compose.yaml

volumes:
  ckan_storage:
  pg_data:
  solr_data:
  jena_data:


services:
  nginx:
    image: nginx:1.27-alpine
    env_file:
      - .env
    depends_on:
      ckan:
        condition: service_healthy
    ports:
      - "0.0.0.0:${NGINX_PORT_HOST}:${NGINX_PORT}"
      - "0.0.0.0:${NGINX_SSLPORT_HOST}:${NGINX_SSLPORT}"
    volumes:
      - ./config/nginx/default.conf.template:/etc/nginx/templates/default.conf.template
      - ./ckan-docker/nginx/setup/nginx.conf:/etc/nginx/nginx.conf
      - ./ckan-docker/nginx/setup/ckan-local.crt:/etc/nginx/certs/ckan-local.crt
      - ./ckan-docker/nginx/setup/ckan-local.key:/etc/nginx/certs/ckan-local.key
    networks:
      - datastack_net
    healthcheck:
      test: [CMD-SHELL, wget -qO /dev/null http://localhost || exit 1]
      interval: 60s
      timeout: 10s
      retries: 5
    restart: unless-stopped

  ckan:
    image: ckan
    build:
      context: ckan-docker/ckan/
      dockerfile: ../../config/ckan/Dockerfile
      args:
        - TZ=${TZ}
    user: 503:502
    environment:
      - DEBUG=false
    env_file:
      - .env
    depends_on:
      db:
        condition: service_healthy
      solr:
        condition: service_healthy
      redis:
        condition: service_healthy
    volumes:
      - ckan_storage:/var/lib/ckan
      - ./config/ckan/start_ckan.sh:/srv/app/start_ckan.sh
      - ./config/ckan/03_harvest.sh:/docker-entrypoint.d/03_harvest.sh
      - ./config/ckan/04_supervisor.sh:/docker-entrypoint.d/04_supervisor.sh
      - ./config/ckan/dcatde_themes.json:/srv/app/dcatde_themes.json
      - ./config/ckan/supervisor-ckan-worker.conf:/etc/supervisor/conf.d/ckan-worker.conf
    restart: unless-stopped
    entrypoint: /srv/app/start_ckan.sh
    healthcheck:
      test: [CMD, wget, -qO, /dev/null, "http://ckan:${CKAN_PORT}"]
      interval: 60s
      start_period: 2m
      start_interval: 10s
    networks:
      - datastack_net

  db:
    build:
      context: ckan-docker/postgresql/
    env_file:
      - .env
    volumes:
      - pg_data:/var/lib/postgresql/data
    restart: unless-stopped
    healthcheck:
      test: [CMD, pg_isready, -U, "${POSTGRES_USER}", -d, "${POSTGRES_DB}"]
    networks:
      - datastack_net

  solr:
    image: "ckan/ckan-solr:${SOLR_IMAGE_VERSION}"
    volumes:
      - solr_data:/var/solr
    restart: unless-stopped
    healthcheck:
      test: [CMD, wget, -qO, /dev/null, http://localhost:8983/solr/]
    networks:
      - datastack_net

  redis:
    image: "redis:${REDIS_VERSION}"
    restart: unless-stopped
    healthcheck:
      test: [CMD, redis-cli, -e, QUIT]
    networks:
      - datastack_net

  fuseki:
    build:
      context: config/fuseki/
    environment:
      - "JAVA_OPTIONS=${FUSEKI_JAVA_OPTS}"
      - "ADMIN_PASSWORD=${CKANINI__CKANEXT__FUSEKI__PASSWORD}"
      - ENABLE_DATA_WRITE=true
      - ENABLE_UPDATE=true
      - ENABLE_UPLOAD=true
      - QUERY_TIMEOUT=60000
    volumes:
      - jena_data:/fuseki-base
    stop_grace_period: 10s
    healthcheck:
      test:
        - CMD-SHELL
        - wget -qO /dev/null http://localhost:3030/$$/ping || exit 1
      interval: 10s
      timeout: 5s
      retries: 3
    restart: unless-stopped
    networks:
      - datastack_net

