This is hard because there is no way to do this in standard SQL (correct me if I’m wrong). SQL is based on relational algebra, and duplicates cannot occur in relational algebra, because duplicates are not allowed in a set. That’s why SQL doesn’t give you tools to solve this problem.
Why is the Common Language Runtime so important for SQL Server? Its programmability. It gives us what we call language-independent stored procedures. Stored procedure language today is T-SQL. It’s just that one language, and it’s not the best language. The SQL guys aren’t necessarily good language people. But the developer team is, and the ability to write in C# stored procedures and be in the IDE [integrated development environment] where developers live is a very powerful feature.
This is a terrible thing to do!
Borland and Microsoft Team to Deliver Enterprise Database Development Solution for the Microsoft .NET Framework
Monday June 2, 8:35 am ET
Borland(R) C#Builder(TM) for the Microsoft(R) .NET Framework to Include Microsoft SQL Server(TM) 2000 Developer Edition
Compare the above to the Yukon discussion here…
I guess I’m not sure of the actual relationship here…
We’ve been conversing since last fall with Randy Holloway — coauthor (with Andrej Kyselica) of an upcoming book “Developing Solutions with Yukon: Beyond Transact-SQL” (Addison Wesley, 2004) – on the ins and outs of Yukon.