From @masak

<masak> rakudo​: class A { multi method foo() { say "foo1" }; multi
method foo($x) { say "foo2"; nextwith() } }; A.new.foo(42)
<p6eval> rakudo f845cc​: OUTPUT«foo2␤Null PMC access in isa()␤in method
A​::foo [...]
<masak> mwhahaha.
* masak submits rakudobug
<masak> rakudo​: class A { multi method foo($x) { say "foo2";
nextwith() } }; A.new.foo(42)
<p6eval> rakudo f845cc​: OUTPUT«foo2␤Null PMC access in isa()␤in method
A​::foo [...]
<jnthn> huh, it blows up when there's nowhere to defer to?
<jnthn> ouch.
<masak> jnthn​: but in the first example, isn't there somewhere to defer to?
<jnthn> masak​: Very debatable.
<masak> why?
<masak> because it isn't in the original list of candidates?
<jnthn> Right.
<masak> understood.
<jnthn> And if we decide to re-generate the list, we get ourselves
into even more "fun".
* masak shudders
<jnthn> I still need to write a mail to p6l on that issue.
<jnthn> And try and get a design decision.

From @moritz

The Null PMC Access is gone, but now it just does nothing, instead of
dying (as I would assume it should).

A test for that can be found in t/spec/S12-methods/defer-next.t.

So I'm handing this ticket back to jnthn++

The RT System itself - Status changed from 'new' to 'open'

From @bbkr

NOM

bbkr​:nom bbkr$ ./perl6 -e 'class A { multi method foo() { say "foo1" };
multi method foo($x) { say "foo2"; nextwith() } }; A.new.foo(42)'
foo2

From @bbkr

Also mentioned test file is passing.

I think it can be closed, but I do not want to steal ticket..

From @jnthn

On Sun Sep 11 06​:18​:58 2011, bbkr wrote​:

Also mentioned test file is passing.

I think it can be closed, but I do not want to steal ticket..

Closing it; thanks.

/jnthn

@jnthn - Status changed from 'open' to 'resolved'