Hi Dan,
Yes, it is a general solution for result sets containing a sequence with
increments of 1.
My comment was concerning imparting significance to a Primary Key ( or
any key actually ) beyond that of row identification.
IMO this is something to avoid if at all possible. As I said however, it
is not an uncommon practice.
> Actually, that is not a general answer.
>
> You are assuming that the FuelID field has some sequence and there
> is no such guarantee. Even using an Autoincrement datatype.
>
> Say you forget to enter the information for a given day, Then enter
> the information for days following, later catch your mistake and
> enter the old days values. The difference between FuelID values
> will be something greater then 1.
>
> That said, it is not an uncommon technique Just recognize that you
> are building in dependencies external from the schema. Fail to
> enter the data in the correct sequence and the results of your
> query will be erroneous.
>
>
> Drew
Actually, it is a general answer. It is a technique of how to work
with sequential records based upon a particular field which contains
the sequence. Granted, anyone entering information in the wrong
sequence will also get erroneous results. But, isn't that where the
expression garbage in garbage out applies? And in my case, the second
query contains two fields: the odometer fields from the table and
the first query next to each other so that any such errors would be
obvious. (I added them since my last email just to view them, but as
you point out they also serve a very important verification function.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
