This document describes the test suite for the ado-aw compiler.

The test suite consists of:

1. Unit tests - Located in src/main.rs (in the tests module)
2. Integration tests - Located in tests/compiler_tests.rs

To run all tests:

cargo test

To run only unit tests:

cargo test --lib

To run only integration tests:

cargo test --test compiler_tests

To run a specific test:

cargo test test_sanitize_filename

To run tests with output:

cargo test -- --nocapture

Unit tests are located in the tests module at the bottom of src/main.rs. These tests cover individual functions:

Tests the filename sanitization function with various inputs including:

• Spaces and special characters
• Numbers and mixed case
• Leading/trailing dashes
• Edge cases

Tests schedule generation for:

• Hourly schedules
• Daily schedules
• Deterministic behavior (same input produces same output)

Tests repository YAML generation for:

• Empty repository list
• Single repository
• Multiple repositories

Tests checkout step generation for:

• Empty repository list
• Multiple repositories

Tests copilot parameter generation for:

• Built-in MCPs (enabled)
• Built-in MCPs (disabled)
• Custom MCPs (should be skipped)

Tests markdown parsing for:

• Valid markdown with front matter
• Missing front matter
• Unclosed front matter
• Invalid YAML in front matter

Tests the default repository reference value.

Integration tests are located in tests/compiler_tests.rs. These tests verify the overall behavior of the compiler:

Tests the basic compilation workflow including:

• Creating temporary test directories
• Writing test input files
• Verifying directory structure

Verifies that the base template contains all required markers:

• {{ repositories }}
• {{ schedule }}
• {{ checkout_repositories }}
• {{ agent }}
• {{ agent_name }}
• {{ engine_run }}

Validates the example file (examples/sample-agent.md) to ensure:

• Proper YAML front matter structure
• Required fields are present
• Closing front matter delimiter exists

Documents expected behavior for various filename sanitization scenarios.

Verifies that all required dependencies are present in Cargo.toml:

• clap
• anyhow
• serde
• serde_yaml

The current test suite provides coverage for:

• ✅ Filename sanitization
• ✅ Schedule generation
• ✅ Repository configuration generation
• ✅ Checkout step generation
• ✅ Copilot parameter generation
• ✅ Markdown parsing
• ✅ Template structure validation
• ✅ Example file validation
• ✅ Dependency verification

When adding new functionality to the compiler:

1. Add unit tests for new helper functions in src/main.rs:

   #[test]
   fn test_new_function() {
       // Test code here
   }

2. Add integration tests for end-to-end functionality in tests/compiler_tests.rs:

   #[test]
   fn test_new_integration_scenario() {
       // Test code here
   }

3. Follow naming conventions:

   • Unit test names: test_<function_name>_<scenario>
   • Integration test names: test_<feature>_<scenario>

4. Include edge cases:

   • Empty inputs
   • Invalid inputs
   • Boundary conditions
   • Error cases

All tests should:

• Be self-contained and not depend on external state
• Clean up any resources they create
• Use descriptive assertion messages
• Follow the Arrange-Act-Assert pattern
• Test both success and failure cases

These tests are designed to run in CI/CD pipelines. Ensure all tests pass before submitting pull requests:

cargo test --all
cargo clippy -- -D warnings
cargo fmt -- --check