R Econometrics MCP Server

#!/usr/bin/env python3 """ Unit tests for formula builder tools. Tests natural language to R formula conversion and validation. """ import sys from pathlib import Path from shutil import which import pytest pytestmark = pytest.mark.skipif( which("R") is None, reason="R binary is required for formula builder tests" ) from rmcp.core.context import Context, LifespanState # noqa: E402 from rmcp.tools.formula_builder import build_formula, validate_formula # noqa: E402 async def create_test_context(): """Create a test context for tool execution.""" lifespan = LifespanState() context = Context.create("test", "test", lifespan) return context class TestFormulaBuilder: """Test natural language formula building.""" @pytest.mark.asyncio async def test_simple_formula_building(self): """Test building a simple regression formula from description.""" context = await create_test_context() result = await build_formula( context, {"description": "predict sales from marketing spend"} ) # build_formula returns the result directly assert "formula" in result formula = result["formula"] # Should generate something like "sales ~ marketing" or similar assert "~" in formula # R formula syntax @pytest.mark.asyncio async def test_multiple_predictor_formula(self): """Test building formula with multiple predictors.""" context = await create_test_context() result = await build_formula( context, { "description": "predict price based on size, location, and age", "target_hint": "price", }, ) # build_formula returns the result directly assert "formula" in result formula = result["formula"] assert "price" in formula assert "~" in formula @pytest.mark.asyncio async def test_validate_valid_formula(self): """Test validating a correct R formula using actual R execution.""" context = await create_test_context() # validate_formula requires both formula and data test_data = { "x1": [1, 2, 3], "x2": [4, 5, 6], "x3": [7, 8, 9], "y": [10, 11, 12], } result = await validate_formula( context, {"formula": "y ~ x1 + x2 + x3", "data": test_data} ) # validate_formula returns validation data directly # Check if it's valid (might be under 'is_valid' or 'valid') assert "is_valid" in result or "valid" in result if "is_valid" in result: assert result["is_valid"] is True else: assert result["valid"] is True @pytest.mark.asyncio async def test_validate_invalid_formula(self): """Test validating an incorrect R formula.""" context = await create_test_context() # validate_formula requires both formula and data test_data = {"x": [1, 2, 3], "y": [4, 5, 6]} result = await validate_formula( context, {"formula": "this is not a formula", "data": test_data} ) # The tool should indicate invalid formula if "is_valid" in result: assert result["is_valid"] is False elif "valid" in result: assert result["valid"] is False # There should be error info assert "error" in result or "message" in result or "errors" in result @pytest.mark.asyncio async def test_formula_with_interactions(self): """Test building formula with interaction terms.""" context = await create_test_context() result = await build_formula( context, { "description": "predict outcome with interaction between treatment and time" }, ) # build_formula returns the result directly assert "formula" in result # Might generate something with : or * for interactions if __name__ == "__main__": pytest.main([__file__, "-v"])