Re: GROUPING - Mailing list pgsql-sql

From Joel Burton
Subject Re: GROUPING
Date
Msg-id Pine.LNX.4.30.0110141131001.6917-100000@temp.joelburton.com
Whole thread Raw
In response to GROUPING  ("Timothy J Hitchens" <tim@hitcho.com.au>)
List pgsql-sql
On Sat, 13 Oct 2001, Timothy J Hitchens wrote:

> It's been a while since I used postgresql but today I have converted one
> of my web apps but with one small problem.  I goto do a group as
> designed and executed in mysql and I get told that this and this must be
> part of the aggreate etc I am puzzled and wonder if someone could bring
> me up to stratch with grouping in postgresql this is my current sql:
>
> SELECT * FROM telemetry WHERE beat > 12 GROUP BY buid;
>
> Result:
>
> Attribute telemetry.rpvuid must be GROUPed or used in an aggregate
> function
>
>
> Oh then if I include rpvuid I get you must include this field and on it
> goes.

Normally, GROUP BY is used to group up records to look at an aggregate.
For example, if you have this table of your friends:

CREATE TABLE friends ( friend TEXT, country_code CHAR(2), income FLOAT,
);

I could get a count of how many friends lived in each country by:

SELECT country_code, COUNT(*) FROM friends GROUP BY country_code;

Or I could get the average amount of money made by friends in each country
with:

SELECT country_code, avg(income) FROM friends GROUP BY country_code;


In other words, when you GROUP BY, you're looking for an aggregate
(a function that is applied to a group and returns a single value,
such as average, minimum, maximum, count, etc.)

Can you be more specific about what you're actually trying to accomplish?




pgsql-sql by date:

Previous
From: "Aasmund Midttun Godal"
Date:
Subject: Re: Indexes
Next
From: missive@frontiernet.net (Lee Harr)
Date:
Subject: Re: to_char()??