Draft proposal: allow the use of relational operators on Comparable classes (original) (raw)

Neal Gafter neal at gafter.com
Tue Mar 10 17:21:28 PDT 2009

Actually, I was referring to issues like the treatment of NaN, negative zero, and the like. Given

Float x1 = NaN; Float x2 = NaN;

currently x1 <= x2 is false. Under this proposal it would be true. That's a breaking change.

-Neal

On Tue, Mar 10, 2009 at 5:03 PM, Reinier Zwitserloot <reinier at zwitserloot.com> wrote:

Reply inline

 --Reinier Zwitserloot

On Mar 10, 2009, at 17:21, Neal Gafter wrote: This proposal is incompatible with the existing behavior of the comparison operators on the boxed value classes, particularly for Float and Double. I'll help Neal out and be a bit more verbose: Existing behaviour unboxes first, which also means that you get type promotion to make parts fit. You can do this: Float x = 10.0; Double y = 12.0; if ( x < y ) /* stuff */;_ _but you can not call x.compareTo(y), or y.compareTo(x). (AFAIK - it_ _would be nice if someone tests this theory. Don't have a javac_ _available at the moment)._ _Easily fixed in the proposal by stating: If the operation would_ _succeed under old unboxing rules, use those._ _After all, there are ambiguous rules regarding == as well (if I == two_ _autoboxed values, is it reference comparison, or an equals check)? A_ _choice was made there too._ _On Tue, Mar 10, 2009 at 9:09 AM, Vilya Harvey_ _<vilya.harvey at gmail.com> wrote: I've attached a draft of a proposal to allow classes which implement the Comparable interface to be used as operands for the relational operators. So for example if you had two Strings, a and b, you would be able to write

if (a < b) {  ... } instead of if (a.compareTo(b) < 0) {  ... } and you could do the same with your own classes as well. Thanks in advance for any feedback, Vil.

More information about the coin-dev mailing list