On Monday 11 July 2005 23:17, Michael G Schwern wrote:
> On Mon, Jul 11, 2005 at 07:38:57PM +0300, Yuval Kogman wrote:
> > > So what should I do to eliminate it?
> >
> > Maybe Just Nothing
> >
> > The issue is that you can't special case get_current_coords to be
> > truish, as far as Devel::Cover is concerned - it might not be.
> >
> > Any fix that could be thought up is inherently problematic.
> >
> > Coverage reporting is not done for the pretty colors - a human reads
> > it, and says "OK, this is logical, get_current_coords always returns
> > a true value". It's not a race for greens and percentages.
>
> While I agree coverage is not a race, I disagree that a human should have
> to disambiguate between real missing coverage and a false negative. At
> least not more than once.
>
> I'll make the same argument "no broken windows" argument here that I do
> about warnings and tests: elminate all warnings, even if they are dubious.
> Ensure all tests pass eliminating all false negatives. Do not leave any
> "expected warnings" or "expected failures" because this erodes the
> confidence in the test suite. Warnings and test failures fail to ring
> alarm bells. One "expected" warning leads to two. Then four. Then
> finally too many to remember which are expected and which are not and you
> ignore them all together.
>
I fully agree with you. The purpose of the test coverage's main report page is
to let me know if I missed anything. And I want to make sure that I missed
nothing. Not to go over the problems one by one and see which one is a false
positive and which one isn't.
I wanted to say something similar along this line before your comment, but you
said it much better. :-)
Regards,
Shlomi Fish
---------------------------------------------------------------------
Shlomi Fish [EMAIL PROTECTED]
Homepage: http://www.shlomifish.org/
Tcl is LISP on drugs. Using strings instead of S-expressions for closures
is Evil with one of those gigantic E's you can find at the beginning of
paragraphs.
