Geospatial Dataset Publishing Dev Environment

This repository contains a minimal Docker Compose setup for running pygeoapi alongside a simple frontend using MapLibre and Chart.js.

Prerequisites

• Docker Desktop with Docker Compose v2+
• (Optional) GDAL >= 3.0 for converting data to other formats

Usage

1. Build and start the services:

   docker compose up --build

2. Access pygeoapi at http://localhost:5000
3. Access the frontend at http://localhost:8080

Deployment

Environment Variables

When deploying in different environments (local, Kubernetes, etc.), the following environment variables should be set:

Backend (pygeoapi):

• PYGEOAPI_SERVER_URL: The external URL where pygeoapi will be accessible. This is used to generate correct links in the API responses.
  • For local development: http://localhost:5000
  • For Kubernetes/production: Your domain, e.g., https://your-domain.com/api

Frontend:

• API_BASE_URL: The URL where the frontend will connect to the backend API.
  • For local development: Not needed (will default to http://localhost:5000)
  • For Docker Compose: http://backend:5000
  • For Kubernetes/production: Use the full URL including protocol and domain, e.g., https://your-domain.com/api (not just the path)

Kubernetes Deployment

To deploy to Kubernetes:

1. Update domain and path settings in k8s-deployment.yaml
2. Apply the configuration:

kubectl apply -f k8s-deployment.yaml

This will deploy both the frontend and backend services with the proper configuration.

The backend Docker image now generates the OpenAPI document at build time and exposes it via the PYGEOAPI_OPENAPI environment variable.

The frontend fetches features from pygeoapi and displays them on a MapLibre map and Chart.js chart.

Data

The sample dataset resides in data/norway/norway-hazard-points.geojson. You can replace this file with your own dataset. Update pygeoapi-config.yml if you change the path or dataset name.

Development

This repository uses Black, isort, and Ruff for Python code. The frontend is formatted with Prettier. Recommended VS Code settings are provided in the .vscode folder. Run pip install -r requirements-dev.txt to install Python tooling. Frontend formatting tools can be installed by running npm install inside the frontend directory. Use npm run format to format frontend code.

CI/CD

A GitHub Actions workflow builds and publishes the frontend Docker image whenever changes are pushed to the frontend/ directory on the main branch. The image is pushed to the GitHub Container Registry under the tag global-tsunami-risk-map-frontend.

Another workflow builds a custom pygeoapi image containing the data from this repository. Whenever files in the data/ or pygeoapi/ folders change, the workflow bumps the repository VERSION file and publishes a new Docker image named pygeoapi-w-global-tsunami-data. The image tag combines the upstream pygeoapi version with the bumped repository version.