Re: No exceptions?

From: paul c <toledobythesea_at_oohay.ac>
Date: Fri, 30 Jun 2006 22:41:30 GMT
Message-ID: <uMhpg.109181$iF6.52435@pd7tw2no>

Bob Badour wrote:
> Jon Heggland wrote:
>

>> paul c wrote:
>>
>>> Let me re-phrase my original question:  Is there a logical flaw in
>>> substituting TABLE_DUM for x in the expression "x join y" when x is not
>>> in the catalogue?
>>
>> I don't know what precisely you mean by "logical flaw", so I'll pass
>> judgement. If something should be substituted for x (a "default value",
>> so to speak), TABLE_DUM does seem the natural choice, though, as it
>> corresponds to false/zero in some sense.

>
> But then, what about expressions like "y minus x" where x is unknown?
> What if y and x were intended to both be very large relations with a
> small difference? The result would go from small to very large.
> ...

Just to be clear, I assume that means (in TTM style):

y <AND> <NOT> TABLE_DUM ==
y <AND> TABLE_DEE ==
y

and yes it would be very large.

> Similarly for "not exists" expressions.

Yes.

>
>

>>> (Assuming that the syntax requires x to be a relation and with the whole
>>> expression's value being TABLE_DUM as well and granting that such a
>>> result might seem surprising to most people.)
>>
>> "TABLE_DUM join y" evaluates to the empty relation with y's heading, not
>> TABLE_DUM (unless y's header is also empty, of course).

>
> Surprising results with no explanation are nowhere near as useful as an
> informative or even instructive error message.

No argument, I just don't want to handle that information in an evaluator, nor prescribe how it should be handled.

p Received on Fri Jun 30 2006 - 17:41:30 CDT