follow up

Hi,
Sorry about that, I'm a newbie. If you forget about my code, how would you write the SQL to do the following:

If you have 1 table with all the data. You want to match rows based on some keys to get a grouping of them and then within the group to identify the 2 rows that should be matched. The 2 items are invoices and reversals. The invoice data should always match to the closest reversal date...

I am trying to write a stored procedure which will match the records from 2 tables and update a column in each saying the records match by using a counter. Both tables are exactly the same. For instance, if a record in table 1 matches a record in table by using a column as a key, then update the Match column in each table to say "1" then incremement the counter. What makes this difficult is that the records have to be matched by the closest...