Dogma JSON Schema

Generates models from JSON schemas.

Build Status Coverage Status Pub Status

Join the chat at

Release Status


Dogma JSON Schema uses the Dogma Codegen library to generate model data for a library. It can also generate unmodifiable model views for immutability as well as converters to handle serialization to and from Map data.

The Dogma JSON Schema library is used by Dogma Swagger and Dogma RAML. If a full RESTful API has been described in either of those formats then the corresponding library should be used instead of Dogma JSON Schema. Dogma JSON Schema is typically meant for cases where just a file format is being described.

To use create a build.dart file in the root of the project. This file can either be run manually, which will trigger a full rebuild of the library, or run when files are changed using the build_system library.

import 'dart:async';
import 'package:dogma_json_schema/build.dart';

Future<Null> main(List<String> args) async {
  await build(
      'my_package_name', // The name of the package within the pubspec
      'lib/models.json'  // The path to the root JSON schema

By default the build function will create unmodifiable model views as well as converters. See the documentation of build for the full options available from the build function. For additional information on how the unmodifiable model views and converters are generated see the documentation of Dogma Codegen.

JSON Schema Extensions

Dogma JSON Schema contains additional fields that can be used to control the code generated off of a schema. These fields are all optional and should only be used when necessary.

Model generation


The x-dart-type value is used to specify a type that isn't present in the JSON schema format. This can be used to specify types outside of the models described in the

As an example the following uses x-dart-type to specify that the value of time is a DateTime within the model definition.

  "Appointment": {
    "properties": {
      "name": {
        "description":"The name of the appointment",
      "time": {
        "description":"When the appointment begins",

This will generate the following definition.

library my_library.src.models.appointment;

class Appointment {
  /// When the appointment begins
  DateTime time;
  /// The name of the appointment
  String name;

When using this field a function to (de)serialize the value needs to be provided. See the documentation of Dogma Codegen for information on specifying custom converters.

Enumeration generation


The x-enum-names value is used to override the generated names. If you have an enumeration whose values are not strings then this field is required.

  "definitions": {
    "ImageSize": {
      "description": "The size of the image to retrieve",
      "type": "string",
      "enum": [
      "x-enum-names": [

This will generate the following definition.

library my_library.src.models.image_size;

import 'package:dogma_data/serialize.dart';

/// The size of the image.
@Serialize.values(const {
  's': ImageSize.small,
  'm': ImageSize.medium,
  'l': ImageSize.large
enum ImageSize { small, medium, large }


The x-enum-descriptions value allows individual enum values to contain code comments.

  "definitions": {
    "AddressType": {
      "type": "string",
      "enum": [
      "x-enum-descriptions": [
        "A home address.",
        "The address of a business."

This will generate the following definition.

library my_library.src.models.address_type;

enum AddressType {
  /// A home address.
  /// The address of a business.

Features and bugs

Please file feature requests and bugs at the issue tracker.