OpenAPI Mocker

An API mocker based in the OpenAPI 3.0 specification.

Installation and usage

Using npm

npm i -g open-api-mocker

open-api-mocker -s my-schema.json -w

open-api-mocker --help # To prompt every available setting.

Using docker

docker run -v "$PWD/myschema.json:/app/schema.json" -p "5000:5000" jormaechea/open-api-mocker

Or to run an specific version

docker run -v "$PWD/myschema.json:/app/schema.json" -p "5000:5000" jormaechea/open-api-mocker:X.Y.Z`

You can set any parameter when running inside a docker container

docker run -v "$PWD/myschema.json:/app/schema.json" -p "3000:3000" jormaechea/open-api-mocker:X.Y.Z -s /app/schema.json -p 3000`

Capabilities

• [x] Read yaml and json OpenAPI v3 schemas.
• [x] Port binding selection
• [x] Request parameters validation
• [x] Request body validation
• [x] Response body and headers generation based on examples or schemas
• [x] Response selection using request header: Prefer: statusCode=XXX or Prefer: example=name
• [x] Request and response logging
• [x] Servers basepath support
• [x] Support x-faker and x-count extension methods to customise generated responses
• [ ] API Authentication

Customizing Generated Responses

The OpenAPI specification allows custom properties to be added to an API definition in the form of x-*. OpenAPI Mocker supports the use of two custom extensions to allow data to be randomised which should allow for more realistic looking data when developing a UI against a mock API for instance.

x-faker

The x-faker extension is valid for use on properties that have a primitive type (e.g. string/integer, etc.) and can be used within an API definition to use one or more methods from the excellent faker library for generating random data.

Given the following API definition:

openapi: '3.0.2'
info:
  title: Cats
  version: '1.0'
servers:
  - url: https://api.cats.test/v1
paths:
  /cat:
    get:
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                type: object
                properties:
                  firstName:
                    type: string
                    x-faker: name.firstName
                  lastName:
                    type: string
                    x-faker: name.lastName
                  fullName:
                    type: string
                    x-faker: '{{name.firstName}} {{name.lastName}}'
                  age:
                    type: string
                    x-faker: 'random.number({ "min": 1, "max": 20 })'

A JSON response similar to the following would be produced:

{
    "firstName": "Ted",
    "lastName": "Kozey",
    "fullName": "Herbert Lowe",
    "age": 12
}

The x-faker extension accepts values in 3 forms:

1. fakerNamespace.method. e.g. random.uuid
2. fakerNamespace.method({ "methodArgs": "in", "json": "format" }). e.g. random.number({ "max": 100 })
3. A mustache template string making use of the 2 forms above. e.g. My name is {{name.firstName}} {{name.lastName}}

NOTE: To avoid new fake data from being generated on every call, up to 10 responses per endpoint are cached based on the incoming query string, request body and headers.

x-count

The x-count extension has effect only when used on an array type property. If encountered, OpenAPI Mocker will return an array with the given number of elements instead of the default of an array with a single item.

For example, the following API definition:

openapi: '3.0.2'
info:
  title: Cats
  version: '1.0'
servers:
  - url: https://api.cats.test/v1
paths:
  /cat:
    get:
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                type: array
                x-count: 5
                items:
                  type: string

Will produce the following response:

[
    "string",
    "string",
    "string",
    "string",
    "string"
]

Advanced usage

See the advanced usage docs to extend or build your own app upon OpenAPI Mocker.

Tests

Simply run npm t

Contributing

Issues and PRs are welcome.