help_text="The color of this kind of fruit",
    )

class Color(models.Model):
    """The hue of your tasty treat"""

    name = models.CharField(
        max_length=20,
        help_text="The color name",
    )

You’ll need to make migrations then migrate:

python manage.py makemigrations
python manage.py migrate

Now use the django shell, the admin, the loaddata command or whatever tool you like to load some fruits and colors. I’ve loaded a red strawberry (predictable, right?!) ready for later.

Define types

Before creating queries, you have to define a type for each model. A type is a fundamental unit of the schema which describes the shape of the data that can be queried from the GraphQL server. Types can represent scalar values (like String, Int, Boolean, Float, and ID), enums, or complex objects that consist of many fields.

Tip

A key feature of strawberry-graphql-django is that it provides helpers to create types from django models, by automatically inferring types (and even documentation!!) from the model fields.

See the fields guide for more information.

import strawberry_django
from strawberry import auto

from . import models

@strawberry_django.type(models.Fruit)
class Fruit:
    id: auto
    name: auto
    category: auto
    color: "Color"  # Strawberry will understand that this refers to the "Color" type that's defined below

@strawberry_django.type(models.Color)
class Color:
    id: auto
    name: auto
    fruits: list[Fruit] # This tells strawberry about the ForeignKey to the Fruit model and how to represent the Fruit instances on that relation

Build the queries and schema

Next we want to assemble the schema from its building block types.

Warning

You’ll notice a familiar statement, fruits: list[Fruit] . We already used this statement in the previous step in types.py . Seeing it twice can be a point of confusion when you’re first getting to grips with graph and strawberry.

The purpose here is similar but subtly different. Previously, the syntax defined that it was possible to make a query that traverses within the graph, from a Color to a list of Fruits. Here, the usage defines a root query (a bit like an entrypoint into the graph).

Tip

We add the DjangoOptimizerExtension here. Don’t worry about why for now, but you’re almost certain to want it.

See the optimizer guide for more information.

import strawberry
from strawberry_django.optimizer import DjangoOptimizerExtension

from .types import Fruit

@strawberry.type
class Query:
    fruits: list[Fruit] = strawberry_django.field()

schema = strawberry.Schema(
    query=Query,
    extensions=[
        DjangoOptimizerExtension,
    ],
)

Serving the API

Now we’re showing off. This isn’t enabled by default, since existing django applications will likely have model docstrings and help text that aren’t user-oriented. But if you’re starting clean (or overhauling existing dosctrings and helptext), setting up the following is super useful for your API users.

If you don’t set these true, you can always provide user-oriented descriptions. See the

STRAWBERRY_DJANGO = {
    "FIELD_DESCRIPTION_FROM_HELP_TEXT": True,
    "TYPE_DESCRIPTION_FROM_MODEL_DOCSTRING": True,
}

from django.urls import include, path
from strawberry.django.views import AsyncGraphQLView

from .schema import schema

urlpatterns = [
    path('graphql', AsyncGraphQLView.as_view(schema=schema)),
]

This generates following schema:

enum FruitCategory {
  CITRUS
  BERRY
}

"""
A tasty treat
"""
type Fruit {
  id: ID!
  name: String!
  category: FruitCategory!
  color: Color
}

type Color {
  id: ID!
  """
  field description
  """
  name: String!
  fruits: [Fruit!]
}

type Query {
  fruits: [Fruit!]!
}

Using the API

Start your server with:

python manage.py runserver

Then visit localhost:8000/graphql in your browser. You should see the graphql explorer being served by django. Using the interactive query tool, you can query for the fruits you added earlier:

GraphiQL with fruit