property-based-testing

Property-based testing finds edge-case bugs that example-based tests miss by turning executable properties into repeatable test specifications. Use it when testing serialization pairs, parsers, normalization logic, pure algorithms, or smart-contract invariants.

To write property-based tests for encode/decode pairs, propose a roundtrip property ensuring decode(encode(x)) equals x, and at least one invariant. Design input strategies using libraries like Hypothesis or fast-check to generate valid inputs for repeatable verification.

Property-based testing works with smart contracts by fuzzing invariants and checking state transitions. Use specialized smart-contract fuzzing tools like Echidna, alongside general libraries like Hypothesis, fast-check, or proptest for multi-language test suite support.

Common property patterns to check include roundtrip, idempotence, invariants, commutativity, associativity, identity, and inverse properties. Prioritize these patterns for testing encode/decode pairs, normalization, validators, sorting, and smart-contract logic.

Analyze failing property-based tests by reproducing and classifying failures using concrete workflows. Check property grounding to avoid tautologies and vacuous tests, then improve shrinking and CI settings to minimize failing inputs and ensure repeatable verification.

Yes, you can use property-based testing patterns to refactor code for testability. Apply decision trees for when to invoke property tests and use refactoring patterns that make serialization, parsing, and algorithmic logic easier to verify with generated inputs.