Add comprehensive SolaraViz testing framework

[Tejasv-Singh]

Comprehensive Testing Framework for Mesa's SolaraViz Components

Overview

This pull request introduces a comprehensive testing framework for Mesa's SolaraViz visualization components, integrating seamlessly into the CI pipeline. The framework addresses GitHub issue #2734 by providing robust testing solutions that extend beyond basic model tests, ensuring SolaraViz components function correctly across all example models.

---

Key Features Implemented

Testing Categories

1. Basic Component Tests

• Validates initialization and rendering of all SolaraViz components
• Tests component properties and attributes
• Ensures components handle edge cases gracefully

2. Integration Tests

• Verifies visualization components correctly interact with model data
• Tests that model changes are properly reflected in visualizations
• Ensures app components can initialize and update models

3. Performance Benchmarks

• Measures rendering performance across components
• Benchmarks model step performance with visualizations attached
• Provides comparison data for performance optimization

4. Regression Tests

• Prevents regressions in visualization functionality
• Verifies component interfaces remain stable
• Tests error handling capabilities

---

Framework Architecture

• Mock Implementation: Created mock Solara components that enable testing without actual Solara dependencies
• Parametrized Tests: Tests run against all example models automatically
• Flexible Fixtures: Shared test fixtures simplify writing new tests
• Web Interface: User-friendly interface for running and monitoring tests
• Command-line Tools: Support for running tests via CLI

---

Technical Implementation

• All tests follow pytest's best practices with proper fixture usage
• Testing framework is isolated and independent, minimizing dependencies
• Comprehensive docstrings explain test purpose and methodology
• Graceful handling of edge cases and failures
• CI integration through GitHub Actions workflow

---

Usage

The framework supports multiple testing methods:

• Web interface for visual test monitoring
• Command-line interface for scripted testing
• Direct pytest execution for development
• Automated execution via GitHub Actions

---

Future Improvements

• Expand visual regression testing with screenshot comparisons
• Add more performance metrics for detailed benchmarking
• Improve test coverage for advanced Solara features
• Enhance documentation with usage examples

---

Related Issues

• Fixes Integrate Visualization Testing in Mesa CI #2734: Add comprehensive tests for SolaraViz components

---

Reviewers

Please focus on:

• Test coverage completeness
• Framework architecture and extensibility
• Integration with Mesa's existing testing infrastructure
• Documentation clarity

[github-actions]

Performance benchmarks:

Model	Size	Init time [95% CI]	Run time [95% CI]
BoltzmannWealth	small	🔵 -0.9% [-2.1%, +0.3%]	🔵 -0.4% [-1.2%, +0.5%]
BoltzmannWealth	large	🔵 +0.7% [-0.7%, +2.0%]	🔵 +0.6% [-0.7%, +1.9%]
Schelling	small	🔵 +0.8% [+0.0%, +1.6%]	🔵 +1.6% [+0.8%, +2.4%]
Schelling	large	🔵 +0.6% [-1.7%, +2.6%]	🔵 +0.5% [-0.6%, +1.7%]
WolfSheep	small	🔵 +0.6% [-0.3%, +1.6%]	🔵 +1.0% [+0.3%, +1.8%]
WolfSheep	large	🔵 +2.3% [+0.3%, +4.4%]	🔵 +2.8% [+1.5%, +4.1%]
BoidFlockers	small	🔵 -2.4% [-4.5%, -0.3%]	🔵 -0.5% [-1.2%, +0.1%]
BoidFlockers	large	🔵 -5.7% [-7.8%, -2.8%]	🟢 -4.8% [-5.8%, -3.8%]

[EwoutH]

Thanks for the PR!

I don't know if you noticed, but 3 days ago @Ya-shh opened a similar PR:

• Added comprehensive testing for visualization components #2737

Have you coordinated with him?

[EwoutH]

Anyway, if you want you can continue working on this one, and we'll merge the one that's best.

Two comments:

1. All other tests are in our tests directory. Is there a reason these are not?
2. I would like all new tests to be added to the CI. In this case, I would recommend adding a new workflow file (in .github/workflows).
3. 1600 lines of tests is a lot. Not necessarily a problem, but could you explain where the bulk is and why that's needed?

[Tejasv-Singh]

Thanks for your review!

Test Location: All test files are in the /tests directory. The supporting files outside (mock_solara_components.py, example_tests.py, etc.) are not tests but necessary mocks, utilities, and web components. If preferred, I can relocate them for consistency.

CI Integration: I've added a GitHub Actions workflow (.github/workflows/viz_tests.yml) to run all SolaraViz tests, generate benchmarks, store artifacts, and report failures.

Test Length (1600 lines): The bulk comes from:

Mocks (300 lines): Needed to run tests without Solara dependencies.

Visualization Coverage (500 lines): Ensures all Grid, Chart, and Network scenarios are tested.

Test Categories (800 lines): Component, integration, performance, and regression tests for full coverage.