Re: insert to projection

From: Walter Mitty <wamitty_at_verizon.net>
Date: Sat, 05 Sep 2009 11:53:07 GMT
Message-ID: <Disom.1190$Jd7.64_at_nwrddc02.gnilink.net>


"Tegiri Nenashi" <tegirinenashi_at_gmail.com> wrote in message news:3f130ae7-bf22-4322-89fb-30392c785620_at_y10g2000prf.googlegroups.com... On Sep 4, 12:12 pm, paul c <toledobythe..._at_oohay.ac> wrote: <quote:>
> Tegiri Nenashi wrote:
> > On Sep 4, 10:52 am, paul c <toledobythe..._at_oohay.ac> wrote:
> >> Tegiri Nenashi wrote:
>
> >> ...
>
> >>> The only RA operation, that seems to allow view updates is renaming.
> >> Since when does RA have views?
>
> > View is a named RA expression.
> > ...
>
> In other words, view is a language device. RA has no ambiguity, so if
> your chosen dbms language allows ambiguous results, it is due to your
> choice of language,

Yes, guilty with sloppy language. The other thread has already established the term "underconstrained" rather than "ambiguous". So my thesis is that all simple relational expressions are underconstrained from view update perspective.

Let's get back to the main point. I'm suggesting that analysis of simple relational expressions update won't be useful. Suppose a view is composition of join, selection, and projection. Each of these can violate some restrictions that you would propose based on analysis of basic relation updates. Yet, the view can still be updatable.

</quote:>

Part of the problem with regard to sloppy language is that the term "view update" is misleading. If view C is defined as A join B and one were to apply an update, let's say MINUS D, what gets updated? If we were, in reality, updating view C, then the update would be really simple. We would update view C by changing its definition. The new definition of C is (A join B) minus D. This might require making D be persistent, so that the view can be applied later. Hey, presto! View C has just been updated!

But that's not what we really mean when we say "update view C". What we mean is "leave view C defined exactly as before, but update A and B such that the effect on C's apparent extension is the same as if the update had been applied to a base relvar whose extension is the same as C's apparent extension." Under this meaning of "update view C" the operation is underconstrained, as has already been noted. Received on Sat Sep 05 2009 - 13:53:07 CEST

Original text of this message