In general, this solution will be useful to anyone who wants to test consistently and in a given order.

It creates a suite, where in the annotation we list the classes in the order of execution:

@RunWith (Suite.class)
@ Suite.SuiteClasses ({OpenConnection.class,
                 GetServerIdTest.class,
                 ModbusStatusCodesTest.class,
                 ModbusSerialTransactionTest.class,
                 CloseConnection.class})
public class OrderedTestSuite {}

Add the plugin to pom.xml:

<plugin>
         <groupId> org.apache.maven.plugins </groupId>
         <artifactId> maven-surefire-plugin </artifactId>
         <configuration>
             <includes>
                 <include> ** / OrderedTestSuite.class </include>
             </includes>
         </configuration>
</plugin>

These exceptions are actually just warnings that the the selenium server was kept alive, this is the timeout protection doing its job.

The console will contain messages like that and its not a problem.

You could remove the log message or filter them from the console if its bothering you.

see here for how to send those log messages to a file https://sites.google.com/a/chromium.org/chromedriver/logging

I have made a website. I visited this website using different mobile browsers and I am getting an error related to visual display and button click event.

Do I need to test with a large number of devices, or will using emulators be enough?

I'm using a baseUrl in the config file and some tests access links that have prepended/appended text to the same url that needs to be verified. How can I verify the clicked url while using the baseUrl?

Example:

baseurl: dev.test.com

Link to be verified

this-dev.test.com/#/1234

'this' will not change but the numeric values will. The environment will change in the baseUrl (dev vs staging, etc).

I am trying to handle the dynamic path of an HTTP request using JMeter. But I am not able to do so. I have tried post processor in a previous HTTP request to get the next HTTP request dynamic path but it is not working. Please guide me.

Dynamic path:

The complete path looks like as below mentioned an example.

abc.c_o_m/en/book/confirm/VZM5O/ARL715?tfk=tfk59771a9df0fbd4.08848159

Post Regular expression:

After execution, I have checked in debug sampler but ${Path} contains only default value i.e. 0.

If you want all tables then just grab the tables like findElements(By.xpath("//table")); and pull all the tables back. You can step through the list and then adjust to the table ID in question and add the tr for a new list of rows like findElements(By.xpath("//table[@id='table_id']//tr")).

In general that looks like what you are trying for. Not all tables have the tbody tag in them so it's best to just xquery search // inside the table for each row/cell (tr/td) that you want and avoid using full xpath.

Any language is a tool to solve certain kind of problems, so (if you allow me to slightly exaggerate) you are talking between which tool you should buy: a hammer or a screwdriver?

Every such selection will have pros and cons, so only you can decide which pros are more important for you.

But in common situation like yours, I would advice to go for Python, without any hesitation.

Both languages are similar in sense they are both dynamically typed with flexible data literals (as compared to Java) and both are widely used for web UI automation (which is more than just automated web regression testing).

Python has more wide use outside of web automation, for different tasks you may encounter around testing, for generic system administration tasks like moving files around, writing cron jobs, and just general text files parsing and manipulation, like generating test data.

OTOH Javascript is THE language for UI development, Angular, Protractor, etc. I think this is smaller niche than Python, because you ALWAYS need to manipulate the files.

Also because Python has more careful design more oriented towards reading programs by humans, Python community finds it easier to collaborate and share the code. With Javascript, it is easier to find low-quality examples of the code. It is not as bad as PHP but close. Javascript is closer to Lisp than to Java, because it is prototype-based, not object-oriented, and it's name is just marketing gimmick.

Python is object-oriented (closer to Java) but not as object-obsessed as Java is (you can also program in procedural style, or functional if you want and need), but there are some subtle differences with Java too.

I am trying to Login to Ebay site using Selenium. I have successfully identified the user name and password Text boxes using Xpath (and validated the xpaths using the Firefox Eval tool in Firebug).

public class LoginTestCase{     
    @Test
    public void LoginEbay() throws InterruptedException 
    {
        System.setProperty("webdriver.gecko.driver", "D:\\selenium_library\\geckodriver-v0.17.0-win64\\geckodriver.exe");
        WebDriver driver = new FirefoxDriver();

        driver.manage().timeouts().implicitlyWait(20, TimeUnit.SECONDS);

        //navigate to Ebay
        driver.get("https://signin.ebay.com/ws/eBayISAPI.dll?SignIn&ru=https%3A%2F%2Fwww.ebay.com%2F");

        Thread.sleep(20000);
        // Enter user name , password and click on Signin button
        driver.findElement(By.xpath("//input[@placeholder='Email or username' AND @type='text']")).sendKeys("####@###");            
        //driver.findElement(By.xpath("//input[@placeholder='Password' AND @class='fld']']")).sendKeys("####");
        driver.findElement(By.id("sgnBt")).click();
    }
}

I am getting the below warning after which the script Fails.

1499976357508   geckodriver INFO    Listening on 127.0.0.1:41809
1499976358783   geckodriver::marionette INFO    Starting browser C:\Program Files (x86)\Mozilla Firefox\firefox.exe with args ["-marionette"]
1499976367184   Marionette  INFO    Listening on port 62686
Jul 13, 2017 4:06:20 PM org.openqa.selenium.remote.ProtocolHandshake createSession
INFO: Detected dialect: W3C
JavaScript warning: https://signin.ebay.com/ws/eBayISAPI.dll?SignIn&ru=https%3A%2F%2Fwww.ebay.com%2F, line 170: unreachable code after return statement

I would go another way.

1. Set test categories for the given tests (e.g. "QA", "DEV", "PROD")
2. Create a CI job for NUnit test execution for given category on a specific environment. 1 category - 1 job.
3. If needed - create a master job which will trigger the child (environment-specific jobs) and this will allow you execute the tests against your environments at once.

I think the code itself should be environment agnostic and condition logic depending on environment variables will reduce the readability and increase maintenance effort. Creating the jobs on the CI is not a silver bullet as well, but there is always a some point of a trade-off.

Pretend you are going to lend a toy (a teddy bear) to a friend 'cos they really want it.

Before you lend the toy you want to make really sure it will make your friend happy.

You have to have dinner first though so you need to make a note of it to remind you to come back to it after dinner.

Dinner is eaten.

You know your friend really likes golden yellow teddy bears, especially those with soft fur.

You check the teddy and it definitely is golden yellow and has soft fur.

You lend it to your friend.

That should cover: requirements, ticket, testing and verification

So the trick is to view the universe from her point of view - and that is a great lesson for the adult world and seeing a business from a customers point of view rather than the businesses.

You may also find the following book helpful:
I am a Bug – July 31, 1999 by Robert Sabourin (Author)
https://www.amazon.com/dp/0968577407

My organization is starting up an automation team, and we are still working on our structures and procedures.

We are using Git to store automation code, and we seem to be of two minds; some feel we should put all our automation code in one repo, and others feel we should separate code into multiple repos.

Here's the wrinkles: Much of the automation we will write is not for AUTs. It is for mission-critical third-party software, the actions of which may induce certain internally designed processes to do certain things. Most of these are integration layers between diverse third party applications. The overall software ecosystem is very large with diverse and mostly third-party applications. As a result, there won't be hardly any shared code between the frameworks for each application.

So far the best argument for single-repository design is that we can run all tests at once from one repo. There is no scenario, though, where we would run all tests at once.

I am proposing separate repos for each application we will automate, with an additional "common" repository for whatever few things we can share.

Also, there will be 6 automation developers, most of the time each will be the only one working on a particular application, for the small ones; but for the larger applications, there will likely be two or more working on it.

Do you agree with my camp of having separate repos? Can you give me pros and cons?