Newsgroups: comp.lang.lisp,comp.os.os2.programmer.oop,comp.object
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!EU.net!sun4nl!hacktic!ijssel!arnoud
From: arnoud@ijssel.xs4all.nl (Arnoud Martens)
Subject: Re: wanted lisp for os/2
References: <78329@sdcc12.ucsd.edu> <ONEEL.95Jan15145046@arupa.gsfc.nasa.gov> <Yj8kDK_00YUz0F5YY7@andrew.cmu.edu>
Organization: IJssel service
Date: Wed, 25 Jan 1995 23:00:20 GMT
Message-ID: <D2zH8K.1s1@ijssel.xs4all.nl>
Reply-To: arnoudm@ijssel.xs4all.nl
Lines: 26
Xref: glinda.oz.cs.cmu.edu comp.lang.lisp:16510 comp.os.os2.programmer.oop:692 comp.object:25648

In article <Yj8kDK_00YUz0F5YY7@andrew.cmu.edu>,
Kenton Shaver  <kenton+@CMU.EDU> wrote:
>I'm wondering what the DSOM bindings for Lisp would look like.
>Xerox's Inter-Language Unification system(*) has Lisp mappings, but it
>is not yet clear to me how close they are to CORBA IDLs.  The DLLs in
>Warp include the DSOM/SOM2 stuff, so it would seem possible to make
>bindings (if so doing is possible in any interpreted language) with a
>compiler and the stuff in those DLLs.   Is this a realistic project?

I am no lisp expert, buut this seems perfectly reasonable to
me. The idea of IDL is to create a langauge indepent description
of an object interface. Mappings (or language bindings) bring
invocations on the object (and access to the ORB) in to the
programming domain of a client. So you can invoke methods on
server objects just like you would invoke methods on local
objects.

In fact the bindings are similar qto the stubs you have to link in
when you are calling functions from a DLL. It is like saying this
function is not available now (==compile time) but it will be at
run-time if you open this dll and call this entry.

Gtx:
-- 
Name: Arnoud Martens, Utrecht, the Netherlands,  tel: +31-30-732679
E-mail: arnoudm@ijssel.xs4all.nl WWW: http://www.xs4all.nl/~arnoudm
