Hey
Evan Ireland wrote:
> It is surprising to me that when the spec so clearly states that this
> class must be used to portably narrow references, that it has been so
> extraordinarily difficult for us to obtain an implementation! The most
> recent thing we were told by one Sun representative was that we could
> implement it ourselves, as apparently some other vendors have done! I
> wouldn't be surprised if this violates our license. Apart from that,
> probably one day some version of the JDK will include this class, so
> we wanted to be sure that our EJB stub architecture would be compatible
> from day one with whatever Sun intended to deliver.
Indeed, it will be part of a standard extension. Did they actually say
that?? Geez..
> We have been told that we are one of only two vendors who have requested
> this class from Sun. If this is true, it shows an appaling lack of
> interest from the vendors in portability of EJB client programs! I am
> very interested in hearing the vendor responses to this point.
I've requested it for EJB implementation purposes, but I guess I don't
count 8-) (todo: gotta get a business card)
> If you have an EJB runtime that allows direct casting without the
> use of PortableRemoteObject.narrow, you will probably find (as I did)
> that PortableRemoteObject.narrow works fine, so the addition of
> PortableRemoteObject.narrow to make your client portable won't break
> your existing stub archtiecture. I hope that Sun will at least commit
> to this continuing to work.
I remember seeing some narrow(..) code somewhere and indeed, if the
given object already is instanceof the needed class/interface then it is
directly returned.
> As you can see by my response, I am not entirely pleased with the
> process by which we have finally obtained the implementation of this
> class, but I am pleased that it is finally available, and I do hope for
> some further discussion of this issue.
Yes, it has been awfully quiet about the RMI/IIOP package. Extremely
hard to get any info. But I guess this is due to the fact that IBM is
running the show (AFAIK), so Sun may not have that much influence on it
(sending a comment to [EMAIL PROTECTED] gives a reply from IBM
engineers).
/Rickard
--
Rickard �berg
Computer Science student@LiTH
@home: +46 13 177937
Email: [EMAIL PROTECTED]
Homepage: http://www-und.ida.liu.se/~ricob684
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
