| From: | Sean Davis <sdavis2(at)mail(dot)nih(dot)gov> |
|---|---|
| To: | pgsql-novice(at)postgresql(dot)org |
| Cc: | Raimon Fernandez <coder(at)montx(dot)com> |
| Subject: | Re: oid or without oid ... |
| Date: | 2007-05-22 09:57:09 |
| Message-ID: | 200705220557.10023.sdavis2@mail.nih.gov |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
On Tuesday 22 May 2007 03:30, Raimon Fernandez wrote:
> Hello again,
>
>
> In my previous databases development, I've been using always a unique
> longint number for identifying each record.
>
> In PostgreSQL I can see that it has the oid, he can do it for you.
>
> Also I see that is an optional parameter, and after surfing the web,
> I could find some people that say never use them, an another ones
> that you can use it ...
>
> Any extra advice would be perfectly before creating the database, if
> it's better for now and future to use oid or simply create an id
> field as a serial or something similar.
In general, the recommendation is not to use OIDs. Use serial columns (or
some other primary key) instead.
Sean
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Raimon Fernandez | 2007-05-22 10:05:46 | Re: oid or without oid ... |
| Previous Message | Raimon Fernandez | 2007-05-22 07:30:34 | oid or without oid ... |