Question

I am new to Oracle, I have two tablespaces one for dev and one for live.

Is there a way that I can have a variable on a Package header that contains a tablespace name, which can then be used from within the procedures defined in the package header?

I need to do this as I have one tableSpace (TableSpaceA) that needs to query tables in TableSpaceB via procedures. The Dev TableSpaceB has a different table space name to live therefore I was hoping I could declare in the Package header a variable like


Tablespace temp = “TableSpaceName”

Then

Select * from temp.TableName


And finally change “TableSpaceName” when i roll my changes out to the live environment?

Answer 1

Tablespaces are an administrative thing: they are a way of logically grouping storage without the need to worry about filepaths or other administrivia.

---

From your revised question it is obvious that Tony is right and what you mean is schema.

A schema is the set of objects owned by a user. When a procedure owned by User A needs to reference an object owned by Schema B there are two ways of doing this. The first way is to hardcode the schema name.

select * 
from user_b.emp;

This is usually fine but will not work in the scenario you describe, where the other schema has a different name different name in other environments (C rather than B).

The way around this problem is to use synonyms.

select * 
from not_my_emp;

In DEV the synonym would be:

create synonym not_my_emp for user_b.emp
/

whereas in production it would be

create synonym not_my_emp for user_c.emp
/

The table could even have a different table in production, it doesn’t matter. The synonym acts as an interface to shield our objects from the untidy details of other schemas.

Note that a synonym does not grant privileges on the underlying object. This means that USER_B has to grant privileges on EMP in Development and USER_C has to grant privileges in Production.