Testkit DSL Documentation

Testkit Usage Documentation.

Testkit Documentation

Introduction

Testkit is a testing tool designed for API manual testing and test automation tasks. It provides a simplified yaml syntax for defining and executing API test scenarios. image

Why a testing DSL?

  • Teams should not be forced to write javascript (Postman), Groovy (Katalon), or Java (Rest-Assured) just because they want to test an API for a web system
  • We should be able to create interactive builders that generate this underlying DSL. So you shouldn’t even need to write this DSL by hand in the future
  • We should be able to use the same script for both individual tests and load testing.
  • We should still be able to persist these tests in our version control and collaborate on them with our peers
  • Inspired by Local-First software principles

Table of Contents

FYI, this table of contents reflects the current sections covered in the documentation based on the information provided so far. It may be expanded or revised as the documentation progresses and more content is added.

Installation

To install the testkit testing tool, follow one of the options below:

Download binaries:

  1. Navigate to the releases page and proceed to install the testkit binary suitable for your operating system (linux, mac and windows). Check releases
  2. After download the binary for your OS, add it to evironment path and run
testkit test --file  <test.yaml>

replace <test.yaml> with your test file name

macOS

To install Testkit on macOS using Homebrew, follow these steps:

  1. Tap the Testkit release repository:

    brew tap apitoolkit/testkit
    

    Tapping the repository adds it as a source for Homebrew formulae.

  2. Install Testkit:

    brew install testkit
    

    This command will download and install Testkit on your macOS system.

  3. Add testkit to your environment path and Run testkit

testkit test --file <test.yaml>

replace <test.yaml> with your test file name

Clone repository:

  1. Ensure you have Rust programming language and Cargo package manager installed on your system. You can download them from the official Rust website: https://www.rust-lang.org/.

  2. Clone the testkit repository from GitHub using the following command:

    git clone https://github.com/testkit/testkit
    
  3. Change into the cloned directory:

    cd testkit
    
  4. Build the project using Cargo:

    cargo build --release
    
  5. Once the build process is complete, you can find the testkit executable file in the target/release directory.

How to Run

To run the testkit testing tool, use the following command:

RUST_LOG=debug cargo run -- test --file ./test.tk.yaml

Make sure to replace ./test.yaml with the path to your YAML test file.

Comparison with Other Testing Libraries

Before delving into the details of testkit and its functionality, let’s compare it to other testing libraries to highlight the elegance, cleanliness, and simplicity of testkit. To illustrate this, we will rewrite a Cypress test using testkit.

By showcasing the differences, you will see how testkit provides a more streamlined and user-friendly approach to API testing. It offers a clean and intuitive syntax that simplifies the process of defining and executing API test scenarios. Let’s explore the rewritten test using testkit.

describe('TODO api testing', () => {
  let todoItem;
  it('fetches Todo items - GET', () => {
    cy.request('/todos/').as('todoRequest');
    cy.get('@todoRequest').then((todos) => {
      todoItem = todos.body[0]['_id'];
      expect(todos.status).to.eq(200);
      assert.isArray(todos.body, 'Todos Response is an array');
    });
  });

  it('deletes Todo items - DELETE', () => {
    cy.request('DELETE', `/todos/${todoItem}`).as('todoRequest');
    cy.get('@todoRequest').then((todos) => {
      expect(todos.status).to.eq(200);
      assert.isString(todos.body, 'todo deleted!');
    });
  });

  it('Adds Todo item - POST', () => {
    cy.request('POST', '/todos/', { task: 'run tests' }).as('todoRequest');
    // adds new Todo item by defining Todo name
    cy.get('@todoRequest').then((todos) => {
      expect(todos.status).to.eq(200);
      cy.wrap(todos.body).should('deep.include', {
        task: 'run tests',
        completed: false,
      });
    });
  });
});
---
    - title: fetches TODO items - GET
      GET: /todos/
      asserts: # Asserts accepts a list of expressions, usually via json_paths to identify the items being refered to.
        ok: $.resp.status == 200  # Rely on an expressions libray for parsing expressions
        array: $.resp.json
      exports: # values which should be accesible to future steps.
        todoItem: $.resp.json[0]._id

