The current implementation of ASTAndNode and ASTOrNode will evaluation both
sides of an expression before returning a result. I think most developers
these days seem to expect the RHS of an AND/OR expression to be ignored if the
LHS is sufficient to determine the result
(e.g. if A=>true, "A or B"=>true, regardless of B).
The patches appear to be trivial - if a committer is interested & thinks it
worthwhile I'll create them.
(btw, the old Turbo Pascal compiler described this optimisation as "short
circuit boolean evaluation"; I'm not sure if that is a common name for it.)
Improvement
Resolved
Minor
[JEXL] Implement short-circuit boolean evaluation
