Re: database design method

On 9 Nov 2002 15:59:11 +0100, hidders_at_REMOVE.THIS.uia.ua.ac.be (Jan Hidders) wrote:

>That is not always enough. For example, if you just take the relational
>algebra (select, project, join, union, difference) and add a while construct
>(you can simulate the if-then-else with that) it is easy to see that you can
>only compute queries that can be computed in polynomial space. Morever, it
>can also be shown that there are some very simple queries, like does this
>relation contain an even number of tuples, that cannot be expressed in it.

With the aggregate operators you can.

In Tutorial D:

(count(R) mod 2) = 0;

Of course it will return a boolean value.

Tutorial D also has scalar types and variables, and variable assignements, thus I think it is clear it is computationally complete having a while construct.

>There's a very big and very interesting body of research literature on this
>subject, and to get an impression of that you might want to take a look at
>the chapters 16, 17 and 18 from "the Alice book":
>
> Serge Abiteboul, Richard Hull, Victor Vianu: Foundations of Databases.
> Addison-Wesley 1995, ISBN 0-201-53771-0

Thanks It was yet in my reading queue, I will try to move it to the top of the queue :-)

>For more information
>and a formal treatment see chapter 2 of "the Alice book".

Ok, thanks

Alfredo Received on Sat Nov 09 2002 - 21:58:17 CET