> The cause is that plan_union_queries() is failing to clear out the
> groupclause before it returns control to union_planner, so the GROUP BY
> gets applied twice, once to the subselect and once (incorrectly) to the
> result of the UNION. (This wouldn't have happened with a less klugy
> representation for UNION parsetrees, but I digress.) You can see this
> happening if you look at the EXPLAIN output; the coredump only happens
> at execution time.
Is going to be similar to LIKE indexing, where everyone beats me up
about it, but the code remains unchanged because no one can think of a
better/cleaner idea? But i digress... :-)
-- Bruce Momjian | http://www.op.net/~candle maillist@candle.pha.pa.us | (610)
853-3000+ If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill,
Pennsylvania19026