# New Ticket Created by "Carl Mäsak"
# Please include the string: [perl #71754]
# in the subject line of all future correspondence about this issue.
# <URL: http://rt.perl.org/rt3/Ticket/Display.html?id=71754 >
<masak> rakudo: multi f(Num $x) { say "Num" }; multi f(Int $x) {
callsame; say "Int" }; f(42)
<p6eval> rakudo db84bc: Null PMC access in clone() [...]
<ash_> masak: after looking at S06 and S12 i don't know how that
should work... *of to check for a spec test to see if those make more
sense*
* arnsholt does not grok the nqp-rx code
<ash_> masak: i think callsame is really for inheritance not for
multi-methods specifically
<TimToady> it's for both
<masak> ash_: well, it's also used when doing calls with .* and .+
<masak> ash_: and for wrappers.
<masak> TimToady: so, what should happen in my multi above?
<TimToady> that should say Num and then Int, I think
<TimToady> it's essentially just thrown a 'next' control exception to
whatever dispatcher is running down a list
<masak> I have a vague feeling I've been discussing this with jnthn++ before...
<masak> the problem, I have a vague feeling he explained, is that the
multi-dispatcher has already done away with the looser candidates.
<masak> but I might mis-remember.
* masak submits rakudobug for now
<TimToady> it can't have, or it's not a Perl 6 dispatcher :)
<masak> :)
<TimToady> at minimum, it has to be able to reproduce the tail of the list
