docker-compose.yml

services:
  # ==========================================
  # Application Services
  # ==========================================

  # FastAPI Backend
  research_api:
    build:
      context: .
      dockerfile: docker/Dockerfile.api
    container_name: research_api
    ports:
      - "8000:8000"
    environment:
      # Application
      APP_NAME: praval_deep_research
      LOG_LEVEL: ${LOG_LEVEL:-INFO}
      ENVIRONMENT: ${ENVIRONMENT:-development}
      API_CORS_ORIGINS: ${API_CORS_ORIGINS:-http://localhost:3000,http://127.0.0.1:3000,http://localhost:3001,http://127.0.0.1:3001}

      # API Keys
      OPENAI_API_KEY: ${OPENAI_API_KEY}

      # Praval
      PRAVAL_DEFAULT_PROVIDER: ${PRAVAL_DEFAULT_PROVIDER:-openai}
      PRAVAL_DEFAULT_MODEL: ${PRAVAL_DEFAULT_MODEL:-gpt-4o-mini}

      # Infrastructure (using service names)
      RABBITMQ_URL: amqp://research_user:research_pass@rabbitmq:5672/research_vhost
      QDRANT_URL: http://qdrant:6333
      QDRANT_COLLECTION_NAME: ${QDRANT_COLLECTION_NAME:-research_vectors}
      MINIO_ENDPOINT: minio:9000
      MINIO_EXTERNAL_ENDPOINT: localhost:9000
      MINIO_ACCESS_KEY: ${MINIO_ACCESS_KEY:-minioadmin}
      MINIO_SECRET_KEY: ${MINIO_SECRET_KEY:-minioadmin}
      MINIO_BUCKET_NAME: ${MINIO_BUCKET_NAME:-research-papers}
      MINIO_SECURE: "false"
      REDIS_URL: redis://redis:6379
      DATABASE_URL: postgresql+asyncpg://research_user:${POSTGRES_PASSWORD:-research_pass}@postgres:5432/praval_research

      # Embedding Configuration
      OPENAI_EMBEDDING_MODEL: ${OPENAI_EMBEDDING_MODEL:-text-embedding-3-small}
      EMBEDDING_DIMENSIONS: ${EMBEDDING_DIMENSIONS:-1536}
      CHUNK_SIZE: ${CHUNK_SIZE:-1000}
      CHUNK_OVERLAP: ${CHUNK_OVERLAP:-200}
      MAX_CHUNKS_PER_PAPER: ${MAX_CHUNKS_PER_PAPER:-50}

      # BM25 Search Configuration (Vajra)
      BM25_INDEX_PATH: /app/data/vajra_indexes
      BM25_K1: ${BM25_K1:-1.5}
      BM25_B: ${BM25_B:-0.75}
      BM25_USE_EAGER: ${BM25_USE_EAGER:-true}

      # LangExtract Configuration
      LANGEXTRACT_ENABLED: ${LANGEXTRACT_ENABLED:-true}
      LANGEXTRACT_PROVIDER: ${LANGEXTRACT_PROVIDER:-gemini}
      LANGEXTRACT_MODEL: ${LANGEXTRACT_MODEL:-gemini-2.5-flash}
      GEMINI_API_KEY: ${GEMINI_API_KEY}
      OLLAMA_BASE_URL: ${OLLAMA_BASE_URL:-http://host.docker.internal:11434}
      OLLAMA_MODEL: ${OLLAMA_MODEL:-gemma2:9b}

      # Knowledge Graph Configuration
      KG_ENABLED: ${KG_ENABLED:-true}
      KG_BUILDER_URL: http://kg_builder:8080
      NEO4J_URI: bolt://neo4j:7687
      NEO4J_USER: neo4j
      NEO4J_PASSWORD: ${NEO4J_PASSWORD:-research_graph}
    volumes:
      - ./data/vajra_indexes:/app/data/vajra_indexes  # Local bind mount for BM25 indexes
    depends_on:
      rabbitmq:
        condition: service_healthy
      qdrant:
        condition: service_started  # Changed from healthy - Qdrant container is minimal
      minio:
        condition: service_healthy
      redis:
        condition: service_healthy
      postgres:
        condition: service_healthy
    networks:
      - research_network
    restart: unless-stopped
    healthcheck:
      test: curl -f http://localhost:8000/health || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 90s  # Increased to allow index initialization

  # Praval Agent Workers
  research_agents:
    build:
      context: .
      dockerfile: docker/Dockerfile.agents
    container_name: research_agents
    environment:
      # Application
      APP_NAME: praval_deep_research_agents
      LOG_LEVEL: ${LOG_LEVEL:-INFO}
      ENVIRONMENT: ${ENVIRONMENT:-development}

      # API Keys
      OPENAI_API_KEY: ${OPENAI_API_KEY}

      # Praval
      PRAVAL_DEFAULT_PROVIDER: ${PRAVAL_DEFAULT_PROVIDER:-openai}
      PRAVAL_DEFAULT_MODEL: ${PRAVAL_DEFAULT_MODEL:-gpt-4o-mini}

      # Infrastructure (using service names)
      RABBITMQ_URL: amqp://research_user:research_pass@rabbitmq:5672/research_vhost
      QDRANT_URL: http://qdrant:6333
      QDRANT_COLLECTION_NAME: ${QDRANT_COLLECTION_NAME:-research_vectors}
      MINIO_ENDPOINT: minio:9000
      MINIO_EXTERNAL_ENDPOINT: localhost:9000
      MINIO_ACCESS_KEY: ${MINIO_ACCESS_KEY:-minioadmin}
      MINIO_SECRET_KEY: ${MINIO_SECRET_KEY:-minioadmin}
      MINIO_BUCKET_NAME: ${MINIO_BUCKET_NAME:-research-papers}
      MINIO_SECURE: "false"
      REDIS_URL: redis://redis:6379

      # Embedding Configuration
      OPENAI_EMBEDDING_MODEL: ${OPENAI_EMBEDDING_MODEL:-text-embedding-3-small}
      EMBEDDING_DIMENSIONS: ${EMBEDDING_DIMENSIONS:-1536}
      CHUNK_SIZE: ${CHUNK_SIZE:-1000}
      CHUNK_OVERLAP: ${CHUNK_OVERLAP:-200}
      MAX_CHUNKS_PER_PAPER: ${MAX_CHUNKS_PER_PAPER:-50}

      # BM25 Search Configuration (Vajra) - shared with API
      BM25_INDEX_PATH: /app/data/vajra_indexes
      BM25_K1: ${BM25_K1:-1.5}
      BM25_B: ${BM25_B:-0.75}
      BM25_USE_EAGER: ${BM25_USE_EAGER:-true}

      # LangExtract Configuration (agents need this for extraction)
      LANGEXTRACT_ENABLED: ${LANGEXTRACT_ENABLED:-true}
      LANGEXTRACT_PROVIDER: ${LANGEXTRACT_PROVIDER:-gemini}
      LANGEXTRACT_MODEL: ${LANGEXTRACT_MODEL:-gemini-2.5-flash}
      GEMINI_API_KEY: ${GEMINI_API_KEY}
      OLLAMA_BASE_URL: ${OLLAMA_BASE_URL:-http://host.docker.internal:11434}
      OLLAMA_MODEL: ${OLLAMA_MODEL:-gemma2:9b}

      # Knowledge Graph Configuration
      KG_ENABLED: ${KG_ENABLED:-true}
      KG_BUILDER_URL: http://kg_builder:8080
      NEO4J_URI: bolt://neo4j:7687
      NEO4J_USER: neo4j
      NEO4J_PASSWORD: ${NEO4J_PASSWORD:-research_graph}
    volumes:
      - ./data/vajra_indexes:/app/data/vajra_indexes  # Local bind mount for BM25 indexes
    depends_on:
      rabbitmq:
        condition: service_healthy
      qdrant:
        condition: service_started  # Changed from healthy - Qdrant container is minimal
      minio:
        condition: service_healthy
      redis:
        condition: service_healthy
      postgres:
        condition: service_healthy
      research_api:
        condition: service_healthy
    networks:
      - research_network
    restart: unless-stopped
    deploy:
      replicas: 1
      resources:
        limits:
          cpus: '1'
          memory: 2G
        reservations:
          cpus: '0.5'
          memory: 1G

  # React + Vite Frontend (frontend-new)
  research_frontend:
    build:
      context: .
      dockerfile: docker/Dockerfile.frontend-new
    container_name: research_frontend
    ports:
      - "3000:80"
    environment:
      # Vite env vars (build-time)
      VITE_API_URL: http://localhost:8000
      # Neo4j connection for neovis.js Knowledge Graph visualization
      VITE_NEO4J_URL: bolt://localhost:7687
      VITE_NEO4J_USER: neo4j
      VITE_NEO4J_PASSWORD: ${NEO4J_PASSWORD:-research_graph}
    depends_on:
      research_api:
        condition: service_healthy
    networks:
      - research_network
    restart: unless-stopped
    healthcheck:
      test: wget --quiet --tries=1 --spider http://localhost/ || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s

  # ==========================================
  # Infrastructure Services
  # ==========================================

  # Message Broker
  rabbitmq:
    image: rabbitmq:3-management
    container_name: research_rabbitmq
    ports:
      - "5672:5672"   # AMQP
      - "15672:15672" # Management UI
    environment:
      RABBITMQ_DEFAULT_USER: research_user
      RABBITMQ_DEFAULT_PASS: research_pass
      RABBITMQ_DEFAULT_VHOST: research_vhost
    volumes:
      - rabbitmq_data:/var/lib/rabbitmq
    healthcheck:
      test: rabbitmq-diagnostics -q ping
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    networks:
      - research_network
    restart: unless-stopped

  # Vector Database
  qdrant:
    image: qdrant/qdrant:v1.7.3
    container_name: research_qdrant
    ports:
      - "6333:6333"
      - "6334:6334"
    volumes:
      - ./data/qdrant:/qdrant/storage  # Local bind mount for vector embeddings
      - ./docker/qdrant_config.yaml:/qdrant/config/production.yaml:ro
    environment:
      QDRANT__SERVICE__HTTP_PORT: 6333
      QDRANT__SERVICE__GRPC_PORT: 6334
      QDRANT__LOG_LEVEL: INFO
    # Health check disabled - container is minimal, use service_started condition instead
    # Qdrant is ready when HTTP responds on port 6333 (verified manually)
    healthcheck:
      disable: true
    networks:
      - research_network
    restart: unless-stopped

