Re: numeric precision when raising one numeric to

On Fri, 20 May 2005, Tom Lane wrote:

> Has anyone bothered to actually look into the code?
>
> regression=# select power(2::numeric,1000);
>                                                                               power
>
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
10715086071862673209484250490600018105614048117055336074437503883703510511249361224931983788156958581275946729175531468251871452856923140435984577574698574803934567774824230985421074605062371141877954182153046474983581941267398767559165543946077062914571196477686542167660429831652624386837205668069376.0000000000000000
> (1 row)
>
> AFAICT the only thing missing is a pg_operator entry linked to the
> function.

It appears fairly limited however given that you rapidly run into the
numeric maximum length for exp.

It also doesn't seem to work terribly well:

sszabo=# select power(0.1::numeric, 15);
       power
--------------------
 0.0000000000000010
(1 row)

sszabo=# select power(0.1::numeric, 16);
       power
--------------------
 0.0000000000000001
(1 row)

sszabo=# select power(0.1::numeric, 17);
       power
--------------------
 0.0000000000000000
(1 row)

sszabo=# select power(0.1::numeric, 17)*100;
      ?column?
--------------------
 0.0000000000000000
(1 row)