https://github.com/koumoul-dev/schema-jtd

JSON Schema and JSON Type Definition transformation utilities (contains the schema2td and td2schema commands)
https://github.com/koumoul-dev/schema-jtd

Last synced: 6 months ago
JSON representation

JSON Schema and JSON Type Definition transformation utilities (contains the schema2td and td2schema commands)

• Host: GitHub
• URL: https://github.com/koumoul-dev/schema-jtd
• Owner: koumoul-dev
• License: mit
• Created: 2023-02-01T07:13:55.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2023-02-24T07:51:06.000Z (over 2 years ago)
• Last Synced: 2025-03-16T21:37:47.210Z (7 months ago)
• Language: JavaScript
• Size: 2.56 MB
• Stars: 2
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# schema-jtd

JSON Schema and JSON Type Definition transformation utilities (contains the schema2td and td2schema commands).

## CLI usage

```

npx --package @koumoul/schema-jtd schema2td --help

```

## Why

  - need help migrating from one format to another ?

  - need to make a tool written for one format work with the other at low cost ?

  - need create an Open API documentation from your JTDs ?

  - need to create code bindings from your schemas ?

## JTD or JSON Schema ?

![](doc/both-the-road-to-el-dorado.gif)

Don't be too fast in chosing one and rejecting the other. JSON schema is good for a self documented API that makes many of its validation rules explicit, JTD is great for non-ambiguous types and generating code bindings.

A possible sequence that includes both:

  1. write an API specification including self documented JSON schemas

  2. use `schema2td` to produce JTD equivalents

  3. use `jtd-codegen` to produce code bindings

  4. use JTD for fast parsing and serializing at both ends of your API endpoints

  5. use a JSON Schema validator to validate the payload with extra rules ignored by JTD

## Limitations

  - `schema2td` will reject schemas with ambiguous typing (for example a oneOf with different types)

    - this might be resolved in the future using a vocabulary to resolve ambiguous typing (for example by flagging a oneOf to be ignored)

  - `td2schema` will be missing some semantics

    - this is partially compoensated by mapping some JTD metadata to JSON Schema attributes