Re: what datatype to use question

I'm not sure I understand perfectly your question, but number(p,s) might be what you're looking for. The "p" is the precision, and the "s" is the scale.

Daniel

Jeff Kish <jeff.kish_at_mro.com> wrote in message news:<vc069v8rp0h985sd1ndh1nagpvce7s2fg7_at_4ax.com>...
> On Tue, 08 Apr 2003 12:45:09 -0400, Jeff Kish <jeff.kish_at_mro.com> wrote:
>
> >Greetings.
> >
> >
> >What datatype should I specify for a column that is to hold image extents?
> >The values can be for raster, pretty small or large integers,
> >and for vectors they can be in ranges like the following:
> >(-21234.5768, 21234.5678)
> >(-0.00332, .14234)
> >etc.
> >
> >
> >I am guessing that as long as I can guarantee 6 or 7 decimal places of accuracy, and hold up to 7 or 8 digits to the
> >left of the decimal place, I should be pretty safe, but I am unsure how to proceed, i.e.:
> >
> >CREATE TABLE TESTDOUBLE (
> > REALONE REAL ,
> > DECIMALONE DECIMAL ,
> > FLOATONE FLOAT ,
> > NUMBERONE NUMBER
> >)
> >Are one of these recommended? I don't "think" I'll have to do where searches on the column, but I can't guarantee that
> >it might happen in some obscure case.
> >
> >Thanks for your recommendations.
> >Jeff
> I'm leaning towards float. How can you tell the scale you get, i.e. number of digits to the right of the decimal point
> with using a certain type? I looked at all_user_tables and did not see anything for scale.. Does this mean it stores
> what it can?
>
> Thanks again,
> Jeff
Received on Tue Apr 08 2003 - 19:07:41 CDT