Re: No exceptions?

> 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?

Is there a logical flaw in substituting the value zero for x in "x + y" if x happens to be undefined ?

Is there a logical flaw in substituting the empty string for x in "x concat y" if x happens to be undefined ?

I think this kind of situation, as perceived by a compiler, points so overwhealmingly to a programmer error, that it is insane to do anything different that raise a compile time error.

Moreover, in the case of "x join y", you are very likely making false deductions wrt the type of the result. The type of TABLE_DUM join y is RELATION {heading of y here}, while it is rather unlikely that this is indeed the intended type of the join. Any further reference to the result of the join is then likely to fail anyway on a type "mismatch". Received on Mon Jul 03 2006 - 12:25:07 CEST