Re: [Chicken-hackers] Regarding the hide declaration, #1376

[Evan Hanson]

Hey all,

On 2017-06-05 11:43, Peter Bex wrote:
> On Mon, Jun 05, 2017 at 06:09:54PM +1200, Evan Hanson wrote:
> > Is that right? Personally, I'd rather make (declare (hide ...)) simply
> > do the right thing -- the right thing being the behaviour you originally
> > expected when filing #1376 -- than add a new type of declaration or
> > module syntax.
> > 
> > I think this is similar to what Peter has said on that ticket, so a
> > patch of that sort would be very welcome.
> 
> This is slightly more complicated due to declare being module-unaware.
> 
> (declare (hide foo))
> 
> (module a * (import chicken scheme) (define foo 1))
> (module b * (import chicken scheme) (define foo 2))
> 
> Which foo should be hidden?  Both, or none?

Personally, I'd expect the following behaviour, depending on where the
declaration appears:

  [1]
  (module a * (import chicken scheme) [2] (define foo 1) [3])
  [4]
  (module b * (import chicken scheme) [5] (define foo 2) [6])
  [7]
  (import a)
  [8]
  (import b)
  [9]

  [1]: foo (a noop as things work currently, or an error as you propose)
  [2]: a#foo
  [3]: a#foo
  [4]: foo
  [5]: b#foo
  [6]: b#foo
  [7]: foo
  [8]: a#foo
  [9]: b#foo

I suspect this is also the current behaviour, but I'd need to check to
confirm.

> I think the sane thing to do here is to make it error out,
> considering there's no top-level foo to hide

I think the issue of whether or not to make declarations error out when
one of the hidden identifiers doesn't exist is a separate one from
whether it should affect module exports. I slightly prefer the current,
more forgiving behaviour, but don't feel too strongly about it. If we do
make a change, however, I think it should affect all declarations that
take identifier names (where it makes sense), not just `hide'.

> make it work like this:
> 
> (module a *
>    (import chicken scheme)
>    (define foo 1))
> (module b *
>    (declare (hide foo))
>    (import chicken scheme) (define foo 2))
> 
> So here a#foo is visible and exported, while b#foo is hidden and
> not exported.

+1, I agree.

Another question is whether this behaviour should apply in eval'd code
as well. I think no, declarations should still be fully ignored in eval.

Cheers,

Evan

signature.asc
Description: PGP signature