[perl #118023] [BUG] typed hashes and metaops and slots, oh my, in Rakudo

[Aleks-Daniel Jakimenko-Aleksejev via RT]

So is the current behavior good or bad?

On 2014-10-15 05:16:27, barto...@gmx.de wrote:
> The current behaviour is:
>
> $ perl6-m -e 'my %h{Any}; %h<a>=Any; %h<a> //= "a"; say %h.perl'
> Hash[Any,Any].new("a" => "a")
>
> $ perl6-m -e 'my %h{Any}; %h<a> //= "a"; say %h.perl'
> Hash[Any,Any].new("a" => "a")
>
> $ perl6-m -e 'my $froms= do { my %h{Any} of Hash; %h }; $froms<a>=Any;
> $froms<a> //= do { my %h{Any} of Hash; %h }; say $froms;' ## formerly
> marked as "works"
> Type check failed in assignment to '%h'; expected 'Hash' but got 'Any'
> in method assign_key at src/gen/m-CORE.setting:10342
> in sub postcircumfix:<{ }> at src/gen/m-CORE.setting:2875
> in block <unit> at -e:1
>
> $ perl6-m -e 'my $froms= do { my %h{Any} of Hash; %h }; $froms<a> //=
> do { my %h{Any} of Hash; %h }; say $froms.perl'
> Hash[Any,Any].new("a" => Hash[Any,Any].new())
>
> $ perl6-m -e '(my %h{Any} of Hash){Any} //= my %; say %h.perl'
> Hash[Any,Any].new(Any => {})