From: Eero Saynatkari <ruby-ml@...>
Date: 2006-10-11T02:16:57+09:00
Subject: Re: [ruby-cvs:18323] ruby: * eval.c (splat_value): use "to_splat" instead of "to_ary" to

--z4+8/lEcDcG5Ke9S
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On 2006.10.10 23:57, dblack@wobblini.net wrote:
> Hi --
>=20
> On Tue, 10 Oct 2006, James Edward Gray II wrote:
>=20
> >On Oct 10, 2006, at 8:43 AM, Yukihiro Matsumoto wrote:
> >
> >>Hi,
> >>
> >>In message "Re: [ruby-cvs:18323] ruby: * eval.c (splat_value): use=20
> >>"to_splat" instead of "to_ary" to"
> >>   on Tue, 10 Oct 2006 22:36:44 +0900, "Rick DeNatale"=20
> >><rick.denatale@gmail.com> writes:
> >>
> >>If it's possible, I'd like to hear your feeling against the changes in
> >>detail.  Can you express?  Why don't you right?
> >
> >I believe I understand where Rick is coming from.
> >
> >He's saying that when we do a *whatever, we are actively asking that=20
> >whatever be split into values.  Because we consciously make that choice,=
=20
> >to_a (a general solution) is fine.  We understand that we are calling to=
_a=20
> >and willingly accept the consequences.  If we want to divide up a string=
=20
> >into arguments that way, that's our business not Ruby's business.
> >
> >At least, that's what I *think* he means.  Correct me if I'm wrong Rick.
>=20
> I've been thinking about the whole thing, and I tend to agree.  I'm
> not sure why we need a whole star-related sub-system, including a
> method hook, independent of the existing array sub-system.  I think
> the star can be understood as syntactic sugar for to_a.

The problem is that * does not just create an Array--in fact, that
seems to be the less-used functionality. In most situations, #to_a
is the exact opposite of what is happening. If

  *5      # =3D> [5]

Then #to_a is correct. This is what can be considered to happen in
method definitions and in the left-hand-side of assignments. What is
more, these are not controlled by the user: the only context where
the programmer can explicitly affect the result of * is this:

 *[1, 2]  # =3D> 1, 2

(And variations thereof)

Here we are definitely NOT doing a #to_a but the exact opposite=20
(whatever it is decreed to be called).=20

I think this is what matz means--correct me if I am wrong :)

--z4+8/lEcDcG5Ke9S
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (FreeBSD)

iD8DBQFFK9T77Nh7RM4TrhIRAqgPAJ4tLEoNdBlv7M5fYB9XAQYILq0/3wCgnxhs
8xrqA8IWRh2mOVCPStDetKA=
=JxGj
-----END PGP SIGNATURE-----

--z4+8/lEcDcG5Ke9S--