Date: Wed, 10 Aug 2005 07:31:57 -0600
From: Michael Fuhr <mike@fuhr.org>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: dracula007@atlas.cz, pgsql-sql@postgresql.org
Subject: Re: **SPAM** Faster count(*)?
Message-ID: <20050810133157.GA46247@winnie.fuhr.org>
References: <000101c59d33$36a4a560$9b00015a@osl.com>
	<196456155.20050810012958@karneval.cz>
	<21626.1123642154@sss.pgh.pa.us>
	<20050810032913.GA40917@winnie.fuhr.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20050810032913.GA40917@winnie.fuhr.org>

On Tue, Aug 09, 2005 at 09:29:13PM -0600, Michael Fuhr wrote:
> On Tue, Aug 09, 2005 at 10:49:14PM -0400, Tom Lane wrote:
> > Current best practice is to run the explain and parse out the "rows"
> > figure using a perl (or axe-of-choice) regexp, though we could be
> > persuaded to supply a simpler API if there's enough demand for it.
> 
> Somebody else requested a row-count-estimate function a couple of
> weeks ago:
> 
> http://archives.postgresql.org/pgsql-admin/2005-07/msg00256.php

Here's a simple example that parses EXPLAIN output.  It should work
in 8.0.2 and later:

CREATE FUNCTION count_estimate(query text) RETURNS integer AS $$
DECLARE
    rec   record;
    rows  integer;
BEGIN
    FOR rec IN EXECUTE 'EXPLAIN ' || query LOOP
        rows := substring(rec."QUERY PLAN" FROM ' rows=([[:digit:]]+)');
        EXIT WHEN rows IS NOT NULL;
    END LOOP;

    RETURN rows;
END;
$$ LANGUAGE plpgsql VOLATILE STRICT;

CREATE TABLE foo (r double precision);
INSERT INTO foo SELECT random() FROM generate_series(1, 1000);
ANALYZE foo;

SELECT count_estimate('SELECT * FROM foo WHERE r < 0.1');
 count_estimate 
----------------
             97
(1 row)

EXPLAIN SELECT * FROM foo WHERE r < 0.1;
                     QUERY PLAN                      
-----------------------------------------------------
 Seq Scan on foo  (cost=0.00..17.50 rows=97 width=8)
   Filter: (r < 0.1::double precision)
(2 rows)

-- 
Michael Fuhr