Working with Test Sets - Best Practices?

[Keith Jones]

We'd like to start using Test Sets, but have some questions on the best way to go about this for our purposes. Any help on any of these items would be greatly appreciated!

 

1. Do you associate multiple Test Sets with a single Test Case? Seems like the system is set up for that, however the last verdict gets overwritten by the last test. So if you have two active Test Sets at the same time, both including the same test case, how do you keep all that straight? (as an example, 2 regression test sets that both utilize a test case for single sign-on).

 

2. Do you copy Test Sets from one sprint to another, clearing out test cases, or do you create a new Test Set for each sprint? 

 

3. Is there any way to assign a Test Case to a Test Set from within the Test Case? Or do you have to go to the Test Set and Choose different Test Cases?

[Sagi Gabay]

Hi Keith,

 

All are good questions. Here is a the help page on Test Sets. It should answer some of your questions:

 

1. You can associate the same Test Case to multiple Test Sets. Yet, as you pointed out, the "Last Verdict" belongs to the test case and will therefore reflect in all test sets. The idea here is that the actual exact test case instance is shared among test sets. You should use it this way only when you intend to run the same Test Case between the test sets. If you only intend to run similar instances but not exactly same then you shall make copies of your test cases in the different test sets, so that each test set has its own instance (copy) of the test case.

 

2. You can copy a Test Set between iterations/sprints. Keep in mind these are copies, it's not sharing the same test set. You can create a copy, then adjust it for you next iteration/sprint as you need.

 

3. Assignment of Test Cases to the Test Set is only done from the Test Set.  On the individual Test Case you have a read-only Test Sets tab that will simply show you the test sets which the test case belongs to.

 

Let us know if that helped.

 

Thanks,

Sagi

[Pat Cousins]

We use the Build field to differentiate between Test Results for parallel test efforts.  It is a freetext field, so the teams do need to be disciplined when filling it out.  You can then create a Custom List of Test Results and filter (or query) on specific Builds.  So the Test Set is primarily used for planning and tracking the work (effort), but we concentrate on the Test Results to know how the testing is actually going (pass/fail).