Re: No exceptions?
Date: Fri, 30 Jun 2006 05:49:59 GMT
> All the attributes in a relation comprise, at least, a superkey. The
> set of attributes that qualify as a candidate key must hold unique
> values and no subset of those attributes must hold unique values. The
> only relations that could have empty candidate keys are those with
> empty headings, right?
Apologies - I'll try to read the rest tomorrow, but right now I'm tired
(from watching a library video - "Guns, Germs and Steel" - the author
seems to be the James Burke of biology, not that I know much about the
latter). I thought a relation with any number of attributes could have
only one value for them if it had an 'empty' set of candidate keys, eg.
a relation that has only one tuple? (and a 'relvar' could have only one
value at any instant in time)?
pReceived on Fri Jun 30 2006 - 07:49:59 CEST