Subject: Re: Combining two SELECTs by same filters
From: Scott Marlowe <smarlowe@g2switchworks.com>
To: Volkan YAZICI <volkan.yazici@gmail.com>
Cc: pgsql-sql@postgresql.org
In-Reply-To: <7104a7370510260816l9262d19j808d23c023e8445e@mail.gmail.com>
References: <7104a7370510260816l9262d19j808d23c023e8445e@mail.gmail.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1130342345.2872.18.camel@state.g2switchworks.com>
Mime-Version: 1.0
Date: Wed, 26 Oct 2005 10:59:05 -0500

On Wed, 2005-10-26 at 10:16, Volkan YAZICI wrote:

> => SELECT
> ->     (SELECT count(id) FROM sales
> ->         WHERE id = 2
> ->         AND date_trunc('hour', dt) = '2005-10-25 21:00:00'),
> ->     (SELECT count(id) FROM sales
> ->         WHERE id = 2
> ->         AND date_trunc('hour', dt) = '2005-10-25 22:00:00');
>  ?column? | ?column?
> ----------+----------
>         6 |        2
> (1 row)
> 
> Isn't it possible to combine these two SELECTs as one. Because one of
> their filters are same: id = 2. I'm just trying to avoid making 2
> scans with nearly same filters.

Do something like this:

select count(id) 
from sales 
where id=2 and 
dt between 'firstdatehere' and 'lastdatehere'
group by date_trunc('hour', dt);