On Wed, May 20, 2026 at 5:30 AM Peter Smith <[email protected]> wrote: > > I agree we should address that \d+ quirk at the same time. > > Thanks for finding/fixing it in v3-0002. > > Your 0002 patch works for me, although I'm thinking those SQL fixes > ought to be made more similar: > 1. Both use "n.oid NOT IN (SELECT pn.pnnspid ..." > 2. Or both use "AND NOT EXISTS (SELECT 1 FROM ... WHERE ...)" > > ~ > > PSA v4, where I have combined the patches, and chosen your SQL style > for my previous \dRp+ fix. >
+1 for the idea. For the describePublications (\dRp+) case, should we add an (sversion >= 150000) guard around the new change, since it accesses pg_publication_namespace, which is only available in PG15 and above? Thoughts? -- Thanks, Nisha
