Re: [Asunit-users] Testing Interfaces

[Paul B. <eye...@gm...>]

I think the confusion here is in the way the article is written...

what I read from it is not that I need to test an interface, but I
need to make sure that when I test a class that implements an
interface I should make sure that the tests confirm the functional
intention of the interface...

The way he does this is to say write an abstract class which
correlates to the *intention* of the interface, and write tests
against that, then with the specific implementation of the class,
write tests against that.

So, I think what John Mark was actually asking is:
"How do run the test methods written against a superclass in by subclass te=
st"

which I think, Peter, you looked at over here (although the example I
found relies on your event driven implementation of asunit)....



On 5/26/06, John Mark Hawley <ma...@ni...> wrote:
> rationale: http://www.placebosoft.com/abstract-test.html
>
> Basically, while the compiler tests that interfaces are correct syntactic=
ally, having a set of tests to run on an interface means that you can verif=
y that each interface method is actually doing what it should be in a coher=
ent, meaningful way.
>
> Without an interface test, I can be sure that HamSandwich implements IFoo=
d, but I can't be sure the IFood method eat() doesn't turn the HamSandwich =
into a tree, when for all other IFoods eat() causes the IFood to dissappear=
.
>
> -john mark hawley
>
> Why are you testing interfaces?
> Perhaps you can identify some kind of value that I can't see?
>
> Peter
>
>
>
>
>
> _______________________________________________
> Asunit-users mailing list
> Asu...@li...
> https://lists.sourceforge.net/lists/listinfo/asunit-users
>