swagger-inline

Generate an OpenAPI/Swagger definition from inline comments.

• Installation
• Usage
  • CLI
  • Library
• Examples

Installation

npm install swagger-inline --save-dev

Usage

CLI

npx swagger-inline [--base] [--format] <inputGlobs ...>

Example

npx swagger-inline "./*.js" --base 'swaggerBase.json' > api.json

Options

The inputGlobs argument is a list of files, or globs, to search for Swagger/OAS comments.

• base: Base API specification to extend. *Required
• format: Output filetype: .json or .yaml (default: .json)
• scope: Matches the scope field defined in each API. For example, if --scope public is supplied, all operations will be generated, if --scope private, only those operations that have a scope: private declaration will be included.

Library

swaggerInline([inputGlobs...], options) => Promise => json | yaml

Example

const swaggerInline = require('swagger-inline');

swaggerInline(['src/**/*.js', 'test/**/*.js'], {
  base: 'swaggerBase.json',
}).then((generatedSwagger) => {
  /* ... */
});

Available options

• base: Base specification to extend. *Required
• format: Output filetype: .json or .yaml (default: .json)
• ignore: An array of globs for files to ignore. (default: ['node_modules/**/*', 'bower_modules/**/*'],
• logger: Function called for logging. (default: empty closure)
• metadata: Add additional annotations to the Swagger file, prefixed with x-si.
• scope: Matches the scope field defined in each API. For example, if --scope public is supplied, all operations will be generated, if --scope private, only those operations that have a scope: private declaration will be included.
• ignoreErrors: Ignore errors due to image files or unknown file types when parsing files. (default: false)

Examples

Standard usage

1) Create a project

swaggerBase.yaml

swagger: "2.0"
host: "petstore.swagger.io"
basePath: "/api"
schemes: ['http']

api.js

/**
 * @api [get] /pets
 * bodyContentType: "application/json"
 * description: "Returns all pets from the system that the user has access to"
 * responses:
 *   "200":
 *     description: "A list of pets."
 *     schema:
 *       type: "String"
 */

api.route('/pets', function() {
    /* Pet code 😺 */
});

/**
 * @schema Pet
 * required:
 *   - id
 *   - name
 * properties:
 *   id:
 *     type: integer
 *     format: int64
 *   name:
 *     type: string
 *   tag:
 *     type: string
 */

// some schema related function

2) Run Command

swagger-inline './*.js' --base './swaggerBase.yaml'

Output:

swagger: '2.0'
host: petstore.swagger.io
basePath: /api
schemes:
  - http
paths:
  /pets:
    get:
      description: Returns all pets from the system that the user has access to
      responses:
        '200':
          description: A list of pets.
          schema:
            type: String
components:
  schemas:
    Pet:
      required:
        - id
        - name
      properties:
        id:
          type: integer
          format: int64
        name:
          type: string
        tag:
          type: string

Scoped compilations

With the --scope parameter, you can compile your files based on a specific target that you define within your inline comments. For example, we have an API with a GET /pets and POST /pets but only the GET operation is public. We can add scope: public to our GET operation documentation to tell swagger-inline what scope it's set under.

/**
 * @api [get] /pets
 * scope: public
 * description: "Returns all pets from the system that the user has access to"
 * responses:
 *   "200":
 *     description: "A list of pets."
 *     schema:
 *       type: "String"
 */

/**
 * @api [post] /pets
 * description: "Creates a new pet
 * responses:
 *   "200":
 *     description: "The created pet."
 */

Now when you run swagger-inline, you can supply --scope public and only the GET /pets operation will be picked up. Omit --scope public and everything will be picked up.