Re: IN vs EXIIST

From: Jochem van Dieten <jochemd(at)oli(dot)tudelft(dot)nl>
To: Jean-Christian Imbeault <jc(at)mega-bucks(dot)co(dot)jp>
Cc: pgsql-general <pgsql-general(at)postgresql(dot)org>
Subject: Re: IN vs EXIIST
Date: 2002-09-19 10:59:55
Message-ID: 1032433195.3d89ae2b96ff5@webmail.oli.tudelft.nl
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Quoting Jean-Christian Imbeault <jc(at)mega-bucks(dot)co(dot)jp>:
>
> The reason being that key1 is not a primary key (key1, key2 is the
> primary key). i.e. I have a table like this
>
> key1 key2 x
> ------------------
> a 1 t
> a 2 t
> a 3 f
> b 1 t
> b 2 t
> b 3 t
> c 3 t
> c 4 f
>
> So basically I want key1 values for which all the X's are true.

SELECT key1, Min(CASE WHEN x THEN 1 ELSE 0 END) AS isTrue
FROM table
GROUP BY key1
HAVING isTrue = 1

> Or is my table schema wrong?

I generally don't design tables with composite keys. I find it to
complicated in many operations. But on occasion I have seen them being
used by others very efficiently.

Jochem

In response to

  • IN vs EXIIST at 2002-09-19 02:40:20 from Jean-Christian Imbeault

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Jean-Christian Imbeault 2002-09-19 11:04:53 Re: IN vs EXIIST
Previous Message Jean-Christian Imbeault 2002-09-19 10:41:20 Re: IN vs EXIIST