(This blog post is part three of a series; there's also a part one and a part two.)

Shortly after I wrote the last post on -n and -p, and how I didn't really understand what a setting was in Perl 6, sorear++ and TimToady++ filled me in on all of the details. So here I am, a third time, to pass the knowledge on.

I wrote last time that I found the term "setting" confusing and overloaded. That's because I thought it was a single, defined thing. In effect, there's no the setting in Perl 6; there can be many at the same time.

A setting is simply something that surrounds your code on the outside. (Haskell has a Prelude, but a prelude only comes before your code. A setting envelopes your code both before and after.) Your code simply finds itself lexically inside some setting or other. In technical parlance, whatever is the OUTER:: of your code is a setting.

So, you can have several settings, just like I wished for. They stack, you see. Or rather peel, like onion layers. One man's setting is another man's code, all the way outwards into the final OUTER:: nothingness of empty space.

And — the final piece of the puzzle — the big default "here, friend, are all of your builtins" setting is called CORE. I'd always wondered why we keep saying both "setting" and CORE. (And why Rakudo calls the directory src/core, not src/setting.) That's why; CORE is just a setting among others.

The discussion — which consists mostly of sorear and TimToady telling how things really are — can be found here.

And now I think I'm finally done writing on how -n and -p work. 哈哈