From @masak

<masak> rakudo​: class A { ... }; class B is A {}; class A {}
<p6eval> rakudo cae7f9​: OUTPUT«===SORRY!===␤Parent isn't a Class.␤»
* masak submits LTA rakudobug
* masak wonders idly what the parent is
<masak> rakudo​: class A { ... }; say A.WHAT
<p6eval> rakudo cae7f9​: OUTPUT«Failure()␤»
<masak> ah.
<masak> rakudo​: class A { ... }; class B is A {}
<p6eval> rakudo cae7f9​: OUTPUT«===SORRY!===␤Null PMC access in can()␤»
<masak> huh.
<masak> that's the first causality-defying Null PMC access. :)

From @coke

On Tue Dec 21 01​:10​:35 2010, masak wrote​:

<masak> rakudo​: class A { ... }; class B is A {}; class A {}
<p6eval> rakudo cae7f9​: OUTPUT«===SORRY!===␤Parent isn't a Class.␤»
* masak submits LTA rakudobug
* masak wonders idly what the parent is
<masak> rakudo​: class A { ... }; say A.WHAT
<p6eval> rakudo cae7f9​: OUTPUT«Failure()␤»
<masak> ah.
<masak> rakudo​: class A { ... }; class B is A {}
<p6eval> rakudo cae7f9​: OUTPUT«===SORRY!===␤Null PMC access in
can()␤»
<masak> huh.
<masak> that's the first causality-defying Null PMC access. :)

This seems fine now​:

09​:18 < [Coke]> rakudo​: class A { ... }; class B is A {}; class A {} #RT
#​81060
09​:18 <+p6eval> rakudo ce5cca​: ( no output )
09​:18 < [Coke]> rakudo​: class A { ... }; say A.WHAT #RT 81060
09​:18 <+p6eval> rakudo ce5cca​: OUTPUT«===SORRY!===␤The following
packages were
  stubbed but not defined​:␤ A␤␤»

Closable with tests.
--
Will "Coke" Coleda

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

From @bbkr

It breaks when parent class body is provided later.

<moritz> r​: class A { ... }; class B is A { }; class A { }; say B.new
<p6eval> rakudo 42a280​: OUTPUT«No such method 'new' for invocant of type 'B'␤ in block at
/tmp/MEw1Loww7z​:1␤␤»

From @coke

On Thu Oct 18 04​:10​:24 2012, bbkr wrote​:

It breaks when parent class body is provided later.

<moritz> r​: class A { ... }; class B is A { }; class A { }; say B.new
<p6eval> rakudo 42a280​: OUTPUT«No such method 'new' for invocant of
type 'B'␤ in block at
/tmp/MEw1Loww7z​:1␤␤»

Fails differently on today's moar​:

class A { ... }; class B is A { }; class A { }; say B.new
Cannot invoke this object (REPR​: Null, cs = 0)

--
Will "Coke" Coleda

From @usev6

This looks right now​:

$ perl6 -e 'class A { ... }; say A.WHAT' # X​::Package​::Stubbed
===SORRY!=== Error while compiling -e
The following packages were stubbed but not defined​:
  A
at -e​:1
------> class A { ... }; say A.WHAT⏏<EOL>
  expecting any of​:
  postfix
  statement end
  statement modifier
  statement modifier loop

$ perl6 -e 'class A { ... }; class B is A {}' # X​::Inheritance​::NotComposed
===SORRY!=== Error while compiling -e
'B' cannot inherit from 'A' because 'A' isn't compose yet (maybe it is stubbed)
at -e​:1
------>

$ perl6 -e 'class A { ... }; class B is A {}; class A {}' # also X​::Inheritance​::NotComposed
===SORRY!=== Error while compiling -e
'B' cannot inherit from 'A' because 'A' isn't compose yet (maybe it is stubbed)
at -e​:1
------>

I added two tests to S12-class/stubs.t with commit Raku/roast@7e89188444

I'm resolving this ticket now.

@usev6 - Status changed from 'open' to 'resolved'