Re: Issue with an SQL having clause

[Pete Helgren <pete@xxxxxxxxxx>]

There are a couple of reasons I haven't headed that direction.  If I put 
them in the select then the 'group by' will also have to have them and 
since they will most likely be discreet values, my totaling logic will 
be lost.  Somehow, referencing them in the select as I have, even though 
they are part of a result field calculation, satisfied the requirement 
because when I run the statement without the extra parens in the 
'having' clause, it runs without error.  I would have expected SQL to 
complain about the missing SQREQ under both scenarios.  It only 
complains when I have the extra parens.  Kinda weird.  I might give 
Joe's CTE approach a whirl as I was leaning that way initially.

Pete Helgren
Value Added Software, Inc
www.asaap.com
www.opensource4i.com


On 2/3/2011 7:02 PM, Pete Hall wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> My old eyes are getting a little weak, but I don't see SQREQ in the list
> of columns in the result set. My understanding is that HAVING runs
> against the result set. What happens if you use openS0 and ordqty in the
> HAVING clause? If that doesn't work, try including sqreq, sqfin, lqord
> and lqshp in the result set. I'm 87.3% certain that will work.
>
> On 2011-02-03 18:57, Pete Helgren wrote:
> > I need to select records where the total of some fields is greater than
> > zero.  The field comparisons are from two different tables.  Here is the
> > sql:
> >
> > select ixprod,ixdesc,extvin,(lqord-lqshp) as ordqty,
> > coalesce(sqreq-sqfin,0) as openSO
> > from fix inner join ecl on ixprod=lprod
> > inner join eclx01 on lord=extord and lline=exline
> > left outer join fso on ixprod=sprod
> > group by ixprod,ixdesc,extvin, (lqord-lqshp),
> > coalesce(sqreq-sqfin,0)
> > having ((coalesce(sqreq-sqfin,0)) - (lqord-lqshp))>  0<---- It doesn't
> > like this line
> >
> > I am running this in an interactive SQL session to test it and it is
> > complaining about the SQREQ column in the having clause.
> >
> > This DOES work:
> >
> > select ixprod,ixdesc,extvin,(lqord-lqshp) as ordqty,
> > coalesce(sqreq-sqfin,0) as openSO
> > from fix inner join ecl on ixprod=lprod
> > inner join eclx01 on lord=extord and lline=exline
> > left outer join fso on ixprod=sprod
> > group by ixprod,ixdesc,extvin, (lqord-lqshp),
> > coalesce(sqreq-sqfin,0)
> > having  coalesce(sqreq-sqfin,0)>  0
> >
> > Yet with addition of one set of parenthesis in the having clause, it
> > will stop working:
> >
> > select ixprod,ixdesc,extvin,(lqord-lqshp) as ordqty,
> > coalesce(sqreq-sqfin,0) as openSO
> > from fix inner join ecl on ixprod=lprod
> > inner join eclx01 on lord=extord and lline=exline
> > left outer join fso on ixprod=sprod
> > group by ixprod,ixdesc,extvin, (lqord-lqshp),
> > coalesce(sqreq-sqfin,0)
> > having  (coalesce(sqreq-sqfin,0))>  0<-----This will not work.
> >
> > The reason for the parens is so that the having clause is properly
> > evaluated, but I must be doing something wrong.
> >
> > What am I missing?
>
> - -- 
> Pete Hall
> pete@xxxxxxxxxxxxxx
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
>
> iEYEARECAAYFAk1LXkAACgkQXczQcKdXKg6kfACeKQ/eFreI0BQ5MqvFLseQ3XFp
> LXoAniA9ux+shLcQi8EOcfhHBBgj5Uho
> =qx8q
> -----END PGP SIGNATURE-----