[omniORB] IR

Sai-Lai Lo S.Lo@uk.research.att.com
05 Apr 2000 10:56:45 +0100


The stubs for the IR is already compiled into the omniORB runtime (libomniDynamic2).
To make the stub header reachable, just define the CPP macro
ENABLE_CLIENT_IR_SUPPORT and include omniORB2/CORBA.h. Then you have access
to the client and server side skeleton.

It is not possible to use unmodified IR stubs generated by omniidl2. This is because
the IR is defined in the CORBA namespace and compilers that don't support
namespace don't like it. I've added an "expert" mode in omniidl2 to generate
stub fragments and include different parts of these fragments at the right
place in CORBA.h. Its ugly but it works on all compilers, old and new.

Sai-Lai

>>>>> Stefan Seefeld writes:

> just out of curiosity I played with the IR code from mico
> and tried to compile the omniidl2 generated skeletons.
> It seems there are a couple of conflicts with predefined
> types in the omniORB CORBA namespace. How is this supposed
> to work ? Is it impossible to build an interface repository
> with omniORB without to be required to do so *within* omniORB ?

> Someone wants to play with COPE (CORBA in perl) to interface 
> our berlin display server. Unfortunately COPE requires an IR
> so I wanted to see how difficult it would be to build one.
> Any ideas ?


-- 
Sai-Lai Lo                                   S.Lo@uk.research.att.com
AT&T Laboratories Cambridge           WWW:   http://www.uk.research.att.com 
24a Trumpington Street                Tel:   +44 1223 343000
Cambridge CB2 1QA                     Fax:   +44 1223 313542
ENGLAND