Igloo MCP

"""Shared fixtures for system tests.""" import uuid from pathlib import Path from typing import Any import pytest from igloo_mcp.cache.query_result_cache import QueryResultCache from igloo_mcp.config import get_config from igloo_mcp.living_reports.service import ReportService from igloo_mcp.logging.query_history import QueryHistory from igloo_mcp.mcp.tools.create_report import CreateReportTool from igloo_mcp.mcp.tools.evolve_report import EvolveReportTool from igloo_mcp.mcp.tools.execute_query import ExecuteQueryTool from igloo_mcp.mcp.tools.render_report import RenderReportTool from tests.helpers.fake_snowflake_connector import FakeQueryPlan, FakeSnowflakeService @pytest.fixture def system_test_env(tmp_path: Path, monkeypatch): """Set up complete environment for system tests. Creates temporary directories for: - Reports storage - Query history - Query cache - Artifacts Returns paths dict for test access. """ reports_root = tmp_path / "reports" history_file = tmp_path / "logs" / "doc.jsonl" cache_root = tmp_path / "cache" artifact_root = tmp_path / "artifacts" # Create necessary directories reports_root.mkdir(parents=True) history_file.parent.mkdir(parents=True) cache_root.mkdir(parents=True) artifact_root.mkdir(parents=True) # Set environment variables env_overrides = { "IGLOO_MCP_QUERY_HISTORY": str(history_file), "IGLOO_MCP_CACHE_ROOT": str(cache_root), "IGLOO_MCP_ARTIFACT_ROOT": str(artifact_root), } for key, value in env_overrides.items(): monkeypatch.setenv(key, value) return { "reports_root": reports_root, "history_file": history_file, "cache_root": cache_root, "artifact_root": artifact_root, "tmp_path": tmp_path, } @pytest.fixture def full_service_stack(system_test_env: dict[str, Any]): """Complete service stack for system tests. Provides all services needed for end-to-end workflows: - ReportService: Living reports management - FakeSnowflakeService: Query execution (mocked) - QueryHistory: Query logging - QueryResultCache: Query caching - Config: System configuration Returns dict with all services and tools. """ config = get_config() # Create services report_service = ReportService(reports_root=system_test_env["reports_root"]) # Create Snowflake service with a dummy initial plan # Tests will add their own plans using add_query_plan method dummy_plan = FakeQueryPlan( statement="SELECT 1", rows=[{"col1": 1}], rowcount=1, duration=0.01, sfqid="dummy_query_id", ) snowflake_service = FakeSnowflakeService([dummy_plan]) # Create query history history = QueryHistory(system_test_env["history_file"]) # Create query cache cache = QueryResultCache.from_env(artifact_root=system_test_env["artifact_root"]) # Create MCP tools create_report_tool = CreateReportTool(config, report_service) evolve_report_tool = EvolveReportTool(config, report_service) render_report_tool = RenderReportTool(config, report_service) execute_query_tool = ExecuteQueryTool( config=config, snowflake_service=snowflake_service, query_service=None, # Not needed for system tests health_monitor=None, # Not needed for system tests ) return { "config": config, "report_service": report_service, "snowflake_service": snowflake_service, "history": history, "cache": cache, "tools": { "create_report": create_report_tool, "evolve_report": evolve_report_tool, "render_report": render_report_tool, "execute_query": execute_query_tool, }, "env": system_test_env, } @pytest.fixture def realistic_query_results(): """Generate realistic query results for testing. Returns list of row dicts suitable for FakeQueryPlan. """ # Q4 2024 revenue data return [ { "month": "2024-10", "total_revenue": 1250000.00, "unique_customers": 450, "avg_order_value": 2777.78, "yoy_growth_pct": 23.5, }, { "month": "2024-11", "total_revenue": 1380000.00, "unique_customers": 485, "avg_order_value": 2845.36, "yoy_growth_pct": 25.2, }, { "month": "2024-12", "total_revenue": 1620000.00, "unique_customers": 520, "avg_order_value": 3115.38, "yoy_growth_pct": 28.1, }, ] @pytest.fixture def realistic_cost_results(): """Generate realistic cost query results.""" return [ { "category": "Engineering", "q4_2024_cost": 850000.00, "q4_2023_cost": 720000.00, "variance_pct": 18.1, }, { "category": "Marketing", "q4_2024_cost": 320000.00, "q4_2023_cost": 280000.00, "variance_pct": 14.3, }, { "category": "Operations", "q4_2024_cost": 450000.00, "q4_2023_cost": 410000.00, "variance_pct": 9.8, }, ] @pytest.fixture def large_report_data(): """Generate data for large report testing (50 sections, 200 insights). Returns dict with sections and insights structure. """ sections = [] insights = [] # Create 50 sections for i in range(50): section_id = str(uuid.uuid4()) sections.append( { "section_id": section_id, "title": f"Section {i + 1}: Analysis Topic {i + 1}", "order": i, "insight_ids": [], # Will be populated below } ) # Create 200 insights, distributed across sections for i in range(200): insight_id = str(uuid.uuid4()) insights.append( { "insight_id": insight_id, "importance": (i % 10) + 1, # 1-10 "summary": f"Insight {i + 1}: Key finding about topic {i + 1}", "supporting_queries": [], "status": "active", } ) # Assign insight to a section (distribute evenly) section_idx = i % 50 sections[section_idx]["insight_ids"].append(insight_id) return { "sections": sections, "insights": insights, }