by drew » Mon Feb 12, 2007 3:22 pm
It's important so that you can track the test cases, between multiple databases, products, and departments. Lets say you have a database of test cases with ID's the way they are assigned in Test Link, and you have a group of people who are working on automating the test cases. But then as you create a new product you have to copy the test cases from the old product to the new one, as the new one will have additional test cases for the new features. Now you have a list of test cases that you ran against version 1 of your product, and a new list with some of the old test cases, some new test cases and some of the old test cases removed for version 2 of the product. Since you now have a completely different ID# for the same test case in two products you have no way of keeping track of them both (other than by name) And now the people automating the test cases don't know if the cases they automated are the ones in the new product, or the old one, or both, or neither. Now when version 3 of the product comes out you are going to have a whole new disconnect between the versions. Because now with the current ID numbering scheme in Test Link you will potentially have 3 copies of the same test case across the products with three different ID# and no way of telling if it's the same test case.
That's why you need to restart the test case numbering for each product, or at least offer that as an option when the user goes to copy the test cases. Either that or a unique field in the test cases itself that will allow to assign test case ID # in a variety of ways.
Bottom line is it offers the people using test link much more flexibility with the product and helps them fit test link to their needs, instead of having them tailor their needs around test link. It seems that this is very commonly requested feature, and I would like to have it too.