GCC site

Setup a development environment

Install docker and docker-compose using the Docker Installation Guide or your distribution's documentation.

You will also need poetry and pre-commit. Those are not strictly needed to develop on this project, but they will become very handy for things like adding dependencies, or pre-commit hooks.

For those using Nix, everything is ready in the flake.nix file. (NOT UPDATED)

Initial setup

$ pushd website && poetry install && popd
$ cd docker/
$ ./gen_secrets.sh # Input the required secrets when asked
$ docker compose -p gccsite up --build # You can use the flag -d if you to detach the containers from your shell

You can then go to http://localhost:8000/.

Inserting fixtures

For the development environment to reproduce normal state, the folder website/fixtures/ contains several files that are used to insert test data in the database. There are users, event centers, events, applications, etc...

To load all the fixtures, run the following command :

$ cd docker/
$ ./manage.sh loaddata fixtures/users.json fixtures/centers.json fixtures/events.json fixtures/profiles.json fixtures/applications.json

/!\ The order is important, as some fixtures references other model fixtures (we need to load users before profiles because profiles references users).

For the users fixture, user passwords are the same as their usernames.

E.g : root@example.com -> root

root@example.com is a superuser

Super User creation

To create a super User :

$ cd docker/
$ ./manage.sh createsuperuser
# Then follow the given instructions.

You can then go to http://localhost:8000/admin and login.

Regular usage

$ cd docker/
$ docker compose up --build

Formatting

We use the following tools to format code:

• ruff for python files
• djlint for Django HTML template files format (and linting)
• eslint for JS files (this is not the case anymore, apparently)

All these tools should be run by pre-commit to ensure the well formatting of your changes. You can install them directly using :

$ pre-commit install

Linting

We also use pylint (via prospector) to check for common Python errors. You can use the following commands to check your code:

$ cd docker/
$ docker-compose exec backend_dev-1 prospector --profile base