From: | Dan Tenenbaum <dandante(at)gmail(dot)com> |
---|---|
To: | pgsql-sql(at)postgresql(dot)org |
Subject: | getting a query column to return 0 if nothing matches |
Date: | 2006-01-03 22:15:25 |
Message-ID: | 65029d820601031415w34794135v307ef13b37d5cd10@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
I have a query like this (this one is simplified to illustrate my question):
select
l.email, p.practice_name, count(lp.provider_key) as patient_count
from
login l, provider p, login_provider lp
where
p.login_key = l.id
and
lp.provider_key = p.provider_key
group by
l.email, p.practice_name
It works fine but only returns rows where the provider key appears in the
login_provider table.
I want to show all rows where the first part of the WHERE clause is true,
regardless of whether the provider_key is in the login_provider table (in
that cause, I want to show 0 in that column).
In other words, currently if I run the above query, it will return 19 rows,
and if I run this one:
select
l.email, p.practice_name
from
login l, provider p
where
p.login_key = l.id
I get 57 rows. What I want is for the first query to also return 57 rows
and just show 0 for patient_count if the second part of the where clause
cannot be satisfied.
I'm sure this is a simple thing, but I'm not sure how do do it. Can anyone
help?
Thanks
From | Date | Subject | |
---|---|---|---|
Next Message | frank church | 2006-01-04 01:32:11 | Loading lots of data in a SQL command |
Previous Message | Jaime Casanova | 2006-01-03 19:18:51 | Re: delete from a using b in postgres 8.1. |