From daniel@ruoso.com

<sbp> not entirely sure what's going on there. what is the use case in
general for declaring that you *don't* want parameters?
<ruoso> sbp, to fail if anyone tries to send one?
<ruoso> rakudo​: -> { say 'live' }.(123);
<p6eval> rakudo 78cb4c​: OUTPUT«live␤»
<ruoso> oops
<ruoso> hmm
<ruoso> rakudo​: -> { say $_ }.(123);
<p6eval> rakudo 78cb4c​: OUTPUT«Use of uninitialized value␤␤»
<ruoso> hmm... for sure a rakudobug

I think -> { ... } implies an empty signature, so it would fail if you
call it with a parameter...

but then in a following test, I realized the problem is not related to
pointy blocks at all...

<ruoso> rakudo​: sub foo() {}; foo(123);
<p6eval> rakudo 78cb4c​: ( no output )

That was supposed to break...

With a further investigation, I found out that...

<ruoso> rakudo​: sub foo() { say @​_[0] }; foo(123)
<p6eval> rakudo 78cb4c​: OUTPUT«123␤»
<ruoso> rakudo​: -> { say @​_[0] }.(123);
<p6eval> rakudo 78cb4c​: OUTPUT«123␤»

Which means that it is actually assuming the default rountine signature,
when both cases imply an empty signature instead...

daniel

From @kyleha

This is an automatically generated mail to inform you that tests are now available in b/t/spec/S06-signature/errors.t

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

From @moritz

Works now, jnthn++, unfudged test, resolving ticket.

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