    - title: deletes TODO items - DELETE
      DELETE: /todos/$.stages[0].todoItem # relative syntax exists: $.stages[-1].todoItem, -1 means one stage before me
      asserts:
        empty: $.resp.json.todos
        string: $.resp.json

    - title: Adds Todo item - POST
      POST: /todos/
      json:
        task: "run tests"
      asserts:
        ok: $.resp.status == 200
        ok: $.resp.json.task == "run tests"
        ok: $.resp.json.completed == false

Test Definition Syntax

testkit uses a YAML-based syntax for defining test scenarios. Each scenario consists of multiple stages, where each stage represents a specific API request and its associated assertions. Below is an example of the YAML syntax for defining API tests:

---
    - title: fetches TODO items - GET
      GET: /todos/
      asserts:
        ok: $.resp.status == 200
        array: $.resp.json
      exports:
        todoItem: $.resp.json[0]._id

    - title: deletes TODO items - DELETE
      DELETE: /todos/$.stages[0].todoItem
      asserts:
        empty: $.resp.json.todos
        string: $.resp.json

    - title: Adds Todo item - POST
      POST: /todos/
      json:
          task: "run tests"
      asserts:
        ok: $.resp.status == 200
        ok: $.resp.json.task == "run tests"
        ok: $.resp.json.completed

The YAML file consists of a list of test scenarios. Each scenario represents an API request and contains the following fields:

  • name (required): A descriptive name for the stage.
  • request (required): Defines the API request to be made. It can include HTTP methods (GET, POST, PUT, DELETE, etc.) and the corresponding request URL or endpoint.
  • asserts (optional): Defines the assertions to be performed on the response. It specifies conditions that must be satisfied for the test to pass.
  • exports (optional): Specifies the values to be captured from the response and made available to future stages.

In the example above, the YAML test file defines three test items fetching TODO items using a GET request, deleting a specific TODO item using a DELETE request, and adding a new TODO item using a POST request. The name field is self explanatory and so we’ll take more about the rest of the fields in detail but before that let’s talk about JSONPath.

What is JSONPath

JSONPath is a powerful query language designed for navigating and extracting data from JSON documents. It provides a concise syntax that allows you to specify paths to specific elements within a JSON structure, facilitating data access and manipulation. In testkit, JSONPath expressions are extensively used to extract data for assertions and exports.

To illustrate how JSONPath works, consider the following examples:

  • $.user.name: This expression retrieves the name of a user from the top-level object in the JSON document.
  • $.todos[0].task: Here, the expression accesses the task property of the first element in an array of todos.
  • $.todos[*].task.description: This expression retrieves the description property of all tasks within the todos array.

The syntax of JSONPath expressions includes several key components:

  • Bracket notation ([]): Used to access elements within an array by providing the index within square brackets.
  • Wildcard (*): Matches any element at the current level, allowing you to retrieve all elements of a particular level.
  • Recursive descent (..): Enables searching for elements at any depth within the JSON structure, including nested objects and arrays.
  • Filters ([?]): Allows applying conditions or filters to select specific elements based on certain criteria.

By employing JSONPath expressions, you can precisely pinpoint the desired data within a JSON structure. These expressions play a vital role in testkit, facilitating the extraction of data for performing assertions and capturing exports during the testing process. learn more about jsonpaths here

request field

The request field in testkit defines the API request to be made and consists of three properties:

  • method: This property specifies the HTTP method for the request, such as GET, POST, PUT, or DELETE. The value of the method property is the request URL path.

    Example:

    # POST request
    - title: Adds Todo item - POST
      POST: /todos/
    
    # GET request
    - title: Fetches Todo items - GET
      GET: /todos/
    
  • headers (optional): This property allows you to include HTTP headers in the request. Headers can be used to pass additional information to the server, such as authentication tokens or content type.

    Example:

    - title: Fetches Todo items - GET with headers
      GET: /todos/
      headers:
        Authorization: Bearer <token>
        Content-Type: application/json
        Allowed-Methods:
          - GET
          - POST
    
  • json The request field in testkit also supports an optional property called json. This property allows you to include request body data in JSON format when necessary. By specifying the json property within the request field, you can provide structured data that needs to be sent along with the API request. Here’s an example illustrating the usage of the json property:

    - title: Create User - POST
      POST: /users/
      json:
        name: John Doe
        age: 25
        email: [email protected]
    

    In the above example, a POST request is made to create a new user. The json property contains the user data in JSON format, including properties such as name, age, and email.

