SQL Query Question

I am a SQL novice, I know enough to do fairly simple things but I'm not sure that I always take full advantage of the power of SQL to solve problems. I currently have a problem that I think I could solve with some simple SQL command but I'm not sure how to go about it. The problem is this, I have a table in an Oracle database which contains information about key customers. Information about customers is collected by an outside source. The outside source provides me a table of customers where each customer is identified by a numeric key. However, there are two types of key which may be used to identify the customer. Either key (or both) may be specified. In the end, I get information of the form:

	Key1	Key2	Customer Attributes
	------------------------------------------------------------
	111111		Foobar
	111112	A-100	Barfoo
		A-101	MooBar

When I get the information, I put it in a table and assign each record a unique id. In the example above, I would assign the ids 1, 2, 3 to the records.

Periodically, I receive new new updates. When these updates come out, customer information is updated. It's possible that where a Key1 wasn't specified there may be a Key1. Where a Key2 wasn't specified, there may be a Key2, there may be new entries, and some existing entries may have been deleted. What I need to be able to do is merge the new data with the old. Wherever I find a matching Key1, or Key2 I want to replace all data with the new data, but I want to keep my assigned identifier. Wherever there are new records, I want to assign a new unique identifier to the record. Finally, I would like to produce a list of all records which have been deleted.

Any tips on a straight-forward way to do this would be appreciated. I

Thanks.

James W. Howe                  		Voice: (734) 669-2244
AppNet Inc.    		 
650 Avis Dr. Suite 100               	mailto:jhowe_at_appnet.net
Ann Arbor, MI  48108         		http://www.appnet.net