Re: SQL to validate data from multiple files

Define small... :)

I know of a cross join that runs weekly, 48 x 5500...giving me 264K rows
out...

But in this case a cross join is exactly what was needed. I'm basically
generating an item catalog with customized pricing for 48 customers.

select cust#, item#, CalcItemPrice(cust#, item#)
from catalogcusts cross join itemfile

FYI: judging from the sequence of the records output, the SQE reads through
the 5500 (well actually there's more like 30K rows, but only about 5500 are
selected) row item file once and loops through the catalog customer file
multiple times. IE. the records look like
item# cust#
1 1
1 2
1 3
<snip>
1 48
2 1
2 2

Charles

On Fri, Aug 9, 2013 at 2:02 PM, CRPence <CRPbottle@xxxxxxxxx> wrote:

Best

to avoid the Cartesian product except when the data is known to be very
small sets.