I thought this would be straightforward, but I've failed.

I'm wanting to populate a database ...but without duplicates.

The simple 'find duplucate query just looks for any erhmm duplicates, but for Amazon Sales Reports (supplied in CSV form), you can have legitimate duplicate order IDs spread down several rows of a spreadsheet (this being where a customer has bought more than one item in an order), for example....

Code:
Order ID            Item
204-0590555-4444342  widget A
204-0590555-4444342  widget B
403-9238068-5549117  widget C
& so on.

If I run a duplicate query against the above data it would match on the order ID, but that's a legitmate occurence of a duplicate.

What I want to trap, is this scenario....
Code:
Order ID            Item
204-0590555-4444342  widget A
204-0590555-4444342  widget B
403-9238068-5549117  widget C
204-0590555-4444342  widget A
204-0590555-4444342  widget B
where clearly the same overall order has been pulled in twice - not good.

Now I reckon a duplicate query that matches on order ID & Item ought to do it....in other words,

if the order ID AND item are duplicated...then that's a duplicate - I need to know about it (knowing about it is the first stage, nexty comes how to remove them automatically).

can anyone help get me out the starting block (becuase all the duplicate queries just seem to go & find the existence of duplication anywhere in the table ...whereas I need need find duplication with an AND)