fast-check — Property-Based Testing Framework for JavaScript and TypeScript

Introduction

fast-check brings property-based testing to the JavaScript ecosystem, inspired by Haskell's QuickCheck. Instead of writing individual test cases with hardcoded inputs, you describe properties that should hold for all valid inputs. fast-check then generates hundreds of random inputs, finds failing cases, and automatically shrinks them to the simplest reproduction.

What fast-check Does

• Generates random test inputs from a rich library of built-in arbitraries (numbers, strings, arrays, objects, dates, and more)
• Tests that properties hold across hundreds of automatically generated cases per run
• Shrinks failing inputs to the smallest example that still triggers the failure
• Replays specific failures deterministically using seed values
• Integrates with Jest, Vitest, Mocha, and any assertion library

Architecture Overview

fast-check centers on two concepts: arbitraries (generators that produce random values with shrinking capabilities) and properties (assertions parameterized by generated values). The runner executes a property by repeatedly sampling from its arbitraries. When a failure is found, the shrinker systematically reduces each input dimension to find a minimal counterexample. Seeds ensure reproducibility across runs.

Self-Hosting & Configuration

• Install: npm install --save-dev fast-check
• Import and use directly in test files alongside your existing test runner
• Configure run parameters: fc.assert(property, { numRuns: 1000, seed: 42 }) for more thorough or reproducible testing
• Create custom arbitraries with fc.record(), fc.oneof(), and fc.chain() for domain-specific data
• Use fc.pre() to filter generated values that do not meet preconditions

Key Features

• Rich built-in arbitrary library covering primitives, collections, recursive structures, and more
• Automatic shrinking produces minimal counterexamples without manual effort
• Seed-based replay lets CI reproduce exact failures from logs
• Model-based testing for stateful systems using fc.commands() and fc.modelRun()
• Async property support for testing promise-based and async/await code

Comparison with Similar Tools

• QuickCheck (Haskell) — The original property-based testing library; fast-check adapts the concept for JavaScript with a modern API
• Hypothesis (Python) — Python's property-based testing; fast-check provides equivalent capabilities for the JS/TS ecosystem
• JSVerify — Earlier JS property testing library; fast-check offers better shrinking, more arbitraries, and active maintenance
• Jest/Vitest — Example-based test runners; fast-check complements them by discovering edge cases you would not write manually
• Faker.js — Generates realistic fake data; fast-check generates structured data specifically designed for shrinking and property verification

FAQ

Q: How is property-based testing different from example-based testing? A: Example-based tests verify specific input-output pairs. Property-based tests verify that a property holds for all valid inputs, automatically discovering edge cases like empty arrays, negative numbers, or Unicode strings.

Q: How many test cases does fast-check run by default? A: 100 runs per property by default. You can increase this with numRuns for more thorough testing or decrease it for faster feedback.

Q: Can I use fast-check with my existing test framework? A: Yes. fast-check works with Jest, Vitest, Mocha, Jasmine, and any runner. Just call fc.assert() inside your test functions.

Q: What is shrinking? A: When fast-check finds a failing input, it automatically tries smaller or simpler values to find the minimal input that still fails. This makes debugging much easier by removing irrelevant complexity.

Sources

• https://github.com/dubzzz/fast-check
• https://fast-check.dev/