Here is a question though now, I have the following:
mdOutput : ModelDiscrepancyOutput( $amt1 : sumRTVQty, $amt2 : sumCmAndRnr)
eval($amt1.doubleValue() != $amt2.doubleValue())
Is there a way for me to eliminate the eval given that it makes the rules
engine sub-optimal? I guess one strategy is to create two separate classes
to house the results, but man that seems messy.
On 4/11/07, Ronald R. DiFrango <ron.difrango(a)gmail.com> wrote:
Michael,
I got it working, but using Salience alone.
Thanks!
Ron
On 4/10/07, Ronald R. DiFrango < ron.difrango(a)gmail.com> wrote:
>
> Michael,
>
> OK, here is my attempt to convert the flow chart I have into words:
>
> Step 1: Sum ObjectA
> Step 2: Sum ObjectB
> Step 3: Sum ObjectC
> Step 4: Perform check on sum's from Steps 1,2 & 3 and do something if it
> fails
> Step 5: Perform an additional check on sum's from Steps 1,2 & 3 and do
> something if it fails
> Step 6: Perform an additional check on sum's from Steps 1,2 & 3 and do
> something if it fails
>
> I literally need it in this order and I was thinking that agenda groups
> along with salience could get me there, but I have never used them before.
>
> Does that help?
>
> Ron
>
> On 4/6/07, Michael Rhoden < mrhoden(a)franklinamerican.com> wrote:
> >
> > Please give a better example or some pseudo code of what you are
> > trying to do.
> >
> > -Michael
> >
> > ------------------------------
> > *From:* rules-users-bounces(a)lists.jboss.org [mailto:
> > rules-users-bounces(a)lists.jboss.org] *On Behalf Of *Ronald R. DiFrango
> > *Sent:* Friday, April 06, 2007 8:15 AM
> > *To:* Rules Users List
> > *Subject:* [rules-users] Rules Design Question
> >
> > All,
> >
> > I have a rules design question for the community. I have 3 lists of
> > different types of objects that I need to sum if a certain attribute is
> > present on each one of them. Then after the summing has taken place execute
> > other rules based upon the results of the summation.
> >
> > Is there a way that I can accomplish this via salience or agenda
> > groups?
> >
> > Thanks in advance,
> >
> > Ron
> >
>
>