As Tore and Borislav have already argued, use Int for whole numbers.

If your data model "used" {CARDINAL below 10**6} to map into (6,0):
There is always a tiny chance model will be "enhanced" by negative numbers in same range
instead of adding a new flag field into the data model(s)
esp. if there are oodles of transport protocols, which now have to be maintained.

If your data model REALLY intended for [-99999..999999], rethink ;-))
A trigger on the integer range in the backend is much better to maintain
and can be elevated to GUI valid checks whenever needed.

my 0.0022 of lazyne$$
thomas

>>>Hi,
>>>
>>>I am trying to write a code (in a .prg) that would change a SQL Server column size from numeric(3,0) to numeric(6,0). But I can't.
>>>Anyone can suggest a code to make such a column size change?
>>
>>ALTER TABLE (?)
>>https://learn.microsoft.com/en-us/sql/t-sql/statements/alter-table-transact-sql
>
>Thank you. I will need to, first, create a t-sql command that will be used to check if the column is numeric(3,0) or (6,0). Then to ALTER the table.