Retry analyzer for failures in Selenium Java - Framework Patterns

Framework Mode - Retry analyzer for failures

Folder Structure

src/
└── test/
    └── java/
        ├── com/
        │   └── example/
        │       ├── pages/          # Page Object classes
        │       │   └── LoginPage.java
        │       ├── tests/          # Test classes
        │       │   └── LoginTest.java
        │       ├── utils/          # Utility classes (e.g., RetryAnalyzer)
        │       │   └── RetryAnalyzer.java
        │       └── config/         # Configuration classes
        │           └── TestConfig.java

Test Framework Layers

• Driver Layer: Manages WebDriver setup and teardown for browser control.
• Page Objects: Encapsulate web page elements and actions (e.g., LoginPage.java).
• Tests: Test classes using TestNG with annotations and assertions (e.g., LoginTest.java).
• Utilities: Helper classes like RetryAnalyzer to retry failed tests automatically.
• Configuration: Holds environment settings, browser types, credentials (e.g., TestConfig.java).

Configuration Patterns

• Environment Properties: Use config.properties or Java classes to store URLs, credentials.
• Browser Selection: Pass browser type as a parameter or system property to run tests on different browsers.
• Retry Analyzer Setup: Configure retry logic in TestNG XML or via annotation @Test(retryAnalyzer = RetryAnalyzer.class).
• Credentials Management: Store sensitive data securely, e.g., environment variables or encrypted files.

Test Reporting and CI/CD Integration

• Use TestNG default reports or integrate with advanced reporters like Allure for detailed test results.
• Reports show retry attempts and final pass/fail status for flaky tests.
• Integrate with CI/CD tools (Jenkins, GitHub Actions) to run tests automatically on code changes.
• Configure build pipelines to fail only if tests fail after retries.

Best Practices for Retry Analyzer Framework

1. Limit Retry Attempts: Avoid infinite retries; set a max retry count (e.g., 2-3) to prevent long test runs.
2. Retry Only on Failures: Retry tests only when failures are due to flaky issues, not assertion failures.
3. Keep Retry Logic Separate: Implement retry logic in a dedicated utility class to keep tests clean.
4. Log Retry Attempts: Provide clear logs when retries happen to help debugging.
5. Use Annotations: Apply retry analyzer via TestNG annotations for easy maintenance.

Self Check

Where in this folder structure would you add a new retry analyzer class to handle test retries?

Key Result

Use a dedicated RetryAnalyzer class in the utils layer to automatically retry failed Selenium tests with TestNG.

Practice

1. What is the main purpose of a RetryAnalyzer in Selenium Java testing?

easy

A. To set up browser configurations before tests

B. To generate test reports after execution

C. To automatically rerun failed tests to reduce false failures

D. To locate web elements on the page

Solution

1. Step 1: Understand the role of RetryAnalyzer

   The RetryAnalyzer reruns tests that fail to handle temporary issues like network delays.

2. Step 2: Compare with other options

   Generating reports, locating elements, and setting browser configs are unrelated to retrying failed tests.

3. Final Answer:

   To automatically rerun failed tests to reduce false failures -> Option C

4. Quick Check:

   RetryAnalyzer reruns failed tests [OK]

Hint: RetryAnalyzer reruns failed tests automatically [OK]

Common Mistakes:

• Confusing RetryAnalyzer with report generation
• Thinking RetryAnalyzer locates elements
• Assuming RetryAnalyzer sets browser options

2. Which method must be overridden when implementing a RetryAnalyzer in Selenium Java?

easy

A. retry(ITestResult result)

B. onTestStart()

C. beforeMethod()

D. afterTest()

Solution

1. Step 1: Identify RetryAnalyzer interface method

   The IRetryAnalyzer interface requires overriding the retry(ITestResult result) method to decide if a test should rerun.

2. Step 2: Exclude unrelated methods

   Methods like onTestStart(), beforeMethod(), and afterTest() belong to other TestNG listeners or annotations, not RetryAnalyzer.

3. Final Answer:

   retry(ITestResult result) -> Option A

4. Quick Check:

   Override retry(ITestResult) method [OK]

Hint: Override retry(ITestResult) to rerun tests [OK]

Common Mistakes:

• Overriding wrong listener methods
• Confusing RetryAnalyzer with TestNG listeners
• Using before/after test methods instead

3. Given this RetryAnalyzer code snippet, what will be the output if the test fails twice and maxRetryCount is 2?

public class Retry implements IRetryAnalyzer {
  private int count = 0;
  private static final int maxRetryCount = 2;

  public boolean retry(ITestResult result) {
    if (count < maxRetryCount) {
      count++;
      return true;
    }
    return false;
  }
}

medium

A. The test will rerun twice after the first failure

B. The test will rerun only once after the first failure

C. The test will not rerun at all

D. The test will rerun three times after the first failure

Solution

1. Step 1: Analyze retry logic and count

   The retry method allows rerun while count is less than maxRetryCount (2). It increments count each retry.

2. Step 2: Count reruns for failures

   On first failure, count=0 < 2, rerun allowed (count=1). On second failure, count=1 < 2, rerun allowed (count=2). On third failure, count=2 not < 2, no rerun.

3. Final Answer:

   The test will rerun twice after the first failure -> Option A

4. Quick Check:

   maxRetryCount=2 means 2 reruns allowed [OK]

Hint: Count < maxRetryCount means rerun allowed [OK]

Common Mistakes:

• Assuming rerun count equals maxRetryCount + 1
• Confusing count increment timing
• Thinking no reruns happen

4. Identify the error in this RetryAnalyzer implementation:

public class Retry implements IRetryAnalyzer {
  private int count = 0;
  private static final int maxRetryCount = 3;

  public boolean retry(ITestResult result) {
    if (count <= maxRetryCount) {
      count++;
      return true;
    }
    return false;
  }
}

medium

A. maxRetryCount should be a non-final variable

B. The count variable should be static

C. The retry method should return void

D. The condition should be count < maxRetryCount, not <=

Solution

1. Step 1: Check retry condition logic

   The condition count <= maxRetryCount allows one extra retry than intended because count starts at 0.

2. Step 2: Correct condition for retry limit

   Using count < maxRetryCount ensures retries happen exactly maxRetryCount times.

3. Final Answer:

   The condition should be count < maxRetryCount, not <= -> Option D

4. Quick Check:

   Use < not <= for retry count limit [OK]

Hint: Use count < maxRetryCount to avoid extra retry [OK]

Common Mistakes:

• Using <= causes one extra retry
• Making count static unnecessarily
• Wrong return type for retry method

5. You want to implement a RetryAnalyzer that retries a failed test only if the failure is caused by a TimeoutException. Which approach is correct?

hard

A. Always return true in retry(ITestResult result) to retry all failures

B. Check the exception type in retry(ITestResult result) and return true only if it is TimeoutException

C. Use a separate listener to rerun tests instead of RetryAnalyzer

D. RetryAnalyzer cannot check exception types; use test annotations instead

Solution

1. Step 1: Understand RetryAnalyzer capabilities

   The retry method receives ITestResult, which contains the exception causing failure.

2. Step 2: Implement conditional retry logic

   Check if result.getThrowable() is instance of TimeoutException. Return true to retry only in that case.

3. Final Answer:

   Check the exception type in retry(ITestResult result) and return true only if it is TimeoutException -> Option B

4. Quick Check:

   Retry only on TimeoutException by checking exception type [OK]

Hint: Check exception type in retry method to conditionally retry [OK]

Common Mistakes:

• Retrying all failures without condition
• Using listeners instead of RetryAnalyzer for retry
• Assuming RetryAnalyzer cannot access exception info