Right way to cope with user error in make test?

[Rob Shearman]

On Fri, May 16, 2008 at 4:49 PM, Jeremy White <jwhite at winehq.org> wrote:
> I think in my case the problem is a bit different; I have the library -
> what I don't have are the development headers.  Thus the linking all
> works, but HAVE_LIBXSLT is not defined, so the transform function
> ends up as a stub, and goes on to fail.
>
> And I still don't know a right answer.
>
> The attached patch is one approach - skip the tests in this case.
>
> Does this seem like a reasonable approach to folks?

Alexandre has said in the past that test failures for an incorrectly
or incompletely built wine tree should result in test failures and I
agree with this.

Still, it would be nice to somehow say in an obvious way "you can
discard these results as the user didn't have the necessary packages
installed."

-- 
Rob Shearman