Wheretofind.me

Server development

You will need:

• Python 3.9 or later (brew install python@3.9)
• PostgreSQL (brew install postgres or Postgres.app)
• yarn (brew install yarn)
• virtualenv

Ensure the development database exists:

createdb wheretofindme

Set the following environment variables in a .env file in the repo root:

export DJANGO_SECRET_KEY=$(uuidgen)
export DJANGO_DEBUG=True
export SENDGRID_API_KEY=INVALID

Set up your virtualenv with your preferred method. For me, using virtualenvwrapper and pyenv, this looks like:

$ mkvirtualenv wheretofindme --python=$(pyenv which python3.9)
$ setvirtualenvproject
$ pip install pip-tools
$ pip-sync requirements/dev.txt requirements/base.txt

You'll have to read the contents of your .env file into your local shell's environment. You may want to set up a system to do this each time you activate this virtualenv. For one way to do this, see :ref:`env-vars` below.

Then get the database into a good state:

$ python manage.py migrate

Then get the frontend building:

$ yarn install
$ yarn serve

Then, you can visit the local site at http://localhost:8000/. The terminal output will say :8080, but ignore it, it's a liar.

Dependencies

To add Python dependencies, add them to requirements/{dev,base}.in, then recompile the locks:

$ pip-compile requirements/base.in && pip-compile requirements/dev.in

To add JavaScript dependencies:

$ yarn add (--dev) <dep name>

Tests

The Python tests are written using pytest, pytest-django, pytest-cov, and pytest-factoryboy. You can run them easily with yarn py:test:unit.

The JavaScript tests are written using Jest and Vue's unit-test helpers. You can run them easily with yarn js:test:unit.

We require 100% test coverage on both parts. We will gladly help you with any test-writing issues you may have.

Setting environment variables automatically

If you use virtualenvwrapper, one way to automatically set your environment variables when you activate your virtualenv is to use the postactivate hook. Having activated your virtualenv, edit $VIRTUAL_ENV/bin/postactivate to contain the following code:

if [[ -e .env ]]; then
    # Strip comment lines out of the .env file, and read
    # the rest in to the environment.
    export $(grep -v "^[ \t]*#" .env)
fi

This will read your .env file into your shell's environment each time you activate the virtualenv. It does no cleanup on deactivation; I must admit I've never found that important enough to be worth the trouble of implementing!