Re: Relieving some ClassLoader contention in PreferredClassProvider

Gregg Wonderly wrote:

>  From an outside view, that's a fairly opaque description.  I'm trying to
> understand whether it's intended to be that opaque or if there are a
> handful of
> known issues that URL.equals() covers which make up the required
> behavior.  The
> DNS related details are perhaps part of the puzzle, but I'm just curious
> if you
> or Peter or someone else has any other specific, historical thoughts to
> clarify
> what might be at issue if an RMIClassLoaderSPI implementation used
> URL.toString().equals() instead of URL.toString(), besides the obvious
> address
> vs name considerations.

Gregg, there is one thing I can come up with that would disqualify plain
string comparison and that is case insensitivity for the host name and
the scheme of the URL. But I'm getting more convinced that replacing
many of the URL used in the current implementation by URIs is the right
thing to do.

I'm very curious to know who is relying on the fact that 'different'
URLs are sometimes considered equal. If it turns out nobody is, I don't
see any harm in changing the spec.
--
Mark

--------------------------------------------------------------------------
Getting Started:     http://www.jini.org/wiki/Category:Getting_Started
Community Web Site:  http://jini.org
jini-users Archive:  http://archives.java.sun.com/archives/jini-users.html
Unsubscribing:       email "signoff JINI-USERS"  to [email protected]