Re: Review items for EXCEPT TABLE publication

On Fri, Sep 11, 2026 at 5:22 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
> On Fri, Sep 11, 2026 at 12:34 PM Peter Smith <smithpb2250(at)gmail(dot)com> wrote:
> >
> > In my earlier example, there are both precept t and f.
> >
> > Won't your v4 new function return a random *isexcept result based on
> > the first one it stumbles upon?
> >
>
> Yes, but how does that matter? We just don't want the operation to
> proceed when the table is part of publication, users anyway need to
> fix it before proceeding.
>

OTOH it doesn't matter because they are both problems the user must
fix. But won't it be better to get a consistent/deterministic error
message for the same problem?

e.g.1
test_pub=# CREATE PUBLICATION pub1 FOR ALL TABLES EXCEPT (TABLE t1);
CREATE PUBLICATION
test_pub=# CREATE PUBLICATION pub2 FOR TABLE t1;
CREATE PUBLICATION
test_pub=# select * from pg_publication_rel;
oid | prpubid | prrelid | prexcept | prqual | prattrs
-------+---------+---------+----------+--------+---------
16392 | 16391 | 16387 | t | |
16394 | 16393 | 16387 | f | |
(2 rows)

e.g.2
test_pub=# DROP publication pub1;
DROP PUBLICATION
test_pub=# DROP publication pub2;
DROP PUBLICATION
test_pub=# CREATE PUBLICATION pub2 FOR TABLE t1;
CREATE PUBLICATION
test_pub=# CREATE PUBLICATION pub1 FOR ALL TABLES EXCEPT (TABLE t1);
CREATE PUBLICATION
test_pub=# select * from pg_publication_rel;
oid | prpubid | prrelid | prexcept | prqual | prattrs
-------+---------+---------+----------+--------+---------
16396 | 16395 | 16387 | f | |
16398 | 16397 | 16387 | t | |
(2 rows)

Everything there is identical. Same pub names, same table
exclsions/inclusions etc. But the order of operations is different, so
the order of pg_publication_rel entries is different, so the *isexcept
returned from that function is different, so the user sees 2 different
error messages for the same problem.

That just seemed a bit strange to me. Won't that kind of
non-determinism also potentially break BF?

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