Re: Mandatory tableOwner property

Hi all

Souri, I think David and Robert are right, the tableOwner should be 
optional, for example in Mysql, just to mention one example, is possible 
to write SQL queries with unqualified table names. The user information 
is only part of the connection properties.

Robert, whenever Souri et al. check this issue and express their final 
decision, I'll remove/keep the tableOwner from the test cases :-)

Boris


On 11/02/2011 15:58, Robert Scanlon wrote:
> I agree with David, this property should be optional.
>
> I would like to see the tableOwner removed from *most* test cases 
> (except property-focused tests specifically meant to test that 
> property).  That provides more flexibility to test implementers to 
> automate the tests, without having to create a _single_ special 
> schema.  (It also allows for test implementers to create a unique 
> schema per testcase if they want, or to try the test cases on 
> different schemas that have different characteristics).
>
> Bob
>
> On Fri, Feb 11, 2011 at 7:54 AM, David McNeil <dmcneil@revelytix.com 
> <mailto:dmcneil@revelytix.com>> wrote:
>
>     Souri-
>
>     I think the tableOwner property should be optional. Using
>     tableOwner and tableName is a short-hand for the SQLQuery:
>
>     SELECT * FROM <tableOwner>.<tableName>
>
>     When writing this query by hand the tableOwner is optional. In the
>     same way I think tableOwner should be optional when using the
>     tableName property. This allows the author of the SQL to either
>     use the tableOwner if it is needed in their context or to omit it
>     if it is not needed. In my experience it is common to write SQL
>     queries with unqualified table names and rely on the user to only
>     be specified as part of the connection properties.
>
>     -David
>
>