This repo contains examples for integrating other frameworks or python packages. By no means is the repo complete or necessarily best-practice, instead it represents a diary of my exploration of Django & Friends.
In particular the repo covers:
There are several ways to use a fully fledged frontend js library like Vue.js with Django.
-
Separate frontend and backend. This is the go-to-way for large applications, but requires rewriting the Django Part to a REST API. Thus the django templating system cannot be used anymore.
-
Integrating a standalone vue frontend in Django templates. This approach allows to use single-file-components (SFC) for the vue side while still using Djangos views. A commonly used package is django-webpack-loader which is not maintained and adds additional dependencies. Thus, a better way of integrating vue.js is to use Djangos template system to integrate the vue cli output. A full description is given in the repo (https://github.com/EugeneDae/django-vue-cli-webpack-demo). This approach is applied to the Django&Friends repo in the folder frontend.
-
Integrate vue.js via CDN. This approach does not allow to use SFC and thus has to rely on Django template splitting. From my experience this can be a first step for using vue.js but is hardly maintainable. A simple example is given in the app
Integration of bokeh and panel is straightforward given the respective documentation. In particular, the panel package provides a nice example. Bokeh documentation is not as simple, but the implementation follows the same route.
Required Packages:
- Django Channels
Django Apps:
- visualization_bokeh
The visualization_chartjs app shows a simple example of integrating Chart.js in a Django app. Besides passing the initial data to the chart, the app shows how to implement an AJAX POST call to your Django view and return data to update the chart. Based on the example more complex callbacks can be easily constructed. Asides from standard Django code this examples requires basic javascript knowledge, but don't worry the code is straightforward and easy to understand.
The GraphQL API examples lives in the "api" app besides the REST API example. Packages in use are graphene for setting up a simple GraphQL API and channels_graphql_ws for adding subscriptions to the API Endpoint.
Whereas setting up Query and Mutation request is very easy, following the graphene documentation, the Subscription via Websockets is a bit more involved. A very nice example can be found here. The Django & Friends Repo follows the same approach, but with the Ingredient/Category models from the graphene documentation
Django signals allow to execute specific actions on model.save() and other methods. The example case is given in the home app, which extends the User model with a Profile and auto-generates Profile instances on each save of a new User instance.
Celery can be used to run background tasks while the Django Server is not blocked. I.e. a request can trigger an expensive calculation but return the response immediately. The example in the current repo is based on the Celery Official Documentation and Stackoverflow. The To interact between the Django Server and the Celery workers a message Broker, i.e. Redis is required. After downloading Redis simple change to the folder and run a '''make'''. To run the celery workers we need
- The running Django App
- Running Redis Server
- Running Celery Queue
celery -A mysite worke -l info - Runniny Celery Beat
celery -A mysite beat -l info
Django crispy forms allow to specify Django form logic and templating inside the python code, i.e. it removes a lot of html code. The crud app shows an extended example of dynamic forms within a Django app, combining javascript, Django crispy forms, Django formsetfactories and dealing with foreign keys in forms.