Test Cases/Test Folder/Test Runs Question/s

[William Antjoule]

I have been a avid user of Rally for many years now and have really got to know the portfolio to task parts very well, but I have not had the opportunity to dive into the Quality part. 

I get the basic concept of a new user story could have a test case/s built and assigned to it, also this test case/s could be added to a test suite/folder. Then when the next user story is created that requires a new test case, this is also added to the test suite/folder.

What i don't get is how you would regress that the previous user story's still working when new story's/features have been built.also it seems that some say to copy a test case/folder which means there duplication of test cases and that does not smell right, as i would expect each validation that a prior user story/feature still works is a Test Run.

Many Thanks

Will

[Mitch Goldman]

Think about updating rather than adding/copying test cases. Then wrap the test cases up into a test set within a sprint.  Then when you need to re-run those (updated) test cases, they can be associated with a new test set. The history of the test set will persist and be visible from the individual test cases. The test set can note which user story/stories are reflected in that testing.

[Tyler Listoe]

I agree that this is the best way to do this and it is really handy to be able to build test sets and copy them to new iterations, and new test cases etc.

Mitch, is there any development in progress or planned to improve the quality side of this a bit more, particularly the "test run" tab of a test set? for those wishing to use this quality functionality while still conforming to ISO standards there it would be very handy to be able to get good controlled sap shots of test plans and test results as they existed at a particular point in time. after many iterations of test case changes it is extremely difficult if not impossible to reconstruct things. I think this is what the test run is meant to do but it is not very friendly to work with, filter, reformat, etc. 

Thanks,
Tyler

[Mitch Goldman]

@Tyler Listoe, that question would be better directed to a Broadcom representative, as I do not work for that firm.  But I agree, that some of the historical snapshot needs are not present in test sets. For example, if a test case was executed as part of a test set, but that test case was later deleted from the system, it will disappear from that historical test set without any noticeable indication that it was ever there.  Additionally, any changes made to a test case, such as to reflect new functionality, will propagate through back to the old test sets with which it was associated despite it not being part of that old testing. ​