From: Pascal Costanza
Subject: ILU?
Date: 
Message-ID: <b483ro$o1a$1@f1node01.rhrz.uni-bonn.de>
Hi,

There are repeatedly complaints about the lack of a standardized FFI in 
CL. Have you also considered ILU? Does anyone have any experience with this?

See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html


Pascal

-- 
Pascal Costanza               University of Bonn
···············@web.de        Institute of Computer Science III
http://www.pascalcostanza.de  R�merstr. 164, D-53117 Bonn (Germany)

From: Marco Antoniotti
Subject: Re: ILU?
Date: 
Message-ID: <paN9a.55$qc.1353@typhoon.nyu.edu>
Pascal Costanza wrote:

> Hi,
>
> There are repeatedly complaints about the lack of a standardized FFI in
> CL. Have you also considered ILU? Does anyone have any experience with 
> this?
>
> See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html


If memory does not fail me (or if new things have happened meanwhile), 
ILU is a prime example of ills of the lack of a standardized FFI for 
Common Lisp.

Cheers

--
Marco Antoniotti
From: Pascal Costanza
Subject: Re: ILU?
Date: 
Message-ID: <b4a6f7$15gi$1@f1node01.rhrz.uni-bonn.de>
Marco Antoniotti wrote:

> If memory does not fail me (or if new things have happened meanwhile), 
> ILU is a prime example of ills of the lack of a standardized FFI for 
> Common Lisp.

You are probably right - I have downloaded ILU yesterday, and they 
actually only support Allegro Common Lisp. I should have checked that 
before posting my question here.


Pascal

-- 
Pascal Costanza               University of Bonn
···············@web.de        Institute of Computer Science III
http://www.pascalcostanza.de  R�merstr. 164, D-53117 Bonn (Germany)
From: Kenny Tilton
Subject: Re: ILU?
Date: 
Message-ID: <3E67B177.8090000@nyc.rr.com>
Pascal Costanza wrote:
> Hi,
> 
> There are repeatedly complaints about the lack of a standardized FFI in 
> CL. 

Not so! Vendors realized years ago that it would be good for Lisp and 
hence good for them to standardize the FFI and got together to....

:)

-- 

  kenny tilton
  clinisys, inc
  http://www.tilton-technology.com/
  ---------------------------------------------------------------
"Cells let us walk, talk, think, make love and realize
  the bath water is cold." -- Lorraine Lee Cudmore
From: Fernando Mato Mira
Subject: Re: ILU?
Date: 
Message-ID: <dc2f1d1b.0303070733.276416a7@posting.google.com>
Pascal Costanza <········@web.de> wrote in message news:<············@f1node01.rhrz.uni-bonn.de>...
> Hi,
> 
> There are repeatedly complaints about the lack of a standardized FFI in 
> CL. Have you also considered ILU? Does anyone have any experience with this?
> 
> See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html

In a previous life, I've used ILU with C++ (gasp) and it was good,
except for some occasional crashes.

But the project has been stuck at version 2.0alpha for years. I guess
it was abandoned.

Omniorb (C++ and python only) was abandoned by AT&T and now lives in
sourceforge. [That's what the people here use for C++]

Although for C++, I'd personally take a look at TAO
(http://www.cs.wustl.edu/~schmidt/TAO.html) I haven't used it, but ACE
is great (if anything written in C++ can be called so).

But who cares about C++? ILU NEEDS TO LIVE ON!!!
From: Pascal Costanza
Subject: Re: ILU?
Date: 
Message-ID: <b4aehm$15lq$1@f1node01.rhrz.uni-bonn.de>
Fernando Mato Mira wrote:
> Pascal Costanza <········@web.de> wrote in message news:<············@f1node01.rhrz.uni-bonn.de>...
> 
>>Hi,
>>
>>There are repeatedly complaints about the lack of a standardized FFI in 
>>CL. Have you also considered ILU? Does anyone have any experience with this?
>>
>>See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html
> 
> 
> In a previous life, I've used ILU with C++ (gasp) and it was good,
> except for some occasional crashes.

Does it carry a lot of overhead, like for example CORBA does, or can it 
reasonably be used for just accessing native code libraries on the same 
platform?


Pascal

-- 
Pascal Costanza               University of Bonn
···············@web.de        Institute of Computer Science III
http://www.pascalcostanza.de  R�merstr. 164, D-53117 Bonn (Germany)
From: Fernando Mato Mira
Subject: Re: ILU?
Date: 
Message-ID: <dc2f1d1b.0303120632.2019e4b9@posting.google.com>
Pascal Costanza <········@web.de> wrote in message news:<·············@f1node01.rhrz.uni-bonn.de>...
> Fernando Mato Mira wrote:
> > Pascal Costanza <········@web.de> wrote in message news:<············@f1node01.rhrz.uni-bonn.de>...
> > 
> >>Hi,
> >>
> >>There are repeatedly complaints about the lack of a standardized FFI in 
> >>CL. Have you also considered ILU? Does anyone have any experience with this?
> >>
> >>See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html
> > 
> > 
> > In a previous life, I've used ILU with C++ (gasp) and it was good,
> > except for some occasional crashes.
> 
> Does it carry a lot of overhead, like for example CORBA does, or can it 
> reasonably be used for just accessing native code libraries on the same 
> platform?

I used it exclusively with C++. In that case, calling methods in the
same address space involves just dispatching to the subclass. No
serialization involved.

If you use it across languages or memory spaces, you'll have to put up
with all the overhead (I haven't timed it, but I've looked at the
code, and it's quite a bit).

I wouldn't have used it for FFI, but, hey, I was doing VR.