Re: [code-review] Please review String::FlexMatch

Fergal Daly <[email protected]> Sun, 7 Sep 2003 23:26:04 +0100
Newsgroups gmane.comp.lang.perl.code-review-ladder
Message-ID <[email protected]>
On Friday 05 September 2003 14:43, Tony Bowden wrote:
> On Fri, Sep 05, 2003 at 11:58:39AM +0200, Marcel Gr?nauer wrote:
> > I completely agree. Do you think the extra line
> > in Test::More::_deep_check(), namely
> >   defined(overload::Method($_, '""')) && $_ = "$_" for $e1, $e2;
> > as discussed in a previous mail would enforce this behaviour? If so,
> > I could write a patch and tests against 5.8.1-RC4 and propose them to
> > p5p.
> 
> Sounds like the wrong place to put it to me ... how does is() cope with
> two overloaded references? is_deeply should do it the same way (ideally
> by just calling is ...)

If you're going to do it, I think Marcel is right.

You can't call is() as this will trigger a full test result. Actually, is() is 
just shorthand for Test::Builder->is_eq, which I think think is a better name 
as it's more explicit.

If you're going to do it, it's a bit more complicated than the above, it needs 
to be asymmetric and you should check for overloading of "", eq and maybe 
even ne and arguably 0+, == and != also...

F