Kohei KaiGai <kaigai(at)kaigai(dot)gr(dot)jp> writes:
> Sorry, are you saying the current (in other words, rte->security_barrier
> stores the state of reloption) approach is not a good idea?
Yes. I think the same as Robert: the way to handle this is to store it
in RelOptInfo for the duration of planning, and pull it from the
catalogs during planner startup (cf plancat.c).
regards, tom lane