data-mocks

Library (written in TypeScript) to mock REST and GraphQL requests

• Why is this library useful
• Setup
  • Integration patterns
    • React
    • Angular
    • No framework (just vanilla JS)
• REST + GraphQL
• Examples
  • Basic mock injection without scenarios
  • Mock injection with scenarios
  • Basic GraphQL mock injection
• Exported types
  • Scenarios
  • HttpMock
  • GraphQLMock
  • Mock
  • Operation
  • MockConfig
• Exported functions
  • injectMocks
  • extractScenarioFromLocation
• Gotchas

Why is this library useful

When developing web applications locally, it is not uncommon to request data from an API. However, the API might not actually exist yet, or we might want to control what the responses are.

Typically, this sort of problem has been solved by either writing a separate mock API service alongside the project (i.e. having a Node service running locally with your application) or creating a development database that replicates staging or production environments. Both of these approaches are heavy and can lead to using incorrect data if schemas are out of sync.

This library aims to allow rapid local development without the dependency of a database or fully implemented APIs.

Setup

• Either npm install data-mocks or yarn add data-mocks
• Optional: extract the scenario from URL using the imported extractScenarioFromLocation() function
  • Or you can just hardcode a string to pass through instead
• Import injectMocks() function into your project with import { injectMocks } from 'data-mocks'
• Create an array of Scenario's you would like to use (see examples)
• Pass array of Scenario's to injectMocks()
• Hooray, all HTTP requests to mocked endpoints will now respond with the mocked data you have specified

Integration patterns

Regardless of framework or CLI tool used to generate your project, integrating data-mocks into your project is easy. Here are how it may look for you:

React

import React from 'react';
import ReactDOM from 'react-dom';
import App from './App';

async function main() {
  if (process.env.NODE_ENV === 'development') {
    const { injectMocks, extractScenarioFromLocation } = await import(
      'data-mocks'
    );

    // You could just define your mocks inline if you didn't want to import them.
    const { getMocks } = await import('./path/to/your/mock/definitions');

    injectMocks(getMocks(), extractScenarioFromLocation(window.location));
  }

  ReactDOM.render(<App />, document.getElementById('root'));
}

main();

Angular

import { enableProdMode } from '@angular/core';
import { platformBrowserDynamic } from '@angular/platform-browser-dynamic';

import { AppModule } from './app/app.module';
import { environment } from './environments/environment';

async function setupMocks() {
    const { injectMocks, extractScenarioFromLocation } = await import(
      'data-mocks'
    );
    // You could just define your mocks inline if you didn't want to import them.
    const { getMocks } = await import('./path/to/your/mock/definitions');

    injectMocks(getMocks(), extractScenarioFromLocation(window.location));
}

async function main() {
  if (environment.production) {
    enableProdMode();
  }

  if (!environment.production) {
    await setupMocks();
  }

  platformBrowserDynamic()
    .bootstrapModule(AppModule)
    .catch((err) => console.error(err));
}

main();

No framework (just vanilla JS)

async function main() {
  if (process.env.NODE_ENV === 'development') {
    const { injectMocks, extractScenarioFromLocation } = await import(
      'data-mocks'
    );

    // You could just define your mocks inline if you didn't want to import them.
    const { getMocks } = await import('./path/to/your/mock/definitions');

    injectMocks(getMocks(), extractScenarioFromLocation(window.location));
  }
}

main();

In these examples, we dynamically import data-mocks and our mock definitions as to not increase production bundle sizes (given that we will never need/want to use these files in production environments).

It is not a requirement to dynamically import data-mocks or your mock definitions - it is just a recommendation :)

REST + GraphQL

data-mocks works with either REST or GraphQL requests. It is also possible to easily mock both in the same application.

See the examples below to see how this is done.

