Write Unit Tests

File naming: Use *.spec.ts for unit tests and *.e2e.spec.ts for end-to-end tests. Place test files alongside the source file they test.

Test framework: Use Jest with ts-jest. Test environment variables are loaded from .env-files/.env.test.

Test data generation:

• Use @faker-js/faker for random synthetic data (names, dates, numbers)
• Use zocker with createStubFromSchema() for generating data from Zod schemas
• Import shared test helpers from @carrot-fndn/shared/testing: stubRuleInput(), stubDocument(), createStubFromSchema()
• NEVER use real data (company names, tax IDs, license plates, addresses, PII)

Test structure:

• Use describe blocks to group related tests
• Use it.each for table-driven tests when testing multiple similar scenarios
• Use expect.objectContaining() for partial assertions on objects
• Keep test descriptions clear and descriptive using should language

Rule processor tests: Follow the standard pattern:

{rule-name}/
├── {rule-name}.processor.spec.ts   # Unit tests for evaluateResult()
├── {rule-name}.lambda.e2e.spec.ts  # E2E tests for the Lambda handler
└── {rule-name}.test-cases.ts       # Shared test data and fixtures

Assertions:

• Test both success and failure paths
• Verify error messages and types, not just that errors are thrown
• For rule processors, test that evaluateResult() returns the expected RuleOutput

Run and verify: Execute pnpm nx test <project-name> to run tests for the specific project. Ensure all tests pass before finishing.