kobotoolbox/kobo-compose

A simplified Docker Compose environment for development

⚠️ ATTENTION - UNSUPPORTED

This repository is used internally by some KoboToolbox developers and shared here in the hope that it may be helpful to others. There is ABSOLUTELY NO SUPPORT offered for this code: not in the issue tracker, not in the discussion forum, not via email, not anywhere! You have been warned. If you are seeking to run your own instance of KoboToolbox, you should look at https://github.com/kobotoolbox/kobo-install/, or, for large installs, consider Kubernetes: https://github.com/kobotoolbox/kobo-helm-chart

First run

Clone kpi and kobocat repos in this directory

1. Build docker compose build --pull
2. Start postgres docker compose up postgres this ensures it has time to initialize
3. Migrate kobocat docker compose run --rm kobocat ./manage.py migrate
4. Migrate kpi docker compose run --rm kpi ./manage.py migrate
5. Make user docker compose run --rm kpi ./manage.py createsuperuser
6. Edit /etc/hosts and add 127.0.0.1 kf.kobo.local kc.kobo.local ee.kobo.local
7. Run npm i --force in the kpi directory.

Start

1. docker compose up
2. Run frontend (in own tab) cd kpi and npm run watch

• KPI http://kf.kobo.local
• EE http://ee.kobo.local:8005 (or localhost:8005)

Run with mailhog for graphical email testing

Mailhog will capture all emails sent, regardless of to address.

1. Start mailhog docker compose -f docker-compose.mailhog.yml up
2. Edit docker-compose.yml (or copy to docker-compose.override.yml to preserve original file) and uncomment EMAIL_BACKEND: 'django.core.mail.backends.smtp.EmailBackend'

Go to http://localhost:8025

Python testing

1. Run from inside a container docker compose exec -it kpi bash (Assumes application is already running)
2. Install dev dependencies if not already done pip install -r dependencies/pip/dev_requirements.txt
3. Run Pytest pytest or to run on a specific directory and reuse the DB (as a speedup) pytest --reuse-db kobo/apps/foo.

Rebuild docker images

If python packages in kpi or kobocat change, you can build like this

docker compose build --pull

Run migrate commands when new migrations are added

Destroy everything and start over

Useful to remake databases.

1. docker compose down

Run first run steps again

Text Editor lint, type checking, and type inference (Optional)

Many editors and cli tools can do type checking and type inference. However, it requires setting up a virtual environment.

1. Install Python 3 dependencies. For Ubuntu this is apt install python3-dev python3-venv
2. Create Python virtual environment python3 -m venv env
3. Activate environment source env/bin/activate
4. Install packages pip install -r dependencies/pip/dev_requirements.txt

These are not required but can improve the experience of editing code. Most text editors cannot read these dependencies from the docker container itself.