On Thu, May 19, 2022 at 2:26 PM Michael Paquier <michael@paquier.xyz> wrote:
>
> On Fri, May 06, 2022 at 02:26:42PM +0900, Masahiko Sawada wrote:
> > DefineOpClass() calls CreateOpFamily() to create the operator family
> > but in CreateOpFamily() we don't report the new object to event
> > triggers. The event by CREATE OPERATOR FAMILY is normally reported by
> > ProcessUtilitySlow(). I've confirmed this happens on all supported
> > branches. I've attached a patch to fix it.
>
> Hmm. It looks like it makes sense to back-patch that.
>
Yeah, I also don't see a reason to not backpatch this.
> That's indeed
> a bit surprising to not have an event trigger inform about both.
>
> +-- CRAETE OPERATOR CLASS without FAMILY clause should report
> +-- both CRAETE OPERATOR FAMILY and CRAETE OPERATOR CLASS
> Got the same typo here, repeated three times.
>
> - tmpAddr = CreateOpFamily(stmt->amname, opcname,
> + tmpAddr = CreateOpFamily(&opfstmt, stmt->amname, opcname,
> namespaceoid, amoid);
> CreateOpFamily() does not need its second argument now that you pass
> down a CreateOpFamilyStmt as first argument, no?
>
Right, apart from the patch looks good to me.
--
With Regards,
Amit Kapila.