From: Tom Lane <tgl@sss.pgh.pa.us>
To: Peter Geoghegan <pg@heroku.com>
cc: =?UTF-8?B?5aSn5aGa5oay5Y+4?= <otsuka.kenji@nttcom.co.jp>,
        pgsql-docs@postgresql.org, Pg Hackers <pgsql-hackers@postgresql.org>
Subject: Re: [HACKERS] The number of bytes is stored in index_size of
 pgstatindex() ?
In-reply-to: 
 <CAM3SWZTUNXgzCfVsYSFP44cRo4K-93BPDC+gEbm_WveFdTKmnA@mail.gmail.com>
References: <56C5B2C7.7010604@nttcom.co.jp> <31925.1455817248@sss.pgh.pa.us>
 <CAM3SWZQtQ2WDGRqa2g_dbfHuu9CebfWvJ_XpXggFoFG1mUYzXw@mail.gmail.com>
 <377.1455818470@sss.pgh.pa.us>
 <CAM3SWZT2UaCTqpd71uhUVQN7b7k4treyzGyQLpv4N=TYU=F7tg@mail.gmail.com>
 <10922.1455821371@sss.pgh.pa.us>
 <CAM3SWZRLfCayENUVE2SqyeAiYKLFO7a7cHK4YbJGj55ZQ2i35w@mail.gmail.com>
 <12435.1455823990@sss.pgh.pa.us> <13291.1455825152@sss.pgh.pa.us>
 <CAM3SWZTUNXgzCfVsYSFP44cRo4K-93BPDC+gEbm_WveFdTKmnA@mail.gmail.com>
Comments: In-reply-to Peter Geoghegan <pg@heroku.com>
	message dated "Thu, 18 Feb 2016 14:56:58 -0800"
Date: Thu, 18 Feb 2016 18:06:32 -0500
Message-ID: <10683.1455836792@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-docs-owner@postgresql.org

Peter Geoghegan <pg@heroku.com> writes:
> I think that the P_ISLEAF() instrumentation of free space and
> fragments might still need to happen for deleted and/or half dead
> pages.

Don't see why; the documentation and field names clearly imply that those
numbers are accumulated only over leaf pages.  I certainly wouldn't expect
the fragmentation measure to include dead pages, for example, since they
would not get traversed by scans.  (Whether the "rightlink points to a
higher page number" rule for fragmentation is actually very useful is a
separate question; but as long as that's the measure, only pages that
are part of the leaf scan sequence should count.)

> Having looked at the 2008 commit d287818eb514d431 myself, ISTM
> that your intent might well have been to have that happen -- why else
> would any reasonable person have changed the order at all?

My best guess is that I was thinking that the tests were independent,
and rearranged them so that the most common case would be tested first.
I'm quite sure I didn't intend to change the statistical behavior, else
I would have updated docs and/or mentioned it in the commit message.

			regards, tom lane


-- 
Sent via pgsql-docs mailing list (pgsql-docs@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs