Re: null-environment

[Daniele Maraschi <maraschi at lirmm dot fr>]

Per Bothner wrote:
> 
> Daniele Maraschi <maraschi@lirmm.fr> writes:
> 
> > #|kawa:1|# (define env (null-environment))
> > #|kawa:2|# env
> > #<procedure null-environment>
> > #|kawa:3|# (eval '(define a 5) env)
> > #|kawa:4|# a
> > 5
> > #|kawa:5|#
> >
> > Probably, the effect is not what it was expected!
> 
> Please read R5RS section 6.5 "Eval".

In the R5RS section 6.5 "Eval" I read literally:

"...with the restriction that 'eval' is not allowed
to create new bindings in the environment associated
with 'null-environment' or 'scheme-report-environment'."

Anyway, I shouldn't debate if Kawa is more or less R5RS compliant!
 
> Kawa should perhaps give an error message, but what you did is clearly
> invalid and meaningless.

I know it, but my intent was to provocate a question
regarding the possibility to implement in Kawa "first-class"
environments.
I know perfectly they are not considered in R5RS paper, but I'd just
like to know if the Kawa community is interested to have them.
By my own, I consider them very important because I use Scheme like an
ACL ( Agent Communication Language ). In fact, I've already implemented
a Kawa extension to manage "first-class" environments and I can give you
if you consider it interesting.

-- 
---------------------
Daniele Maraschi
PhD Student
L.I.R.M.M. ( http://www.lirmm.fr )
161, rue Ada
34392 Montpellier Cedex 5
Tel. +(33)(4)67418670
Fax. +(33)(4)67418500
e-mail: maraschi@lirmm.fr
e-mail(Italy): maraschi@tiscalinet.it
begin:vcard 
n:Maraschi;Daniele
tel;fax:+33 (4) 67418500
tel;work:+33 (4) 67418670
x-mozilla-html:FALSE
org:L.I.R.M.M.;Arc
adr:;;;Montpellier;France;34392;France
version:2.1
email;internet:maraschi@lirmm.fr
title:PhD Student
x-mozilla-cpt:;0
fn:Daniele Maraschi
end:vcard