Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.misc -> Re: float storage

Re: float storage

From: sybrandb <sybrandb_at_gmail.com>
Date: Thu, 07 Jun 2007 08:05:36 -0700
Message-ID: <1181228736.766703.261940@n4g2000hsb.googlegroups.com>


On Jun 7, 4:40 pm, Jeff Kish <jeff.k..._at_mro.com> wrote:
> Greetings.
>
> I need to convert some columns of type number(12) to hold floating point
> information scale and precision I can't determine in advance (customer data
> can vary wildly) so I wanted to use the datatype that offers the maximum scale
> and precision..
>
> the customers can have anywhere from Oracle 9i onwards.
>
> I used the oracle 9i docs to drive my decision.
>
> I noticed that the number type offers "maximum scale and precision", but
> the float type offers the same precision, but no limit on the scale.
>
> I'm a bit confused as to which offers superior storage given that description.
> Then I figured, the chances are that converting to either 'number' or 'float'
> would in all likelyhood be good enough. (comment?)
>
> So then I wondered, since these two columns are part of a primary key, is one
> significantly better with respect to columns that are indexed?
>
> thanks
> Jeff
> Jeff Kish

Datatype won't matter here. Whatever you choose, it is *VERY* unusual (not to say undesirable) to have non-integer data constitute a PK.

--
Sybrand Bakker
Senior Oracle DBA
Received on Thu Jun 07 2007 - 10:05:36 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US