Message-ID: <021a9dcef9be53a70a73251b8fe9080a@news.teranews.com>
Date: Tue, 18 Nov 2003 18:41:31 GMT
Lines: 33
From: Jeff Boes <jboes@nexcerpt.com>
Organization: Nexcerpt, Inc.
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20030925
MIME-Version: 1.0
Subject: Need query to separate rows by one field's value
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
To: pgsql-sql@postgresql.org

Here's a query challenge for you. Given:

table foo (
   a integer primary key,
   b integer check (b > 0))

with a limited number of rows (say, <= 10). Values in column "a" are 
unique (obviously). Values in column "b" are not.

I want to find a set of values for "a" such that the sum of "b" values 
is as close to one-half of the total sum of "b" over all rows. That is,

select 2.0*sum(b) from foo
where a in (...);

would be approximately the same as

select sum(b) from foo;

Likewise,

select 2.0*sum(b) from foo
where a NOT in (...);

would be approximately that value, too.


-- 
Jeff Boes                                      vox 269.226.9550 ext 24
Database Engineer                                     fax 269.349.9076
Nexcerpt, Inc.                                 http://www.nexcerpt.com
            ...Nexcerpt... Extend your Expertise