RE: Data is copied twice when specifying both child and parent table in publication

On Wed, Feb 8, 2023 4:29 AM Andres Freund <andres(at)anarazel(dot)de> wrote:
> Hi,
>
> On 2022-11-16 08:58:31 +0000, wangw(dot)fnst(at)fujitsu(dot)com wrote:
> > Attach the new patch set.
>
> This patch causes several of the tests to fail. See e.g.:
>
> https://cirrus-ci.com/task/6587624765259776
>
> Most of the failures appear to be due to the main regression tests failing:
> https://api.cirrus-
> ci.com/v1/artifact/task/6587624765259776/testrun/build/testrun/regress/regres
> s/regression.diffs
>
> diff -U3 /tmp/cirrus-ci-build/src/test/regress/expected/publication.out
> /tmp/cirrus-ci-build/build/testrun/regress/regress/results/publication.out
> --- /tmp/cirrus-ci-build/src/test/regress/expected/publication.out 2023-02-
> 07 20:19:34.318018729 +0000
> +++ /tmp/cirrus-ci-build/build/testrun/regress/regress/results/publication.out
> 2023-02-07 20:22:53.545223026 +0000
> @@ -1657,7 +1657,7 @@
> SELECT * FROM pg_publication_tables;
> pubname | schemaname | tablename | attnames | rowfilter
> ---------+------------+------------+----------+-----------
> - pub | sch2 | tbl1_part1 | {a} |
> + pub | sch2 | tbl1_part1 | |
> (1 row)
>
> DROP PUBLICATION pub;

Thanks for your kind reminder and analysis.

I think this failure is caused by the recently commit (b7ae039) in the current
HEAD. Rebased the patch set and attach them.

Regards,
Wang Wei