Re: Possible problems with Date & McGoveran View Updating

Mikito Harakiri wrote:
> "Jan Hidders" <jan.hidders_at_pandora.be> wrote in message
> news:s7n9b.21899$%C3.1400873_at_phobos.telenet-ops.be...

>> > I would rather start with "invertible"
>> >
>> > #1 Transformation Q is locally invertible for database instance X and
>> > set
>> > #of
>> > views Y if there exists Q^(-1) such that:
>> >
>> > Q * X = Y <=> X = Y * Q^(-1)
>>
>> I assume you meant "Q * X = Y <=> X = Q^(-1) * Y" ?
>>
>> That's a trivial property: every transformation is locally invertible in
>> that sense. Note that because of the way you formulated it I only have to
>> show that given an X an Y there is a Q^-1. An example would the function
>> that maps every relation to X.

It is a constant function and therefore independent of everything. The way you formulated your definition all that I have to show is that if you give me a Q and X such that Q(X) is defined then I can construct a transformation Q^-1 that maps Q(X) to X. So if you give me X = { (a=1,b=2), (a=1,b=3) } I give you the following transformation:

  (SELECT 1 AS a, 2 AS b) UNION (SELECT 1 AS a, 3 AS b)

As you can see there is no reference to an instance, and not even to the view. Since I can do this for any X this demonstrates that for any Q and X it holds that Q is locally invertable for X.

Again, I know that this is not what you meant to say, but it is what your definition says and I'm not able to guess what it is that you really want to say.

> Please disregard the nested variables nonsence comment in my previous
> message.

Ok. I will.

• Jan Hidders