Re: Some more questions about defsystem modules

On 4/27/26 12:42 PM, Robert Dodier wrote:

> On Mon, Apr 27, 2026 at 10:30 AM Raymond Toy<[email protected]> wrote:
>> According to maxima.system, we still compile and load sloop.lisp. But I think the last use of sloop was removed many, many years ago? Can we remove that from the defsystem?
>>
>> Likewise, gcl includes optimize.lisp. Do we still need this? I guess it’s telling gcl to optimize (inline) some simple operations. I’m not familiar with this. Maybe leave it?
>>
>> At the very beginning we have a gcl module named proclaim that just loads (no compiling) a bunch of files. Not sure why this needs to be done. Couldn’t find any notes on this, but I didn’t look very hard.
> Assuming that SLOOP was successfully expunged (some years ago), I
> think it's OK to remove sloop.lisp. I suppose there might possibly be
> some user code somewhere which uses SLOOP. If so we will hear about
> it. I am inclined to say that any code still using SLOOP is the
> author's responsibility at this point.
My thinking right now was to remove it from defsystem and move 
sloop.lisp to share or something so that some it’s still available in 
case anyone wants it.
> I am inclined to remove optimize.lisp. Inlining makes it harder or
> impossible to trace functions, which I believe outweighs the speed
> gain.
Hopefully, Camm can provide some insight on this.
> About the module "proclaim", dunno what's up with that. Maybe stuff
> got moved into that in order to quiet some warnings or something like
> that? Maybe try to do without "proclaim" and remove the #-gcl
> conditionals for those items elsewhere in maxima.system. On the whole
> I think reducing implementation dependent stuff is a good idea.

I’ll try to look up some history on this soon. I don’t build or test 
with gcl very often.

&#8203;

_______________________________________________
Maxima-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/maxima-discuss