RE: [Axiom-developer] Re: learning Lisp

[C Y]

--- Bill Page <address@hidden> wrote:

> Because we are talking specifically about lisp in the context
> of Axiom and at present there is no recommended way to use
> "modern" tools like SLIME in Axiom, I would be content to 
> discuss just the use of ')fin' and ')set input mode lisp'.

Funny you should mention SLIME :-).

While I cannot yet attest to all the utility this might have, I was
able (with a lot of help) to combine SLIME and Maxima in a rather
interesting way a while back.  The main problem was it required
threading, and I don't know where GCL stands on threads.

In CMUCL with Maxima, here was what I did:
http://www.math.utexas.edu/pipermail/maxima/2005/010325.html

Here is a screenshot showing the final setup I wound up trying:
http://maxima.sourceforge.net/debugging_maxima_and_lisp_code.png

On the left is a window with Maxima code, and below it is a Maxima
REPL.  On the left is a lisp source file related to the .mac file
(needed some low level definitions, IIRC) and below it a SLIME Lisp
REPL connected to the same running Maxima image that was displaying the
Maxima REPL in the left window.  So I could:

a) input commands in the Maxima REPL and investigate the changed image
with SLIME
b) input lisp commands into the SLIME REPL and use the Maxima REPL to
test the results at a high level without having to reload or anything
c) Presumably I could have sent code blocks of either mac or lisp code
to the image from the files, although I don't think I tried that

I'm not an expert SLIME user (I too need to sit down and watch the
movie) so there is doubtless a lot more I can do.  CMUCL is SLIME's
best Lisp, IIRC - advanced things like xref work best there.  I don't
think GCL was a major target for them yet due to its ANSI issues:
http://common-lisp.net/pipermail/slime-devel/2005-April/003392.html

Supported lisps are listed here:
http://common-lisp.net/project/slime/doc/html/slime_3.html#SEC3

So SLIME will make more sense when we are ANSI compliant, since we can
then run on CMUCL and other well supported Lisps.

For Windows+SLIME, Bill Clementson has some instructions in his blog:
http://bc.tech.coop/blog/040306.html

Postulating a powerful Axiom Emacs mode, this might be a really really
nice way to debug both Aldor/Spad code and the Lisp code, all at once
in real time :-).

BTY, anybody had any news on the state of Aldor?

Cheers,
Cliff  (trying to break the CY habit ;-)


                
__________________________________________ 
Yahoo! DSL – Something to write home about. 
Just $16.99/mo. or less. 
dsl.yahoo.com