On Thu, Sep 23, 2021 at 5:02 PM houzj.fnst@fujitsu.com
<houzj.fnst@fujitsu.com> wrote:
>
> > Sounds like a good idea.
> > Is it possible to incorporate the existing
> > CheckPublicationAlterTables() and CheckPublicationAlterSchemas() functions
> > into your suggested update?
> > I think it might tidy up the error-checking a bit.
>
> I agreed we can put the check about ALL TABLE and superuser into a function
> like what the v30-patchset did. But I have some hesitations about the code
> related to CheckObjSchemaNotAlreadyInPublication(). Currently, we need to open
> and lock the table before invoking the CheckObjxxx function, ISTM we'd better
> open the table in function AlterPublicationTables. Maybe we can wait for the
> author's(Vignesh) opinion.
>
Yes, I think you're right, the code related to
CheckObjSchemaNotAlreadyInPublication() should be left where it is
(according to your schema refactor patch).
Regards,
Greg Nancarrow
Fujitsu Australia