Functionality question

[Conor]

As part of my review of Telerik Test Studio (trying to determine which automated testing suite to recommend to the company), I have a question about functionality that I'm unable to answer using the user guide or forums (though perhaps I missed it, the documentation is pretty comprehensive).

I have certain tasks that will become part of a LOT of our tests. Logging in, entering notes on a standard window, etc.

I'm concerned, that if these steps are part of so many tests, and we change the functionality, I'll have a lot of places that the scripts need to be changed.

Is it possible to have scripts linked together (following each other) so that I can have common tasks recorded/coded once, and call them from within multiple tests?

For example, call the generic login script, call a script that does common navigation, call a "this test specific test", call the common "create a note", then call a a "run a report specific to this test", then the common log off / close browser test.

So far, I've only been able to create tests that start with the browser closed and end with the browser closed (our application doesn't allow two instances from the same computer, so I always have to log out before starting the next test)

Any help you can provide would be appreciated!

Thanks,

[Cody]

Hello Conor,

Thank you for taking the time to evaluate Test Studio!

"I have certain tasks that will become part of a LOT of our tests. Logging in, entering notes on a standard window, etc."

This is precisely the reason we added our Test-as-step feature a long time ago. This allows you to modularize your tests, much like functions/subroutines of a software program. Use this feature to create smaller subtests that do just the individual task (logon, enter note, etc.) and then call that module from the parent test. Now if your login sequence changes or the note page changes, you have just one test to update it all the parent tests will automatically use the updated subtest.

There's also no technical limit on how deep you can go i.e. a subtest can call another subtest which can another subtest and so on. For practical reasons (being able to keep track of the design of your test project) it's a good idea to limit the number of levels to about 3. More than that and the complexity becomes more than most people (myself included) can keep track of.

I hope that answers your questions.

Regards,
Cody
the Telerik team

[Conor]

Excellent, thanks so much!