Usage Examples
Runnable, scenario-driven walkthroughs that combine multiple json-tology APIs to solve one real problem end to end. Each page below assumes the core concepts and the bookstore domain; use this index to jump to the scenario that matches what you're building.
Custom format validators
Use this page when you need to validate a field format json-tology doesn't ship (an ISBN checksum, a slug pattern) or replace a built-in format validator (date, email, uuid) with your own.
Transform recipes
Use this page when you need a ready-made Transform.create (or Transform.chain) recipe for a common wire-to-runtime conversion — ISO date-time to Date, epoch milliseconds to Date, Temporal PlainDate, and other everyday decode/encode pairs.
Class hydration
Use this page when you want jt.instantiate to return a real class instance (not a plain object) — methods, getters, instanceof — and need to choose between the lift strategies (Reflect.construct, fromPlain, prototype swap) and understand their tradeoffs.
Class hydration: ORM recipes
Use this page when the class you're hydrating into is an ORM entity or a DDD value object — TypeORM, Prisma, Mikro-ORM, Drizzle, Active Record — and you need the recipe for that specific pattern rather than the general lift strategies.
Sub-schema patterns
Use this page when you're working with registered $ref sub-schemas and need to see how validation, defaults, coercion, Transform decoders, and TBox emission each behave across a reference — including self-referential cycles.
Bookstore OWL taxonomy
Use this page when you're modeling an OWL class hierarchy on top of the bookstore domain — subClassOf, disjointWith, complementOf, property restrictions, invariants, and sameAs — and want to see how each axiom shows up in the TBox/ABox graph.
Multi-format ETL
Use this page when you need to ingest the same domain from several source shapes (multiple external APIs or message formats) into one canonical schema, and re-encode back out to a source shape — a fan-in/fan-out pivot-codec pipeline with owl:sameAs for cross-source provenance.
Related
- Bookstore domain - the schemas every example builds on
- Picking a method - choosing between
instantiate,validate,materialize, and friends Transform.createandjt.encode- the underlying decode/encode API used throughout this section