Re: [PATCH 1/2 v2] mscoree: add stub for CorIsLatestSvc to hint user at broken .NET installation

[Vincent Povirk]

We should only have the E_POINTER test IMO. There's no way to know
what the function will do given a real pointer.

On Tue, Jul 10, 2018 at 10:07 AM, Louis Lenders
<xerox.xerox2000x(a)gmail.com> wrote:
> Well, there was a comment in the bugreport that it might be a BOOL, so I
> added test to show that it is a HRESULT;  that was the only reason to add
> the test
>
> Regards
>
> 2018-07-10 16:44 GMT+02:00 Vincent Povirk <vincent(a)codeweavers.com>:
> > > +    hr =  pCorIsLatestSvc(&unk1, &unk2);
> > > +    ok(hr == S_OK, "CorIsLatestSvc returned %08x\n", hr);
> >
> > Does it really make sense to test this, given that we have no way of
> > knowing the true function prototype and no reason to implement it?