Chattr: A simple live chat website

---

Running the app

This app runs on docker-compose, which you can install from here.

You can then simply run docker-compose up in order to start the app.

Setting up a superuser and creating a new room

Once all the containers have started, log in to daphne_production with the following command:

docker exec -ti daphne_production /bin/bash.

You then need to run python manage.py createsuperuser.

Finally, go the the /admin page, (eg: localhost/admin) and create a new room.

Developing

In order to set up a development environment, you should follow these steps:

1. Install pip requirements:

You can do so using pip install -r requirements.txt (I would recommend using a virtual environment).

2. Export environment variables:

There are many ways to do this, but the one I found the most simple and easy was using this command:

export $(egrep -vh '^#' .env development.env | xargs -d '\n') (source).

If you'd like to unset these variables, you can use this command:

unset $(egrep -oh '^[^#].+=' .env development.env | xargs -d '=')

It's important .env goes before development.env, since it overrides some variables in order to achieve a simpler transition from a production environment to a development environment without breaking anything.

I'd recommend setting up a virtual environment so that these variables are automatically set/unset each time you activate/deactivate it, but there are plenty of other ways to do this.

3. Run a redis docker container:

You can use docker run -p $REDIS_PORT:$REDIS_PORT -d redis:5

(Use 6379 as the port number if the appropiate environment variable isn't set).

This container will be used when you run python manage.py runserver and will allow for websocket handling.

4. Install geckodriver/chromedriver:

In order to run automated tests, the app uses selenium along with geckodriver. If you'd like to, you can use chromedriver too, but remember to change the corresponding settings in the chat/tests.py file.

Testing

Once you've followed all these steps, you can run tests using python manage.py test.

You can also run the development server using python manage.py runserver.

If you want to directly run the docker containers, you'll have to rebuild them with docker-compose up --build.

CI

There is also a CI workflow set up to connect to my personal server, update the source code, and restart the app.

docker-compose layout

This app is composed of four services: nginx, daphne, redis and postgres:

• nginx works as a reverse proxy which serves static files and routes websocket traffic to daphne. Static files are located in nginx/collected_static (inside the container).

• The app runs on daphne, which handles websocket requests and communicates with the database. Websocket connections are set on port 8000.

• redis is an in-memory data structure used as a database, cache, and message broker.

• postgres is the database in which all data (eg: admin accounts) is stored.

Most of the settings of these containers are defined as environment variables, which can be found inside the .env file.

Services who do not need to know about each other are isolated through the use of two networks:

• frontend (nginx - daphne)
• backend (daphne - redis - postgres).

---

Notes

This is not meant to be a real and working web application. It's just a simple personal project with the goal of trying out a new framework, and learning how to set up a server.