Web Test Automation Using Python, Pytest, and Selenium

Introduction

Automation testing is a critical part of software quality assurance, ensuring applications work

as expected across different environments. In this blog, I will share insights into our

Selenium test automation framework using Python and Pytest, with a focus on:

 Pytest adoption

 Page Object Model (POM) implementation

 Data-driven testing

 Custom reporting

Pytest Adoption for Flexible Test Execution

We leveraged Pytest to create a dynamic and scalable test framework. Some key benefits

include:

Dynamic Execution

 Test cases run based on predefined conditions, allowing flexibility.

Command-Line Configurations

 We can pass browser type (Chrome, Edge, Firefox) and environment (staging,

production, etc.) as arguments, making the test execution process more adaptable.

Conditional Execution

 Test cases can be enabled, skipped, or executed with different configurations

depending on runtime parameters.

Page Object Model (POM) for Better Maintainability

To enhance test structure and maintainability, we implemented Page Object Model (POM),

which promotes code reusability and separation of concerns.

Reusable Utility Layer

 A base layer handles core interactions with web elements, reducing redundancy.

Page-Specific Methods

 Each web page has a dedicated Python file that encapsulates UI interactions and

business logic.

Cross-File Method Usage

 Some methods are shared across multiple pages and test cases, ensuring

consistency.

Structured Test Execution

 Test scripts in the test folder call functions from the pages folder, keeping test logic

organized and improving maintainability.

Implementing Data-Driven Testing with Excel

To increase test coverage and reusability, we incorporated data-driven testing by

fetching test data dynamically from an external source.

Excel-Based Test Data Storage

 Test inputs are stored in an Excel sheet, making it easy to modify data without

altering the test script.

Reusable Data Extraction Method

 A common function extracts data, ensuring it can be used across multiple test cases.

Dynamic Execution Based on Input Data

 By modifying the Excel file, new test scenarios can be executed without updating the

script.

Custom Reporting for Better Test Insights

Pytest’s built-in reporting is minimal, so we implemented a custom reporting system for

improved visibility into test results.

Capturing Detailed Test Outcomes

 We store individual test scenario results separately for better tracking.

Storing Results in JSON Format

 Passed and failed test cases are recorded in a structured JSON file.

Converting JSON to HTML Reports

 Extracted data is formatted into an interactive HTML report for easy review.

Interactive HTML Report Features

Graphical Insights – Pie charts visualize the pass/fail ratio.

Module-Based Summaries – The report categorizes tests by modules to track progress.

Drill-Down Test Details – Clicking on test counts reveals specific test case results,

making debugging easier.

Optimized Test Execution with Parallelism

By enabling parallel execution in Pytest, we significantly reduced test execution time and

improved efficiency.

Challenges and Learnings

Handling Dynamic Web Elements

 Implemented robust wait mechanisms to improve test stability.

Cross-Browser Execution

 Enabled tests to run on multiple browsers via command-line arguments.

Scalability & Maintainability

 The POM structure and data-driven approach reduced maintenance efforts and

improved efficiency.

Multi-Instance Browser Testing

We implemented a unique multi-instance testing approach where:

1️⃣ Instance 1 launches Browser 1, modifies certain conditions.

2️⃣ Instance 2 launches Browser 2, validates the changes, and then closes.

3️⃣ Instance 1 continues testing with Browser 1, ensuring smooth validation across different

instances.

Conclusion

Our Python-Pytest-Selenium automation framework has become a key part of our

release validation process, ensuring reliable, efficient, and scalable testing.

Pytest adoption enabled dynamic execution, browser selection, and environment

configuration from the CLI.

Page Object Model (POM) improved test maintainability by separating UI interactions

from test logic.

Data-driven testing with Excel enhanced flexibility and test coverage.

Custom reporting provided structured, interactive insights for better debugging and

tracking.

Multi-instance browser testing allowed for cross-instance validation of changes

dynamically.

By leveraging these strategies, we have streamlined automation workflows, improved

reporting accuracy, and enhanced test efficiency.