Re: equality



Date view Thread view Subject view Author view Attachment view

From: Tim Klinger (klinger@cs.nyu.edu)
Date: Sun 23 Nov 2003 - 14:49:14 GMT


Bernhard,

Let me try to clarify what I mean. First, I do think object identity is a useful equality operation. I was under the impression that the spec prescribed object identity as the _only_ form of equality. Looking at the spec again I realize that the intention is probably that it be overriden in user classes, giving it the same semantics as Object::equal() in java. However, it is not overriden in the primitive data type classes leading to the (I think) unintuitive result that 1 <> 1.0.

--Tim

Date view Thread view Subject view Author view Attachment view