The upcoming release of TestSprite v1.0 is scheduled for Oct. 2024. Read more →
Backend (APIs) Testing

TestSprite Documentation: Back-End (APIs) Testing

Overview

Welcome to TestSprite's back-end testing feature! Our goal is to simplify API testing and help you efficiently cover a wide range of scenarios. With TestSprite, you can easily perform end-to-end tests for your RESTful APIs, ensuring reliability and performance without the usual headaches. Whether you are a small team or an enterprise, our tool is designed to meet your testing needs with speed and precision.

Key Features

  • Fast and Easy Setup: Get started quickly with minimal documentation—no need for extensive prompts or full codebases.
  • Comprehensive Test Types: Supports a wide range of tests, including functional testing, error handling, security, authorization & authentication, boundary testing, load & performance testing, edge case testing, response content validation, and concurrency testing.
  • Concise Reports: Delivers detailed outputs, including error types, potential root causes, and recommended fixes.
  • Natural Language Feedback: Easily provide feedback on test results and adjustments using natural language.

Getting Started

To begin using TestSprite for back-end testing, follow these steps:

Step 1: Set Up Your API Testing Environment

  1. Log in to your TestSprite account and go to the Overview section.
  2. Click on Create New Test, which will automatically proceed to the Back-End (APIs) Testing step.
  3. Select the appropriate Authentication Type for your API and enter the API endpoint/URL.
  4. Provide any necessary Credentials/Keys, if applicable.
  5. Upload your API documentation or use natural language to describe the API you wish to test, including details like input and output formats (API docs serve the same purpose).

Example API documentation: https://docs.genrex.com/docs/1.0/api-reference/request-generation (opens in a new tab)

Step 2: Review Generated Test Plans

  1. Review AI Generated Test Plan:

Our AI testing agent generates a comprehensive test plan that includes a variety of test types, such as:

  • Functional Testing
  • Error Handling Testing
  • Security Testing
  • Authorization & Authentication Testing
  • Boundary Testing
  • Load & Performance Testing
  • Edge Case Testing
  • Response Content Testing
  • Concurrency Testing
  • And more...

You can expand each test type to see the detailed scenarios that have been created specifically for your software application.

{
  "example_test_plan": [
    {
      "category": "Authorization Tests",
      "scenarios": [
        "POST request without 'gx-key' header",
        "POST request with expired timestamp in 'gx-signature'",
        "POST request without 'gx-signature' header"
      ]
    },
    {
      "category": "Basic Functional Tests",
      "scenarios": [
        "POST request with valid 'duration' and 'text' parameters",
        "POST request with minimum allowed 'duration' (5 seconds)",
        "POST request with maximum allowed 'duration' (60 seconds)"
      ]
    },
    {
      "category": "Boundary Tests",
      "scenarios": [
        "POST request with empty 'text' field",
        "POST request with 'duration' set to 61 seconds (above maximum threshold)",
        "POST request with 'duration' set to 1 second (below minimum threshold)"
      ]
    },
    {
      "category": "Error Handling Tests",
      "scenarios": [
        "POST request leading to a 500 Internal Server Error",
        "POST request handling non-JSON response format",
        "POST request expecting a network timeout"
      ]
    },
    {
      "category": "Format Tests",
      "scenarios": [
        "POST request with 'Content-Type' other than 'application/json'",
        "POST request with additional unexpected parameters in payload",
        "POST request with incorrectly formatted JSON payload"
      ]
    },
    {
      "category": "Payload Validation Tests",
      "scenarios": [
        "POST request with missing 'text' parameter",
        "POST request with missing 'duration' parameter",
        "POST request with 'text' exceeding 250 characters"
      ]
    },
    {
      "category": "Signature Validation Tests",
      "scenarios": [
        "POST request with tampered payload leading to invalid 'gx-signature'",
        "POST request with 'gx-signature' generated from incorrect API secret",
        "POST request with invalid 'gx-signature' format"
      ]
    }
  ]
}
  1. Select Test Categories and Scenarios:
    Based on your needs, you can choose the test categories and cases you want the AI to implement and execute for your APIs. You can also manually add test cases using natural language, but we recommend selecting all test cases to ensure comprehensive coverage for your software.

Step 3: Run Your Tests

  1. Click Next to initiate the execution of your selected test plan.
  2. TestSprite will automatically generate the required test code, execute them, and analyze the results.

Step 4: Review Test Results

After running your tests, TestSprite displays a detailed execution report filled with insights and actionable recommendations to help you refine your software. You can also interact with our integrated chatbot to provide feedback, request adjustments to your test code, or ask any questions about the results.

Each failed test case includes a Detailed Analysis section with four key tabs that offer deeper insights:

  • Error: A clear description of the issue encountered during the test, helping you understand what went wrong.
  • Trace: A full code stack traceback of the error, allowing you to see exactly where and how the issue occurred.
  • Cause: A thorough analysis of potential causes for the failure, using AI to identify root causes and contributing factors.
  • Fix: AI-generated suggestions for resolving the issue, including code snippets or configuration adjustments to help you quickly address the problem.

Step 5: Generate Test Reports

TestSprite automatically generates a comprehensive test report from the executed test cases, providing actionable recommendations to optimize your application. The report is easy to share with your team, helping to enhance future testing efforts.

Advanced Configuration

Using Natural Language for Feedback

You can provide feedback using natural language to refine your test results—no specific format is required. For example:

"Test POST /orders with invalid parameters and expect a 400 error code."

TestSprite will automatically interpret and update the corresponding test case, making the testing process smoother and more efficient.

Examples

Example 1: Generated Test Code

import hashlib
import hmac
import json
import pytest
import requests
import time
 
# Define the API URL and credentials (use environment variables for added security)
api_url = "https://your-api-url.com/v1/text2music/generateMusic"
api_key = "hide_for_privacy_protection"
api_secret = "hide_for_privacy_protection"
 
def create_signature(api_secret, data_to_sign):
    return hmac.new(api_secret.encode(), data_to_sign.encode(), hashlib.sha256).hexdigest()
 
def test_invalid_gx_signature():
    # Construct the payload
    payload = {
        "duration": 10,
        "text": "intense EDM",
    }
    payload_json = json.dumps(payload, separators=(",", ":"))
 
    # Create correct signature
    timestamp = str(int(time.time() * 1000))
    data_to_sign = f"{timestamp}.{payload_json}"
    correct_signature = create_signature(api_secret, data_to_sign)
 
    # Tamper the payload
    tampered_payload = payload_json.replace("intense EDM", "soft jazz")
 
    # Use correct timestamp and an intentionally incorrect signature
    tampered_signature = create_signature(api_secret, f"{timestamp}.{tampered_payload}")
 
    # Create headers with tampered payload
    headers = {
        "gx-key": api_key,
        "gx-signature": f"t={timestamp},v={tampered_signature}",
        "Content-Type": "application/json",
    }
 
    # Send POST request with tampered payload
    response = requests.post(api_url, data=tampered_payload, headers=headers)
    
    # Parse the response
    response_data = response.json()
 
    # Assertions
    assert "statusCode" in response_data, "Expected 'statusCode' in the response"
    assert response_data["statusCode"] == 400, f"Expected statusCode 400, got {response_data['statusCode']}"
 
test_invalid_gx_signature()
Get StartedFrontend (UI) Testing