多浏览器/多Driver测试场景下：基于ITestResult监听器识别失败Driver、定位故障浏览器及最近访问Driver的方法咨询

Great question! When dealing with multiple WebDrivers in a single test method—like your scenario where two browsers handle different user actions—pinpointing failures and capturing relevant screenshots can feel tricky. Let’s break down actionable solutions for each of your needs:

The core challenge here is that ITestResult doesn’t natively track WebDrivers used in the test. To fix this, you need to explicitly bind your drivers to the test context so the listener can access them later. Here’s how:

Option A: Store Drivers in TestNG’s TestContext #

TestNG’s TestContext lets you attach custom attributes to your test. You can use this to store all active drivers along with their metadata (browser name, user credentials):

@Test
public void multiBrowserApprovalTest() {
    // Initialize drivers with wrapped metadata (more on this below)
    WrappedWebDriver chromeDriver = new WrappedWebDriver(new ChromeOptions(), "Chrome", "user1@example.com");
    WrappedWebDriver firefoxDriver = new WrappedWebDriver(new FirefoxOptions(), "Firefox", "user2@example.com");

    // Perform test actions: user1 submits, user2 approves
    chromeDriver.get("https://your-app.com/submit");
    // ... submit logic ...
    firefoxDriver.get("https://your-app.com/approve");
    // ... approval logic ...

    // Store drivers in TestContext for listener access
    Reporter.getCurrentTestResult().getTestContext()
            .setAttribute("activeDrivers", Arrays.asList(chromeDriver, firefoxDriver));
}

Then in your ITestListener, retrieve the drivers and inspect their metadata:

public class TestFailureListener implements ITestListener {
    @Override
    public void onTestFailure(ITestResult result) {
        List<WrappedWebDriver> drivers = (List<WrappedWebDriver>) result.getTestContext()
                .getAttribute("activeDrivers");

        if (drivers != null) {
            for (WrappedWebDriver driver : drivers) {
                // You can log or capture screenshots for all drivers, or target the faulty one
                System.out.println("Checking driver: " + driver.getBrowserName() + " (user: " + driver.getUserCredentials() + ")");
                // Capture screenshot with metadata in filename
                captureScreenshot(driver);
            }
        }
    }

    private void captureScreenshot(WrappedWebDriver driver) {
        try {
            File screenshot = ((TakesScreenshot) driver).getScreenshotAs(OutputType.FILE);
            String fileName = String.format("failure_%s_%s.png", driver.getBrowserName(), driver.getUserCredentials());
            FileUtils.copyFile(screenshot, new File(fileName));
        } catch (IOException e) {
            e.printStackTrace();
        }
    }
}

Option B: Use a Custom Wrapped Driver Class #

Create a wrapper around WebDriver to embed metadata directly into the driver instance. This makes it easy to pull browser/user info whenever you need it:

public class WrappedWebDriver extends ChromeDriver { // Or extend RemoteWebDriver for grid
    private final String browserName;
    private final String userCredentials;

    public WrappedWebDriver(Capabilities capabilities, String browserName, String userCredentials) {
        super(capabilities);
        this.browserName = browserName;
        this.userCredentials = userCredentials;
    }

    // Getters for metadata
    public String getBrowserName() { return browserName; }
    public String getUserCredentials() { return userCredentials; }
}

To know exactly which driver caused the failure, you have two reliable approaches:

Track the Last Accessed Driver #

Maintain a thread-local reference to the most recently used driver. Every time you perform an action with a driver, update this reference. When a failure occurs, this reference points directly to the driver that was in use:

public class DriverTracker {
    private static final ThreadLocal<WrappedWebDriver> lastAccessedDriver = new ThreadLocal<>();

    public static void setLastAccessed(WrappedWebDriver driver) {
        lastAccessedDriver.set(driver);
    }

    public static WrappedWebDriver getLastAccessed() {
        return lastAccessedDriver.get();
    }
}

In your test, update the tracker every time you switch drivers:

// Using Chrome for user1
DriverTracker.setLastAccessed(chromeDriver);
chromeDriver.findElement(By.id("submit-btn")).click();

// Switching to Firefox for user2
DriverTracker.setLastAccessed(firefoxDriver);
firefoxDriver.findElement(By.id("approve-btn")).click();

Now in your listener, you can directly target the last accessed driver for screenshots and logging:

@Override
public void onTestFailure(ITestResult result) {
    WrappedWebDriver faultyDriver = DriverTracker.getLastAccessed();
    if (faultyDriver != null) {
        System.out.println("Failure occurred in: " + faultyDriver.getBrowserName() + " (user: " + faultyDriver.getUserCredentials() + ")");
        captureScreenshot(faultyDriver);
    }
}

Log Driver Context with Every Action #

Add log statements that include the driver’s metadata every time you perform an action. When a failure occurs, check the test logs to see which driver was active right before the error:

// Example log statement
Logger.info(String.format("Performing action with %s (user: %s)", chromeDriver.getBrowserName(), chromeDriver.getUserCredentials()));

• Clean Up Drivers Properly: Always quit all drivers in @AfterMethod to avoid leaks, even if a test fails.
• Use Concurrent Maps for Parallel Tests: If running tests in parallel, use ConcurrentHashMap to track drivers across threads safely.
• Leverage TestNG’s IInvokedMethodListener: For more granular control, use this listener to track driver usage before/after each method invocation.

内容的提问来源于stack exchange，提问作者Anand