OpenSAML user discussion

["Scott Cantor" <>]

> > but the unit test in the OpenSAML 1.1 head branch correctly do an
> > equality comparison on the string serialized format.
> 
> Well, this doesn't always work.  If one nameid has no format attribute
> and another nameid is explicitly set to 'unspecified', the two
> elements are comparable, aren't they?  (I've written an equals method
> for SAMLNameIdentifier that takes this into account.)

Comparable, certainly. Equal? Depends on who you ask and why. They sure
aren't equal in the eyes of a digital signature.

In an extensible data model, comparison can be extremely complicated, and
leaving it outside the library allows applications to define what
equivalence means for a given use case. It has the downside of not enabling
certain kinds of algorithms, if those algorithms depend on Object.equals().

You don't have to agree with the trade-off I made, but I'm explaining that
the contract works the way I intended it to, that no objects are considered
"equal" unless they are the same object.

There may be a way I wasn't aware of in Java to disallow comparison of
objects that way. Sorry if I didn't use it.

I don't know whether Chad had planned to change this behavior or not. I'd
have to see how it would work in the general case before I would change my
mind.

-- Scott