DataSet Object

The last class we discuss in the context of the XmlSerializer is the DataSet. We have learned all about how to store and access XML data in a DataSet object in chapter 8. Now it’s time to learn how the XmlSerializer serializes and deserializes DataSet objects, because it’s different from anything else we have learned so far. Fasten your seat-belt, here we go.

The .NET architects wanted the DataSet class to become the preferred vehicle to transmit data through WebServices. Since efficiency is an important aspect in the transmission of data they designed a special (XML based) format to transfer DataSets data, which only transmits the changes since a DataSet was loaded. The format is called a DiffGram. Since ASP.NET WebServices create and parse SOAP messages with the XmlSerializer, the serializer was now on the hook to serialize DataSet object to the DiffGram format instead of simply serializing the data exposed by the public properties. Furthermore, it needed to support the reverse operation and deserialize DataSets from DiffGrams as well.

The following example serializes a DataSet loaded from a simple XML document. Note that we do not set up anything different from the way we serialized objects throughout the past two chapters to enable serialization to the DiffGram format.

18        Listing 10.14 Serializing a DataSet containing simple contacts file

<?xml version="1.0" encoding="utf-8" ?>

<contacts>

    <contact>

        <name>Roger Dolph</name>

        <address>100 Washington Ave, Atlanta, GA</address>

        <phone>123-4568</phone>

   </contact>

</contacts>

static void SerializeContactDataSet()

{

  DataSet   dataset = new DataSet();

  dataset.ReadXml("contacts.xml");

  XmlSerializer ser = new XmlSerializer( typeof( DataSet ) );

  XmlTextWriter writer = new XmlTextWriter( "dataset.xml",

    System.Text.Encoding.UTF8 );

  writer.Formatting = Formatting.Indented;

  ser.Serialize( writer, dataset );

  writer.Close();

}

When we examine the output of the XmlSerializer we find two distinct sections. One is the DiffGram we expected to find. The other one is an XML schema describing the data structure and the relationships inside the DataSet in addition to the data in the various tables. The schema information is necessary to accurately deserialize the DataSet later on, since the DiffGram does not include relationship information. Listing 10.15 shows the complete output of the method above.

Listing 10.15 A serialized DataSet 

<?xml version="1.0" encoding="utf-8"?>

<DataSet>

  <xs:schema id="contacts" xmlns="" xmlns:xs="http://www.w3.org/2001/XMLSchema"

xmlns:msdata="urn:schemas-microsoft-com:xml-msdata">

    <xs:element name="contacts" msdata:IsDataSet="true">

      <xs:complexType>

        <xs:choice maxOccurs="unbounded">

          <xs:element name="contact">

            <xs:complexType>

              <xs:sequence>

                <xs:element name="name" type="xs:string" minOccurs="0" />

                <xs:element name="address" type="xs:string" minOccurs="0" />

                <xs:element name="phone" type="xs:string" minOccurs="0" />

              </xs:sequence>

            </xs:complexType>

          </xs:element>

        </xs:choice>

      </xs:complexType>

    </xs:element>

  </xs:schema>

  <diffgr:diffgram xmlns:msdata="urn:schemas-microsoft-com:xml-msdata" xmlns:diffgr="urn:schemas-microsoft-com:xml-diffgram-v1">

    <contacts>

      <contact diffgr:id="contact1" msdata:rowOrder="0" diffgr:hasChanges="inserted">

        <name>Roger Dolph</name>

        <address>100 Washington Ave, Atlanta, GA</address>

        <phone>123-4568</phone>

      </contact>

    </contacts>

  </diffgr:diffgram>

</DataSet>