Re: Avoid orphaned objects dependencies, take 3

Hi,

On Thu, Jun 13, 2024 at 10:49:34AM -0400, Robert Haas wrote:
> On Fri, Jun 7, 2024 at 4:41 AM Bertrand Drouvot
> <bertranddrouvot(dot)pg(at)gmail(dot)com> wrote:
> > Do you still find the code hard to maintain with v9?
>
> I don't think it substantially changes my concerns as compared with
> the earlier version.

Thanks for the feedback, I'll give it more thoughts.

>
> > > but we're not similarly careful about other operations e.g.
> > > ConstraintSetParentConstraint is called by DefineIndex which calls
> > > table_open(childRelId, ...) first, but there's no logic in DefineIndex
> > > to lock the constraint.
> >
> > table_open(childRelId, ...) would lock any "ALTER TABLE <childRelId> DROP CONSTRAINT"
> > already. Not sure I understand your concern here.
>
> I believe this is not true. This would take a lock on the table, not
> the constraint itself.

I agree that it would not lock the constraint itself. What I meant to say is that
, nevertheless, the constraint can not be dropped. Indeed, the "ALTER TABLE"
necessary to drop the constraint (ALTER TABLE <childRelId> DROP CONSTRAINT) would
be locked by the table_open(childRelId, ...).

That's why I don't understand your concern with this particular example. But
anyway, I'll double check your related concern:

+ if (object->classId == RelationRelationId || object->classId ==
AuthMemRelationId)
+ return;

in depLockAndCheckObject().

Regards,

--
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com