On Tue, 19 Aug 2003, Peter Eisentraut wrote:
> Christopher Kings-Lynne writes:
>
> > > > It's a constraint name. IIRC, it happens to affect all such named
> > > > constraints currently. We should probably allow <tablename>.<constraint>
> > > > (and <schema>.<tablename>.<constraint>) as well. Too late for 7.4, but
> > > > this can happen for 7.5 if there aren't any objections.
> > >
> > > I object.
> >
> > Thanks for the helpful objection. To what do you object specifically and
> > why?
>
> I object to creating gratuitous incompatibilities with the SQL standard,
> which will obstruct legitimate features down the road. The SQL standard
> says it is <schema>.<constraint>.
We *already* have the incompatibility with the SQL standard because of the
fact we allow non-unique constraint names in the same schema;
<schema>.<constraint> does not uniquely identify a constraint in
PostgreSQL currently.
Tom objected to following the spec on this regard back when the issue was
brought up for adding checks on the constraint names on the grounds that
table based constraint names were more natural.