Thomas, I was curious about this and tried to set up a minimal sample, and as far as I can tell, this is not possible. Based on my testing, even if $a, $b, and $c ARE present, you cannot do this, because the evaluation short-circuits if $a is found, and therefore $b and $c are never defined. Even putting null-checks around the do($a) lines won't work, because the error happens before the RHS is evaluated. David Faulkner david.faulkner@ametnra.com<mailto:david.faulkner@ametnra.com> ________________________________ From: rules-users-bounces(a)lists.jboss.org [rules-users-bounces(a)lists.jboss.org] On Behalf Of thomas.polzin(a)gmail.com [thomas.polzin(a)gmail.com] Sent: Thursday, February 03, 2011 3:55 PM To: rules-users(a)lists.jboss.org Subject: [rules-users] Simple Disjunction Question Hi there Assume we have a rule like this: when $a : A(a==1) or ($b : B(b==1) and $c : C(c==1)) then do($a) do($b) do($c) end Assuming also, that for some reason I do not want to split this rule into two. Then, I get a NullPointer exception because either $a or $b and $c are not defined. Is there a way to catch or check for this. Thanks so much for any help Thomas

Given the different structure of the terms of the disjunction, there is no valid reason for not splitting such a rule except to avoid code repetition. But: (a) If the disjunction is at the end of a sequence of patterns, the "extend" feature can be used. (b) If the consequence is very similar (obviously, it can't be quite the same if the bound variables should serve any purpose) , common parts of the RHS can be written as functions. -W 2011/2/3 <thomas.polzin(a)gmail.com&gt;: