@har-sdk/openapi-sampler

This is a tool that generates samples based on OpenAPI payload/response schema.

• Supports a wide range of schema types and formats, including allOf, additionalProperties, and common string formats such as email, password, date-time, and more.
• Infers schema type automatically following the same rules as json-schema-faker.
• Supports $ref resolving.
• Produces deterministic and predictable output for a given input.
• Makes it easier and faster to generate sample data for OpenAPI schemas.
• Reduces errors and increases code quality by ensuring that generated samples follow the defined schema.

Using the @har-sdk/openapi-sampler package can save you a lot of time when you need to generate sample data for your OpenAPI schemas. With just a few lines of code, you can create realistic data objects that you can use for testing or documentation purposes.

Installation

To install the package using npm, run the following command:

$ npm i --save @har-sdk/openapi-sampler

If you prefer using yarn, you can use the following command instead:

$ yarn add @har-sdk/openapi-sampler

Usage

After installing, you can import the package in your code using the following import statement:

import { sample } from '@har-sdk/openapi-sampler';

The sample() function is the main API provided by the package, and it takes three parameters:

Option	Description
schema	OpenAPI Schema Object that specifies the structure of the data you want to generate samples for.
options	Provides additional options for the sampler, such as skipping non-required properties, read-only properties, or write-only properties. You can find a full list of available options in the code example below.
spec	Entire OpenAPI specification that the schema is taken from. This parameter is only required if the schema contains $ref references to external schemas.

Here's an example of how to use the sample() function:

import { sample } from '@har-sdk/openapi-sampler';

sample({
  type: 'object',
  properties: {
    a: { type: 'integer', minimum: 10 },
    b: { type: 'string', format: 'password', minLength: 10 },
    c: { type: 'boolean' }
  }
});
// { a: 10, b: 'pa$$word_q', c: true }

In this example, it is generating a sample data object based on a schema that contains three properties (a, b, and c).

Note, when $ref is used, the spec object must be provided so that the sampler can resolve the reference:

import { sample } from '@har-sdk/openapi-sampler';

const spec = {
  // ...
  components: {
    schemas: {
      Pet: {
        type: 'object',
        properties: {
          id: {
            type: 'integer',
            format: 'int64'
          },
          name: {
            type: 'string'
          }
        }
      }
    }
  }
};

sample({ $ref: '#/components/schemas/Pet' }, {}, spec);
// { id: 42, name: 'lorem' }

By default, the sampler generates values for all properties defined in a schema, regardless of whether they are required or not. If you want to exclude non-required object properties, you can use the skipNonRequired option as follows:

sample(
  {
    type: 'object',
    properties: {
      a: { type: 'string' },
      b: { type: 'string' }
    },
    required: ['b']
  },
  { skipNonRequired: true }
);
// { b: 'lorem' }

By default, all properties, including those marked as read or write-only, are included in the generated sample. However, if you want to exclude readOnly or writeOnly properties, you can use the skipReadOnly or skipWriteOnly options, respectively, as shown below:

sample(
  {
    type: 'object',
    properties: {
      a: { type: 'string' },
      b: { type: 'string', readOnly: true }
    }
  },
  { skipReadOnly: true }
);
// { a: 'lorem' }

Note, the library recursively generates the entire sample object tree up to a maximum depth of 2 level to prevent infinite recursion. If you want to increase or decrease this maximum depth, you can use the maxSampleDepth option as follows:

sample(schema, { maxSampleDepth: 20 });

Also, the library logs console warning messages when it encounters an unsupported schema. If you want to suppress these warning messages, you can use the quiet option as follows:

sample(schema, { quiet: true });

When the schema comes from the specification which does not allow the example node to exist e.g. OAS 2.0 parameter definition, some vendors may provide such schema example value in OAS vendor extension nodes namely x-example or x-examples. If you want to include such kind of example values into the output, you can use the includeVendorExamples as shown below:

sample(
  {
    'type': 'string',
    'x-example': 'some_value'
  },
  { includeVendorExamples: true }
);
// some_value

License

Copyright © 2023 Bright Security.

This project is licensed under the MIT License - see the LICENSE file for details.