property-based-testing

Property-based testing validates universal correctness claims by generating many inputs across a domain and shrinking failures to minimal counterexamples. Unlike example-based tests checking specific expectations, it automatically explores rich input domains to challenge invariants.

Use property-based testing for combinatorial correctness, invariants, and round-trip specifications. Use fuzz testing for crash-focused exploration, mutation testing to measure suite quality, and example-based tests for specific behaviors, ensuring clear separation between approaches.

Express a universal round-trip property, design generators to produce representative edge-covering inputs, and run trials to verify the property holds. When a failure occurs, the shrinker automatically reduces it to a minimal counterexample for fast debugging.

Yes, property-based testing requires generator and shrinker discipline to produce representative edge-covering inputs and shrink failures to minimal counterexamples. You must also set appropriate trial budgets to effectively validate the universal property across the input domain.

Property-based testing applies to unit-level and component-level verification for invariants, oracles, and round-trip specifications. It requires clear separation from state-machine modeling, meaning you must define clearly universal and falsifiable properties for your specific domain.