[E-Lang] Remote Object E Question

[Nathan Combs]

  As sort of a 0 level experiment to assist my "groking E", i'm trying to 
do the following:


1.) Given a distributed system of java processes (not important for this 
discussion, but these are "agents" in the sense of 
http://www.cougaar.org/)  which wish to communicate via backchannel E.

2.) Right now, i have it set up so they pass in generated script to 
org.erights.e.elang.interp.Interp  (thanks Mark for previous help). Since 
they can share URIs  via agent infrastructure - these E sessions can synch 
up URIs.

Okay, this is just an experiment, but what i'd like to try to do to see 
what it looks like is:

A.) Have these java processes (agents) essentially invoke a method on a 
remote object (from a collection of objects at another agent) via E.    The 
question here is then how do I/ can I  bind an object from the runtime to 
the E script interpretation?   I found documentation on what seems like 
java object creation and invocation - using "E call..."  - but in my case I 
want to take a "live" object, import it into my E session, obtain a URI and 
offer it up for access by others via E.

B.) the other question is this:  presuming I can bind a live object to my E 
session, is there any way that I can "wrap" it in E?   So for example, I 
may want to hide it behind a "transparent forwarder" (per example given at 
bottom of http://www.skyhunter.com/marcs/ewalnut.html#SEC14)  so that the 
owning process may choose to revoke its capabilities should it choose?

thanks, -n.