    Including the json property in the request field enables you to pass structured data to the API endpoint, facilitating actions such as creating or updating resources on the server.

These properties in the request field provide flexibility and control over the API requests made during testing. You can specify the HTTP method and include headers as needed to interact with the API endpoints effectively.

asserts field

The asserts field in testkit plays a crucial role in defining assertions or validations to be performed on the API response. It allows you to specify conditions that must be met for the test to pass successfully.

The asserts field accepts a collection of key-value pairs, where the keys represent the type of assertion and the values define the corresponding expressions or conditions to be evaluated.

Here’s an example to demonstrate the usage of the asserts field:

- title: Fetches Todo items - GET
  GET: /todos/
  asserts:
    ok: $.resp.status == 200
    array: $.resp.json
    ok: $.resp.json[0].task == "run tests"

The .json tells testkit to convert the response into JSON format. This allows you to access properties of the response JSON using JSONPath expressions.

In the above example, we have defined three assertions:

  1. ok: This assertion checks whether the response status code is equal to 200. The expression $.resp.status == 200 is evaluated, and if it returns true, the assertion is considered successful.

  2. array: This assertion verifies that the response body is an array. The expression $.resp.json is evaluated, and if the result is an array, the assertion passes.

You can include multiple assertions within the asserts field to perform various validations on different aspects of the API response, such as checking specific properties, verifying the presence of certain data, or comparing values.

By utilizing the asserts field effectively, you can ensure that the API response meets the expected criteria, providing confidence in the correctness and reliability of your API. All possible assertions you could use in the asserts field of testkit are as follows:

  • ok: Checks if the provided expression evaluates to true.
  • empty: Checks if a value is empty (e.g., an empty array, string, or null).
  • array: Checks if a value is an array.
  • string: Checks if a value is a string.
  • number: Checks if a value is a number.
  • boolean: Checks if a value is a boolean.
  • null: Checks if a value is null.
  • exists: Check if a value exists
  • date: Checks if a value is a valid date string

These assertions provide a wide range of options to validate different aspects of the API response, allowing you to ensure the correctness and integrity of the data and behavior. You can select the appropriate assertion based on the specific validation requirements of your API test scenario.

exports

The exports field in testkit allows you to capture and store values from the API response of a stage for future reference within the test scenario. It provides a convenient way to extract specific data and make it accessible in subsequent stages of the test.

To use the exports field, you define key-value pairs where the keys represent the names of the exports (think of it as a variable), and the values define the JSON paths or expressions used to extract the desired data from the response.

Here’s an example that demonstrates the usage of the exports field:

- title: Fetches Todo items - GET
  GET: /todos/
  exports:
    todoItem: $.resp.json[0]._id

In the above example, the exports field captures the value of the _id property from the first element of the API response array. It assigns this value to the todoItem export.

By capturing the _id value in the todoItem exports, you can access it in subsequent stages of the test scenario. This allows you to use the extracted data for further API requests, assertions, or any other necessary operations.

The exports field enables you to create a bridge between different stages within the test scenario, providing a way to pass relevant data between them. This can be particularly useful when you need to refer to specific values or dynamically generate inputs for subsequent API requests.

Using the exports field, you can enhance the flexibility and modularity of your API tests, making them more robust and adaptable to different scenarios.

Referencing Values and Dynamic Inputs for Subsequent API Requests

The exports field in testkit not only allows you to capture values from the API response but also provides a powerful mechanism for referencing those values and dynamically generating inputs for subsequent API requests.

By capturing relevant data using the exports field, you can store it as an export and easily refer to it in later stages of your test scenario. This capability becomes particularly useful when you need to access specific values extracted from the response and utilize them in subsequent API requests.

For example, let’s say you retrieve an ID from an API response in one stage using the exports field:

- title: Fetch User - GET
  GET: /users/1
  exports:
    userId: $.resp.body.id

To reference this userId export in a subsequent API request, you can use the $.stages[n].<VAL> syntax:

- title: Update User - PUT
  PUT: /users/$.stages[0].userId
  json:
    name: 'John Doe'

In the above example, the userId captured in the first stage is accessed using the syntax $.stages[0].userId. testkit understands that it should substitute the reference with the corresponding value during execution.

