meep-testing-generation

name	meep-testing-generation
description	Standards, workflows, and helper scripts for generating unit tests and mock data to test the Multi-Explore Experience Pattern (MEEP) in multiExploreUtils.ts.

Instructions

1. Overview

This skill tracks the iterative generation and maintenance of test cases and mock data for the Multi-Explore Experience Pattern (MEEP).

As we iterate on MEEP functionality (such as field layering, group relabeling, timeline linking, and exclusions), we follow a recurring workflow to update our mock Explores and write automated Jest unit tests.

Core Testing Files

1. LookML Model: meep_test.model.lkml - Contains the raw LookML Explore definitions used for testing.
2. Implementation: multiExploreUtils.ts - The core frontend logic that processes Explores, dimensions, and measures.
3. Test Suite: multiExploreUtils.test.ts - The Jest unit test suite covering MEEP features.
4. Mock Data: mockExplores.ts - The generated TypeScript array of ILookmlModelExplore objects.

2. Workflow: Regenerating Mock Explores

Whenever changes are made to meep_test.model.lkml or underlying LookML views, we must regenerate the mock Explores in mockExplores.ts.

Script Reference

We maintain MCP executable Python scripts in the ./scripts folder. Each script contains exactly one code block designed to be run via the @mcp:lkr_dev_cli_codemode:run_python_code tool.

• Script: fetch_meep_test_explores.md - Refreshes mock explores.
• Script: test_json_bi.md - Runs sample Looker queries with the json_bi output format to inspect row and pivot metadata.

Execution Steps

To refresh the mock Explores:

1. Open fetch_meep_test_explores.md and extract the Python code block.
2. Run the code via call_mcp_tool using the lkr_dev_cli_codemode server and run_python_code tool.
3. Copy the returned TypeScript code directly into mockExplores.ts.

To test query execution and check json_bi formats:

1. Open test_json_bi.md and extract the Python code block.
2. Run the code via call_mcp_tool with the lkr_dev_cli_codemode server and run_python_code tool.

3. Test Case Strategy

We will build unit tests in multiExploreUtils.test.ts and meepQueryBuilder.test.ts. Specifically, for client-side queries merging (full outer joining on timelines/dimensions/pivots), we mock standard Looker json_bi response objects and assert that mergeMeepResults correctly joins them on overlapping keys.