Re: Review: Non-inheritable check constraints

Excerpts from Greg Smith's message of vie dic 16 15:02:20 -0300 2011:
> On 12/04/2011 02:22 AM, Nikhil Sontakke wrote:
> >
> >     Is it okay to modify an existing constraint to mark it as "only", even
> >     if it was originally inheritable?  This is not clear to me.  Maybe the
> >     safest course of action is to raise an error.  Or maybe I'm misreading
> >     what it does (because I haven't compiled it yet).
> >
> >
> > Hmmm, good question. IIRC, the patch will pass is_only as true only if
> > it going to be a locally defined, non-inheritable constraint. So I
> > went by the logic that since it was ok to merge and mark a constraint
> > as locally defined, it should be ok to mark it non-inheritable from
> > this moment on with that new local definition?

I think I misread what that was trying to do.  I thought it would turn
on the "is only" bit on a constraint that a child had inherited from a
previous parent, but that was obviously wrong now that I think about it
again.

> With this open question, this looks like it's back in Alvaro's hands
> again to me.  This one started the CF as "Ready for Committer" and seems
> stalled there for now.  I'm not going to touch its status, just pointing
> this fact out.

Yeah.  Nikhil, Alex, this is the merged patch.  Have a look that it
still works for you (particularly the pg_dump bits) and I'll commit it.
I adjusted the regression test a bit too.

Thanks.

--
Álvaro Herrera <alvherre@commandprompt.com>
The PostgreSQL Company - Command Prompt, Inc.
PostgreSQL Replication, Consulting, Custom Development, 24x7 support