| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
Home -> Community -> Usenet -> comp.databases.theory -> Re: Cardinality "highly unusual"
Hugo Kornelis wrote:
...
> Based on your description, you've chosen the right cardinality. I don't
> know if the statement of MS Visio EA about this type of cardinality being
> rare is true or not. In any case, I wouldn't worry about it. It doesn't
> sound like an error message - more like a warning. Consider it a reminder:
> the program tells you "this is unusual - you might want to double-check if
> you didn't make an error here".
Hi Hugo
Thanks, you've helped to set my mind at rest!
> Is the text you're working from specifically written for your situation
> (rare books, that are unique and can therefor be sold only once). I'm
> asking because the solution from the book is exactly what a reular book
> store would use (though I'd personally prefer to call the table "Titles"
> instead of "Books" in a regular book store). The BookOrder table is used
> to store the many-to-many relationship between Order and Book (title),
> since in a regular book, each order has one or more books and each title
> can be sold in zero or more orders.
Right, you would use a composite entity to resolve a n:m relationship. That makes sense if the bookshop sells new books. But my text (actually an MS book) deals explicitly with rare and unique books, and uses a composite entity to resolve a 1:m relationship between order and book. This strikes me as logically and conceptually wrong, but if it is the preferred way of doing things then so be it.
Try as I might, I simply cannot generate this sort of schema from an ORM source model (my preferred modeling method). Instead, I have to do a lot of tweaking of the logical diagram mapped from the ORM source model.
Matt Received on Tue Feb 15 2005 - 06:33:36 CST
![]() |
![]() |