Programming Visual Basic .NET

Typed DataSets

There is nothing syntactically wrong with this line of code:

Dim dt As System.Data.DataTable = ds.Tables("Custumers")

However, “Custumers” is misspelled. If it were the name of a variable, property, or method, it would cause a compile-time error (assuming the declaration were not similarly misspelled). However, because the compiler has no way of knowing that the DataSet ds will not hold a table called Custumers, this typographical error will go unnoticed until runtime. If this code path is not common, the error may go unnoticed for a long time, perhaps until after the software is delivered and running on thousands of client machines. It would be better to catch such errors at compile time.

Microsoft has provided a tool for creating customized DataSet-derived classes. Such classes expose additional properties based on the specific schema of the data that an object of this class is expected to hold. Data access is done through these additional properties rather than through the generic Item properties. Because the additional properties are declared and typed, the Visual Basic .NET compiler can perform compile-time checking to ensure that they are used correctly. Because the class is derived from the DataSet class, an object of this class can do everything that a regular DataSet object can do, and it can be used in any context in which a DataSet object is expected.

Consider again Example 8-1, shown earlier in this chapter. This fragment of code displays the names of ...