Django Scalar

Django-Scalar provides an easy way to serve your API docs with Scalar.

Note: This project is still in its early stages. Pull requests are welcome.

Installation

The python package to install is django-scalar. You can install it with pip install django-scalar or uv add django-scalar based on your preferred python package manager.

Integrate in your Django app

At this stage, the view expects that you have your OpenAPI schema published with DRF-Spectacular under the api/schema/ endpoint. See below under Customize Django Scalar for how you can customize this.

To integrate django scalar include django-scalar.views.scalar_viewer in your urls.py.

from django.urls import path
from django_scalar.views import scalar_viewer
from drf_spectacular.views import SpectacularAPIView

urlpatterns = [
    # ...existing URL patterns...
    path("api/schema/", SpectacularAPIView.as_view(), name="schema"),
    # Add this:
    path("api-docs/", scalar_viewer, name="scalar-viewer"),
]

Feel free to change api-docs/ to your liking. We've seen api/docs etc. in the wild.

Customize Django Scalar

ℹ️ Note we have plans to make this easier soon.

The scalar_viewer view will pass on some hardcoded defaults to the template. You can override the template and change everything to match your environment.

Requirements:

There are two main requirements for this project:

• drf-spectacular
• django-filters (optional)

This project implicitly depends on Django and Django-Rest-Framework.

Testing

For information on running tests, see the tests README.

Documentation example:

    #views.py
    @extend_schema(
        responses={200: UserSerializer},
        tags=["User"],
        description="Retrieve a list of all users",
        methods=["GET"],
        parameters=get_filter_parameters(UserFilter), ## this get_filter_parameters is using the django-filters base to create the necessary parameters automatically.
    )
    def list(self, request: Request, *args: Any, **kwargs: Any) -> Response:
        return super().list(request, *args, **kwargs)
  #filters/user.py

Using get_filter_parameters

You need to create a basic or complex django-filter class and then use the script from get_filter_parameters the auto-parse the class to a valid OpenApiParameter.

  #(app)/filters/users.py
  import django_filters
  from django.conf import settings


  class UserFilter(django_filters.FilterSet):
      # Filter by username (case-insensitive exact match)
      username = django_filters.CharFilter(lookup_expr="iexact")

      # Filter by email (case-insensitive exact match)
      email = django_filters.CharFilter(lookup_expr="iexact")

      # Filter by full name (case-insensitive partial match)
      full_name = django_filters.CharFilter(
          field_name="first_name", lookup_expr="icontains", label="Full Name"
      )

      # Filter by is_active (boolean filter)
      is_active = django_filters.BooleanFilter()

      # Filter by date joined (range filter for datetime)
      date_joined_start = django_filters.DateTimeFilter(
          field_name="date_joined", lookup_expr="gte", label="Joined After"
      )
      date_joined_end = django_filters.DateTimeFilter(
          field_name="date_joined", lookup_expr="lte", label="Joined Before"
      )

      # Filter by genres (many-to-many relationship)
      genres = django_filters.ModelMultipleChoiceFilter(
          queryset=User.objects.all(), field_name="genres__name", label="Genres"
      )

      # Filter by trusty (boolean filter)
      is_trusty = django_filters.BooleanFilter()

      # Filter by fee range (decimal filter)
      min_fee = django_filters.NumberFilter(
          field_name="fee", lookup_expr="gte", label="Min Fee"
      )
      max_fee = django_filters.NumberFilter(
          field_name="fee", lookup_expr="lte", label="Max Fee"
      )
      type = django_filters.ChoiceFilter(
          choices=[
              ("music", "Musico"),
              ("establishment", "Estabelecimento"),
          ],
      )

      class Meta:
          model = settings.AUTH_USER_MODEL,
          fields = [
              "username",
              "email",
              "is_active",
              "is_trusty",
              "genres",
              "date_joined_start",
              "date_joined_end",
              "min_fee",
              "max_fee",
              "type",
          ]