Jie Liang wrote:
> e.g.
> Try:
>
> CREATE TABLE emp (
> id int4 primary key,
> empname text,
> salary int4,
> last_date datetime,
> last_user name);
>
> CREATE FUNCTION emp_stamp () RETURNS OPAQUE AS
> BEGIN
> update emp set last_date=''now''::timestamp where id=NEW.id;
> RETURN NEW;
> END;
> ' LANGUAGE 'plpgsql';
>
> CREATE TRIGGER emp_stamp BEFORE UPDATE ON emp
> FOR EACH ROW EXECUTE PROCEDURE emp_stamp();
Not sure if it works this way at all, but the update on emp
table is definitely a wasted scan. And the 'now'::timestamp
will be evaluated at the first function call - never again;
not sure if he wanted that behaviour either.
CREATE FUNCTION emp_stamp () RETURNS opaque AS '
BEGIN
new.last_date := now();
RETURN new;
END;'
LANGUAGE 'plpgsql';
Is the correct trigger for this purpose.
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== [EMAIL PROTECTED] #
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
