REST API for accessing information related to waterbodies that was extracted from Earth observation data.
The docker compose file includes two containers
- server - includes a Python based FastAPI application and all dependencies needed for the waterbodies-api application
- db-postgres - includes a PostgreSQL server with the PostGIS extension
Clone the repo
git clone https://github.com/digitalearthafrica/waterbodies-api.git
cd waterbodies-api
Create an environment file that includes env vars needed by each of the docker containers. If required edit the .env file to connect to alternate databases, the sample file includes all details needed to connect to the DB container specified in the docker compose file.
cp .env.sample .env
Build the waterbodies API docker containers
docker compose build
Start the waterbodies API
docker compose up
The application should then be accessible on localhost:8080. A simple connection check request handler can be used to ensure the web server and database are running as expected, this can be accessed at http://localhost:8080/check-connection
FastAPI automatically generates documentation for request handlers added to the application, this can be accessed from the following link.
When using the docker compose based development environment code within the ./server/app folder is mounted into the server container. The compose yaml also includes the --reload uvicorn command like argument which means that any change to the Python source code will automatically restart the application with updated code (this is shown in docker logs). You do not need to rebuild, or stop/start the server container during development.
Changes to the requirements.txt file will require the application to be rebuilt (eg; stop the container and run docker compose build)
The local ./data/db folder is mounted into the db-postgres container as /data. Database dumps should be placed in this folder.
The following command can be used to restore a postgres database dump.
docker compose exec db-postgres /bin/bash -c "psql -U postgres -h localhost -d waterbodies < /data/waterbodies_dump.psql"
The Waterbodies API Docker image is built using a GitHub workflow. New images are deployed to target environments using Flux CD.
| Environment | Trigger | Build Branch | Image Tag | Container Registry | Deployment Interval |
|---|---|---|---|---|---|
| Development | PR: create/sync | PR branch | Latest release, commit ID, timestamp (e.g. v0.0.1-81541a0-1715142069) |
Docker Hub and ECR | 5 min |
| Production | Release: publish | Main branch | Release tag (e.g. 1.0.0) |
Docker Hub and ECR | 5 min |