Momentum Classroom

This Django application is used to conduct self-evaluations for Momentum Learning.

Set up for development

This application uses Pipenv to manage dependencies. To install all dependencies, run:

pipenv install

Pipenv requires several environment variables to be set. These are all set by running:

pipenv shell

You must do this before starting this application.

For deployment purposes, we also need a requirements.txt file. After installing any new packages via Pipenv, please run make requirements.txt.

Database

This application requires PostgreSQL as its database. For development, the default settings are a database with the name classroom, owned by a user with the name classroom and no password. To set this up locally, run:

createuser -d classroom
createdb -U classroom classroom

Seeding skills data

Skills are listed in versioned markdown files and can be seeded to the database with the custom management command load_skills followed by the path to the file.

The naming convention is important as the script will assign versions to the skills based on the filename; the file should start with a lowercase v followed by a version number. For instance, a file named v2-skills.md will have skills saved with a version number 2.

Versioning the skills lets us keep backwards compatible data as we update skill sets on which we want to evaluate students.

python ./manage.py load_skills skills.md

.env

This application is configured via environment variables. In lieu of using environment variables, you can copy the file config/.env.sample to config/.env and then edit this file. Note that by default, debug mode is off, and that you need to set the environment variable DEBUG or use the config/.env file to turn debug mode on.

If you add new configuration that might change on environment, please use the environ package (see the top of config/settings.py) and environment variables to set up that configuration.

Deployment

This application is currently deployed on Heroku under the admin@momentumlearn.com account. Automatic deploys are enabled through GitHub, so every push to main will also deploy to Heroku.