[Openmcl-devel] What is the new context argument in all the compiler functions?

[Ron Garret]

I figured out a workaround for this that should be robust against future changes in the API.

In case anyone is interested, the code is here:

http://www.flownet.com/ron/lisp/combination-hook.lisp

It lets you do things like compute a factorial using an inline Y-combinator:

? (((λ (f) ((λ (g) (g g)) (λ (h) (λ (x) ((f (h h)) x)))))
  (λ (f) (λ (n) (if (zerop n) 1 (* n (f (1- n)))))))
 15)
1307674368000

Or compute factorial of 6 using Church numerals:

? ((((λ (f) ((λ (g) (g g)) (λ (h) (λ (x) ((f (h h)) x)))))
   (λ (f)
      (λ (n)
         (if ((λ (p a b) (p a b)) ((λ (n) (n (λ (x) (λ (x y) y)) (λ (x y) x))) n) t nil)
           ((λ (n) (λ (f x) (f (n f x)))) (λ (f x) x))
           ((λ (m n) (m ((λ (m n) (λ (f x) (m f (n f x)))) n) (λ (f x) x)))
            n (f ((λ (n) (λ (f x) (n (λ (g h) (h (g f))) (λ (u) x) (λ (u) u)))) n)))))))
  ((λ (n) (λ (f x) (f (n f x))))
   ((λ (n) (λ (f x) (f (n f x))))
    ((λ (n) (λ (f x) (f (n f x))))
     ((λ (n) (λ (f x) (f (n f x))))
      ((λ (n) (λ (f x) (f (n f x))))
       ((λ (n) (λ (f x) (f (n f x))))
        (λ (f x) x))))))))
 '1+ 0)
720

Reasonable people could disagree, but I think this is pretty cool.

rg


On Apr 23, 2012, at 12:59 PM, Ron Garret wrote:

> 
> On Apr 23, 2012, at 11:55 AM, Gary Byers wrote:
> 
>> I'm spmewhat dismayed that changes like this don't happen more often
>> than they do.  There are lots of reasons for that, but worrying about
>> whether someone will be "somwhat dismayed" when private internal interfaces
>> change isn't and can't be one of them.
> 
> I completely agree, but there is an extenuating circumstance in this case.
> 
> One of my interests is pedagogy.  When writing code to illustrate how Lisp is derived from the lambda calculus I often find it useful to use ((...) ...) as a shorthand for (funcall (...) ...).  So a while back I requested that the last line of nx1-combination be changed from:
> 
> 	(nx-error "In the form ~S, ~S is not a symbol or lambda expression." form sym)))))
> 
> to something like:
> 
> ...
> 	(nx-combination-hook form sym)))))
> 
> (defun nx-combination-hook (form sym)
>  (nx-error "In the form ~S, ~S is not a symbol or lambda expression." form sym))
> 
> so that a user could change this behavior.
> 
> You denied this request (as of course is your prerogative) and instead suggested that I use ADVISE to change the behavior of NX1-COMBINATION, which I did, and everyone went home happy.  But changing the interface to NX1-COMBINATION breaks my ADVISE code (because it has to call back in to nx1-combination).
> 
> I'm not saying that you should feel obligated to do anything about this.  I was just surprised that this part of the compiler was changed and wondered why it happened.  I also wanted you to be aware of the situation in case you were feeling charitable, or had another suggestion.
> 
> rg
> 
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> http://clozure.com/mailman/listinfo/openmcl-devel