Testing Related Groups of Tables - Gnorion/BizVR GitHub Wiki

You might for example decide that you want to test these two tables as a group since there is a dependency between them

You can think of this as a self contained "min-decision" that just has two inputs: A and B

If you run the validation specifying only the inputs A and B then you will get a much smaller set of test case combinations (since C and D are irrelevant) Now the test cases are

Based on the rules in table T1 ——-need image—-

However there is a much deeper issue with this "mini-decision". The DRG seems to say that T2 depends only on T3 since the only arrow going into T2 is from T3.

Unfortunately the DRG is wrong. The author forgot to add some other dependencies to the DRG.

If you look inside table T2 you can see the following

T2 actually uses three variables - one from T3 but also variables that come from T5 and T11

These dependencies are easily seen in the BVR view of the same decision:

But wait... there's more. T2 is testing T11 as a condition but where does it come from? Its not defined as an input and there is no table that creates it as an output.

So there is something seriously wrong with the model.

In fact you can see in the BVR view that variable T11 is also tested in two other tables: T10 and T11

Most likely the rule author wrote the wrong variable names in his tables - so that has to be fixed before you can even begin testing.

Oh and one more thing... Table T12 doesn't seem to contribute anything to the overall decision despite what the DRG says. In the BVR diagram its not connected to anything.

Follow this link to see a version that resolves these issues (TODO)