Comparison of numerical types

Comparison of numerical types

[Panicz Maciej Godek]

[-- Attachment #1: Type: text/plain, Size: 504 bytes --]

The equality and comparison predicates on gnu.math.UByte, gnu.math.UInt and
gnu.math.ULong require that they are compared with objects of the same type.

This results in situations, where numbers with identical print
representations are not equal (in the sense of both "equal?" and "=") to
each other, and where the "less than" operator behaves counter-intuitively:


> (define b0 (gnu.math.UByte:valueOf (as byte 241)))
> (= b0 241)
#f
> (< b0 242)
#f

Is there any particular reason for such behavior?

Re: Comparison of numerical types

[Per Bothner]

On 1/21/24 02:11, Panicz Maciej Godek via Kawa wrote:
> The equality and comparison predicates on gnu.math.UByte, gnu.math.UInt and
> gnu.math.ULong require that they are compared with objects of the same type.

no, that is not the case - or at least I don't think it should be
> 
> This results in situations, where numbers with identical print
> representations are not equal (in the sense of both "equal?" and "=") to
> each other, and where the "less than" operator behaves counter-intuitively:

>> (define b0 (gnu.math.UByte:valueOf (as byte 241)))
>> (= b0 241)
> #f
>> (< b0 242)
> #f
> 
> Is there any particular reason for such behavior?

Just a bug. Fixed now.

My guess is this was a copy-paste error - I copied code from the UShort
implementation into the UByte implementation, incorrectly.
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/