# from David Cantrell
# on Thursday 30 October 2008 09:11:
>> It would certainly be for the best if those distributions contained
>> as much useful data as possible, and that certainly does include
>> author tests. I seem to remember that putting them in xt/ was the
>> consensus, and that the cpantesters tools would ignore them.
>
>The cpantesters tools are just wrappers around the normal test
>harnesses. It's those that would need to ignore xt/. Which, I
> believe, is what they do anyway.
Yes. To be precise, they ignore everything which isn't 't/'.
(Though, one day, I would like for them to selectively pay attention to
the contents of xt/ based on some sort of standard declaration of
requirements for "eXtra Testing".)
--Eric
--
"...our schools have been scientifically designed to
prevent overeducation from happening."
--William Troy Harris
---------------------------------------------------
http://scratchcomputing.com
---------------------------------------------------
