Re: Unique constraint and NULL values

<ctcgag_at_hotmail.com> wrote in message
news:20041021122301.861$C2_at_newsreader.com...
| "Mark C. Stock" <mcstockX_at_Xenquery .com> wrote:
|
| > yes, in theory, the SQL Server approach is more 'pure', but in practice,
| > the Oracle approach is more practical (i'm starting to sound like yogi
| > berra!)
| >
| > if i'm creating a UNIQUE constraint, my goal is to disallow values that
| > are known to be equal to other known values -- so, it's entirely
| > appropriate to not disallow (sorry for the double-negs) unknown values
| > (triple-negs?) since there is no known basis for excluding them
|
| But then why does it exclude them when there are other columns present?
|
| Xho
|
| --

i suppose because if you concatenate the values, the results are equivalent

perhaps someone can weigh in on how this maps to the SQL standard?

++ mcs

(by the way -- let's keep this thread in just one group, no need to cross-post) Received on Thu Oct 21 2004 - 11:28:22 CDT