Re: Are one row, one column tables "acceptable"?

From: Ed Prochak <edprochak_at_adelphia.net>
Date: Mon, 07 Apr 2003 02:34:09 GMT
Message-ID: <3E90E786.2060503@adelphia.net>

Niall Litchfield wrote:
> "Hans Forbrich" <forbrich_at_telusplanet.net> wrote in message
> news:3E8FCE58.DA81CCE1_at_telusplanet.net...
>

>>Thomas T wrote:
>>
>>
>>>How acceptable are one row, one column tables?  I've heard of people
>>

> storing
>

>>>multiple system variables in two-column, multi-row tables, with column
>>

> names
>

>>>such as "parameter_name" and "parameter_value".  But is using a one
>>

> row/one
>

>>>column table common?  Or is it a way that works, but of bad design (like
>>>creating multiple public rollback segments in the System tablespace)?
>>
>>A 1-row, 1-column table is not a bad thing.  It tends to look silly, and

>
> many
>

>>purists will argue against it.  But, remember the most famous table of all
>>"DUAL"!

>
>
> I won't argue against it, it depends on the design. However if bringing dual
> into the conversation be aware of the following (in a play area not a
> production system that wuld be silly)
>
> sqlplus /nolog
> shutdown
> startup nomount
> select * from dual;
>
>

And some how some day, someone will insert something into your one row table. then your system fails because your PL/SQL has lines like:

SELECT myvalue INTO mypackagevar from my1rowtable;

which throws an exception for multiple rows returned.

(Yes this can even happen with DUAL!)

So I vote to avoid this kind of table.

-- 
Ed Prochak
running: http://www.faqs.org/faqs/running-faq/
family:  http://web.magicinterface.com/~collins
--
"Two roads diverged in a wood and I
I took the one less travelled by
and that has made all the difference."
robert frost

Received on Sun Apr 06 2003 - 21:34:09 CDT