Well, in the second code snippet you are trying to set an unexistent variable called test2. For setting a global variable you need to define it first in the drl file. Greetings! - CTO @ http://www.plugtree.com - MyJourney @ http://salaboy.wordpress.com - Co-Founder @ http://www.jbug.com.ar - Mauricio "Salaboy" Salatino - On Jun 22, 2010, at 4:32, Samuli Saarinen <samuli.saarinen(a)remion.com&gt; wrote: > Hello, > > I'm new to drools and have problems with globals or with the different > behaviour of drools depending on how the globals are specified. > > I have the following simple rule: > > global String test > > rule "Global" > when > then > System.out.println(test); > end > > If i run the rule with: > > KnowledgeBase kb = getKnowledgeBase("global.drl"); > StatelessKnowledgeSession ses = kb.newStatelessKnowledgeSession(); > ses.setGlobal("test2", "foo"); > ses.execute(""); > > I get the expected result of null printed > > but if I use commands: > > KnowledgeBase kb = getKnowledgeBase("global.drl"); > StatelessKnowledgeSession ses = kb.newStatelessKnowledgeSession(); > List<Command<?>> cmds = new ArrayList<Command<?>>(); > cmds.add(CommandFactory.newSetGlobal("test2", "bar")); > cmds.add(CommandFactory.newInsert("")); > ses.execute(CommandFactory.newBatchExecution(cmds)); > > I get java.lang.RuntimeException: Unexpected global [test2] > > Why is the behaviour different and which one is correct? > > Regards, > > Samuli > > > -- > Remion Oy Etävalvontajärjestelmät liiketoiminnan > Samuli Saarinen tehostamiseen > gsm +358 (0)50 3560075 > fax +358 (0)3 2125064 www.remion.com > _______________________________________________ > rules-users mailing list > rules-users(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/rules-users _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users -- STOP Virus, STOP SPAM, SAVE Bandwidth! http://www.safentrix.com/adlink?cid=0 -

yes you are right.. did you try removing the insert cmd from the second code snippet? just to have the same scenario? The first code should fail as the second. With the message Unexpected global [test2] On Tue, Jun 22, 2010 at 8:40 AM, Samuli Saarinen <samuli.saarinen(a)remion.com <mailto:samuli.saarinen@remion.com>> wrote: Actually I am trying to set an unexistent variable (test2) in both snippets. But for some reason only the second one produces the exception. Samuli Salaboy wrote: > Well, in the second code snippet you are trying to set an unexistent > variable called test2. For setting a global variable you need to > define it first in the drl file. Greetings! > > - CTO @ http://www.plugtree.com > - MyJourney @ http://salaboy.wordpress.com > - Co-Founder @ http://www.jbug.com.ar > - Mauricio "Salaboy" Salatino - > > On Jun 22, 2010, at 4:32, Samuli Saarinen <samuli.saarinen(a)remion.com <mailto:samuli.saarinen@remion.com>> > wrote: > >> Hello, >> >> I'm new to drools and have problems with globals or with the different >> behaviour of drools depending on how the globals are specified. >> >> I have the following simple rule: >> >> global String test >> >> rule "Global" >> when >> then >> System.out.println(test); >> end >> >> If i run the rule with: >> >> KnowledgeBase kb = getKnowledgeBase("global.drl"); >> StatelessKnowledgeSession ses = kb.newStatelessKnowledgeSession(); >> ses.setGlobal("test2", "foo"); >> ses.execute(""); >> >> I get the expected result of null printed >> >> but if I use commands: >> >> KnowledgeBase kb = getKnowledgeBase("global.drl"); >> StatelessKnowledgeSession ses = kb.newStatelessKnowledgeSession(); >> List<Command<?>> cmds = new ArrayList<Command<?>>(); >> cmds.add(CommandFactory.newSetGlobal("test2", "bar")); >> cmds.add(CommandFactory.newInsert("")); >> ses.execute(CommandFactory.newBatchExecution(cmds)); >> >> I get java.lang.RuntimeException: Unexpected global [test2] >> >> Why is the behaviour different and which one is correct? >> >> Regards, >> >> Samuli >> >> >> -- >> Remion Oy Etävalvontajärjestelmät liiketoiminnan >> Samuli Saarinen tehostamiseen >> gsm +358 (0)50 3560075 >> fax +358 (0)3 2125064 www.remion.com <http://www.remion.com> >> _______________________________________________ >> rules-users mailing list >> rules-users(a)lists.jboss.org <mailto:rules-users@lists.jboss.org> >> https://lists.jboss.org/mailman/listinfo/rules-users > > _______________________________________________ > rules-users mailing list > rules-users(a)lists.jboss.org <mailto:rules-users@lists.jboss.org> > https://lists.jboss.org/mailman/listinfo/rules-users > > -- > STOP Virus, STOP SPAM, SAVE Bandwidth! > http://www.safentrix.com/adlink?cid=0 > - -- Remion Oy Etävalvontajärjestelmät liiketoiminnan Samuli Saarinen tehostamiseen gsm +358 (0)50 3560075 fax +358 (0)3 2125064 www.remion.com <http://www.remion.com> _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org <mailto:rules-users@lists.jboss.org> https://lists.jboss.org/mailman/listinfo/rules-users -- - CTO @ http://www.plugtree.com - MyJourney @ http://salaboy.wordpress.com - Co-Founder @ http://www.jbug.com.ar - Salatino "Salaboy" Mauricio - ------------------------------------------------------------------------ STOP Virus, STOP SPAM, SAVE Bandwidth! www.safentrix.com <http://www.safentrix.com/adlink?cid=0> ------------------------------------------------------------------------ ------------------------------------------------------------------------ _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users

