From @masak

<masak> rakudo​: module A { our sub foo { say "A​::foo" } }; module B is
A { our sub foo { say "B​::foo"; nextsame } }; B​::foo
<p6eval> rakudo a95c1d​: OUTPUT«B​::foo␤Null PMC access in clone()␤ in
'B​::foo' [...]
<masak> rakudo​: module B is floozey {}; say "alive"
<p6eval> rakudo a95c1d​: OUTPUT«alive␤»
* masak submits rakudobug

The Null PMC access in question has been reported before. What I'd
like to report as an error here is the fact that an 'is' trait on the
module doesn't result in a (compile-time) error. See
<http://perl6advent.wordpress.com/2010/12/14/day-14-nextsame-and-its-cousins/#comment-542>
for when this would have helped to catch an error early.

From @kyleha

This is an automatically generated mail to inform you that tests are now available in S10-packages/basic.t

commit fef3dccf5d7492b427068fcdc979abcf528a3ceb
Author​: Kyle Hasselbacher <kyleha@​gmail.com>
Date​: Thu Dec 23 10​:39​:50 2010 -0600

  [S10-packages] Test for RT 80856

diff --git a/S10-packages/basic.t b/S10-packages/basic.t
index aafc5c4..89286b1 100644
--- a/S10-packages/basic.t
+++ b/S10-packages/basic.t
@@ -4,7 +4,7 @@ use v6;
 
 use Test;
 
-plan 52;
+plan 53;
 
 my regex fairly_conclusive_platform_error {:i ^\N*<<Null?>>}
 
@@ -219,4 +219,8 @@ eval_lives_ok q' module MapTester { (1, 2, 3).map: { $_ } } ',
         'package Foo; is indicator for Perl 5 code';
 }
 
+#?rakudo todo 'RT 80856'
+eval_dies_ok 'module RT80856 is not_RT80856 {}',
+             'die if module "is" a nonexistent';
+
 # vim: ft=perl6

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

From @jnthn

On Thu Dec 16 00​:56​:30 2010, masak wrote​:

<masak> rakudo​: module A { our sub foo { say "A​::foo" } }; module B is
A { our sub foo { say "B​::foo"; nextsame } }; B​::foo
<p6eval> rakudo a95c1d​: OUTPUT«B​::foo␤Null PMC access in clone()␤
in
'B​::foo' [...]
<masak> rakudo​: module B is floozey {}; say "alive"
<p6eval> rakudo a95c1d​: OUTPUT«alive␤»
* masak submits rakudobug

The Null PMC access in question has been reported before. What I'd
like to report as an error here is the fact that an 'is' trait on the
module doesn't result in a (compile-time) error. See
<http://perl6advent.wordpress.com/2010/12/14/day-14-nextsame-and-its-
cousins/#comment-542>
for when this would have helped to catch an error early.

Now gives​:

===SORRY!===
'RT80856' cannot inherit from 'not_RT80856' because it is unknown.

And the test attached to this RT is already passing, so resolving.

/jnthn

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