Examples

Basic mock injection without scenarios

import { injectMocks } from 'data-mocks';
import axios from 'axios';

const scenarios = {
  default: [
    {
      url: /login/,
      method: 'POST',
      response: { some: 'good response' },
      responseCode: 200,
    },
    {
      url: /some-other-endpoint/,
      method: 'GET',
      response: { another: 'response' },
      responseCode: 200,
      delay: 1000,
    },
    {
      url: /endpoint-with-headers/,
      method: 'GET',
      response: { same: 'response' },
      responseHeaders: { token: 'mock-token' },
      responseCode: 200,
    },
  ],
};

injectMocks(scenarios);

fetch('http://foo.com/login', { method: 'POST', body: {} })
  .then((response) => response.json())
  .then((myJson) => console.log(myJson)); // resolves with { some: 'good response' } after a 200ms delay

fetch('http://foo.com/some-other-endpoint')
  .then((response) => response.json())
  .then((myJson) => console.log(myJson)); // resolves with { another: 'response' } after a 1 second delay

axios
  .post('http://foo.com/login', {})
  .then((response) => console.log(response)); // resolves with { another: 'response' } after a 200ms delay

axios
  .get('http://foo.com/some-other-endpoint')
  .then((response) => console.log(response)); // resolves with { another: 'response' } after a 1 second delay

In this example, we define a default scenario in our scenarios map. The Scenario objects are described by the Scenario interface. We then inject the scenarios into our application via the injectMocks() function. Finally, when we use fetch / XHR to make a request to endpoints that match our scenario objects, the mocked responses are returned.

N.B

In the above example we are using axios as our XHR library of choice. However data-mocks will work with any library that uses XMLHttpRequest under the hood.

Mock injection with scenarios

import { injectMocks, extractScenarioFromLocation } from 'data-mocks';
import axios from 'axios';

const scenarios = {
  default: [
    {
      url: /login/,
      method: 'POST',
      response: { some: 'good response' },
      responseCode: 200,
    },
    {
      url: /some-other-endpoint/,
      method: 'GET',
      response: { another: 'response' },
      responseCode: 200,
      delay: 1000,
    },
  ],
  failedLogin: [
    {
      url: /login/,
      method: 'POST',
      response: { some: 'bad things happened' },
      responseCode: 401,
    },
  ],
};

injectMocks(scenarios, 'failedLogin');

fetch('http://foo.com/login', { method: 'POST', body: {} }).then((response) =>
  console.log(response)
); // resolves with a 401 after a 200ms delay

fetch('http://foo.com/some-other-endpoint').then((response) =>
  console.log(response)
); // resolves with { another: 'response' } after a 1 second delay

axios
  .post('http://foo.com/login', {})
  .then((response) => console.log(response)); // resolves with a 401 after a 200ms delay

axios
  .get('http://foo.com/some-other-endpoint')
  .then((response) => console.log(response)); // resolves with { another: 'response' } after a 1 second delay

In this example, if we load our site up with ?scenario=failedLogin in the querystring and then attempt to hit the login endpoint, it will fail with a 401. However, the some-other-endpoint endpoint will still respond with the response in the default scenario as we have not provided one in the failedLogin scenario.

Basic GraphQL mock injection

Here, we have a React application using urql as a GraphQL client. This shows how GraphQL queries work and it can be assumed that if you want to use REST mocks in this application, you can do so as you normally would (see examples above).

In reality, the mock definitions would live at a higher level (like the entrypoint into the application) where they could be injected only if we were in development mode.

import React from 'react';
import { injectMocks, extractScenarioFromLocation } from 'data-mocks';
import gql from 'graphql-tag';

const mocks = {
  default: [
    {
      url: /graphql/,
      method: 'GRAPHQL',
      operations: [
        {
          operationName: 'Query',
          type: 'query',
          response: { data: { test: 'test' } },
        },
        {
          operationName: 'Mutation',
          type: 'mutation',
          response: { data: { test: 'test' } },
        },
      ],
    },
  ],
};

injectMocks(mocks, extractScenarioFromLocation(window.location));

const Component = () => {
  const [result] = useQuery({ query: Query }); // result will be { data: { test: 'test' } }

  return <>{result.data.test}</>;
};

Exported types

Scenarios

HttpMock

GraphQLMock

Mock

Union type of HttpMock and GraphQLMock.

Operation

MockConfig

Exported functions

injectMocks

extractScenarioFromLocation

Gotchas

• GraphQL operations must have an operation name.