Code Coverage Analysis

Check code coverage for the current branch and identify areas that need additional test cases.

Step 1: Run Coverage Analysis

Full Coverage Report

uv run pytest --cov=custom_components/haeo --cov-branch --cov-report=term-missing

This provides:

• Overall coverage percentage
• Branch coverage
• Line-by-line coverage with missing lines identified

HTML Coverage Report

For detailed visual analysis:

uv run pytest --cov=custom_components/haeo --cov-branch --cov-report=html

Open htmlcov/index.html in a browser to see:

• File-by-file coverage
• Line-by-line highlighting
• Branch coverage visualization

Coverage for Changed Files Only

To focus on changes in the current branch:

# List changed test files
git diff main...HEAD --name-only -z --diff-filter=AM -- 'tests/**/*.py' | xargs -0 -n1

# Run coverage for changed test files only (safe from shell injection)
git diff main...HEAD --name-only -z --diff-filter=AM -- 'tests/**/*.py' | \
    xargs -0 uv run pytest --cov=custom_components/haeo --cov-branch --cov-report=term-missing

Step 2: Analyze Coverage Results

Coverage Requirements

• CI Requirement: Coverage ≥ 95% overall
• Codecov Enforcement: Coverage must not decrease from main on changed lines
• Focus: Test behavior and edge cases, not arbitrary percentages

Identify Missing Coverage

For each file with low coverage:

1. Review untested lines: Check which lines are not covered

2. Consider simplification first: Before adding test cases, evaluate if the code can be simplified:

   • Simplify logic: If possible, refactor to remove unnecessary branches and conditional paths
   • Reduce complexity: Simpler code with fewer branches is easier to test and maintain
   • Prefer simplification over testing: Removing code that needs testing is better than adding tests for complex logic
   • Example: Instead of testing multiple conditional branches, refactor to use a single, simpler approach

3. Determine if coverage is needed:

   • Unreachable code: If lines cannot be covered by exercising input data, they may be unreachable and should be removed
   • Edge cases: Missing coverage often indicates untested edge cases
   • Error paths: Ensure error handling is tested
   • Branch coverage: Check that both true/false branches of conditionals are tested

4. Prioritize by importance:

   • Critical business logic (optimization, constraints, cost functions)
   • Error handling paths
   • Edge cases and boundary conditions
   • New features added in this branch

Step 3: Add Test Cases

Test Style Guidelines

Follow HAEO testing standards from the testing skill:

• Function-style tests: Use def test_...() not class-based tests
• Parametrized tests: Use @pytest.mark.parametrize for data-driven tests
• Test data modules: Add cases to core/model/tests/test_data/ for model element tests
• Direct property access: Access properties directly without None checks when you've created the entities

Where to Add Tests

1. Model elements: Add test cases to core/model/tests/test_data/{element}_cases.py

   • Add to VALID_CASES or INVALID_CASES lists
   • Cases are aggregated in __init__.py for parametrized tests

2. Element adapters: Add tests to core/adapters/elements/tests/

3. Config flows: Add test data to flows/tests/test_data/

4. Coordinator/data loading: Add tests to coordinator/tests/

5. New functionality: Create appropriate test files following existing patterns

Example: Adding Model Element Test Cases

# In core/model/tests/test_data/battery_cases.py
VALID_CASES = [
    # ... existing cases ...
    {
        "description": "Battery at minimum SOC with zero power",
        "factory": create_battery,
        "data": {"capacity": 10.0, "soc_min": 0.2, "soc_max": 0.9},
        "expected_outputs": {...},
    },
]

Step 4: Verify Coverage Improvement

After adding tests:

1. Re-run coverage:

   uv run pytest --cov=custom_components/haeo --cov-branch --cov-report=term-missing
   

2. Check coverage for changed lines:

   • Ensure new code has adequate coverage
   • Verify coverage hasn't decreased for modified code
   • Note: Codecov may report changed lines that are not directly part of your changeset due to indirect changes (e.g., refactoring that affects other code paths, or changes in dependencies that trigger different execution paths)

3. Run tests to ensure they pass:

   uv run pytest
   

Step 5: Summary

Provide a summary of:

• Current overall coverage percentage
• Coverage for changed files (if applicable)
• Files with low coverage that need attention
• Test cases added to improve coverage
• Any unreachable code identified
• Confirmation that coverage requirements are met

Notes

• Coverage philosophy: Focus on testing behavior and edge cases, not achieving arbitrary percentages
• Simplification preferred: When encountering untested code, first consider if the logic can be simplified to remove branches and lines that need testing
• Codecov: Enforces that coverage does not decrease from main on changed lines
• Indirect changes: Codecov may report changed lines that are not directly part of your changeset due to indirect changes (refactoring effects, dependency changes, or different execution paths)
• Unreachable code: If lines cannot be covered by exercising input data, they may be unreachable and should be removed
• Branch coverage: Use --cov-branch to ensure both branches of conditionals are tested
• CI requirement: Overall coverage must be ≥ 95%
• Changed lines: Codecov focuses on coverage of changed lines, not overall project coverage