Message-Id: <3.0.6.32.19990309183251.007fd2a0@mtl.pl>
Date: Tue, 09 Mar 1999 18:32:51 -0800
To: pgsql-general@hub.org
From: Marcin Grondecki <ojciec@mtl.pl>
Subject: Re: [GENERAL] Re: [INTERFACES] the last row in a table
In-Reply-To: <36E53C2E.C2E59616@alumni.caltech.edu>
References: <36E163D7.CD95C55A@pd.jaring.my>
	<87emn1fwsq.fsf@tolikus.hq.aaanet.ru>
	<36E50F4A.CB527B78@pd.jaring.my>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

At 15:20 99-03-09 +0000, you wrote:
>> I am developing a database with PostgreSQL in Linux, and using ecpg to 
>> write the C program to communicate with the database.
>> One of my application need to get the data from the last row in a 
>> table. Is there any function or command that can tract which is the 
>> last row of data in a table?
>
>SQL is a set-oriented language. One should probably not assign any
>significance to a storage order of rows.
>
>It *is* possible to order the results of a query, and then your
>first/last qualities do have meaning. I would suggest ordering your
>query (perhaps on a "row update time" if by "the last row" you mean "the
>more recently entered row") using the "DESC" qualifier in the ORDER BY
>clause, then use a cursor to pick up the first row returned.
>
There is better answer, i think. Simply read smth about cursors - declaring,
using, droping. I think this is what u need.
In case of troubles with understanding, i'd serve with some examples in Perl.

Marcin Grondecki
ojciec@mtl.pl
+48(604)468725
***** I'm not a complete idiot, some parts are missing...