Request: allow method calls on constants

[marcandre]

Currently methods & predicates can be called on self:

(_ #predicate?(:arg, %1) ...)
(_ #method(:arg, %1) ...)

Request is to allow calling them on constants:

(_ #Foo.predicate?(:arg, %1) ...)
(_ #Foo::Bar.method(:arg, %1) ...)
(_ #::Global::X.method(:arg, %1) ...)

[pirj]

Just for the reference, the idea was to allow usage of #rspec.example_groups.regular call chains, for namespacing purposes. Since this would only create complication in NP parser, and since Ruby has namespacing capabilities built-in, I completely support RSpec::ExampleGroups.regular syntax instead.

Just to make sure, is it correct that given:

module A
  module B
    def self.a_method
       # ...
    end
  end
end

module RuboCop
  module Cop
    class SomeCop < Base
      include A
      def_node_matcher :a, '( #B.a_method ...)'
    end
  end
end

would be able to call A::B.a_method?

It would be a nice way to structure configurable DSL helpers.

[marcandre]

Just for the reference, the idea was to allow usage of #rspec.example_groups.regular call chains, for namespacing purposes. Since this would only create complication in NP parser

I don't think it would be difficult tbh. I just wonder what is the best to have, the most useful in general. Maybe even both forms...

and since Ruby has namespacing capabilities built-in, I completely support RSpec::ExampleGroups.regular syntax instead.

Just to make sure, is it correct that given:

module A
  module B
    def self.a_method
       # ...
    end
  end
end

module RuboCop
  module Cop
    class SomeCop < Base
      include A
      def_node_matcher :a, '( #B.a_method ...)'
    end
  end
end

would be able to call A::B.a_method?

Yes.

So the question is: should I implement both, or only one (and which one) 😅

[pirj]

If the namespace is relative, i.e. you don't have to specify #A::B.a_method when A is included, I'd go with constants, no method call chaining.

[marcandre]

Released as v1.1.0

[pirj]

@marcandre You're like a ⚡-man! 💯
This is just in time for the 2.0 RuboCop-RSpec release, thanks a lot!