To simplify development including creating migrations and running the devserver, the bundled manage.py file can be used with the settings and configuration in the config module.:

./manage.py makemigrations orb
./manage.py migrate
./manage.py runserver

All custom settings should be placed in config/local_settings.py. This file is not source controlled. It should be written in the same format as any other Django settings file. The primary settings file will try a naked import from this file at the end of the main settings file, overriding any other settings specified and adding any new ones.

The settings included are designed to be as simple and sensible for baseline development as possible. It is highly recommended that the local settings file be used to specify database settings.

The Docker container system is mostly there, and at this point allows basic use and development using Docker Compose (using version 2 config).

With Docker and Docker Compose installed, add this line to the .env file in your project root (adding it if absent):

COMPOSE_FILE=dev.yml

Add any seed data in the form of a SQL dump to local_data/mysql-init/. All SQL files found here will be executed when the container is first created.

Then run:

docker-compose up

This will bring up three containers:

1. MySQL
2. Solr
3. Django application

The Solr container will be up but for reasons of configuration to be completed is not a working search backend yet.

To shutdown the containers run:

docker-compose down

If the Django application is unresponsive because of a missing database then run:

docker-compose restart django

It may be the case that the container finished loading before the MySQL container and service were available.

To run a Django command, use the following:

docker-compose run django python manage.py <command>

The tests for the COVID-19 Library platform can be run using the current Python path (or currently activated virtual environment) using the command:

make test-django

The project is also configured to use tox for isolated testing and test environment matrices. This ensures that the full test environment can be created outside and apart from the development environment and that tests can be run against the application with different dependency versions, including Django and even different Python versions. To use tox first install it:

pip install tox

And then run the command:

tox

Using the -r flag will recreate the test environment from scratch.:

tox -r
tox -e py36-django2.2 -r

For running a large matrix of tests, consider using the detox tool which parallelizes tests based on the number of available cores.:

pip install detox
detox

The flake8 environment is limited to running flake8 for linting and basic static analysis.

The deployed environment uses the frozen requirements file to ensure that every dependency is pinned to the same requirements used to deploy the application. (Note that this still requires settings updates to ensure that the same database backend is used.)

The remaining environments are designed to work around different versions of Django, to allow testing against future Django versions. The requirements here should be loose enough to allow newer versions of the dependency to be used.

All JS-based application development based on VueJS is done in the local vue directory. On initial environment, run make vue-deps. This will download the needed node and js files to the local vue directory.

Vue and its dependencies are bundled with the application JS file, course-builder.js. This is done so that all JS downloads at one time and is cached, so that if a user goes offline, they will not need to download the files again.

The following commands are useful for development:

`make vue-deps`: install vue and node dependencies `make vue-test`: run the vue unit tests once `make vue-test-dev`: run the vue unit tests in watch mode for TDD `make vue-dev`: run the webpack vue builder for development in browser, file are watched 'make vue-build`: run the webpack build to minimize and select correct vue files for production