Front page | perl.perl5.porters |
Postings from November 2010
On Mon, Nov 29, 2010 at 09:37:07PM +0000, Zefram wrote:
> Jesse Vincent wrote:
> >safe-dereferencing arrow
>
> I think this name for it is problematic. There are many types of safety.
> For me, actually, I don't think of the semantics I'm looking for as
> safety at all: I think of them as a shorthand.
>
> >"If the dereference of the left hand side succeeds,
>
> Which dereference, and what constitutes success? For $lhs->[0], there's
> an array dereference, and it obviously fails if $lhs is not a reference
> to an array or something with @{} overloading. But if there is such
> overloading, does that constitute success? If the @{} overload method
> dies, was that a failure of dereferencing? But that's the easy case.
> For $lhs->method, is the dereferencing a success if $lhs is a reference
> to a blessed object? Or does the method lookup constitute part of the
> dereferencing too? Or the method execution?
>
> >I suspect that ->? is going to be more interestingly useful,
>
> I suspect that it would hide a lot of bugs.
This is my suspicion too.
I was assuming that the operator was
SvGETMAGIC(sv);
SvOK(sv)
? ... /* do the appropriate dereference, calling overloading if necessary */
: sv;
(given that sv at that point will be undef)
and that anything that is defined-but-not-a-reference will explode under
strict 'refs', and be treated as a symbolic reference (or explode) when
not under strict refs.
Nicholas Clark
Thread Previous
|
Thread Next