In my previous article, I showed you how to write a very simple function in Visual Basic .NET and then call it from T-SQL code in SQL Server "Yukon." But that's only part of the CLR integration story for the next version of SQL Server. One important part of the story that I left out is the in-process managed provider, an ADO.NET provider that CLR functions can use to talk directly to the instance of SQL Server that invoked them. In this article, I'll show you some of the basics of using this plumhing.
A word of caution before I begin, though: Microsoft is giving us a public look at Yukon at a very early stage in its development. Though it's been demonstrated at the PDC and copies are in the hands of many beta testers, this is far from final code. Likely many details will change on the way to the final product, including namespaces, attribute names, and so on. But even though I would be astounded if the code from this article were to run with the release version of the product, it seems likely that the general patterns of working with SQL Server and the CLR will remain intact. It's time to start thinking about what you could do with this, not time to start writing code for production.
The article continues at