run a test many times

[wissam ab]

basic question: how can I tell the tool to run the test many times and generate report result.
thanks

[Cody]

Hello wissam ab,

Thanks for asking. Just to better understand your goals, can you explain your purpose, what you hope to accomplish or learn by making the same test run multiple times?

The easiest method I can think of is to bind your test to a data source. Even though you may not actually be using any data from the data source, the test will run once/row that exists in the data source. You can use the internal data table for this, instead of an external data file (like a .CSV file). So, let's say there are 5 rows in the data source, the test will run 5 times in a row and report results for each iteration separately.

Regards,
Cody
the Telerik team

[wissam ab]

sorry, I wasn't clear maybe.
I don't need to bind data, since all what I'm doing is just the following scenario:
login webpage
go to the appropriate menu
click on a button (it call a webservice)
test if the result is ok (label)

my goal is to run this test 10,20,30 times to see if I have all the time ok from the webservice.
that's all.

thanks a lot

[wissam ab]

i forgot to say that this point is very very important to us.

[Cody]

Hi wissam ab,

Thanks for explaining in more detail.

In a data bound test, it is not required to actually use the data from your data source. It is perfectly OK to ignore the data and just let the test run N times (once per row in the data) while not actually using the data for anything. It's just a quick and easy way of making the test run N times. All you need to do is bind the test to any data source, put N rows of any data (junk/dummy data) in the data source then let the test run. If you want to run 30 times, put 30 rows of data in your data source. It doesn't matter what the data actually is since your test isn't going to actually use it.

Does that help clear up my idea of using a data bound test?

Another idea I just thought of is to use the "Test As Step" feature. In this approach create a new test (let's call it Test A) in which is nothing but 10, 20, 30 "Test As Step" test steps calling your real test (let's call it Test B). When you run Test A it will simply call your Test B however many times you repeat the Test As Step in Test A. Did I explain that clearly?

Sincerely yours,
Cody
the Telerik team