Re: currval question

Поиск
Список
Период
Сортировка
Искать
От
Garo Hussenjian
Тема
Re: currval question
Дата
Msg-id
B9ABAA6D.35CE%garo@xapnet.com
Ответ на
currval question (Andy Kriger)
Список
Дерево обсуждения
currval question "Andy Kriger" <akriger@greaterthanone.com>
Re: currval question Tino Wildenhain <tino@wildenhain.de>
Re: currval question Jason Earl <jason.earl@simplot.com>
Re: currval question Garo Hussenjian <garo@xapnet.com>
Re: currval question "Andy Kriger" <akriger@greaterthanone.com>
Andy, 

I don't know exactly what's under the hood, but I know that currval() won't
even return a result unless you've used nextval() in your session. If
currval() did not operate solely within the scope of the session, it stands
to reason that it would return a result... It's behavior (from the driver's
seat) is consistent with a well conceived concurrency model. This should be
easy enough to test with a couple of terminals. My money is on Postgres! :)

Regards,
Garo. 

on 9/16/02 3:14 PM, Andy Kriger at akriger@greaterthanone.com wrote:

> I am trying to get the last value updated by an column auto-incrementing
> with nextval(). In MySQL, you'd use LAST_INSERT_ID() - in Postgre, currval()
> appears to do the trick.
> 
> Is this maintained on a per-connection basis? For example, user A inserts
> and the nextval() updates to 5, user B does 2 inserts, updating nextval() to
> 7. When user A calls currval() they should get 5 if the updates are per-cnx.
> What does psql do under the hood here?
> 
> thx
> a
> 
> 
> 
> ---------------------------(end of broadcast)---------------------------
> TIP 5: Have you checked our extensive FAQ?
> 
> http://www.postgresql.org/users-lounge/docs/faq.html
> 

=-=-==-=-=-==

Xapnet Internet Solutions
1501 Powell St., Suite N
Emeryville, CA 94608

Tel - (510) 655-9771
Fax - (510) 655-9775
Web - http://www.xapnet.com

В списке pgsql-general по дате отправления
От: Mihai Gheorghiu
Дата:
Сообщение: Performance again
От: Cindy
Дата:
Сообщение: plperl.so
FAQ