Date: Tue, 8 Oct 2002 17:20:47 +1000
From: Martijn van Oosterhout <kleptog@svana.org>
To: Shridhar Daithankar <shridhar_daithankar@persistent.co.in>
Cc: pgsql-hackers@postgresql.org,
 pgsql-general <pgsql-general@postgresql.org>,
	pgsql-performance@postgresql.org
Subject: Re: [pgsql-performance] Large databases, performance
Message-ID: <20021008072047.GB558@svana.org>
Reply-To: Martijn van Oosterhout <kleptog@svana.org>
Mail-Followup-To: Shridhar Daithankar <shridhar_daithankar@persistent.co.in>,
	pgsql-hackers@postgresql.org,
	pgsql-general <pgsql-general@postgresql.org>,
	pgsql-performance@postgresql.org
References: <3DA1E50F.4394.107FF85D@localhost>
	<3DA2BE03.3674.13CF7DD1@localhost>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3DA2BE03.3674.13CF7DD1@localhost>
User-Agent: Mutt/1.3.28i

On Tue, Oct 08, 2002 at 11:14:11AM +0530, Shridhar Daithankar wrote:
> On 7 Oct 2002 at 11:21, Tom Lane wrote:
> 
> > "Shridhar Daithankar" <shridhar_daithankar@persistent.co.in> writes:
> > > I say if it's a char field, there should be no indicator of length as
> > > it's not required. Just store those many characters straight ahead..
> > 
> > Your assumption fails when considering UNICODE or other multibyte
> > character encodings.
> 
> Correct but is it possible to have real char string when database is not 
> unicode or when locale defines size of char, to be exact?
> 
> In my case varchar does not make sense as all strings are guaranteed to be of 
> defined length. While the argument you have put is correct, it's causing a disk 
> space leak, to say so.

Well, maybe. But since 7.1 or so char() and varchar() simply became text
with some length restrictions. This was one of the reasons. It also
simplified a lot of code.
-- 
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> There are 10 kinds of people in the world, those that can do binary
> arithmetic and those that can't.