On Oct 4, 2004, at 9:58 PM, Brent 'Dax' Royal-Gordon wrote:
Tim Bunce <[EMAIL PROTECTED]> wrote:Now, with that out of the way, let's talk about overlaid namespaces.
I don't think I ever read a description of what the purpose of this was.
I get the "what" but not the "why". Without the "why" it's hard to
critique the "how".
Clearly I'm not Dan, but I think the idea here is that, for example, in the following code:
module Foo::Bar { class Baz { $quux } }
You can have "the current namespace" actually be [ ::Foo::Bar::Baz, ::Foo::Bar, ::* ] (or, for the last one, whatever the namespace that @*ARGS and friends are in is called), so that the search for $quux can be done very easily.
This may have changed for Perl6, but at least for Perl5, non-lexicals are only ever looked for in the current package, the top-level (main) namespace, plus the pre-defined variables (don't know if these are handled at compile-time), according to the Camel. So having an arbitrary search list might not be used in the Perl case.
JEff
