To: pgsql-sql@postgresql.org
From: Thomas Kellerer <spam_eater@gmx.net>
Subject: Re: Query question
Date: Thu, 8 Mar 2018 20:48:35 +0100
Lines: 29
Message-ID: <p7s3uj$dg0$1@blaine.gmane.org>
References: <415803152.48510563.1520535505737.JavaMail.zimbra@rglholdings.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.21)
 Gecko/20090302 Thunderbird/2.0.0.21 Mnenhy/0.7.5.666
In-Reply-To: 
 <415803152.48510563.1520535505737.JavaMail.zimbra@rglholdings.com>
Content-Language: de-DE
Precedence: bulk

Stanton Schmidt schrieb am 08.03.2018 um 19:58:
> My question is:
> I have a table that has log events for pieces of equipment.  For each piece of equipment this table may contain 1 or more (hundreds potentially).
> I need to write a query that will return only the last 5 log events for each (and every) piece of equipment.
> 
> log_table (
> equipment_id character(30),
> log_date date,
> log_time time,
> event_desc text
> )

Queries like that are typically solved using window functions:

     select *
     from (
       select equipment_id,
              log_date,
              log_time,
              event_desc,
              row_number() over (partition by equipment_id order by log_date desc, log_time desc) as rn
       from log_table
     ) t
     where rn <= 5;


Unrelated, but: why aren't you storing "log_date_time" in a single timestamp?