bigger integers

bigger integers

[Joe Corneli]

This fails on big numbers like 12.

 (defun factorial (num)
   (if (<= num 0)
       1
     (* num (factorial (1- num)))))

Suggestions on how to make it work for bigger numbers?

Re: bigger integers

[Kevin Rodgers]

Joe Corneli wrote:
 > Suggestions on how to make it work for bigger numbers?

Get a bigger computer :-)

(defun factorial (num)
   (if (<= num 0)
       1.0
     (* num (factorial (1- num)))))

-- 
Kevin Rodgers

Re: bigger integers

[Pascal Bourguignon]

Joe Corneli <jcorneli@math.utexas.edu> writes:

> This fails on big numbers like 12.
> 
>  (defun factorial (num)
>    (if (<= num 0)
>        1
>      (* num (factorial (1- num)))))
> 
> Suggestions on how to make it work for bigger numbers?

Run it on a Common-Lisp implementation instead of emacs lisp.

After all, emacs lisp is designed to count characters in a file, not
to compute factorials.

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

Our enemies are innovative and resourceful, and so are we. They never
stop thinking about new ways to harm our country and our people, and
neither do we.

Re: bigger integers

[Phillip Lord]

>>>>> "Joe" == Joe Corneli <jcorneli@math.utexas.edu> writes:

  Joe> This fails on big numbers like 12.

  Joe>  (defun factorial (num)
  Joe>    (if (<= num 0)
  Joe>        1
  Joe>      (* num (factorial (1- num)))))

  Joe> Suggestions on how to make it work for bigger numbers?


I think calc can cope with arbitrary precision. 


http://www.synaptics.com/people/daveg/


Cheers

Phil

Re: bigger integers

[Joe Corneli]

   After all, emacs lisp is designed to count characters in a file,
   not to compute factorials.

 "Emacs is the extensible, customizable, self-documenting real-time
 display editor."

If I can't extend or customize it to compute 12! then there seems to
be something missing.  Kevin's factorial works though so I guess
that's good enough for me... oh wait a second now it stopped working.

  Debugger entered--Lisp error: (error "Lisp nesting exceeds max-lisp-eval-depth")
  (factorial (1- num))
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(7)
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(8)
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(9)
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(10)
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(11)
  (* num (factorial (1- num)))
  (if (<= num 0) 1.0 (* num (factorial ...)))
  factorial(12)
  eval((factorial 12))

(This same computation was fine just a minute ago, then I tried to
compute 100! and now it doesn't want to do 12! any more.)

   I think calc can cope with arbitrary precision.

I guess I'll check that out.

Re: bigger integers

[Pascal Bourguignon]

Joe Corneli <jcorneli@math.utexas.edu> writes:

>    After all, emacs lisp is designed to count characters in a file,
>    not to compute factorials.
> 
>  "Emacs is the extensible, customizable, self-documenting real-time
>  display editor."
> 
> If I can't extend or customize it to compute 12! then there seems to
> be something missing.  Kevin's factorial works though so I guess
> that's good enough for me... oh wait a second now it stopped working.

Oh yes, it IS extensible and customizable.
The question is: is it worth the effort to extend and customize it?

And what does 12! have to do with editing files anyway?

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

Our enemies are innovative and resourceful, and so are we. They never
stop thinking about new ways to harm our country and our people, and
neither do we.

Re: bigger integers

[Alan Mackenzie]

Joe Corneli <jcorneli@math.utexas.edu> wrote on Thu, 02 Sep 2004 17:28:17
-0500:

> This fails on big numbers like 12.

>  (defun factorial (num)
>    (if (<= num 0)
>        1
>      (* num (factorial (1- num)))))

> Suggestions on how to make it work for bigger numbers?

Why on earth do you want to write a program for a calculation than just
about anybody could do in his head?

