CLIPS with Java GUI

[jwar...@gmail.com]

I am working on an expert system utilizing CLIPS that I would like to
interface with Java to handle my user interaction. I have looked at
JClips (http://www.cs.vu.nl/~mrmenken/jclips/) and briefly looked at
JESS (http://herzberg.ca.sandia.gov/). I would like to make my
project cross platform working with Windows/MacOSX/UNIX/Linux and
JClips says that they have only confirmed to work on Windows and UNIX,
has anyone been successful on getting it to work on MacOSX (it is a
UNIX based system after all)? Another question I have is if I go with
JESS how similar is it syntactically and functionally to CLIPS?

Thanks,
Jordan

[Denis.Berthier]

I've not tried JClips
But I've developed a large project that runs both in JESS and CLIPS
with no major compatibility problem.
There's currently a restriction on using tests in logicals in JESS
but this should be corrected in a couple weeks.

Le 20 oct. 07 à 20:49, jwar...@gmail.com a écrit :

[CLIPS Support]

Here's the prior thread discussing how to get jclips running on Mac OS
X: http://groups.google.com/group/CLIPSESG/browse_thread/thread/f01ccbb4a7de3293/60c794f74db27b81?lnk=gst&q=jclips#60c794f74db27b81.
You can also find it by searching the group for jclips.

Within the next few days I'm going to post some examples of how to
integrate CLIPS with Java, including 3 examples of a front end written
in Swing interfacing with a CLIPS program.

[jwar...@gmail.com]

I would greatly appreciate that, thanks!

On Oct 21, 2:07 am, CLIPS Support <gdronline2...@swbell.net> wrote:
> Here's the prior thread discussing how to get jclips running on Mac OS

> X:http://groups.google.com/group/CLIPSESG/browse_thread/thread/f01ccbb4....

[Johan Lindberg]

> [...] Another question I have is if I go with

> JESS how similar is it syntactically and functionally to CLIPS?

The modify function works differently in Clips and Jess.

In Jess, a modify that doesn't actually change any slots in a fact is
not given a new index (I don't know if it's avoiding the retract and
re-assert altogether or if it's just given the same index again, but
that doesn't matter). Whereas in Clips, if you modify a fact it will
be retracted, re-asserted and given a new fact-index and that will
make it activatable again!

This means that in Jess you can do:

CLIPS> (deftemplate foo
(slot bar))
CLIPS> (assert (foo (bar some_value)))
<Fact-4>
CLIPS> (defrule clear-foo
?foo <- (foo)
=>
(modify ?foo (bar nil)))

but in Clips, you get an infinite loop. So if you want to have code
that works in both environments you must do:

CLIPS> (defrule clear-foo
?foo <- (foo (bar ~nil))
=>
(modify ?foo (bar nil)))

BR
Johan Lindberg
jo...@pulp.se