A Schema is all you need!
summary
Claims that a single runtime description generates everything needed: FS schema definitions produce decoders, encoders, async validators, fast-check arbitraries and JSON Schema, so types, runtime checks and test data stay in sync. Demonstrates a real bug (using toString instead of toJSON) that corrupts storage, verifies encoders with property‑based round‑trip tests, shows promise‑based validation for async checks, and builds a custom equality interpreter from the schema that actually works.
AI-written summary · about
chapters (11)
- 0:00Schemas and data validation
- 7:17Problem with repeated definitions
- 9:04F Schema and its benefits
- 10:49Simple schema for a to-do
- 13:39Validators from schema definitions
- 18:25Encoding data back to form
- 21:55Asynchronous validation with F Schema
- 25:29Generating random values
- 28:49JSON Schema from F Schema
- 31:16Advanced use cases
- 36:58Conclusion and Q&A
related talks
30mDynamic & Proud
24mEffective API schemas testing
30mThe Fun of Runtime Type Systems (or "Funtime Type Systems")
07m"Using schemas in your API Automation"
31mStop Guessing A11y: Auto-Generate Playwright Tests from Your GraphQL Schema
25mSophisticated Schema Mocking
29mContract Testing in Python for a Data Science API
33mHow to quickly and painlessly close an API regression from PyTest + JSON Schema