Re: remove_useless_joins vs. bug #19560 - Mailing list pgsql-hackers

Hi,

On Sun, 27 Sept 2026 at 22:02, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> I wrote:
> > Concretely, about like this.  This fixes the given reproducer.
> > I've not put thought into what sort of test cases we might want
> > to add.
>
> Now with test cases, a commit message explaining why do it like
> this, and a bit of additional comment-polishing.

Thanks for the patch.

I went through v1.  Grouping the SJE candidates by checkAsUser
looks right to me, including the unset case.  I didn't spot any
issues.  LGTM.

Regards,
Ayush