Crawl4AI+SearXNG MCP Server

""" Integration tests for Crawl4AI MCP. This package contains integration tests that use real services running in Docker containers to validate the complete system behavior and performance characteristics. Test Structure: - conftest.py: Integration test fixtures and configuration - test_e2e_workflows.py: End-to-end workflow tests for the complete RAG pipeline - test_database_integration.py: Database integration tests with real Qdrant/Supabase - test_performance_benchmarks.py: Performance and scalability benchmarks Running Integration Tests: ```bash # Start Docker services first make dev # Run all integration tests pytest tests/integration/ -m integration # Run specific test categories pytest tests/integration/ -m e2e pytest tests/integration/ -m performance # Run with verbose output pytest tests/integration/ -v -s # Skip slow tests pytest tests/integration/ -m "integration and not slow" ``` Requirements: - Docker and Docker Compose must be available - Services (Qdrant, SearXNG) must be running via `make dev` - Sufficient system resources for concurrent operations - Network access for downloading dependencies Performance Targets: - E2E workflow: < 20 seconds - Single crawl: < 5 seconds - Batch crawl (20 URLs): < 15 seconds - Search latency: < 2 seconds - Storage latency: < 1 second - Throughput: > 2 URLs/sec crawling, > 5 searches/sec """