Files
NScrapy/docker-compose.yml
zheng elendil f53d6b78dc feat(cli): add nscrapy CLI tool with new and run commands
- Add NScrapy.Cli project using Spectre.Console.Cli
- Implement 'nscrapy new' command for generating spider projects from templates
  - Support basic and distributed spider templates
  - Configurable output directory and force overwrite
- Implement 'nscrapy run' command for spider execution
  - Support single, spider, and downloader roles
  - Distributed mode with Redis configuration via CLI args
  - Configuration priority: CLI args > ENV > JSON
- Update Dockerfiles to package CLI instead of raw DLLs
- Update docker-compose.yml to use CLI commands
2026-03-29 23:34:07 +08:00

114 lines
3.7 KiB
YAML

# =============================================================================
# NScrapy Docker Compose - Three Deployment Modes
# =============================================================================
# Mode 1 - local-only:
# Redis disabled. Single spider runs locally without Redis.
# Usage: docker compose up -d spider
#
# Mode 2 - local-redis:
# Uses a local Redis container for distributed scraping.
# All services on the same Docker network.
# Usage: docker compose --profile local-redis up -d
#
# Mode 3 - managed-redis:
# Uses an external Redis (e.g., Azure Redis Cache, ElastiCache).
# Configure REDIS_HOST, REDIS_PASSWORD, REDIS_PORT, REDIS_USESSL via .env.
# Usage: docker compose up -d
# =============================================================================
services:
# ---------------------------------------------------------------------------
# Spider Service
# - Runs the NScrapy CLI spider node
# - Command format: nscrapy run <spider-name> [options]
# - Environment variables override CLI args (lowest priority)
# - Scale with: docker-compose up -d --scale spider=3
# ---------------------------------------------------------------------------
spider:
build:
context: .
dockerfile: Dockerfile.Spider
container_name: nscrapy-spider
command: run ${SPIDER_NAME:-MySpider} --role spider --distributed
environment:
# Redis configuration
- REDIS_HOST=${REDIS_HOST:-redis}
- REDIS_PORT=${REDIS_PORT:-6379}
- REDIS_PASSWORD=${REDIS_PASSWORD:-}
- REDIS_USESSL=${REDIS_USESSL:-false}
# Spider settings
- NSCRAPY_CONCURRENCY=${CONCURRENCY:-5}
- NSCRAPY_RECEIVER_QUEUE=${RECEIVER_QUEUE:-nscrapy:requests}
- NSCRAPY_RESPONSE_QUEUE=${RESPONSE_QUEUE:-nscrapy:responses}
# CLI will set these automatically when --distributed is used
- REDIS_ENABLED=true
volumes:
# Mount config file (optional, CLI args take priority)
- ./appsettings.spider.json:/app/appsettings.json:ro
depends_on:
- redis
networks:
- nscrapy-net
restart: unless-stopped
profiles:
- local-redis
# ---------------------------------------------------------------------------
# Downloader Service
# - Runs the NScrapy CLI downloader node
# - Command format: nscrapy run --role downloader
# - Scale with: docker-compose up -d --scale downloader=4
# ---------------------------------------------------------------------------
downloader:
build:
context: .
dockerfile: Dockerfile.Downloader
container_name: nscrapy-downloader
command: run --role downloader
environment:
# Redis configuration
- REDIS_HOST=${REDIS_HOST:-redis}
- REDIS_PORT=${REDIS_PORT:-6379}
- REDIS_PASSWORD=${REDIS_PASSWORD:-}
- REDIS_USESSL=${REDIS_USESSL:-false}
# CLI settings
- NSCRAPY_RECEIVER_QUEUE=${RECEIVER_QUEUE:-nscrapy:requests}
- NSCRAPY_RESPONSE_QUEUE=${RESPONSE_QUEUE:-nscrapy:responses}
- REDIS_ENABLED=true
volumes:
- ./appsettings.downloader.json:/app/appsettings.json:ro
depends_on:
- redis
networks:
- nscrapy-net
restart: unless-stopped
profiles:
- local-redis
deploy:
replicas: 1
# ---------------------------------------------------------------------------
# Local Redis (only used in local-redis mode)
# ---------------------------------------------------------------------------
redis:
image: redis:7-alpine
container_name: nscrapy-redis
ports:
- "6379:6379"
volumes:
- redis-data:/data
command: redis-server --appendonly yes
networks:
- nscrapy-net
restart: unless-stopped
profiles:
- local-redis
networks:
nscrapy-net:
driver: bridge
volumes:
redis-data: