Re: Default setting of NAMEDATALEN

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: "jochen(dot)westland(at)invigo(dot)de" <jochen(dot)westland(at)invigo(dot)de>
Cc: "'pgsql-hackers(at)postgresql(dot)org'" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Default setting of NAMEDATALEN
Date: 2002-10-21 15:29:22
Message-ID: 200210211529.g9LFTM122576@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


New NAMEDATALEN default will be 64 in 7.3. We are in beta now. We did
see performance hit with values greater than 64. It is too hard to make
it changeable after compilation.

---------------------------------------------------------------------------

Jochen Westland wrote:
> Hi there,
> my question is short, quite simple and
> definately a hacker thing (hopefully).
>
> FIRST of all:
> POSTGRES is a great database system! Thanks
> a lot for your perfect work.
>
> Now my Question:
> Wouldn't it be possible to change the
> default setting of NAMEDATALEN in
> all distributions to a higher value,
> lets say 128?
>
> Or a bit better:
> Get the length of NAMEDATALEN by a
> select statement for all applications
> using postgres beeing a bit more flexible?
>
> Reason:
> I think that there are several systems out
> there running with higher values to make
> reading (and understanding) table- and
> rownames a bit easier.
>
> All linux packages are compiled with that standard
> value as well as all applications (like psqlodbc eg)
> are actually designed - also by default - for a length
> of 32, so changing NAMEDATALEN and recompiling
> postgreql does'nt solve anything, afterwards you
> have to contact odbc-developers, tool-developers
> and so on for a hint how to make their systems
> cope with that new value.
>
> Regarding decreasing cpu-speed and increasing hardware
> cost, shouldn't it be possible to set a higher value
> for NAMEDATALEN by default, so that especially complex
> postgres-databases can be developed on a standard compilation,
> and the development of applications would be made a bit more
> comfortable?
>
> Regards and thanks for your time
>
> jochen
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddressHere" to majordomo(at)postgresql(dot)org)
>

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2002-10-21 15:35:33 Re: Please help
Previous Message Doug McNaught 2002-10-21 15:27:19 Re: Postgresql and multithreading