Re: Archive records with relationships

"Ole Laurisch" <OIe.Laurisch_at_NOSPAMt-systems.com> wrote:
> Hi at all,
>
> I have a complex database structure with many tables. But let's simplify
> the whole thing. Let us presume that we have this three tables A, B and
> C.
>
> [A]-1--------n-[B]-1---------n-[C]
>
> where "1----------n" are 1:n relations.
>
> Now the application will add thousands of records per day in these
> tables. Therefor we need to archive records that are no longer needed.
>
> Now the user can specify to archive some records (by a filter) from the
> table B.
>
> How could we handle the problem,

Well, that's a business problem, not a technical problem.

> that there are references now pointing
> from a record in table A to a record in table B, which might not be
> existing any more?

If I understand your diagram, A is the one and B is the many. In that case A does not point to B, B points to A. If records in B are removed, there are no dangling pointers left behind, because the pointers were also removed.

> One opportunity is not to delete the records in table B, but replacing
> them with null-values (all except the key values). But that could neither
> be the only nor the best solution, could it?

What would be the point? "Nyah nyah Nyah Nyah nyah, your data is gone but I just want to rub your face in it?"

Xho

-- 
-------------------- http://NewsReader.Com/ --------------------
Usenet Newsgroup Service              New Rate! $9.95/Month 50GB