Use of nextval, currval

From: Matt Clark <mclark(at)rushe(dot)aero(dot)org>
To: pgsql-novice(at)postgresql(dot)org
Subject: Use of nextval, currval
Date: 2002-11-07 23:15:56
Message-ID: 200211072315.gA7NFuF15378@rushe.aero.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-novice

I'm attempting to convert a code which uses Oracle to Postgres.

I am concerned about a couple of points in the SQL, which I should be able to do
in Postgres.

Say I have defined a sequence DataId;

The Oracle uses the sequence as follows:

insert into TABLE (id1,.......,id2)
values (DataId.NextVal,......,DataId.Currval);

I believe that I can replace this in Postgres with:

insert into TABLE (id1,.......,id2)
values (nextval('DataId'),.....,currval('DataId');


What I think is expected is that the values of id1 and id2 will be equal.
I have been told that Oracle will guarantee this; will that be assumption be
true for Postgres, also? (My concern is that, with multiple/many users of the
same database, there may be instances where DataId increments before the currval
happens, such that it's possible that id1 != id2, above.)

Is there a simple way to write the SQL such that I can only use nextval once,
keeping its value for both id1 and id2? (I have no idea why the values are
duplicated.....)

For that matter, is that a realistic assumption for Oracle's treatment of
sequences?

Thanks,

Matt

Responses

Browse pgsql-novice by date

  From Date Subject
Next Message Josh Berkus 2002-11-08 00:53:32 Re: Use of nextval, currval
Previous Message Michael Paesold 2002-11-07 18:15:40 Re: enforcing alphabetic/numeric content in "char" column