Re: more on delete from join

From: Bob Badour <>
Date: Tue, 25 Aug 2009 13:46:11 -0300
Message-ID: <4a941558$0$23773$>

paul c wrote:

> Mr. Scott wrote:

>> "paul c" <> wrote in message
>> news:IPWjm.40275$Db2.30224_at_edtnps83...
> ...

>>> If A' and B' stand for the resulting values of the base relvars,
>>> A' = (J MINUS D} {hA} UNION (A MINUS J{hA})
>>> B' = {J MINUS D} {hB} UNION (B MINUS J{hB}).
>> I don't think this is right. It assumes that a delete from J
>> translates into a delete from both A and B, when a delete from either
>> A or B would suffice. The predicate of a join view is the conjunction
>> of the predicates of the tables being joined. For there to be a row
>> in J requires that there be a row in both A and B, so the fact
>> represented by a row in J is the conjunction of the facts represented
>> by the corresponding rows in A and B. Denial of at least one of the
>> facts represented in A or B denies the fact represented in J;
>> conversely, denial of the fact represented in J denies at least one of
>> the facts represented in A or B, but not necessarily both.
>> ...
> Yes, it does make that assumption, consequence of logical independence I 
> think.

Preservation of symmetry as well. However, I see no theory to support either choice, which is why I prefer the pragmatism of allowing an expert user to specify how the delete should operate in this situation. Received on Tue Aug 25 2009 - 18:46:11 CEST

Original text of this message