[vtk-developers] Use of std::cout in tests

[Bill Lorensen]

I agree.

On Thu, Sep 20, 2012 at 12:02 PM, David Gobbi <david.gobbi at gmail.com> wrote:

> I have to say that I agree with Marcus.  Tests must check for correct
> behavior.  That is really the purpose of a unit test, isn't it?
>
> Having a test that will pass even when the results are incorrect is
> bad, bad, bad.  It leads to false confidence in the code.
>
> If we allow tests that provide full coverage and zero valgrind
> defects, but which will pass when the code is giving incorrect
> results, then coverage becomes a meaningless metric.
>
> I consider an incorrect result to be a more serious defect than a segfault.
>
>  - David
>
>
> On Thu, Sep 20, 2012 at 9:41 AM, Bill Lorensen <bill.lorensen at gmail.com>
> wrote:
> > But, it could segfault for have valgrind defects...  It is a form of
> testing
> > to check for memory access and leaks.
>



-- 
Unpaid intern in BillsBasement at noware dot com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20120920/afe002e1/attachment.html>