Daniel Staal wrote
> --As of August 10, 2014 1:40:11 PM -0500, JORGE MALDONADO is alleged to
> have said:
>
>> I have a table which will contain a set of pre-defined records defined by
>> the government of my country. The fields are ID and DESCRIPTION. The ID
>> field is composed of 3 characters that are always numbers. This means
>> that I can set such a field as a char(3), varchar(3), smallint or
>> integer, for example, and all of them will provide the correct
>> functionallity. My question is, why should I define such a field as a
>> character based type or a numeric based type? What is the best choice if
>> this field will always be a 3-digit number?
>
> --As for the rest, it is mine.
>
> My opinion: Unless you are doing arithmetic, it's better to stick to a
> char
> field of some type. In this case especially, the fact that the ID is only
> numeric is incidental - at some future point someone could decide they
> need
> more than 1000 ID fields, and start using letters. (They could also
> decide
> to lengthen it instead.)
>
> The only reason to store them as numbers is that it would be slightly more
> space-efficient, but it would be *very* minor. My recommendation is to
> use
> varchar(3).
I would use "varchar" and put the rest of the business logic into a check
constraint.
David J.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Clue-to-define-a-field-data-type-tp5814371p5814406.html
Sent from the PostgreSQL - novice mailing list archive at Nabble.com.