Re: Distinguish publication exclusions in object addresses

On Wed, Sep 16, 2026 at 5:41 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
> On Wed, Sep 16, 2026 at 12:23 PM Peter Smith <smithpb2250(at)gmail(dot)com> wrote:
> >
> > On Wed, Sep 16, 2026 at 4:46 PM Chao Li <li(dot)evan(dot)chao(at)gmail(dot)com> wrote:
> > >
> > >
> > >
> > > > On Sep 16, 2026, at 13:50, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> > > >
> > > > --
> > > > With Regards,
> > > > Amit Kapila.
> > > > <v4-0001-Distinguish-publication-exclusions-in-object-addr.patch>
> > >
> > > V4 overall looks sold to me. Just a couple of small comments:
> > >
> > > 1
> > > ```
> > > -- No entry of either kind. testpub_default publishes nothing.
> > > SELECT pg_get_object_address('publication excluded relation',
> > > '{public, testpub_tbl1}', '{testpub_default}');
> > > ERROR: publication relation "testpub_tbl1" in publication "testpub_default" does not exist
> > > ```
> > >
> > > For this new test, the error message is a little surprising to me. Since the requested object type is "publication excluded relation", I would expect the error message to say something like:
> > > ```
> > > publication excluded relation "testpub_tbl1" in publication "testpub_default" does not exist
> > > ```
> >
> > +1 this because maybe the reported table name really *does* exist and
> > is part of the publication, even though the message is saying the
> > exact opposite.
> >
>
> Can you share an example of such a case?
>

My example was using 'publication relation' (not 'publication excluded
relation') and I've already shared it.

But that bug (?) already exists because this logic appears only aware
of PUBLICATIONRELMAP rels and is not accounting for FOR ALL TABLES
etc, so it reports an error like
ERROR: "t1" is not a published relation of publication "pub"
even though table public.t1 does exist and really is a published
relation of pub.

======
Kind Regards,
Peter Smith.
Fujitsu Australia