| From: | James Olin Oden <joden(at)lee(dot)k12(dot)nc(dot)us> |
|---|---|
| To: | "Jackson, DeJuan" <djackson(at)cpsgroup(dot)com> |
| Cc: | pgsql-general(at)postgreSQL(dot)org |
| Subject: | Re: [GENERAL] getting the currval of a sequence |
| Date: | 1998-08-14 15:30:16 |
| Message-ID: | 81Aug13.101844edt.35713@gateway.lee.k12.nc.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
> > > I think you should just go with what Vadim suggested and use:
> > > select currval('my_sequence');
> > > This will be the same within a transaction, even if there is another
> > > transaction using the same sequence.
> >
> > Sorry, I wasn't thinking of it happening in the same transaction. In
> > that
> > case that would be the thing to do. If on the other hand you needed
> > that
> > number over multiple transactions, wouldn't what I had suggested work?
> >
> > ...james
> >
> Yes it would, but your suggestion introduces problems of it's own. The
> whole point of a transaction is encapsulate a logical unit of work, such
> that it all succeeds together or it all fails together. This would
> include the incrementing of the sequence. The way that you are
> suggestion you would always increment the sequence even if you never
> used it, it just seems like a waste to me
Yikes, I didn't think of that. I suppose with 4 billion sequence numbers
you could waste a few, but...things always seem grow beyon their intended
purposes. I guess your absolutely right on this one...james
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kapoor, Nishikant X | 1998-08-14 17:03:48 | No DB connection in Netscape ? Please help. |
| Previous Message | Jackson, DeJuan | 1998-08-14 15:05:15 | RE: [GENERAL] getting the currval of a sequence |