Message-ID: <4A8EBA03.4010309@gmail.com>
Date: Fri, 21 Aug 2009 09:15:15 -0600
From: Rob Sargent <robjsargent@gmail.com>
User-Agent: Thunderbird 2.0.0.22 (X11/20090608)
MIME-Version: 1.0
To: tripplowe <tripplowe@gmail.com>
CC: pgsql-sql@postgresql.org
Subject: Re: Rewrite multiple joins...
References: <25073534.post@talk.nabble.com>
In-Reply-To: <25073534.post@talk.nabble.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Well indexing (or lack thereof) could be the real problem but you could 
try "chaining" the tables

select * from sale s, taxes t, property p, buyer b
where s.id = t.id and t.id = p.id and p.id = b.id

and see if that (or other combination) changes the query plan appreciably.


(I would have to wonder if all those records are really unique per 
sale.id? e.g. multiple buyers per unit sold or school and property tax 
in separate records. And wondering further why, if you have sql access, 
do you need a single table. Maybe you want a view for convenience (e.g. 
select * from view_of_sales where lower(county) like 'a%' )?

tripplowe wrote:
> Hey Folks,
>
> I have a query that seems like it will never end.  I'm hoping you can help
> me rewrite it.
>
> I have 4 tables that contain information about house sales within the state
> (sale, taxes, property, buyer).  Each table has an 'id' field that links
> each record.  I am just trying to flatten these files so I will have all of
> the information in a single file.  Below is my attempt to do so.  Is there a
> more efficient way to rewrite the joins?
>
> select * from sale, taxes, property, buyer 
> where sale.id = taxes.id and sale.id = property.id and sale.id = buyer.id
>
> I'm using postgresql 8.3 on a Windows 7 machine.  
>
> Thank You,
> -Tripp
>