OK, let's reset. We are discussing dynamically created bag indexes to
hold indexing requests for types without a covering index. In your
problem scenario above:

-TypeSystem: t2 and t3 derive from t1
-User defined sorted index exists for t3
-fs of type t2 added to index -> causes bag index for t2 to be created
-fs of type t1 added to index -> causes bag index for t1 to be created.

What should happen here?

Indexes for t2 and t1 were added dynamically to support the indexing
of FS without a user defined index. It could make sense that for these
types there is no reason for an iterator over t1 to include instances
of t2, which would be the case for your alternative solution. So when
the bag index for t1 is created dynamically, it should not create
additional indexes for subtypes and link to them.

Eddie

On 2/8/07, Thilo Goetz <[EMAIL PROTECTED]> wrote:
So what if t3 has a user defined index, e.g., a sorted index?  Then the
bag index for t1 must make sure not to mention t3 at all.

Eddie Epstein wrote:
> Assuming t3 is also derived from t1, and t3 already has a bag index
> defined, couldn't the t1 bag index link to the existing bag index for
> t3 instead of creating a new one?

Reply via email to