Skip to content

B-Alica/django-superset-integration

Repository files navigation

django-superset-integration

PyPI - Version GitHub Issues or Pull Requests

Monthly Downloads Total Downloads

django-superset-integration is a Django app to integration Apache Superset dashboards into a Django application.

Quick start

  1. Add django_superset_integration to your INSTALLED_APPS setting like this:
INSTALLED_APPS = [
    ...,
    "django_superset_integration",
    ...,
]
  1. Include the superset-integration URLconf in your project urls.py like this:
path("superset_integration/", include("django_superset_integration.urls")),
  1. You will need a cryptography Fernet key, so you need to install cryptography:
pip install cryptography
  1. Generate a Fernet key in a python terminal:
from cryptography.fernet import Fernet
FERNET_KEY = Fernet.generate_key()
  1. The result is a bytestring like b'jozEHFGLKJHEFUIHEZ4'. Copy ONLY the content of the string, not the b nor the quotation marks

  2. In your env variables, create a variable FERNET_KEY with the copied content as value

  3. Add a variable ENCRYPTION_KEY in your settings.py referencing your env variable FERNET_KEY:

ENCRYPTION_KEY = os.environ.get("FERNET_KEY", "").encode()
  1. By default, all dashboard data will be displayed. You can override this by creating your own filtering function and adding it in your settings.py:
RLS_FUNCTION = "my_app.my_module.create_rls_clause"

Your function must take a parameter user and return a SQL rls clause like this : [{"clause": "1=1"}] (where you replace 1=1 by the clause you want).

See Superset documentation for more information

  1. Make sure that your Superset instance parameter GUEST_TOKEN_JWT_EXP_SECONDS is more than 300 (5 minutes). Otherwise it will expire before it can be refreshed. For example, set it to 600 (10 minutes).

  2. In the template where you want to integrate the dashboard, add the following at the emplacement where you want the dashboard:

{% load static %}

...

{% include "django_superset_integration/superset-integration.html" %}
  1. Run python manage.py migrate to create the models.

  2. Start the development server and visit the admin site to create a SupersetInstance object.

    • address: the address of your Superset instance
    • username: the username that allows to connect via api to your instance. By default : superset_api You need to have a service account with minimal permissions to embed dashboards. See Superset documentation for more info.
    • password: the password that allows to connect via api to your instance.
  3. After you have created a SupersetInstance object, create a SupersetDashboard object.

    • integration_id: the integration id given by Superset to integrate your dashboard
    • name: a name for your dashboard
    • domain: (foreign key) the SupersetInstance object corresponding to the instance where the dashboard is
    • comment: (optional) a plain text comment
    • superset_link: (optional) the link to your dashboard in Superset
  4. In the view where you want to integrate the dashboard, in get_context_data, add the following:

from django_superset_integration.models import SupersetDashboard

def get_context_data(self, **kwargs):
    context = super().get_context_data(**kwargs)

    ...

    dashboard_name = "THE NAME OF YOUR DASHBOARD FROM STEP 13"
    dashboard = SupersetDashboard.objects.get(name__iexact=dashboard_name)
    context["dashboard_integration_id"] = dashboard.integration_id
    context["dashboard_id"] = dashboard.id
    context["superset_domain"] = dashboard.domain.address

    ...

    return context
  1. You can personalize the buttons "Fullscreen" and "Quit fullscreen" by giving class names in your view's get_context_data:
def get_context_data(self, **kwargs):
    context = super().get_context_data(**kwargs)

    ...

    context["button_fullscreen_classes"] = "myClass1 myClass2"
    context["button_quit_fullscreen_classes"] = "myClass1 myClass2"

    ...

    return context
  1. That should be it!

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Packages

No packages published