Re: View Updates Problem Statement

"Mikito Harakiri" <mikharakiri_at_ywho.com> wrote in message news:v6b7b.31$741.153_at_news.oracle.com...
>
> "Mikito Harakiri" <mikharakiri_at_ywho.com> wrote in message
> news:BBa7b.29$741.241_at_news.oracle.com...
> > Database A has relations RA1, RA2, RA3, ...
> > Database B has relations RB1, RB2, ...
> >
> > Views V1, V2, V3, ... establish the mapping between the 2 databases:
> >
> > V1*RA1 = RB1
> > V2*RA2 = RB2
> > ...
>
> Terminology sucks. Since now on by "view" I assume an operator
transforming
> input relation into output relation (while the literature calls a view
just
> an output relation variable). That definition is more consistent with
> algebraic view update ideas that I'm trying to convey.

I agree your terminology sucks. Instead of view, why not use "transform" and leave view to mean an unstored named derived relvar? Thus the expression defining a view is a transform of one or more relation variables to another relation variable, and a view update inverts the transform.

> Next correction involves the equation above. Clearly V1 can be applied to
> any number of input relation, not to RA1 only. Therefore, it has to be
>
> V1*(RA1, RA2, ...) = RB1
> V2*(RA1, RA2, ...) = RB2
>
> Note nice symmetry: we have an input "vector" of relational variables
(RA1,
> RA2, ...) transformed by the system of views V1,V2,... into the output
> vector of relational variables (RB1, RB2, ...)

You have a transform matrix. Received on Tue Sep 09 2003 - 17:15:39 CEST