Re: constraints in algebra instead of calculus

From: paul c <toledobythesea_at_oohay.ac>
Date: Fri, 15 Jun 2007 16:26:02 GMT
Message-ID: <u4zci.29317$NV3.11158@pd7urf2no>

Jan Hidders wrote:
> On 15 jun, 16:32, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:
>

>>Jan Hidders wrote:
>>
...

>>
>>That constraint looks like a tautology to me. Can you explain how any
>>relation with a B attribute could fail the constraint?

>
>
> I'm not that well versed in the TTM / Tutorial D notation so I may
> have abused the notation a bit. To clarify:
> - B is a set of attributes (the non-key attributes)
> - R{B} denotes the projection of R on the attributes in B
> - R GROUP {B} AS C groups the attributes in B and names the resulting
> set-valued attribute C
>
> That probably clears it up, but I'll give a small example anyway
> (note: here B is not a set of attributes but a single attribute):
>
> Assume R = { (A:1, B:2), (A:1, B:3) }
>
> R{B} = { (B:2}, (B:3) }
> R1 = R{B} GROUP {B} AS gB = { (gB:{ (B:2} }), (gB:{ (B:3) }) }
> ...

I take it that R1 here has two tuples? (This seems crucial to me, assuming that TTM/TD allows grouping on all attributes even though it doesn't seem to mention that case.)

p Received on Fri Jun 15 2007 - 11:26:02 CDT