I spoke too soon. The from will match items in the two Lists that are not at
the same index.
As for asserting ArrayList( ) facts -- in my actual application I have facts
that have Collections attributes for which I need this behavior. I just made
the problem simpler for this post.
On 8/7/07, Aaron Dixon <atdixon(a)gmail.com> wrote:
The "from" conditional element solves part of my problem:
rule "Find matching entry in parallel arrays (alternate)"
when
$first : ArrayList( )
$second : ArrayList( this != $first )
$fi : Object() from $first
$si : Object() from $second
eval( $fi == $si )
then
System.out.println ( " Found match at index ? : " + $fi + "!"
);
end
I do want reference equality (==) in my eval. But I understand that object
equality would look just like:
eval( $fi .equals( $si ) )
Thanks!
Aaron
On 8/7/07, Mike D <mike.dalaker(a)kewill.com> wrote:
>
>
> Mark,
>
> Just to be sure I read this correctly:
>
> eval( $fi == $si )
>
> means it's checking objects, not values within the object?
> If a compare of values is needed, another eval would have to be written
> to
> compare temp value.
>
> Thanks,
> Mike
>
>
> Oh, and can we change the documenatation examples please?
>
> Somthing like...
>
> rule
> when
> Team( $team : name == "ManU" )
> $person : Person( favouriteTeam == $team )
> then
> System.out.println( $person.getName() + " likes Manchester" );
> end
>
> rule
> when
> Team( $team : name == "Chelsea" )
> $person : Person( favouriteTeam != $team )
> then
> System.out.println( $person.getName() + " does not like Chelsea
> at
> all" );
> end
>
>
>
> Mark Proctor wrote:
> >
> > We don't really recommend you assert ArrayLists etc as facts, as they
> > have no contextual meaning
> >
> > $first : ArrayList( )
> > $second : ArrayList( this != $first )
> > $fi Object() from $first
> > $si Object from $second
> > eval( $fi == $si )
> >
> > 'from' allows iteration of lists, so you could potentially do it this
> > way, although it doesn't provide an index number for you.... You could
> > possibly have a global that as part of a function in eval get
> > incremented and that global is available in the consequence. But
> you'll
> > have to be very careful with concurrency.... Might be easier to have a
> > hashmap of indexes where the key is made up of $first and $second. Btw
>
> > == checks for same instances, not an equality check, is that what you
> > wanted?
> >
> > Mark
> >
> > Aaron Dixon wrote:
> >> My problem boils down to finding matching entries in parallel arrays.
>
> >>
> >> Here is a rule that succesfully does just that:
> >>
> >> rule "Find matching entry in parallel arrays"
> >> when
> >> $first : ArrayList( )
> >> $second : ArrayList( this != $first )
> >> $i : Integer( this < $first.size )
> >> eval( $first.get($i) == $second.get($i) )
> >> then
> >> System.out.println ( "Found match at index " + $i +
"!" );
> >> end
> >>
> >>
> >> To execute this rule, I must insert (assert) the two ArrayList facts
> >> as well as at least as many Integer facts as there are items in the
> >> ArrayLists.
> >>
> >> What I don't like about this rule is that
> >>
> >> (1) I have to assert the Integers
> >> (2) I can't support arbitrarily-sized lists in my rules (without
> >> asserting that many Integer facts)
> >>
> >> So -- Is there a better way?
> >>
> >> What if a future version of Drools supported implicit Number facts
> >> that allowed for these kinds of indexing rules?
> >>
> >> Aaron
> >>
> >>
> >>
> >>
> >>
> ------------------------------------------------------------------------
> >>
> >> _______________________________________________
> >> 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
> >
> >
>
> --
> View this message in context:
http://www.nabble.com/finding-matching-entries-in-parallel-arrays-tf42251...
>
> Sent from the drools - user mailing list archive at
Nabble.com.
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/rules-users
>