Looking for code

[Dan Lieuwen]

I'm a Ph.D student with Dr. DeWitt of the University of Wisconsin-Madison.
I'm looking for some code of a particular style.  You can modify it
so I can't tell what it does and still have it be useful to me though
not as useful.  Being able to see the form of the code is useful in and
of itself.  Being able to understand it helps, but isn't essential.

Do you have any embedded SQL that computes aggregates not supported
by the system?  I'm especially interested in cases where the aggregate
is being computed for a join and some grouping is desired, for instance
(in pseudo-embedded-SQL)

select(*) from Department {
  init variable used to compute standard deviation
  select(*) from Employee where Employee.dept = Deptartment.id {
      compute standard deviation;
  }
  use standard deviation;
}

Does anyone actually do this sort of thing?  Dr. DeWitt wants some practical
examples of where my work might be useful.  Your help would be most
appreciated.

Dan