Posted by David Portas on 10/01/53 11:28

I would have to ask why you want to return the duplicate data in the
first place. I originally assumed that this was just for display on the
screen, in which case the client application is surely the best place
to solve the problem. Duplicate data in the database is bad news as I'm
sure you know and adding an arbitrary integer value obviously doesn't
solve the problem of redundancy, except maybe in the case of an
intermediate "staging" table. In a staging table you would probably
want to use IDENTITY to generate the row numbers and then filter the
result based on some logical key.

In SQL 2000 there is no generic method to generate row numbers in a
query. In SQL Server 2005 we have the ROW_NUMBER function to do this.

--
David Portas
SQL Server MVP
--

[Back to original message]