Arithmetic issues Michael Sperber (18 Oct 2005 06:03 UTC)
Re: Arithmetic issues felix winkelmann (18 Oct 2005 07:00 UTC)
Re: Arithmetic issues John.Cowan (18 Oct 2005 17:36 UTC)
Re: Arithmetic issues Aubrey Jaffer (19 Oct 2005 18:13 UTC)
Re: Arithmetic issues John.Cowan (19 Oct 2005 18:21 UTC)
Re: Arithmetic issues bear (18 Oct 2005 19:52 UTC)
Re: Arithmetic issues John.Cowan (18 Oct 2005 21:12 UTC)
Re: Arithmetic issues bear (19 Oct 2005 02:13 UTC)
Re: Arithmetic issues John.Cowan (19 Oct 2005 02:19 UTC)
Re: Arithmetic issues bear (19 Oct 2005 03:23 UTC)
Re: Arithmetic issues Andre van Tonder (19 Oct 2005 11:47 UTC)
Re: Arithmetic issues Aubrey Jaffer (19 Oct 2005 14:14 UTC)
Re: Arithmetic issues Andre van Tonder (19 Oct 2005 16:00 UTC)

Re: Arithmetic issues Aubrey Jaffer 19 Oct 2005 18:13 UTC

 | Date: Tue, 18 Oct 2005 13:36:39 -0400
 | From: "John.Cowan" <xxxxxx@reutershealth.com>
 |
 | Michael Sperber scripsit:
 | ...
 | > The fixnum, flonum, and inexact arithmetic come with a full deck of
 | > operations, including some that are defined in terms of integers (such
 | > as quotient+remainder, gcd and lcm), and others that are easily abused
 | > (such as fxabs). Should these be pruned?
 |
 | Yes, absolutely.  (I'm not clear what the problem with fxabs might be.)

In twos-complement fixnums there is one more negative number than
there are positive numbers.

Thus (abs most-negative-fixnum) and (- most-negative-fixnum) can only
return a bignum: (+ 1 most-positive-fixnum)