On Thu, Dec 13, 2018 at 03:34:56PM +0900, Amit Langote wrote: > Thank you Michael for taking care of this. I agree with the consensus > that instead of throwing an error for unsupported relkinds, > pg_partition_tree should rather return single row containing NULL for all > columns.
Cool, thanks for confirming! If there are complains on the matter, let's of course tune it again if necessary before the release. There is still time for that. -- Michael