ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Larry V. Streepy, Jr." <stre...@healthlanguage.com>
Subject Re: PATCH to Rmic.java for IIOP
Date Tue, 26 Jun 2001 15:07:52 GMT
After some discussion with our weblogic guru here, it seems that we're
not processing any of the implementation classes (which would generate
_Tie.class files) because the weblogic server "handles" this internally.
It doesn't need iiop specific classes for the server, just the clients.

So, that explains why I wasn't seeing the _Tie files being generated.

As for the rest, it does appear that reflection is the safest way to go
for determining the generated file names.

1. When processing an interface (like a Home interface), the _Stub class
name will be the same as the interface.  E.g., VersionHome becomes
_VersionHome_Stub.  No tie class is created.

2. When processing a concrete class, the tie file is named from the
class and the stub is named from the implemented interface.  From
Stefan's example, public class SomeImpl implements Some (and Some
extends Remote), results in _Some_Stub and _SomeImpl_Tie.

Does everyone agree that this seems to capture the name generation
accurately?

Stefan, where are you at on this?  SHould I wait for your patches, or
try to implement the above myself?

Thanks.

Stefan Bodewig wrote:
> 
> Larry V. Streepy, Jr. <streepy@healthlanguage.com> wrote:
> 
> > In case anyone wants the rmic task to properly handle the IIOP
> > stub naming conventions, I'm posting this patch to the released
> > version 1.3 Rmic.java file.
> 
> In case people haven't been following this thread but do know what to
> do here.
> 
> My own little experiments show a naming convention that is a little
> difficult to follow.  If my class SomeImpl implements interface Some
> and Some extends Remote, rmic -iiop will result in _SomeImpl_Tie.class
> and _Some_Stub.class - that is the stub file name will be taken from
> the name of the interface, not the implementation.
> 
> So - to determine the name of the stub file, Ant will need to use
> reflection on the class to determine the name of the interface.  Am I
> missing something?
> 
> Stefan

-- 
Larry V. Streepy, Jr.
Chief Technical Officer and VP of Engineering

Health Language, Inc.  -- "We speak the language of healthcare"

970/626-5028 (office)           mailto:streepy@healthlanguage.com
970/626-4425 (fax)              http://www.healthlanguage.com

Mime
View raw message