I'd like some answers to the following questions which was raised by my team lead, we're seriously considering purchasing telerik test studio but would first like to clarify on these points:
- Can a project be divided into several files to make management of the test suite “smaller”. For example, one “module” (Assets, Service Centre) per file, and then a master project brings them all together. The analogous concept in development is a library.
- What happens if the tests runs the test on a monitor with a different resolution to the one the tests were conducted with. For example, Talal’s second monitor is a 19 or 22 inch wide-screen whereas my monitor is a 17 inch non-widescreen, and both would have different resolutions and therefore the app would appear differently on each. I can see this having implications for concepts such as Image compare.
- We need to investigate ways to ensure some things are not present on the screen. If the automated tests are going to target those kinds of things we would want to check that certain menus (for example) have not appeared, or certain buttons have not appeared on a dialog, etc.
- Parameterisation of certain test sequences so they are reusable – the example was the login sequence where we could supply the name & password as parameters and therefore re-use the same sequence for many different logins – because multiple logins would be needed to test different permissions scenarios..
- I have a situation where I need to login, if the login succeeds the next screen is loaded isntantly, if it fails the next step of the test is to inform the user that the login step failed. How should we go about testing for a failed login, currently what we do is we check for a textblock being created that says the loginf ailed, however if the login succeeds that test stil executes and sicne we're in a different screen now the test effectively becomes broken.
5 Answers, 1 is accepted
Thank you for taking the time to evaluate Test Studio. I will endeavor to answer all your questions.
- Can a project be divided into several files...? Definitely and we highly encourage this! What you want is our Test-as-step feature.
- What happens if the tests runs the test on a monitor with a different resolution...? Nothing significant. We do not use screen coordinates for locating elements and clicking. We're using element properties and find expressions to locate elements. As a result the test automatically adjust to varying screen resolutions.
- We need to investigate ways to ensure some things are not present on the screen. You will need to use our Wait For (or Verify) Not Exists feature. The approach is to record this step when the element does exist, then make sure your test is correctly updated so that the element does not exist at the correct step of your test.
- Parameterisation of certain test sequences so they are reusable... You have two options for this:
a) InheritParentDataSource This allows the data bound to a parent test to automatically flow into a child test.
b) Extract an element property to a variable and bind the file to a follow on test step.
- I have a situation where I need to login... I would turn your verification around i.e. verify something like the logout button appears. This verifies the login is successful. If the login was not successful the logout button will not be present and the test will correctly fail.
I hope these answers are satisfactory to you. Don't hesitate to contact again if you have followup questions.
Cody
the Telerik team
Rank 1
I did what the linked described but am not able to convert 'Wait - element does not exist' to a verification as the image shows.
Is the Set As Wait (as shown in this screen shot) disabled or something similar? Can you send me a screen shot of what you do see in that context menu? Also can you send me that test (the .tstest, .resx and if applicable the .cs/.vb file) so I can try and repro the problem here and figure out what may be causing it?
Regards,Cody
the Telerik team
Test Studio Trainings
Rank 1
I've attached screenshots of what I'm seeing.
You have just uncovered a bug for us, thanks! I've created a bug report on this here. I have granted you some Telerik points for being the first to report this issue.
The good news is that it's not really important in this instance. Leaving that test step as a "Wait For" instead of "Verification" is just as useful. The difference is:
- Wait For will keep trying for up to 10 seconds and then give up, abort and fail the test
- Verification will try just one time. If the verification fails the test will immediately abort and the test will be set to Fail.
Cody
the Telerik team
Test Studio Trainings