[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gcl-devel] Re: probably a very dumb question

From: Camm Maguire
Subject: [Gcl-devel] Re: probably a very dumb question
Date: 31 Mar 2006 13:23:38 -0500
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2


Robert Boyer <address@hidden> writes:

> Why doesn't GCL, at the beginning of the compilation of each top level form,
> call the function walker::macroexpand-all on the form so that one does not
> have to worry about any further macro expansion?
> Why not make the function DEFUN also call walker::macroexpand before building
> the #<interpreted-function ...>?
> One advantage of doing such macro expansion early and permanently would be
> that it would not have to happen over and over again in interpreted code, so
> some things involving macro expansion, e.g., loop, would run a lot faster
> interpreted.
> This is all uninformed, idle speculation.  I've never heard of
> walker::macroexpand-all before today, and since its code in pcl_walk.c does
> not *macroexpand-hook*, I suspect it may not be up-to-date.

Not dumb at all.  Some considerations:

1) We need to support a 'traditional' non-ansi image without pcl.

2) compiler-macros.  Come to thing of it, these might be handled via
   the hook, no?

Reducing code to normal form facilitates all kinds of analysis, not to
mention making the code analyzing code more concise and intelligible.

I've never heard of the hook before your last email, so it is likely
it is not being handled correctly.

> Bob

Camm Maguire                                            address@hidden
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah

reply via email to

[Prev in Thread] Current Thread [Next in Thread]