Recipe 2.18 Controlling the Names Used in a Strongly Typed DataSet

Problem

You want to assign your own names to the classes and properties for strongly typed DataSet classes.

Solution

Use annotations in the XML schema to control the names of classes and properties in strongly typed DataSet classes.

The sample uses one XSD file:

CategoriesDS_AnnotatedName.xsd

The schema used to generate the strongly typed DataSet. The schema is annotated so that you can access the collection of rows in the table by using the Categorys property of the DataSet rather than categories, each row by using the Category property of the row collection rather than CategoriesRow, and the CategoryName field by using the Name property of the row rather than CategoryName. The annotations are marked in bold in the Example 2-23.

Example 2-23. File: TypedDataSets\CategoriesDS_AnnotatedName.xsd

<?xml version="1.0" standalone="yes" ?>
<xs:schema id="CategoriesDS_AnnotatedName"
    targetNamespace=
    "http://www.tempuri.org/CategoriesDS_AnnotatedName.xsd"
    xmlns:mstns="http://www.tempuri.org/CategoriesDS_AnnotatedName.xsd"
    xmlns="http://www.tempuri.org/CategoriesDS_AnnotatedName.xsd"
    xmlns:xs="http://www.w3.org/2001/XMLSchema"
    xmlns:msdata="urn:schemas-microsoft-com:xml-msdata"
    xmlns:codegen="urn:schemas-microsoft-com:xml-msprop"
    attributeFormDefault="qualified" elementFormDefault="qualified">
    <xs:element name="CategoriesDS_AnnotatedName" msdata:IsDataSet="true">
        <xs:complexType>
            <xs:choice maxOccurs="unbounded">
                <xs:element name="Categories"
                    codegen:typedName="Category"
                    codegen:typedPlural="Categorys">
                    <xs:complexType>
                        <xs:sequence>
                            <xs:element
                            name="CategoryID"
                            msdata:ReadOnly="true"
                            msdata:AutoIncrement="true"
                            type="xs:int" />
                            <xs:element
                            name="CategoryName"
                            type="xs:string"
                            codegen:typedName="Name" />
                            <xs:element
                            name="Description"
                            type="xs:string"
                            minOccurs="0" />
                        </xs:sequence>
                    </xs:complexType>
                </xs:element>
            </xs:choice>
        </xs:complexType>
        <xs:unique name="Constraint1" msdata:PrimaryKey="true">
            <xs:selector xpath=".//mstns:Categories" />
            <xs:field xpath="mstns:CategoryID" />
        </xs:unique>
    </xs:element>
</xs:schema>

The sample code creates a strongly typed DataSet based on the Categories table in Northwind. The user specifies whether the one based on the default or annotated schema file is used. In either case, data is loaded into the DataSet and the collections of rows and columns in the DataSet are iterated over to display the data and to demonstrate the effect of the schema annotations.

The C# code is shown in Example 2-24.

Example 2-24. File: TypedDataSetNamesForm.cs

// Namespaces, variables, and constants
using System;
using System.Configuration;
using System.Text;
using System.Data;
using System.Data.SqlClient;

// Table name constants
private const String CATEGORIES_TABLE = "Categories";

//  . . . 

StringBuilder result = new StringBuilder( );

// Create the DataAdapter.
SqlDataAdapter da = new SqlDataAdapter("SELECT * FROM Categories",
    ConfigurationSettings.AppSettings["Sql_ConnectString"]);

if (annotatedRadioButton.Checked)
{
    // Create the typed DataSet with name annotations.
    CategoriesDS_AnnotatedName ds = new CategoriesDS_AnnotatedName( );
    // Fill the Categories table within DataSet.
    da.Fill(ds, CATEGORIES_TABLE);

    result.Append("Annotated Names" + Environment.NewLine +
        Environment.NewLine);
    // Iterate over the rows collection and display columns.
    // Note that the row collection is Categorys
    // and that each row is Category.
    foreach(CategoriesDS_AnnotatedName.Category row in ds.Categorys)
    {
        // Note that the CategoryName field is referred to simply as Name.
        result.Append(row.CategoryID + "\t" + row.Name + "\t" +
            row.Description + Environment.NewLine);
    }
}
else
{
    // Create the typed DataSet without name annotations.
    CategoriesDS ds = new CategoriesDS( );
    da.Fill(ds, CATEGORIES_TABLE);

    result.Append("Default" + Environment.NewLine + Environment.NewLine);
    // Iterate over the rows collection and display columns.
    foreach(CategoriesDS.CategoriesRow row in ds.Categories)
    {
        result.Append(row.CategoryID + "\t" + row.CategoryName + "\t" +
            row.Description + Environment.NewLine);
    }
}

resultTextBox.Text = result.ToString( );

Discussion

Annotations are modifications to the XSD schema used to generate a strongly typed DataSet that allows the names of elements in the strongly typed DataSet to be customized without changing the underlying schema. This allows more meaningful element names to be used resulting in code that is easier to read, use, and maintain. Table 2-16 lists available annotations.

Table 2-16. Available annotations

Annotation	Description
typedChildren	Name of the method that returns objects from a child data relation.
typedName	Name of the object.
typedParent	Name of the method that returns an object from a parent data relation.
typedPlural	Name of the collection of objects.
nullValue	Value or behavior if the underlying value is DBNull. Table 2-17 lists possible values for this annotation. The default value is _throw.

Table 2-17 describes possible values for the nullValue annotation.

Table 2-17. Values for nullValue annotation

nullValue	Description
Replacement Value	A value having the same type as the element to be returned
_empty	Return String.Empty for a StringReturn an object created from an empty constructor for other objectsThrow an exception for primitive types
_null	Return a null reference for objectsThrow an exception for primitive types
_throw	Raise an exception

Table 2-18 lists the different objects in a strongly typed DataSet and the default names and available annotations for each.

Table 2-18. Default values and available annotations for elements of strongly typed DataSet objects

Element	Default name	Annotation
DataTable	TableNameDataTable	typedPlural
DataTable methods	NewTableNameRowAddTableNameRowDeleteTableNameRow	typedName
DataRowCollection	TableName	typedPlural
DataRow	TableNameRow	typedName
DataColumn	DataTable.ColumnNameColumnDataRow.ColumnName	typedName
Property	PropertyName	typedName
Child accessor	GetChildTableNameRows	typedChildren
Parent accessor	TableNameRow	typedParent
DataSet events	TableNameRowChangeEventTableNameRowChangeEventHandler	typedName

The use annotations, a reference to the codegen namespace, must be included in the XSD schema, as shown:

xmlns:codegen="urn:schemas-microsoft-com:xml-msprop"

The codegen namespace allows the names of methods, properties, relations, constraints, and events in the strongly typed DataSet to be customized.