RE: SQL Question

Thank you all for the responses... Charles hit the nail on the head. This is certainly non-trivial.

Our "Box Contents File" contains shipments for many years (11 million rows). My thinking was that (smaller) orders would repeat - people tend to buy the same widgets together.
Rather than trying to figure out what box the items will fit into (based on their dimensions), I was considering finding an order that matched the contents.

Our file contains a row for each widget and quantity inside a box AS WELL AS the box "code" it was actually packed/shipped in. So I would almost need to find a box that contained the exact item & quantity combinations and nothing else.
We also ran into some other issues with this process that makes it even more complicated.

So for right now, we're going to rely on the "human" computer (i.e. the picker) to know what box to use.


-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of Charles Wilt
Sent: Tuesday, August 30, 2022 5:56 PM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Subject: Re: SQL Question

I'm not sure the structure of the data you have is enough to answer your
question...

How do you know that
boxB, 1, widget2
boxB, 1, widget3

Held both types of widgets at the same time? Seems just as likely that at
some point boxB was used to ship 1xwidget2 and at another time was used for
1xwidget3.

It's an interesting challenge, certainly seems set based. So I'd think SQL
should be able to provide an answer.
But I'd definitely call the solution non-trivial...

Possible paths,
using the INTERSECT of two results sets, this would give you all boxes
that had contained at least the desired objects. But you'd have to then
discard the boxes with extras.

Using LISTAGG, to build a string 'widget2@1|wudget3@1' that you could
search for directly. You'd have to be careful about how you combined qty &
sku. But this might be the best performance as you could pre-calculate and
store the historical data.

Charles