From @masak

<masak> is this a known bug?
<masak> rakudo​: class C { has %!p; submethod BUILD(​:%!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«Null PMC access in invoke() [...]
<moritz> nom​: class C { has %!p; submethod BUILD(​:%!p) {} }; C.new; say 'alive'
<p6eval> nom​: OUTPUT«alive␤»
* masak submits rakudobug
<masak> rakudo​: class C { has $!p; submethod BUILD(​:$!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»
<masak> rakudo​: class C { has @​!p; submethod BUILD(​:@​!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»
<masak> rakudo​: class C { has &!p; submethod BUILD(​:&!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»

So, only fails for hashes; works in nom.

From @jnthn

On Thu Jul 21 14​:06​:49 2011, masak wrote​:

<masak> is this a known bug?
<masak> rakudo​: class C { has %!p; submethod BUILD(​:%!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«Null PMC access in invoke() [...]
<moritz> nom​: class C { has %!p; submethod BUILD(​:%!p) {} }; C.new;
say 'alive'
<p6eval> nom​: OUTPUT«alive␤»
* masak submits rakudobug
<masak> rakudo​: class C { has $!p; submethod BUILD(​:$!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»
<masak> rakudo​: class C { has @​!p; submethod BUILD(​:@​!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»
<masak> rakudo​: class C { has &!p; submethod BUILD(​:&!p) {} }; C.new;
say "alive"
<p6eval> rakudo 922500​: OUTPUT«alive␤»

So, only fails for hashes; works in nom.

Confirmed still working in nom; tagging testneeded.

class C { has %!p; submethod BUILD(​:%!p) {} }; C.new;
C.new()

/jnthn

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

From @moritz

Now tested in S12-construction/BUILD.t

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