You can also use relative references like $.stages[-n] which refers to the exports of the nth stage before the current stage. Example:

- title: deletes TODO items - DELETE
  DELETE: /todos/$.stages[-1].todoItem #-1 means one stage before me
  asserts:
    string: $.resp.json.task
    ok: $.resp.json.id == $.stages[-1].todoItem

By referencing specific values captured in previous stages, you can establish dependencies between different API requests and ensure seamless data flow throughout your test scenario. This flexibility allows you to build more comprehensive and realistic tests, simulating complex user interactions or workflows.

Date assertions

To make date assertions in Testkit you’ll need to provided the date string and the date format Example:

- title: Get User Profile - GET
  GET: /user/jon_doe
  asserts:
    date: $.resp.json.createdAt %Y-%m-%d %H:%M:%S %Z

As you can we first provide a json path to the date followed by the date’s format.

More on date format

Testkit uses the chrono crate’s formatting tokens to represent different components of a date. Here are some commonly used formatting tokens:

  • %Y: Year with century as a decimal number (e.g., 2023).
  • %m: Month as a zero-padded decimal number (e.g., 07 for July).
  • %b or %h: Abbreviated month name (e.g., Jul).
  • %B: Full month name (e.g., July).
  • %d: Day of the month as a zero-padded decimal number (e.g., 03).
  • %A: Full weekday name (e.g., Monday).
  • %a: Abbreviated weekday name (e.g., Mon).
  • %H: Hour (00-23).
  • %I: Hour (01-12).
  • %M: Minute (00-59).
  • %S: Second (00-59).
  • %p: AM/PM designation for 12-hour clock (e.g., AM or PM).
  • %Z: Timezone offset or name.

Examples dates and their formats

Here’s some example dates and their correct formats:

Date String Format
2023-07-26 %Y-%m-%d
2023-07-26 12:34:56 UTC %Y-%m-%d %H:%M:%S %Z
15 August, 1995, 03:45 PM UTC %d %B, %Y, %I:%M %p %Z
Mon, 05 Dec 2022 11:05:30 UTC %a, %d %b %Y %H:%M:%S %Z
January 01, 2000 - 00:00:00 UTC %B %d, %Y - %H:%M:%S %Z
1987/03/10 06:30 AM UTC %Y/%m/%d %I:%M %p %Z

In this table, the “Date String” column represents the example date string, and the “Format” column contains the corresponding format string to parse the given date string.

Using environment variables

Testkit supports environment variables in two ways: using a .env file or directly setting environment variables. These approaches allow users to configure and customize their test scripts without exposing sensitive data and making it easier to switch between different environments and scenarios seamlessly. Here’s how each method works:

Using a .env file involves creating a text file named .env in the test script’s directory and defining KEY=VALUE pairs for each environment variable. Testkit automatically loads these variables from the .env file during test execution.

Example .env file:

APIURL=https://api.example.com
EMAIL=[email protected]
PASSWORD=mysecretpassword
USERNAME=myusername
APIKEY=mysecretapikey

Setting environment variables directly is done via the command-line or the test environment.

Example command-line usage:

APIKEY=SECRETAPIKEY cargo run -- --file test.yaml

To utilize environment variables in Testkit, you can access them using the following syntax: $.env.<VAL>, where <VAL> represents the name of the specific environment variable you want to use. This allows you to easily reference and incorporate the values of these environment variables within your test scripts, enabling greater flexibility and adaptability without hardcoding sensitive information or configuration details.

Example:

- title: Register
  POST: '$.env.APIURL/users'
  headers:
    Content-Type: application/json
    X-Requested-With: XMLHttpRequest
  json: '{"user":{"email":"$.env.EMAIL", "password":"$.env.PASSWORD", "username":"$.env.USERNAME"}}'
  asserts:
    exists: $.resp.json.user
    exists: $.resp.json.user.email
    exists: $.resp.json.user.username
    exists: $.resp.json.user.bio
    exists: $.resp.json.user.image
    exists: $.resp.json.user.token

In this example, Testkit performs a POST request to the API URL specified in the environment variable APIURL. The user information for registration is taken from the environment variables EMAIL, PASSWORD, and USERNAME, allowing for easy customization and reusability of the test script across different environments.