dandi-publish

Develop with Docker (recommended)

This is the simplest configuration for developers to start with.

Initial Setup

1. Run ./dev/init-minio.sh
2. Run docker-compose run web ./manage.py migrate
3. Run docker-compose run web ./manage.py createsuperuser and follow the prompts to create your own user

Run Application

1. Run docker-compose up
2. When finished, use Ctrl+C

Develop natively (advanced)

This configuration still uses Docker to run attached services in the background, but allows developers to run the Python code on their native system.

Initial Setup

1. Run ./dev/init-minio.sh
2. Run docker-compose -f ./docker-compose.yml up -d
3. Install Python 3.8
4. Install psycopg2 build prerequisites
5. Create and activate a new Python virtualenv
6. Run pip install -e .
7. Run source ./dev/.env-docker-compose-native.sh
8. Run ./manage.py migrate
9. Run ./manage.py createsuperuser and follow the prompts to create your own user

Run Application

1. Run (in separate windows) both:
   1. ./manage.py runserver
   2. celery worker --app dandi.celery --loglevel info --without-heartbeat
2. When finished, run docker-compose stop