12! is, of course, 478,002,600.  ;-) 

-- 
Alan Mackenzie (Munich, Germany)
Email: aacm@muuc.dee; to decode, wherever there is a repeated letter
(like "aa"), remove half of them (leaving, say, "a").

Re: bigger integers

[Neil Woods]

>>>>> "Alan" == Alan Mackenzie <acm@muc.de> writes:

    >> Suggestions on how to make it work for bigger numbers?

    Alan> Why on earth do you want to write a program for a calculation
    Alan> than just about anybody could do in his head?

    Alan> 12! is, of course, 478,002,600.  ;-)

ITYM 479,001,600.

Besides, as has been already mentioned, the calc package does all this
and more.

-- 
,---------------------------------------------------------------.
|   Neil Woods   |   Do you ever get that feeling of deja vu?   |
`---------------------------------------------------------------'

Re: bigger integers

[Miles Bader]

Joe Corneli <jcorneli@math.utexas.edu> writes:
> Suggestions on how to make it work for bigger numbers?

You might want to look into the calc/math package (it's part of emacs in
CVS), which implements its own bignum support.  It doesn't seem to have
all that much in the way of clean programmatic interfaces for external
use, but the functionality is there:

E.g.

   (defun factorial-work (num)
     (if (math-lessp num '(float 1 0))
         '(float 1 0)
       (math-mul num (factorial (math-sub num '(float 1 0))))))

Works if you use its bignum notation:

   (factorial-work '(float 12 0))
   => (float (bigpos 16 790 4) 2)

I'm not sure what you need to require to use these routines; maybe
(require 'calc) would be enough?

I looked for conversion routines, e.g., to/from strings, for the calc
bignum format; they surely exist, as calc must use them for its own I/O,
but I couldn't find them in my brief search (calc's code is
unfortunately a bit hard to follow in many places).

-Miles
-- 
[|nurgle|]  ddt- demonic? so quake will have an evil kinda setting? one that
            will  make every christian in the world foamm at the mouth?
[iddt]      nurg, that's the goal

Re: bigger integers

[Kevin Rodgers]

Joe Corneli wrote:
 >    After all, emacs lisp is designed to count characters in a file,
 >    not to compute factorials.
 >
 >  "Emacs is the extensible, customizable, self-documenting real-time
 >  display editor."
 >
 > If I can't extend or customize it to compute 12! then there seems to
 > be something missing.  Kevin's factorial works though so I guess
 > that's good enough for me... oh wait a second now it stopped working.
 >
 >   Debugger entered--Lisp error: (error "Lisp nesting exceeds max-lisp-eval-depth")

The obvious pseudo-fix is something like:

(setq max-lisp-eval-depth (* 42 max-lisp-eval-depth))

 >   (factorial (1- num))
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(7)
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(8)
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(9)
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(10)
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(11)
 >   (* num (factorial (1- num)))
 >   (if (<= num 0) 1.0 (* num (factorial ...)))
 >   factorial(12)
 >   eval((factorial 12))
 >
 > (This same computation was fine just a minute ago, then I tried to
 > compute 100! and now it doesn't want to do 12! any more.)

That's weird, it sounds like the stack has been corrupted.

-- 
Kevin Rodgers

Re: bigger integers

[giacomo boffi]

Joe Corneli <jcorneli@math.utexas.edu> writes:

> This fails on big numbers like 12.
>
>  (defun factorial (num)
>    (if (<= num 0)
>        1
>      (* num (factorial (1- num)))))
>
> Suggestions on how to make it work for bigger numbers?

from my *scratch*:

 (defun factorial (num) 
   (if (<= num 0) 
       1 
     (* num (factorial (1- num))))) 

factorial
(factorial 12)

479001600

ah, yes, (insert (emacs-version))

XEmacs 21.5 (beta17) "chayote" (+CVS-20040721) [Lucid] (i686-pc-linux)
of Thu Sep 16 2004 on boffi95

-- 
ATTACKED BY SALESMEN. TORPEDOS. LEAVE BOAT. U-941.