[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Internal defines
From: |
Tom Lord |
Subject: |
Re: Internal defines |
Date: |
Mon, 10 Nov 2003 20:22:51 -0800 (PST) |
> From: David Van Horn <address@hidden>
> Tom Lord wrote:
> > But I think it is crystal clear (once the subtleties understodd) that
if the
> > [R5RS] denotational semantics do not support [R5RS] 5.3 then the
> > denotational semantics have a bug.
> The denotational semantics given by R5RS are for primitive expressions and
> selected built-in procedures of the language that the macro system must
> transform programs into; it makes no sense for the DS to make mention of
> macros and their errors.
> The fact that there is no formal semantics for the syntax transformation
> language is sad and would be a welcome addition to R6RS, IMO.
One way to do that is to add sufficiently to the set of primitive
expressions and built-in procedures so that macros can be explained
operationally (as a translation into those core elements carried out
by an explicit expand-phase that precedes evaluation).
If you'll forgive the self-promotion, please watch for the upcoming
Pika Scheme announcement on c.l.s. (sometime this month) which does
exactly that. In a not unrelated fashion, Pika provides a foundation
for module systems.
-t