Summary

In this chapter, we looked primarily at ADO.NET data access as applied to .NET Web services. You saw how the various ADO.NET techniques fit in with Web services, and you saw tips on how to use them to expose data via Web methods.

You learned how to use DataReader objects for fast, read-only access, and how to use them to shape data into simpler forms, which can be advantageous for both .NET and non-.NET clients. Next we turned to DataSet objects, discussing where to use them or avoid them, as well as looking in a little more depth at how DiffGrams work and why you should use typed DataSet objects wherever possible.

In addition, you saw how ADO.NET can be extended such that legacy or custom data can be exposed via a Web service in a standardized way. This was illustrated with a custom ADO.NET provider that can be used to query file system data. This was a read-only implementation, but the basic principles behind custom data providers are clear.

We also had a brief tour through the topic of XML data. You saw how XML data sources can be queried directly and how ADO.NET DataSet objects can be treated (and queried) as XML. For SQL Server users, we also looked at a shortcut—SQLXML. And for hardcore programmers, we briefly discussed customized ways to treat other data sources as XML.