On Sun, Aug 30, 2026 at 07:21:07PM -0500, Sami Imseih wrote:
>> Claude advises me that removing the redundant filter clause was a mistake.
>> See the attached patch.
>
> Yes, that was a miss. I did a quick confirmation of the execution plans
> with and without the patch and the change LGTM.
The extra time taken would really hurt when retrieving statistics.
There may be an argument in simply reverting 4b5ba0c4ca52, as one of
the reasons for it was to remove the qual you are re-adding, but I
cannot get excited about using two sets of catalog names for tablename
and spcname vs a single OID array approach in the long run.
--
Michael
