From: Brock Weaver <brockweaver@...>
Date: 2005-08-11T04:37:59+09:00
Subject: Re: new block notation

------=_Part_137_3626752.1123702676293
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

I always thought that blocks (aka proc objects) should be in #{}, just like=
=20
they are in strings for substitution purposes.

That would differentiate {} for grouping lines of code (if, while, etc) and=
=20
#{} for grouping a set of code (proc / block / whatever). But that's just m=
y=20
$0.02.

On 8/10/05, James Edward Gray II <james@grayproductions.net> wrote:
>=20
> On Aug 10, 2005, at 2:09 PM, Eric Mahurin wrote:
>=20
> > I'd rather keep the {}
> > for blocks/lambdas. But, I don't think we should move the
> > ambiguity from blocks vs. hashes (overloading {}) to arrays vs.
> > hashes (overloading []). It would be better for hashes to just
> > have their own unambiguous syntax.
>=20
> We have the current Array[...]/Hash[...] option, which has been
> growing on me lately.
>=20
> James Edward Gray II
>=20
>=20


--=20
Brock Weaver
http://www.circaware.com

------=_Part_137_3626752.1123702676293--