And as Daniel mentioned earlier, it's not at all obvious what we mean by "bits used" when it comes to negative numbers. GMP pretends that any negative number has infinite bits in the two's-complement representation. Should we count the highest _unset_ bit when the number is negative?

Or to put it another way, what invariants should the proposed bit-counting function satisfy?

On Thu, Aug 25, 2011 at 6:32 PM, Albert Y. C. Lai <trebla@vex.net> wrote:
On 11-08-25 01:57 PM, Andrew Coppin wrote:
Does anybody else think it would be *far* more useful if bitSize applied
to an Integer would tell you how many bits that particular Integer is
using? Especially given that it can vary?

It is useful to know the number of bits used in a value.

It is useful to know the maximum number of bits storable in a type.

It is useless to conflate the two into one single method.

The name "bitSize" is already taken. Come up with another name, and I will agree with you.


_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe