bfg-schemautil 0.1.3

A set of utilities to generate JSON Schema from simple JSON objects

schemautil

A set of utilities to generate JSON Schema from simple JSON objects.

Example

Basic Object Schema

Given the following object schema:

{
  "type": "object",
  "properties": {
    "a": { "type": "string" },
    "b": { "type": "string" }
  },
  "required": ["a", "b"],
  "additionalProperties": false
}

You can generate this schema using:

object_schema({"a": "string", "b": "string"})

Enum Example

For an enum field:

object_schema({"status": ["enum", "pending", "approved", "rejected"]})

Generates:

{
  "type": "object",
  "properties": {
    "status": {
      "type": "string",
      "enum": ["pending", "approved", "rejected"]
    }
  },
  "required": ["status"],
  "additionalProperties": false
}

Rules

• All fields are required and additionalProperties is always set to false.

• To make a field nullable, add a ? to the end of its name. Example: {"a": "string", "b?": "string"} (b is nullable)

• To define an array, wrap the type in brackets. Example: {"a": "string", "b": ["string"]} (b is an array of strings) Note: There should be only one type in the array.

• To define a field that can be one of multiple types, list the types in brackets. Example: {"a": "string", "b": ["string", "number"]} (b can be a string or a number)

• To define an enum (restricted set of string values), use ["enum", ...values]. Example: {"status": ["enum", "pending", "approved", "rejected"]} (status must be one of these specific strings)

  Note: The library distinguishes between type unions and enums:

  • ["string", "number"] → Type union (can be any string or any number)
  • ["enum", "pending", "approved"] → String enum (must be exactly one of these values)