Peter,

>><survey>
>>  Have any regulars actually tried AltRMI yet?
>>  Thouhts?
>></survey>
>>
>
>I tried it a bit back but haven't had a chance to try it again ;)
>
Well it has changed a little.  More transports, more robust.
This might help for those that have little time ( pasted from a private 
email ) :

Proposal : 
http://cvs.apache.org/viewcvs/jakarta-commons-sandbox/altrmi/PROPOSAL

If you (like me) are example orientated, then this should help:
Consider an interface.........
 
http://cvs.apache.org/viewcvs/jakarta-commons-sandbox/altrmi/src/java/org/apache/commons/altrmi/test/TestInterface.java

And the serverside class that implements it......
 
http://cvs.apache.org/viewcvs/jakarta-commons-sandbox/altrmi/src/java/org/apache/commons/altrmi/test/TestInterfaceImpl.java

Now it's use on the client side..........
 
http://cvs.apache.org/viewcvs/jakarta-commons-sandbox/altrmi/src/java/org/apache/commons/altrmi/test/SocketClientTest.java

( The the smallest code would be ) ........

 ArmiFactory af = new ServerClassAltrmiFactory(false);
 af.setHostContext(new SocketObjectStreamHostContext("127.0.0.1", 1234));
 TestInterface ti = (TestInterface) af.lookup("Hello");
 new TestClient(ti);  // client performs tests and is blind to where the 
impl is.

The crucial bit is the interface.  It is not special.  It does not 
extends Remote.  It does not have "throws RemoteException" on every 
method.  It is just a normal* Java interface.  * Should, but not has to 
be designed along facade pattern lines.

Regards,

- Paul


--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to