Re: Object-oriented thinking in SQL context?
From: Brian Selzer <brian_at_selzer-software.com>
Date: Thu, 18 Jun 2009 20:12:00 -0400
Message-ID: <kJA_l.145$OF1.63_at_nlpi069.nbdc.sbc.com>
Date: Thu, 18 Jun 2009 20:12:00 -0400
Message-ID: <kJA_l.145$OF1.63_at_nlpi069.nbdc.sbc.com>
"rpost" <rpost_at_pcwin518.campus.tue.nl> wrote in message
news:h1eass$2imf$4_at_mud.stack.nl...
> Brian Selzer wrote:
>
> [...]
>
>>[...] The view update problem is the result of the ambiguity inherent
>>in the relational expressions that define views.
>
> Perhaps it helps to state the cause of the ambiguity: when an expression
> defines a non-injective transformation, there is no (unique) inverse
> transformation, hence the ambiguity.
>
Yes, that's it. Whenever the mapping between the base relation(s) and the derived relation is onto but not one-to-one, that is, a surjection that is not also an injection, there is no (unique) inverse transformation.
> --
> Reinier
Received on Fri Jun 19 2009 - 02:12:00 CEST