Namespace hack

[Steven D'Aprano]

Pardon me for breaking threading, but Daniel's response is not available 
on my ISP's news server, and I only discovered it by accident.

On Thu May 24 15:04:34 CEST 2012, Daniel Fetchinson <fetchinson at 
googlemail.com> wrote:

> > On Thu, 24 May 2012 08:50:59 +0000, Steven D'Aprano wrote:
> > From the Zen of Python ("import this"):
> > Namespaces are one honking great idea -- let's do more of those!
[...]

> Funny, you got to the last line of "import this" but apparently
> skipped the second line:
> 
> Explicit is better than implicit.
> 
> And you didn't even post your message on April 1 so no, I can't laugh
> even though I'd like to.

Do you object to the ability to write standard Python modules?

# module.py
def spam(obj, n):
    return len(obj) + n

def ham(obj):
    return spam(obj, 23)


By your apparent misunderstanding of the Zen, you think that this should 
be written with oodles of more explicitness, 'cos explicit is always 
better, right?

keyword.def globals.spam(locals.obj, locals.n):
    keyword.return builtin.len(locals.obj) + locals.n

keyword.def globals.ham(locals.obj):
    keyword.return globals.spam(locals.obj, 23)


Python, like most (all?) non-trivial languages, has scoping rules so that 
you can refer to names without explicitly specifying which namespace they 
are in. So long as this is unambiguous, Explicit vs Implicit is 
irrelevant if not outright wrong.

My namespace decorator simply applies a slightly different set of scoping 
rules to the ones you are already used to in modules. It's no worse than 
nested functions (hardly a surprise, because it is built on nested 
functions!) or module-level scoping rules.


-- 
Steven