Well, this could just be one facet of the strange and wonderful world
of multislot matching. This is a grey area that will be clarified
somewhat in Jess 7; this has come up in the past. It's hard to
explain, but basically, ?a1 and ?a2 aren't necessarily references to
the facts in working memory; they might just be objects that *look*
like those facts. When the real facts are modified, ?a1 and ?a2 won't
match perfectly anymore.
To get ahold of the actual facts in working memory so you can send
them to the client, you can call "getIcon()" on the facts themselves
(this is a horrible method name, I know.) So doing something like
this:
(new-cluster (create$ ?*globcid*) (?a1 getIcon) (?a2 getIcon))
should solve the problem. Please let me know if this is it.
I think Nik Joshi wrote:
> The IDs are just data, not really relevant to the problem, the "list of IDs"
> is just a multislot of integers. In my JESS code, I basically have a rule
> whose LHS matches a fact. I take that fact on the RHS and I modify the
> contents of the multislot of IDs. I then (still in the RHS) send the newly
> modified fact to the userfunction in my Java code. When I print out the
> values in the userfunction, they are correct. The userfunction then creates
> an object with the fact (and some other data), serializes the object and
> sends it to the client process. The client deserializes the object, and
> when it prints out the values in the multislot of IDs (using null as the
> context), the values are incorrect.
>
> Here are some snippets of my code:
>
>
> JESS:
> -----
>
> ?a1 <- (alert (name ?name1) (ID ?id1) (requires $?) (provides $? ?pro1 $?)
> (globfacts $?) (rollups ?) (timestamp ?t1) (cluster-ids $?cids1) (depth ?d1)
> (retirement ?rm1))
>
> =>
> (modify ?a2 (globfacts (insert$ $?gf2 1 ?a1)) (depth (+ ?d1 1)))
>
> ; create a new cluster
> (modify ?a1 (cluster-ids (union$ $?cids1 (create$ ?*globcid*))))
> (modify ?a2 (cluster-ids (union$ $?cids2 (create$ ?*globcid*))))
>
> (new-cluster (create$ ?*globcid*) ?a1 ?a2)
>
>
> Server Side Java:
> -----------------
>
> public class NewCluster implements Userfunction {
>
> // params are: cluster-id, fact1, fact2
> public String getName() { return ("new-cluster"); }
>
> public Value call (ValueVector vv, Context context) throws
> JessException {
>
> ValueVector cid = vv.get(1).listValue(context);
> Fact f1 = vv.get(2).factValue(context);
> Fact f2 = vv.get(3).factValue(context);
>
> RPData rpd = new RPData (RPData.NEW, cid, f1, f2);
>
> try {
> clientout.writeObject (rpd);
> } catch (IOException ioe) {System.out.println ("got ioe: " +
> ioe);}
>
> return (new Value (true));
> }
> }
>
>
>
>
> Client Side Java (after deserializing object):
> ----------------------------------------------
>
> public void new_cluster (ValueVector cids, Fact f1, Fact f2, Vector
> tempcids1, Vector tempcids2) throws JessException {
>
> int cid = cids.get(0).intValue(null);
> int depth = f1.getSlotValue ("depth").intValue(null) + 1;
> String alertid = f2.getSlotValue ("name").stringValue(null);
> String category = f2.getSlotValue ("category").stringValue(null);
> double av = ((f1.getSlotValue ("av").numericValue(null) >
> f2.getSlotValue
> ("av").numericValue(null)) ?
> f1.getSlotValue ("av").numericValue(null) :
> f2.getSlotValue
> ("av").numericValue(null));
> boolean hashyp = ((f1.getSlotValue ("ID").intValue(null) < 0 ||
> f2.getSlotValue ("ID").intValue(null) < 0) ? true : false);
> ValueVector ips1 = f1.getSlotValue ("ips").listValue(null);
> ValueVector subnets1 = f1.getSlotValue ("subnets").listValue(null);
> ValueVector ips2 = f2.getSlotValue ("ips").listValue(null);
> ValueVector subnets2 = f2.getSlotValue ("subnets").listValue(null);
>
>
> ClusterState cs = new ClusterState (cid, depth, alertid, category, av,
> hashyp);
>
>
>
>
> Any ideas?
>
> - Nik.
>
>
>
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of [EMAIL PROTECTED]
> Sent: Friday, December 12, 2003 6:00 AM
> To: [EMAIL PROTECTED]
> Subject: Re: JESS: problems with serialization
>
>
> I think Nik Joshi wrote:
>
> > Can anyone shed any light on this, or do I need to give more detail?
>
> I don't know what IDs are in this context, and I don't know if a
> "list" is the contents of a multislot or a Java List, so it's hard to
> narrow it down. The only thing I can think of is to point out that
> when a multislot is modified, it may or may not be that the
> ValueVector object in the slot is a different one than before the
> modification, so that if you fetch a ValueVector using
> fact-slot-value, then modify the fact, the contents of that
> ValueVector may, or may not, change.
>
> If this doesn't help, then you probably need to show us some code.
>
>
>
> ---------------------------------------------------------
> Ernest Friedman-Hill
> Science and Engineering PSEs Phone: (925) 294-2154
> Sandia National Labs FAX: (925) 294-2234
> PO Box 969, MS 9012 [EMAIL PROTECTED]
> Livermore, CA 94550 http://herzberg.ca.sandia.gov
>
> --------------------------------------------------------------------
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list
> (use your own address!) List problems? Notify [EMAIL PROTECTED]
> --------------------------------------------------------------------
>
> --------------------------------------------------------------------
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list
> (use your own address!) List problems? Notify [EMAIL PROTECTED]
> --------------------------------------------------------------------
>
---------------------------------------------------------
Ernest Friedman-Hill
Science and Engineering PSEs Phone: (925) 294-2154
Sandia National Labs FAX: (925) 294-2234
PO Box 969, MS 9012 [EMAIL PROTECTED]
Livermore, CA 94550 http://herzberg.ca.sandia.gov
--------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]
--------------------------------------------------------------------