  csvtocsvw:
    image: ghcr.io/mat-o-lab/csvtocsvw:latest
    environment:
      APP_MODE: production
      PORT: "${CSVTOCSVW_APP_PORT}"
      SSL_VERIFY: "False"
    restart: unless-stopped
    entrypoint: [uvicorn, app:app, --host, 0.0.0.0, --port, "${CSVTOCSVW_APP_PORT}", --workers, "1", --proxy-headers]
    healthcheck:
      test: [CMD, curl, -f, "http://localhost:${CSVTOCSVW_APP_PORT}/info"]
      interval: 60s
      timeout: 10s
      retries: 5
    networks:
      - datastack_net

  maptomethod:
    # TODO: pin to a specific release tag — https://github.com/Mat-O-Lab/MapToMethod/releases
    image: ghcr.io/mat-o-lab/maptomethod:latest
    environment:
      APP_NAME: MapToMethod
      PORT: "${MAPTOMETHOD_APP_PORT}"
      SSL_VERIFY: "${CKANINI__CKANEXT__CSVWMAPANDTRANSFORM__SSL_VERIFY}"
      SERVER_URL: "${CKAN_SITE_URL}/maptomethod"
    restart: unless-stopped
    entrypoint: [uvicorn, app:app, --host, 0.0.0.0, --port, "${MAPTOMETHOD_APP_PORT}", --workers, "1", --proxy-headers, --root-path, /maptomethod]
    healthcheck:
      test: [CMD, curl, -f, "http://localhost:${MAPTOMETHOD_APP_PORT}/info"]
      interval: 60s
      timeout: 10s
      retries: 5
    networks:
      - datastack_net

  yarrrml-parser:
    # TODO: pin to a specific release tag — https://github.com/Mat-O-Lab/yarrrml-parser/releases
    image: ghcr.io/mat-o-lab/yarrrml-parser:latest
    environment:
      - "PORT=${PARSER_PORT}"
    restart: unless-stopped
    healthcheck:
      test: [CMD-SHELL, "nc -z localhost ${PARSER_PORT}"]
      interval: 30s
      timeout: 5s
      retries: 3
    networks:
      - datastack_net

  rmlmapper:
    # Note: changing PORT does not work due to an upstream limitation; port is effectively fixed at 4000.
    # TODO: pin to a specific release tag — https://github.com/Mat-O-Lab/RMLMapper/releases
    image: ghcr.io/mat-o-lab/rmlmapper-webapi:latest
    environment:
      - "PORT=${MAPPER_PORT}"
    restart: unless-stopped
    healthcheck:
      test: [CMD-SHELL, wget -qO /dev/null http://localhost:4000/ || exit 1]
      interval: 30s
      timeout: 5s
      retries: 3
    networks:
      - datastack_net

  rdfconverter:
    # TODO: pin to a specific release tag — https://github.com/Mat-O-Lab/RDFConverter/releases
    image: ghcr.io/mat-o-lab/rdfconverter:latest
    depends_on:
      yarrrml-parser:
        condition: service_healthy
      rmlmapper:
        condition: service_healthy
    environment:
      PARSER_PORT: "${PARSER_PORT}"
      MAPPER_PORT: "${MAPPER_PORT}"
      PORT: "${RDFCONVERTER_APP_PORT}"
      YARRRML_URL: "http://yarrrml-parser:${PARSER_PORT}"
      MAPPER_URL: "http://rmlmapper:${MAPPER_PORT}"
      APP_NAME: RDFConverter
      APP_MODE: production
      SSL_VERIFY: "False"
    restart: unless-stopped
    entrypoint: [uvicorn, app:app, --host, 0.0.0.0, --port, "${RDFCONVERTER_APP_PORT}", --workers, "1", --proxy-headers]
    healthcheck:
      test: [CMD, curl, -f, "http://localhost:${RDFCONVERTER_APP_PORT}/info"]
      interval: 60s
      timeout: 10s
      retries: 5
    networks:
      - datastack_net

  sparklis:
    # TODO: pin to a specific release tag — https://github.com/sebferre/sparklis/releases
    image: sferre/sparklis:latest
    restart: unless-stopped
    healthcheck:
      test: [CMD-SHELL, wget -qO /dev/null http://localhost:8080/ || exit 1]
      interval: 60s
      timeout: 10s
      retries: 5
    networks:
      - datastack_net

networks:
  datastack_net: