Content-Type: text/plain;
  charset="us-ascii"
From: Gary Stainburn <gary.stainburn@ringways.co.uk>
Organization: Ringways Garages Ltd
To: pgsql-sql@postgresql.org
Subject: efficient count/join query
Date: Fri, 7 Feb 2003 10:07:28 +0000
User-Agent: KMail/1.4.3
MIME-Version: 1.0
Message-Id: <200302071007.28697.gary.stainburn@ringways.co.uk>
Content-Transfer-Encoding: quoted-printable

Hi folks,

I've got two tables, first a history table containing tallies for staff/job=
s=20
prior to going live, and second a roster table showing date, diagram, job=
=20
with one record per person per job per day.  the tables are:

create table history (
hsid	int4 not null references staff(sid),
hjid	int4 not null references jobs(jid),
hcount	int4,
primary key (hsid,hjid));

create table roster (
rodate	date not null,
rogid	int4 not null references diagrams(gid),
rojid	int4 not null references jobs(jid),
rosid	int4 references staff(sid),
primary key (rodate, rogid, rojid));

What's the best/quickest/cheapest way to create a view in the format of the=
=20
history table but including the details from the roster table for all recor=
ds=20
prior to today.

I've been looking at some form of sub-select/join scheme but as some will o=
nly=20
exist on the history and some will only exist on the roster while many will=
=20
exist on both.
--=20
Gary Stainburn
=20
This email does not contain private or confidential material as it
may be snooped on by interested government parties for unknown
and undisclosed purposes - Regulation of Investigatory Powers Act, 2000=20=
=20=20=20=20