2011/9/26 Tom Lane <tgl@sss.pgh.pa.us>:
> Kohei KaiGai <kaigai@kaigai.gr.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).
>
Hmm. If so, it seems to me worthwhile to investigate an alternative
approach that stores relation-id of the view on rte->relid if rtekind is
RTE_SUBQUERY and pull the "security_barrier" flag from the catalog
during planner stage.
Thanks,
--
KaiGai Kohei <kaigai@kaigai.gr.jp>