Hi Ole,
I would be very interested in your code.
Itens 1) and 3) below are pretty much what I was planning to do. I was not
thinking about setting up a CORBA IR, though. My impression is that
IDL-generated stubs are the most common way of writing CORBA clients.
Agree with you that a fairly complete CORBA implementation for Java is
required. More specifically: POA, portable interceptors, and OBV
support. The ORB in jdk 1.3 has neither of these. JacORB already supports
POA and portable interceptors, and should be a perfect choice by the
end of August, when OBV support is expected. Meanwhile, the options are
commercial products like Orbix or Visibroker, or the ORB in jdk 1.4.
Did some tests with jdk 1.4 and it seems suitable, at least for
development. Hacked a very simple CORBA server which implemented RMI/IIOP
objects using a stream-based dynamic skeleton (without idl or rmic
compilation at all). Was able to call this server both from a pure
CORBA client and from an RMI client built with rmic-gererated stubs.
I think a compile-less approach makes perfect sense on the server side,
but statically generated stubs would still be preferred for clients.
RMI/IIOP Java clients would be mostly built with rmic-generated
stubs, CORBA clients written in other languages would typically use IDL
stubs. I also believe it would be important to support code downloading of
Java stubs (even though this requires rmic compilation at deployment
time, which otherwise would not be needed).
Were you able to test IIOP-based container invocations with JacORB?
Best regards,
Francisco
On Sat, 28 Jul 2001, Ole Husgaard wrote:
> Hi,
>
> I am currently hacking a bit on some IIOP-related code
> for JBoss:
>
> 0) In package org.jboss.iiop, I have written a simple MBean
> that serves an ORB and POA in JNDI "java:" for JBoss to
> use.
>
> 1) In package org.jboss.rmi, I have written some code for
> analysing java classes to get the metadata needed for
> a compile-less RMI-IIOP mapping.
>
> 2) In package org.iiop.rmi.ir, I have written the code for
> a CORBA Interface Repository that can be easily filled
> in with some remote interfaces. This IR is meant to
> publish the metadata of the RMI-IIOP mapping in the
> usual CORBA IR format.
>
> 3) In package org.iiop.rmi.server, I have written some code
> for supporting an IIOP-based container invoker for JBoss.
> This contains code for unmarshaling invocation arguments,
> and marshaling return values and exceptions thrown.
>
> This is all very unstable/unfinished/untested, and I am not
> yet really sure what way this is going, so I have committed
> nothing yet.
> But with sufficient luck, maybe JBoss will end up supporting
> IIOP with the same ease as it currently supports JRMP.
>
> It needs a fairly new CORBA implementation for Java to run,
> and it seems that the one from Sun it not even complete
> enough to compile.
> Currently, I am testing my code on JacORB (www.jacorb.org),
> with a little hack I recently posted to the JacORB mailing
> list for support of the value type TypeCodes, and support
> for Value ValueMember and ValueBox IR entry types.
> Unfortunately, JacORB will not be able to support the
> transfer of value types over IIOP until late August, and
> that is needed for our bean invocations over IIOP.
>
> >From manual inspection of published test interface
> repositories from 2) above using the JacORB irbrowser, it
> looks like 1) and 2) are pretty complete. (Still some
> confusion about abstract IDL interfaces: Looks like OMG
> did some last-minute spec-changes, and that they are
> currently publishing IDL that does not conform fully to
> their own spec.)
> But 3) is only half-finished.
> I haven't even thought about the client side of things,
> but the interface repository is so complete that it should
> be possible (in theory) for a client to work with no code
> downloading at all.
> For non-Java clients some kind of bootstrap (CORBA naming
> service?) would be needed. Probably it would also be nice
> if the IR could print out IDL so that the metadata is
> available in IDL format for non-Java clients.
>
> I _could_ commit this half-baked code, but then I would
> have to also commit a hacked version of jacorb.jar to
> JBoss. Probably it would be better to wait until JacORB
> officially supports this (planned for the end of August).
> If anybody is interested, it would probably be better if
> I uploaded the code to the web.
>
>
> Best Regards,
>
> Ole Husgaard.
>
>
> Francisco Reverbel wrote:
> >
> > Hello JBoss-ers,
> >
> > Is there anybody working on an IIOP plugin for JBoss?
> >
> > I have some ideas on this line and would like to know if somebody is
> > already working on it.
> >
> > Thanks for such a great open-source EJB server!
> >
> > Francisco
> > --
> > Francisco Reverbel, Assistant Professor
> > Department of Computer Science [EMAIL PROTECTED]
> > University of Sao Paulo, Brazil http://www.ime.usp.br/~reverbel/
> >
> > _______________________________________________
> > Jboss-development mailing list
> > [EMAIL PROTECTED]
> > http://lists.sourceforge.net/lists/listinfo/jboss-development
>
> _______________________________________________
> Jboss-development mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/jboss-development
>
_______________________________________________
Jboss-development mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-development
