Skip to content

Latest commit

 

History

History
130 lines (97 loc) · 5.28 KB

README.md

File metadata and controls

130 lines (97 loc) · 5.28 KB

REST API Test Automation

This repository contains my own studies and tests on different public APIs. In this project I use JavaScript, Mocha, Chai and Request Promise libraries to send HTTP requests and validate api responses. I also practiced general structure and tools used in API test automation to streamline and enhance testing process. If you find examples here useful, feel free to clone repo and build on top of this framework to test your own projects as well.

Features

  • Organized structure: All tests, data, and API wrapper classes are oganized within dedicated directories.
  • Random data: Project includes Faker api to creative random and flexible test data.
  • Parallel execution: Utilize JavaScript promise concepts to send bulk requests and optimize test cleanup.
  • Public APIs: All examples are created using public and free API endpoints for practice purpose.
  • JSON schema validation: Along with Chai assertions, I added Ajv schema validation utilities.
  • Reporting: Easily generate custom reports for all or certain test suites.

How to Use

Prerequisites

  • Install Node.js and NPM

Installation

  1. Clone the repository:

    git clone https://github.com/gtsereteli/rest-api-automation-framework.git

  2. Install dependencies:

    npm install

  3. Copy .env-template file and rename it to .env

Running tests

  1. You can run sample test for api which does not require authorization

    npm run test:cards

  2. To run all other tests, you will need to generate key/tokens and add these to .env file. In the scripts block of package.json you will find all custom commands which you can use to quickly run different suites. You can write npm run {command-name} to run custom script.

API authorization

Some of the APIs in this project require key and token. You will need to provide values in .env file. You can use provided tests and APIs or build on top of this project and create tests on other endpoints. If you want to run all scenarios present in /test you will need to create authorization for following APIs

Writing tests

Below is an example of making a HTTP GET request:

describe("Can get GitHub user info", function () {
  it("Can get user data by username", async function () {
    const response = await request.get({
      url: `https://api.github.com/users/gtsereteli`,
      json: true,
      headers: {
        "Content-Type": "application/json",
        "User-Agent": "Awesome-Octocat-App",
      },
    });
  });
});

Verifying responses

In this project I used Chai and Ajv libraries for all assertions. Chai provides a set of expressive and readable assertions to verify he expected outcomes of code. Ajv helps with JSON schema validations and I added helper function in utils folder which you can use for all schema assertions.

Below is an example of Chai and Ajv assertions:

describe("Can get GitHub user info", function () {
  it("Can get user by username", async function () {
    const username = "gtsereteli";
    const response = await api.getUser(username);
    const isSchemaValid = validateJsonSchema(response, getUserSchema);

    expect(isSchemaValid).to.be.true;
    expect(response).to.have.property("login", username);
    expect(response).to.have.property("name", "Giorgi Tsereteli");
    expect(response).to.have.property("type", "User");
    expect(response).to.have.property("site_admin", false);
  });

JSON schema validation

When you know the expected Json structure, you can create schema and save it in schemas directory as schemaName.json. Use free online tool such as https://transform.tools/json-to-json-schema to copy/paste json object and convert it into schema which you can then donwload and use.

Use validateJsonSchema function to pass api response and your schema for validation

validateJsonSchema(response, getUserSchema);

Test data

For test data I created factory folder and use Faker api to generate different data sets. It also provides methods to fetch random elements from arrays and objects. For more details, you can visit their page: https://fakerjs.dev

const user = () => {
  return {
    name: `${faker.name.firstName()} ${faker.name.lastName()}`,
    gender: `${faker.helpers.arrayElement(["female", "male"])}`,
    email: `${qaPrefix}${randomString()}@example.com`,
    status: `${faker.helpers.arrayElement(["active", "inactive"])}`,
  };
};

Reports

I added simple mochawesome and mochawesome-report-generator libraries for creating HTML reports. You can use following flags when running mocha tests or research and add your own options for reporting.

mocha './**/*.test.js' --reporter mochawesome --reporter-options reportDir=reports

Thanks for checking out my repo!

As mentioned earlier, this repo is just for practice and not meant for professional use. Feel free to fork it, play around to learn something new, and apply it to your own projects. Contributions, pull requests, and questions are welcome.