From @coke

S02-types/declare.t contains a test which relies on this working​:

13​:51 < [Coke]> m​: my Complex $namcu =1.3; say $namcu.WHAT;
13​:51 <+camelia> rakudo-moar bbdcfd​: OUTPUT«Type check failed in assignment to
  '$namcu'; expected 'Complex' but got 'Rat'␤ in block at
  /tmp/RwaUOyDQ9D​:1␤␤»

--
Will "Coke" Coleda

From @coke

On Sun Jul 20 10​:54​:43 2014, coke wrote​:

S02-types/declare.t contains a test which relies on this working​:

13​:51 < [Coke]> m​: my Complex $namcu =1.3; say $namcu.WHAT;
13​:51 <+camelia> rakudo-moar bbdcfd​: OUTPUT«Type check failed in
assignment to
'$namcu'; expected 'Complex' but got 'Rat'␤ in block
at
/tmp/RwaUOyDQ9D​:1␤␤»

15​:28 < TimToady> [Coke]​: I think it's fair for that to fail, and require the
  (NYI) Complex(Cool) type for that to work
15​:29 < TimToady> but I could perhaps be argued into special casing that to work

--
Will "Coke" Coleda

From @jnthn

On Thu Aug 21 12​:33​:48 2014, coke wrote​:

On Sun Jul 20 10​:54​:43 2014, coke wrote​:

S02-types/declare.t contains a test which relies on this working​:

13​:51 < [Coke]> m​: my Complex $namcu =1.3; say $namcu.WHAT;
13​:51 <+camelia> rakudo-moar bbdcfd​: OUTPUT«Type check failed in
assignment to
'$namcu'; expected 'Complex' but got 'Rat'␤ in
block
at
/tmp/RwaUOyDQ9D​:1␤␤»

15​:28 < TimToady> [Coke]​: I think it's fair for that to fail, and
require the
(NYI) Complex(Cool) type for that to work
15​:29 < TimToady> but I could perhaps be argued into special casing
that to work

By contrast, I can't be argued into making that work. :-) Type constraints should be constraints, or we can't trust them. Coercion types are spec'd for coercing, and in this case a complex literal is just​:

my Complex $namcu = 1.3 + 0i; say $namcu.WHAT
(Complex)

Turns out the test isn't actually interested in coercion at all, though. It just wants to check you can assign a complex number to a variable with the Complex type - or at least, all the tests around it are doing that kind of thing. So, I got the test in line with those around it, and am rejecting the ticket.

Thanks,

Jonathan

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

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