From @masak

<dalek> specs​: a56ffc3 | larry++ | S (2 files)​:
<dalek> specs​: make ?/* semantics consistently Nil/()
<dalek> specs​: review​: Raku/old-design-docs@a56ffc33ac
<TimToady> rakudo​: 'x' ~~ /(y)? (z)*/; say $0.defined, $1.defined;
<p6eval> rakudo 684fdf​: OUTPUT«Bool​::TrueBool​::True␤»
<TimToady> that should be False True
<TimToady> by the new rules
* masak submits rakudobug
<TimToady> and actually, the first was supposed to be False already by
the old spec

From @bbkr

2011.09 build

bbkr​:rakudo-2011.09 bbkr$ ./perl6 -e '"x" ~~ /(y)? (z)*/; say
$0.defined, $1.defined;'
Bool​::FalseBool​::False

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

From @coke

On Sat Oct 01 02​:02​:50 2011, bbkr wrote​:

2011.09 build

bbkr​:rakudo-2011.09 bbkr$ ./perl6 -e '"x" ~~ /(y)? (z)*/; say
$0.defined, $1.defined;'
Bool​::FalseBool​::False

Changed again, but still wrong per Larry's comment​:

./perl6 -e '"x" ~~ /(y)? (z)*/; say 0.defined, $1.defined;'
TrueTrue

--
Will "Coke" Coleda

From @timo

It gives the right output nowadays​:

"x" ~~ /(y)? (z)*/; say $0.defined; say $1.defined;
False
True

I'll mark this testneeded, but there's probably tests in roast that cover this.

From @coke

On Sun Dec 29 09​:38​:52 2013, timo wrote​:

It gives the right output nowadays​:

"x" ~~ /(y)? (z)*/; say $0.defined; say $1.defined;
False
True

I'll mark this testneeded, but there's probably tests in roast that
cover this.

... And now it gives the wrong output again, TrueTrue.

--
Will "Coke" Coleda

From @usev6

This works now (again)​:

$ perl6-m -e '"x" ~~ /(y)? (z)*/; say $0.defined; say $1.defined;'
False
True

I added a test to S05-match/capturing-contexts.t with the following commit​: Raku/roast@ae245a65f6

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