  # Object Storage
  minio:
    image: minio/minio:latest
    container_name: research_minio
    ports:
      - "9000:9000"
      - "9001:9001"
    environment:
      MINIO_ROOT_USER: ${MINIO_ACCESS_KEY:-minioadmin}
      MINIO_ROOT_PASSWORD: ${MINIO_SECRET_KEY:-minioadmin}
      MINIO_BROWSER_REDIRECT_URL: http://localhost:9001
    command: server /data --console-address ":9001"
    volumes:
      - ./data/minio:/data  # Local bind mount - papers browsable at ./data/minio/research-papers/
    healthcheck:
      test: curl -f http://localhost:9000/minio/health/live || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    networks:
      - research_network
    restart: unless-stopped

  # Cache & Session Store
  redis:
    image: redis:7-alpine
    container_name: research_redis
    ports:
      - "6379:6379"
    command: >
      redis-server 
      --appendonly yes 
      --maxmemory 1gb 
      --maxmemory-policy allkeys-lru
    volumes:
      - redis_data:/data
    healthcheck:
      test: redis-cli ping || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s
    networks:
      - research_network
    restart: unless-stopped

  # ==========================================
  # Knowledge Graph Services
  # ==========================================

  # Neo4j Graph Database
  neo4j:
    image: neo4j:5-community
    container_name: research_neo4j
    ports:
      - "7474:7474"  # Browser UI
      - "7687:7687"  # Bolt protocol
    environment:
      NEO4J_AUTH: neo4j/${NEO4J_PASSWORD:-research_graph}
      NEO4J_PLUGINS: '["apoc"]'
      NEO4J_dbms_memory_heap_initial__size: 512m
      NEO4J_dbms_memory_heap_max__size: 1G
    volumes:
      - ./data/neo4j/data:/data  # Local bind mount for graph data
      - ./data/neo4j/logs:/logs  # Local bind mount for logs
    healthcheck:
      test: wget --quiet --tries=1 --spider http://localhost:7474 || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 60s
    networks:
      - research_network
    restart: unless-stopped

  # Kay-Gee-Go Knowledge Graph Builder
  kg_builder:
    build:
      context: ./kay-gee-go
      dockerfile: Dockerfile
    container_name: research_kg_builder
    ports:
      - "8081:8080"
    environment:
      NEO4J_URI: bolt://neo4j:7687
      NEO4J_USER: neo4j
      NEO4J_PASSWORD: ${NEO4J_PASSWORD:-research_graph}
      LOG_LEVEL: ${LOG_LEVEL:-INFO}
    depends_on:
      neo4j:
        condition: service_healthy
    networks:
      - research_network
    restart: unless-stopped
    healthcheck:
      test: wget --quiet --tries=1 --spider http://localhost:8080/health || exit 1
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 30s
    profiles:
      - knowledge-graph

  # Development Tools
  adminer:
    image: adminer:latest
    container_name: research_adminer
    ports:
      - "8080:8080"
    environment:
      ADMINER_DEFAULT_SERVER: research_postgres
    networks:
      - research_network
    restart: unless-stopped
    profiles:
      - dev-tools

  # PostgreSQL for chat history storage
  postgres:
    image: postgres:16-alpine
    container_name: research_postgres
    ports:
      - "5432:5432"
    environment:
      POSTGRES_DB: praval_research
      POSTGRES_USER: research_user
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-research_pass}
    volumes:
      - ./data/postgres:/var/lib/postgresql/data  # Local bind mount for database
    healthcheck:
      test: pg_isready -U research_user -d praval_research
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    networks:
      - research_network
    restart: unless-stopped

  # Monitoring
  prometheus:
    image: prom/prometheus:latest
    container_name: research_prometheus
    ports:
      - "9090:9090"
    volumes:
      - ./docker/prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - prometheus_data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
      - '--web.console.libraries=/usr/share/prometheus/console_libraries'
      - '--web.console.templates=/usr/share/prometheus/consoles'
      - '--web.enable-lifecycle'
      - '--web.external-url=http://localhost:9090'
    networks:
      - research_network
    restart: unless-stopped
    profiles:
      - monitoring

volumes:
  # Only keeping volumes for services that don't need local filesystem access
  rabbitmq_data:
    driver: local
  redis_data:
    driver: local
  prometheus_data:
    driver: local
  # The following are now bind-mounted to ./data/ for local access:
  # - minio -> ./data/minio (research papers)
  # - qdrant -> ./data/qdrant (vector embeddings)
  # - vajra -> ./data/vajra_indexes (BM25 search indexes)
  # - postgres -> ./data/postgres (chat history, metadata)
  # - neo4j -> ./data/neo4j (knowledge graph)

networks:
  research_network:
    driver: bridge