Date: Wed, 1 Nov 2006 16:07:21 -0500
From: Andrew Sullivan <ajs@crankycanuck.ca>
To: Volkan YAZICI <yazicivo@ttnet.net.tr>
Cc: Andrew Sullivan <ajs@crankycanuck.ca>, pgsql-sql@postgresql.org
Subject: Re: record datatype comparisons
Message-ID: <20061101210721.GI25856@phlogiston.dyndns.org>
References: <55c095e90610311135l101f6a91s433e6d7e5b90cb7c@mail.gmail.com>
	<8C5B026B51B6854CBE88121DBF097A8655CBDD@ehost010-33.exch010.intermedia.net>
	<20061031212253.GD20809@phlogiston.dyndns.org>
	<20061031214500.GC1911@alamut>
	<20061101142833.GA25856@phlogiston.dyndns.org>
	<20061101210004.GA3321@alamut>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20061101210004.GA3321@alamut>
User-Agent: Mutt/1.5.9i

On Wed, Nov 01, 2006 at 11:00:04PM +0200, Volkan YAZICI wrote:
> But there should be a way to reach the types of the attributes of the
> record data type. Otherwise, how would it be possible to place it into
> another tuple as an attribute. 

Well, surely there is, but that has to happen at assignment time,
AFAICT.  So you can decide what the record is on the basis of the
datatype, but you can't write an equality operator, because you don't
know what the general rule is going to be in advance.  At least,
that's as far as I'm able to understand how to do this.  But I'm
really past my Latin at this point, because even if I looked at the
code that supports the record type, I wouldn't understand it.

A

-- 
Andrew Sullivan  | ajs@crankycanuck.ca
The whole tendency of modern prose is away from concreteness.
		--George Orwell