Sender: wieck
Message-ID: <3870E982.B123DF8A@debis.com>
Date: Mon, 03 Jan 2000 19:25:06 +0100
From: Jan Wieck <wieck@debis.com>
MIME-Version: 1.0
To: Tom Lane <tgl@sss.pgh.pa.us>
CC: Bruce Momjian <pgman@candle.pha.pa.us>, pgsql-hackers@postgreSQL.org
Subject: Index toasting (was: Re: [HACKERS] Error "vacuum pg_proc")
References: <199912270336.WAA28667@candle.pha.pa.us>
	<25243.946268065@sss.pgh.pa.us> <25283.946268363@sss.pgh.pa.us>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Tom Lane wrote:

> I wrote:
> > And that'll still crash and burn for multicolumn indexes.
>
> Not to mention functional indexes, which typically store values that
> don't appear in the referenced tuple at all.
>
> Basically, indexes have to have their own toasters.  There's no other
> way.

     You're right.

     I think it's best to delay index toasting until we have some
     experience with normal, main tuple attribute toasting. It'd be
     nice if the solution had covered huge values to be indexed
     automatically (what it doesn't any more). But I think most
     people can live with a database, that cannot index huge
     values, but is capable to store and retrieve them for now.


Jan

--

#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me.                                  #
#========================================= wieck@debis.com (Jan Wieck) #