Sender: hannu@kodu.home.ee
Message-ID: <36BF564C.D1A9F8FA@trust.ee>
Date: Mon, 08 Feb 1999 23:25:32 +0200
From: Hannu Krosing <hannu@trust.ee>
Organization: Trust-O-Matic =?iso-8859-1?Q?O=DC?=
MIME-Version: 1.0
To: Matthew Hagerty <matthew@venux.net>
CC: pgsql-interfaces@postgreSQL.org
Subject: Re: [INTERFACES] Dealing with large query results
References: <4.1.19990208145905.00a0b8d0@mail.venux.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Matthew Hagerty wrote:
> 
> Greetings,
> 
> I keep seeing reference to an 8K limit on query results.  So how to deal
> with query results greater than 8K?  Am I under the wrong impression that
> when you run a query from, say libpq, that each retch row is returned only
> when you execute a fetchrow?  Or is the whole query stuffed in to a buffer
> which has an 8K limit?

the current 8K limit is for queries (and unrelatedly single rows), 
not query _results_
 
> I have to deal with some queries that create a huge result, and/or doing
> reports requires a large result set.  How can I deal with these larger results?

If you dont want to digest a huge result, you can do a 

DECLARE CURSOR - FETCH -FETCH - FETCH - CLOSE CURSOR thing to get it in
pieces

(needs to be inside BEGIN TRANSACTION - END TRANSACTION)
 
> I'll probably get flamed for asking this one but, how can I step through a
> table?  Hey, sometimes you just have to do it.

See above.

Actually postgres is quite happy to return gigabytes of data, if you
have 
the resources (disk space, memory, time ;)

--------------
Hannu