How to Rerun Failed Test Cases in Selenium using Cucumber Framework

Introduction 

One of the essential aspects of test automation is the ability to rerun failed test cases automatically. In the Cucumber Framework, rerunning failed tests helps in improving test efficiency and reducing manual intervention. In this blog will help readers understand how to efficiently rerun failed test cases, improving their automation testing process.

Why Rerun Failed Test Cases

Test cases may fail due to intermittent issues like:

- Temporary environmental problems

- Network latency

- Third-party service downtime

Rerunning only the failed test cases allows you to:

- Save time by not running the entire test suite again.

- Focus on analyzing the failed scenarios.

Steps to Rerun Failed Test Cases in Cucumber Framework

Step 1: Add Rerun Configuration in @CucumberOptions

To rerun failed tests, you need to configure the @CucumberOptions annotation in your Test Runner class. This configuration will allow you to capture failed test cases in a file.

To learn more, explore the tutorial on how to configure the @CucumberOptions annotation in your automation framework  

 Step 2: Create a New Runner for Failed Tests

Once the failed scenarios are captured in the `failed_scenarios.txt` file, create a new Test Runner class to rerun those failed test cases.

Here’s how you can create a separate runner for the failed scenarios:

import io.cucumber.junit.Cucumber;
import io.cucumber.junit.CucumberOptions;
import org.junit.runner.RunWith;

@RunWith(Cucumber.class)
@CucumberOptions(
        // Path to failed scenarios file
        features = "@target/failed_scenarios.txt",
        glue = "stepDefinitions",
        plugin = {
                "pretty",
                "html:target/cucumber-reports-failed.html"
        }

        public class FailedTest Runner extends AbstractTestNGCucumberTests{
        @Override
        @DataProvider(parallel = false)
        publicObject[][]scenarios(){
        returnsuper.scenarios();
        }
        }
        

rerun: This option points to a file (`target/failed_scenarios.txt`) where failed scenarios will be stored

plugin: The `plugin` option is used to generate a file (`failed_scenarios.txt`) that contains all the failed scenarios after running the tests.

features = "@target/failed_scenarios.txt": This ensures that only the failed scenarios are picked up and rerun.

scenarios() method in the FailedTestRunner class is an overridden method from the AbstractTestNGCucumberTests class. This method is used as a @DataProvider for TestNG. It returns a two-dimensional array of Object (Object[][]), where each Object[] can be seen as the scenario to be executed and its data. 

@DataProvider(parallel = false) annotation indicates that the scenarios should not be run in parallel. 

Separate Reports: Generating a separate report (`cucumber-reports-failed.html`) helps you analyze the failures effectively.

Step 3: Execute the Tests

Now that you have both the original test runner (`TestRunner`) and the failed test runner (`FailedTestRunner`), here’s the execution process:

1.Run the TestRunner class: This will execute all the scenarios, and if any tests fail, they will be logged in the `failed_scenarios.txt` file.

2.Run the FailedTestRunner class: This will pick up the failed scenarios from the `failed_scenarios.txt` file and rerun them.

Conclusion 

rerunning failed test cases in the Selenium Cucumber framework is an important way to improve test reliability and ensure better coverage. Using tools like @CucumberOptions, TestNG, or JUnit retry features, you can manage flaky tests and make your automation more stable. This approach helps catch temporary issues, reduce false test failures, and keep your test results accurate and reliable.