To: Greg Smith <greg@2ndquadrant.com>
cc: pgsql-docs@postgresql.org
Subject: Re: Explain docs and the regression database 
In-reply-to: <4C01CE67.8010709@2ndquadrant.com> 
References: <4C01CE67.8010709@2ndquadrant.com>
Comments: In-reply-to Greg Smith <greg@2ndquadrant.com>
	message dated "Sat, 29 May 2010 22:33:11 -0400"
Date: Sat, 29 May 2010 23:37:40 -0400
Message-ID: <28482.1275190660@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Greg Smith <greg@2ndquadrant.com> writes:
> The introduction to using EXPLAIN:  
> http://developer.postgresql.org/pgdocs/postgres/using-explain.html 
> includes a footnote at the bottom suggesting you might follow along with 
> these examples because they come from the regression test database.  
> This would really be a lot more useful if it gave some sort of clue just 
> how exactly a user might get that database populated with that 
> regression data.

You run the regression tests.

> I'm not even quite sure myself exactly what point in 
> the regression testing cycle these examples came from.

The ending state.

> Another open 
> question in my head is whether enough pieces of the regression set are 
> shipped in the most common packaging sets that it's ever practical to 
> expect users to run the tests at all.  Thoughts?

In Fedora/RHEL, and I believe also the PGDG RPMs, you'd install the
postgresql-test RPM and run it per directions.

			regards, tom lane