Greetings all, I have a requirement to compare a set of facts against each other, calculate a value that indicates the amount of difference (Levenshtein distance) between a single attribute on each fact. So I load all of my facts into working memory then what? Do I execute a rule first that creates a new set of facts which is a Cartesian product of the original set of facts, each new fact containing a reference to two original facts and the diff between them? Looked at using a custom accumulate function but I don't think that's going to do what I need since it is intended to go through a set of facts and return a single value. Any and all related help is appreciated. Earnie! _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users

Greg, Thanks for the example. This should do what I need. Can you point me to some documentation on how to create custom operators? Earnie! ------------------------------ *From:* Greg Barton <greg_barton(a)yahoo.com&gt; *To:* Rules Users List <rules-users(a)lists.jboss.org&gt; *Sent:* Tue, June 22, 2010 4:29:50 PM *Subject:* Re: [rules-users] Inter-fact comparison A regular old accumulate could do this. Doesn't even have to be custom: rule "many bogus attempts" when $login : Login( $baseName : name) $total : Number( intValue > 5 ) from accumulate( Login( this != login, eval(levenshtein(name, baseName) ), init( int total = 0; ), action( total++; ), reverse( total--; ), result( total ) ) then //ALARM! end This basically says "accumulate all Logins where the name is within a levenshtein distance of a given Login and freak out if there's more than X." This rule isn't all fusiony, but you get the idea. Also, the eval in the accumulate could be a custom operator for readability. --- On *Tue, 6/22/10, Earnest Dyke <earniedyke(a)yahoo.com&gt;* wrote: From: Earnest Dyke <earniedyke(a)yahoo.com&gt; Subject: Re: [rules-users] Inter-fact comparison To: "Rules Users List" <rules-users(a)lists.jboss.org&gt; Date: Tuesday, June 22, 2010, 2:57 PM Thanks for the quick reply. Yes, the original facts are already coming in via a stream (Fusion) to an existing set of rules so I was hoping to build on that. What I am trying to do is evaluate logins information as users are attempting to login trying to identify hack attempts by a person trying to login with a slightly different user name over a given period of time. Earnie! ------------------------------ *From:* David Sinclair <dsinclair(a)chariotsolutions.com&gt; *To:* Rules Users List <rules-users(a)lists.jboss.org&gt; *Sent:* Tue, June 22, 2010 3:47:36 PM *Subject:* Re: [rules-users] Inter-fact comparison Hi Earnie, What you are explaining seems pretty straight forward and would more likely than not be overkill to use a rule engine. Are there other aspects of the system you are leaving out that warrant the use of Drools? dave 2010/6/22 Earnest Dyke <earniedyke(a)yahoo.com&gt; > Greetings all, > > I have a requirement to compare a set of facts against each other, > calculate a value that indicates the amount of difference (Levenshtein > distance) between a single attribute on each fact. So I load all of my facts > into working memory then what? Do I execute a rule first that creates a new > set of facts which is a Cartesian product of the original set of facts, each > new fact containing a reference to two original facts and the diff between > them? Looked at using a custom accumulate function but I don't think that's > going to do what I need since it is intended to go through a set of facts > and return a single value. > > Any and all related help is appreciated. > > Earnie! > > _______________________________________________ > rules-users mailing list > rules-users(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/rules-users > > -----Inline Attachment Follows----- _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users