Corncrowe wrote:
>
> >We can't really help you if we don't have an error message to >decipher.
>
> I was implying structure and flow rather than a specific error message.
>
> >> delete from temp_aging where ord_id is not null;
> >> commit;
>
> The delete temp_aging and commit statement...there is a cursor at the beginning
> of the transaction. The cursor is populated with a select statement from
> temp_aging, Later in the code fragment a delete statement is encountered.
> What is the purpose of the delete temp_aging at this point in the flow?
>
> I guess the question is to find mistakes in structure and point out how this
> stored procedure can be written differently. Think of it as an exercise of
> sorts.
>
> Thanks,
>
> Jon
The only thing I see questionable is the update of
temp_aging.
Every row is to be updated so perhaps a straight
update with subselect might be more efficient than
the cursor select, ordvol select, and temp_aging update
approach. A straight update statement would process
temp_aging row by row without reselecting for each row.
That's obvious I guess and I suspect you used the cursor
approach because a single update statement just didn't
perform as well.
Hope this helps, even though it probably doesn't.
Terry
Received on Tue Oct 21 1997 - 00:00:00 CDT
