Skip to content

Latest commit

 

History

History
123 lines (84 loc) · 4.04 KB

CONTRIBUTING.md

File metadata and controls

123 lines (84 loc) · 4.04 KB
Raw

Introduction

This repository contains a development environment for the Mittaridatapumppu project.

It is based on the microservices architecture and consists of the following services:

Local Kafka is used as a message broker between the services. For demo purposes there is also an InfluxDB V2 database for storing the parsed data.

Included docker-compose.yml file is for running the whole pipeline locally.

There are also some scripts for populating the database with real data.

Developer's guide

Clone the repo

git clone --recurse-submodules git@github.com:City-of-Helsinki/mittaridatapumppu.git

Branches

Create applicable branches in sub-repositories where needed while developing, e.g. feature/new-feature or bugfix/bugfix-description.

Developing with Docker

Set up the environment by building the images and loading initial data for development.

docker compose up --build
docker compose exec deviceregistry python manage.py loaddata devices/fixtures/auth.json devices/fixtures/authtoken.json devices/fixtures/endpoints.json

Later on you can start the services with

docker compose up

Developing without Docker

Do not use this for now. It is not tested with the latest code and probably does not work. 
dropdb deviceregistry
createdb deviceregistry
psql deviceregistry -c "create extension postgis;commit;"
export DEBUG=1
export DATABASE_LOCAL=1
python manage.py makemigrations
python manage.py migrate
python manage.py loaddata devices/fixtures/auth.json devices/fixtures/authtoken.json devices/fixtures/endpoints.json

Log in to Django admin

Go to http://localhost:8000/admin/ and login with root and ruutruut. Check that there are some endpoints defined.

Verify that the hosts endpoint returns localhost and at least 2 endpoints.

Populate database with real data

You will need to have the device registry service (deviceregistry) running and an Excel file containing data of devices and locations and uirasmeta.py, which contains additional metadata of UiRaS sensors.

python mittaridatapumppu/devices/scripts/populate_db_via_api.py --api-url http://127.0.0.1:8000/api/v1/ --api-token abcdef1234567890abcdef1234567890abcdef12 --excel-file Mittaridatapumppu-initial-data.xlsx
python mittaridatapumppu/devices/scripts/populate_db_via_api.py --api-url http://127.0.0.1:8000/api/v1/ --api-token abcdef1234567890abcdef1234567890abcdef12 --uirasmeta

# Try it out with httpie (https://httpie.org/):
http GET http://127.0.0.1:8000/api/v1/devices/70B3D57050011422/ "Authorization:Token abcdef1234567890abcdef1234567890abcdef12"
# The same with curl:
curl http://127.0.0.1:8000/api/v1/devices/70B3D57050011422/ -H "Authorization:Token abcdef1234567890abcdef1234567890abcdef12" | jq

Send data to the pipeline

python mittaridatapumppu-endpoint/tests/test_api2.py

docker run --rm -e API_BASE_URL -e API_TOKEN ghcr.io/city-of-helsinki/mittaridatapumppu-endpoint python tests/test_api2.py

Check docker logs what happens.

Check the data in InfluxDB

Go to http://localhost:18086/ and login with root and ruutruut.

Check the data in Django admin

Go to http://localhost:8000/admin/ and login with root and ruutruut.

Running the tests

Currently, there are both Django tests and pytest tests.

Run tests:

docker compose exec deviceregistry python manage.py test devices/tests
docker compose exec deviceregistry pytest devices/tests/test_api_devicetype.py
docker compose exec deviceregistry pytest devices/tests/test_api_device.py

Later only pytest tests will be used.

Code Formatting

Linting

Pre-commit hook