Robert,

<< The above would probably be a subquery, with the main query selecting
name, phone number, whatever. >>

That's what I thought you would say.  The only time this is an issue is with
sub-queries, just like the original example with COUNT(*).  Any other time,
it is of no consequence to include the SUM() expression in the SELECT list.
In fact, I would assume that in most cases one would want it anyways.

<< How else would you give the year end bonuses at Elevate? >>

We don't get bonuses - we have to be happy with a paycheck, period.

--
Tim Young
Elevate Software
www.elevatesoft.com

"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:504A657E-0E82-4B1C-BCB3-0C91C0820C6A@news.elevatesoft.com...
> Robert,
>
> << The above would probably be a subquery, with the main query selecting
> name, phone number, whatever. >>
>
> That's what I thought you would say.  The only time this is an issue is
with
> sub-queries, just like the original example with COUNT(*).  Any other
time,
> it is of no consequence to include the SUM() expression in the SELECT
list.
> In fact, I would assume that in most cases one would want it anyways.

Sure. It is really no big deal, it was mostly an academic discussion. I very
seldom use subqueries anyway, since with DBISAM a script selecting the
aggregate into a memory table, indexing the memory table and then using a
join in the final query is sooo much faster.

It just seemed that since you can use WHERE field = xx without selecting
field, you should also be able to prevent the selecting of the aggregate
number, and just use it in the HAVING clause.

I guess we'll have to live with this "serious flaw" sigh

Robert

PS: Thanks for taking care of my customer and his ODBC problem, as usual Tim
performing above and beyond the call of duty.

Robert,

<< Sure. It is really no big deal, it was mostly an academic discussion. I
very seldom use subqueries anyway, since with DBISAM a script selecting the
aggregate into a memory table, indexing the memory table and then using a
join in the final query is sooo much faster.

It just seemed that since you can use WHERE field = xx without selecting
field, you should also be able to prevent the selecting of the aggregate
number, and just use it in the HAVING clause. >>

Normally, yes.

<< I guess we'll have to live with this "serious flaw" sigh >>

Yes, at least for now.

<< PS: Thanks for taking care of my customer and his ODBC problem, as usual
Tim performing above and beyond the call of duty. >>

No problem.  He's always very nice when he calls, so I'm always more than
happy to help him out.

--
Tim Young
Elevate Software
www.elevatesoft.com