docker-compose.yml

# =============================================================================
# M-flow — Docker Compose Stack
# Usage:
#   docker compose up                      # backend only
#   docker compose --profile ui up         # backend + frontend
#   docker compose --profile neo4j up      # backend + Neo4j
#   docker compose --profile postgres up   # backend + Postgres
#   docker compose --profile mcp up        # backend + MCP server
#   docker compose --profile playground up # backend + face recognition
# =============================================================================

x-common-env: &common-env
  HOST: "0.0.0.0"
  ENVIRONMENT: local
  LOG_LEVEL: INFO

services:
  # ---------------------------------------------------------------------------
  # Core: M-flow Backend API
  # ---------------------------------------------------------------------------
  mflow-api:
    container_name: m_flow
    build:
      context: .
      dockerfile: Dockerfile
    volumes:
      - .env:/opt/m_flow/.env
    environment:
      <<: *common-env
      DEBUG: "false"  # Set "true" to enable debugpy
    extra_hosts:
      - "host.docker.internal:host-gateway"
    ports:
      - "8000:8000"
      - "9230:9230"   # debugpy
    networks:
      - m_flow-network
    deploy:
      resources:
        limits:
          cpus: "4.0"
          memory: 8GB
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
      interval: 30s
      timeout: 10s
      retries: 3

  # ---------------------------------------------------------------------------
  # MCP Server — Model Context Protocol for IDE integration
  # ---------------------------------------------------------------------------
  mflow-mcp:
    container_name: m_flow-mcp
    profiles: [mcp]
    build:
      context: .
      dockerfile: m_flow-mcp/Dockerfile
    volumes:
      - .env:/opt/m_flow/.env
    environment:
      <<: *common-env
      DEBUG: "false"
      TRANSPORT_MODE: sse
      # Database — should mirror the main backend service
      DB_PROVIDER: "${DB_PROVIDER:-sqlite}"
      DB_HOST: "${DB_HOST:-host.docker.internal}"
      DB_PORT: "${DB_PORT:-5432}"
      DB_NAME: "${DB_NAME:-mflow_store}"
      DB_USERNAME: "${DB_USERNAME:-m_flow}"
      DB_PASSWORD: "${DB_PASSWORD:-m_flow}"
      MCP_LOG_LEVEL: INFO
      PYTHONUNBUFFERED: "1"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    ports:
      - "8001:8000"   # MCP port (offset to avoid clash with backend)
      - "9231:9230"   # MCP debugger (offset from backend's 9230)
    networks:
      - m_flow-network
    deploy:
      resources:
        limits:
          cpus: "2.0"
          memory: 6GB

  # ---------------------------------------------------------------------------
  # Frontend — Next.js Console UI  (profile: ui)
  # ---------------------------------------------------------------------------
  frontend:
    container_name: m_flow-frontend
    profiles: [ui]
    build:
      context: ./m_flow-frontend
      dockerfile: Dockerfile
    volumes:
      - ./m_flow-frontend/src:/frontend/src
      - ./m_flow-frontend/public:/frontend/public
    ports:
      - "3000:3000"
    networks:
      - m_flow-network
    depends_on:
      - mflow-api

  # ---------------------------------------------------------------------------
  # Infrastructure: Graph DB  (profile: neo4j)
  # ---------------------------------------------------------------------------
  neo4j:
    image: neo4j:5
    container_name: m_flow-neo4j
    profiles: [neo4j]
    environment:
      NEO4J_AUTH: neo4j/mflow-dev-neo4j
      NEO4J_PLUGINS: '["apoc", "graph-data-science"]'
    ports:
      - "7474:7474"
      - "7687:7687"
    networks:
      - m_flow-network
    volumes:
      - neo4j_data:/data

  # ---------------------------------------------------------------------------
  # Infrastructure: Vector-capable Postgres  (profile: postgres)
  # ---------------------------------------------------------------------------
  postgres:
    image: pgvector/pgvector:pg17
    container_name: m_flow-postgres
    profiles: [postgres]
    environment:
      POSTGRES_USER: m_flow
      POSTGRES_PASSWORD: m_flow
      POSTGRES_DB: mflow_store
    ports:
      - "5432:5432"
    networks:
      - m_flow-network
    volumes:
      - postgres_data:/var/lib/postgresql/data

  # ---------------------------------------------------------------------------
  # Infrastructure: ChromaDB  (profile: chromadb)
  # ---------------------------------------------------------------------------
  chromadb:
    image: chromadb/chroma:0.6.3
    container_name: m_flow-chromadb
    profiles: [chromadb]
    environment:
      IS_PERSISTENT: "TRUE"
      CHROMA_SERVER_AUTH_CREDENTIALS: "${VECTOR_DB_KEY}"
      CHROMA_SERVER_AUTH_CREDENTIALS_PROVIDER: chromadb.auth.token.TokenConfigServerAuthCredentialsProvider
      CHROMA_SERVER_AUTH_TOKEN_TRANSPORT_HEADER: AUTHORIZATION
      CHROMA_SERVER_AUTH_PROVIDER: chromadb.auth.token.TokenAuthServerProvider
    volumes:
      - chromadb_data:/chroma/chroma/
    networks:
      - m_flow-network
    ports:
      - "3002:8000"

  # ---------------------------------------------------------------------------
  # Infrastructure: Redis  (profile: redis)
  # ---------------------------------------------------------------------------
  redis:
    image: redis:7-alpine
    container_name: m_flow-redis
    profiles: [redis]
    command: ["redis-server", "--appendonly", "yes"]
    ports:
      - "6379:6379"
    networks:
      - m_flow-network
    volumes:
      - redis_data:/data

  redisinsight:
    image: redislabs/redisinsight:latest
    container_name: m_flow-redisinsight
    profiles: [redis]
    restart: unless-stopped
    ports:
      - "5540:5540"
    networks:
      - m_flow-network

  # ---------------------------------------------------------------------------
  # Playground: Face Recognition  (profile: playground)
  # ---------------------------------------------------------------------------
  # Requires: git clone https://github.com/FlowElement-ai/fanjing-face-recognition.git
  #           in the parent directory (../fanjing-face-recognition).
  # Models: download via scripts/download_model.py etc., or copy into ./models/
  # Shared secret: set FACE_API_KEY in .env (same key for both services).
  fanjing-face:
    container_name: fanjing-face-recognition
    profiles: [playground]
    build:
      context: ../fanjing-face-recognition
      dockerfile: Dockerfile
    volumes:
      - ../fanjing-face-recognition/models:/app/models
      - ../fanjing-face-recognition/data:/app/data
    environment:
      HOST: "0.0.0.0"
      PORT: "5001"
      FACE_API_KEY: "${FACE_API_KEY:-}"
      PYTHONUNBUFFERED: "1"
    ports:
      - "5001:5001"
    networks:
      - m_flow-network
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:5001/"]
      interval: 30s
      timeout: 10s
      retries: 3

# =============================================================================
networks:
  m_flow-network:
    name: m_flow-network

volumes:
  neo4j_data:
  postgres_data:
  chromadb_data:
  redis_data: