Version franaise
Home About Download Resources Contact us
Browse thread
The Implicit Accumulator: a design pattern using optional arguments
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: -- (:)
From: Bill Wood <william.wood3@c...>
Subject: Re: [Caml-list] The Implicit Accumulator: a design pattern using optional arguments
On Wed, 2007-06-27 at 14:55 +0100, Thomas Fischbacher wrote:
   . . .
> One thing OCaml can do better than, say, CMU CL, is to define globally
> visible functions that depend on some otherwise inaccessible context,
> as in the following example:
> 
> let float_factorial =
>    let _known_factorials = ref [|1.0;1.0;2.0;6.0;24.0;120.0;720.0|] in
>    (fun n ->
>      let known_factorials = !_known_factorials in
>      let nr_known = Array.length known_factorials in
>      if n < nr_known
>      then
>        known_factorials.(n)
>      else
>        let new_known_factorials = Array.make (n+1) 0.0 in
>        begin
> 	for i=0 to nr_known-1 do
> 	  new_known_factorials.(i) <- known_factorials.(i)
> 	done;
> 	(let rec fill f_pos pos =
> 	  if pos > n then ()
> 	  else
> 	    let () = new_known_factorials.(pos) <- f_pos in
> 	    fill (f_pos *. (float_of_int (pos+1))) (pos+1)
> 	in
> 	fill (known_factorials.(nr_known-1)*.(float_of_int nr_known)) nr_known);
> 	_known_factorials := new_known_factorials;
> 	new_known_factorials.(n)
>        end)
> ;;
> 
> A corresponding
> 
> (let ((buffer (make-array ...)))
>   (defun float-factorial (n)
>      ...))
> 
> just plainly does not work with CMU CL/SBCL. :-(

Just to set the record straight on this, Thomas and I did a little
experimentation and discovered that change happens -- CMU CL (at least)
now does the Right Thing here.  You can indeed define stateful
functions, like the OCaml one above, in CMU Common Lisp using 'defun'.