Thanks!

ST

Actually I wrote data in both instances.  An example is below

-- Approach 1

CREATE TABLE FOO (iid int identity, ky int)

INSERT FOO(ky) SELECT ky FROM myTable ORDER BY ky

-- Approach 2

CREATE...

Thanks for the responses, everyone.  Yes, once again tripped up by floats.  And fooled by the fact that for most of the numbers the result worked out how we wanted. ...

Yes, this was happening on an Excel import process.  Users entered percentage values from 1-100.  When they get to the server they are presented as floats by the C# import...

that's some awesome coding there!  Thanks for the improvements 🙂

ST

In looking at it further I think I'm barking up the wrong tree.

Thanks

Great idea, I'll look at the execution plan.  Thanks for the tip.

Yes, The reason for the cartesian join is to simulate every possible calculation in the current data set to see of the result is the same when handled as an...

Thanks for the reply.  What are your concerns with the cartesian join?

I think this will work but I don't know that it really saves you much.  The subquery that you used doesn't go back to the same table again, it just...

I agree with you J. Drew Allen.  It's not the MOST efficient way to do it but it demonstrates the mechanics of how to do it.  The approach you mention is...

Here's one way to do it

-- Use a common table expression to find the max values for records and records that are duplicated
-- This won't...

I see what you mean.  If I look at the data from the POSS perspective instead of the Sales perspective that makes sense.  Thanks, now to apply this to the...

If you were to use inner join and the Mary sales record had 2 entries, each entry would join to the POSS table causing duplication.  I don't think inner join...

DoH.  I can use exists and 'not exists'.  I guess that's the better option.

SELECT person, category, item, Qty FROM Sales
UNION
SELECT person, category, item,...