D & R reprot difference

[Kenneth Brody]

Charles Day wrote:
[...]
> > > -----
> > > If:
> > > Then: qn(9,.0)=3
> > > ----
> > > If: 2 lt  "100"
> > > Then: qn=""
> > > ----
> > > If:
> > > Then: qo(9,.0)=qo+qn
> > > ----
> > > If:
> > > Then:end
> > > ----
> > >
> > > Two:
> > > ------
> > > If: 2 ge "100"
> > > Then: qn(9,.0)=3;  qo(9,.0)=qo+qn
> >
> > Why move field 3 into qn, and then total qn?  Why not just total
> > field 3?
> 
> field 3 is only 4 length and the grand total sometimes will be much larger than
> 9999.

So?  As long as the destination field qo(9,.0) is big enough, there is no
need for an intermediate field.  And if qo isn't big enough, the intermediate
field won't help anyway.

[...]
> > Are you overriding automatic processing on output?  If so, is field
> > qn or qo defined in the default automatic, but not the one used at
> > output, or vice versa?  Have you complied the output processing using
> > the same overridden automatic?
> 
> yes, I am using both  -z  and  -y tables and all are defined in those tables.
> The extra count depended on which record was last selected.  If field 3 was 0 in
> last record the total was one greater than actual.  I still do not understand
> why it did it, but it did.  .
> 
> I got it all working by making the the second/larger dummy field global.

Which "second/larger dummy field" are you refering to?

Also, the code as originally posted should work just fine as-is.  To
repeat the second and third parts of my previous question:

Is field qn or qo defined in the default automatic, but not the one used at
output, or vice versa?  Have you complied the output processing using the
same overridden automatic?

-- 
+-------------------------+--------------------+-----------------------------+
| Kenneth J. Brody        | www.hvcomputer.com |                             |
| kenbrody/at\spamcop.net | www.fptech.com     | #include <std_disclaimer.h> |
+-------------------------+--------------------+-----------------------------+