A question about role-private attributes
I am trying to understand the following small portion from S12, and it
seems slightly ambiguous to me:
=== from S12:
You may wish to declare an attribute that is hidden even from the
class; a completely private role attribute may be declared like this:
C
The name of such a private attribute is always considered lexically
scoped. If a role declares private lexical items, those items are
private to the role due to the nature of lexical scoping.
===
I was unable to figure out what this piece of code would print:
role A {
my $!spleen;
method set_spleen($x) { $!spleen = $x; return self }
method say_spleen() { say $!spleen }
}
class B does A {}
class C does A {}
B.new.set_spleen(3).say_spleen();
B.new.say_spleen();
C.new.say_spleen();
==
How many 3s will that print? Just a single 3 would indicate that every
instance of every class that composed A has its own copy. If 3 is
printed twice, that would indicate that not every instance but every
class has its own copy. All three 3s will indicate only a single copy
for the entire role.
My guess is that the first interpretation was intended. Is that guess
correct? I would also guess that C would give us the third
interpretation. But I do not know what syntax would result in a role-
private class attribute.
Thanks,
Abhijit
Re: A question about role-private attributes
[EMAIL PROTECTED] wrote:
I am trying to understand the following small portion from S12, and it
seems slightly ambiguous to me:
=== from S12:
You may wish to declare an attribute that is hidden even from the
class; a completely private role attribute may be declared like this:
C
The name of such a private attribute is always considered lexically
scoped. If a role declares private lexical items, those items are
private to the role due to the nature of lexical scoping.
===
It's an attribute - thus it's per instance of the class(es) the role was
composed into. The lexical stuff is just about the visibility of the
attribute, not saying it's like a lexical variable in that it's one of
them existing for the block (which isn't even true for blocks in
general, since there can be closures).
I was unable to figure out what this piece of code would print:
role A {
my $!spleen;
method set_spleen($x) { $!spleen = $x; return self }
method say_spleen() { say $!spleen }
}
class B does A {}
class C does A {}
B.new.set_spleen(3).say_spleen();
B.new.say_spleen();
C.new.say_spleen();
==
How many 3s will that print? Just a single 3 would indicate that every
instance of every class that composed A has its own copy. If 3 is
printed twice, that would indicate that not every instance but every
class has its own copy. All three 3s will indicate only a single copy
for the entire role.
I believe it would print one three. I expected this would work in Rakudo
too - will look into why it doesn't.
Thanks,
Jonathan
pointy blocks on repeat
S04 says that pointy blocks are allowed on repeat while blocks
when the conditional is on either side of the block. STD.pm
currently allows only the case where the conditional is in
front -- i.e., STD.pm doesn't currently recognize the form
repeat -> $thing { ... } while something();
Is this just an oversight in STD.pm or is the second form not
allowed?
Thanks,
Pm
