>
> If an INT has a fixed range, then what is the point of giving it scale? As
> in, "int(12)".
>
> In Oracle, a NUMBER(12) indicates how many digits you could have (in this
> case, 99999999999 would be the max value).
>
> Would an int(2) allow -99 to 99, or -2147483648 to 2147483647?
That's because in Oracle NUMBER is a primitive or internal type and has 38
digits of precision. Everything else including INTEGER is a subtype of NUMBER.
So in Oracle, you must limit the subtype or you'll have an INTEGER that can
store a number up to like 10^125. For example, the subtype BINARY_INTEGER is a
signed int with range -2147483648 to 2147483647.
----------------------------------------------------------------------
Michael Brando
Senior Manager of Engineering
Applied Biosystems
3833 North First Street
San Jose, CA 95134-1701