Partager via

DiffGrams

  • Article

A DiffGram is an XML format that is used to identify current and original versions of data elements. The DataSet uses the DiffGram format to load and persist its contents, and to serialize its contents for transport across a network connection. When a DataSet is written as a DiffGram, it populates the DiffGram with all the necessary information to accurately recreate the contents, though not the schema, of the DataSet, including column values from both the Original and Current row versions, row error information, and row order.

The DiffGram format that is used by the .NET Framework can also be used by other platforms to send and receive information to a .NET Framework application.

When sending and retrieving a DataSet from an XML Web service, the DiffGram format is implicitly used. Additionally, when loading the contents of a DataSet from XML using the ReadXml method, or when writing the contents of a DataSet in XML using the WriteXml method, you can select that the contents be read or written as a DiffGram. For more information, see Loading a DataSet from XML and Writing a DataSet as XML Data.

While the DiffGram format is primarily used by the .NET Framework as a serialization format for the contents of a DataSet, you can also use DiffGrams to modify data in tables in a Microsoft SQL Server™ 2000 database. For more information, see the XML for SQL Server 2000 Web Release 2 (WR2) located at https://msdn.microsoft.com.

DiffGram Format

The DiffGram format is divided into three sections: the current data, the original (or "before") data, and an errors section, as shown in the following example.

<?xml version="1.0"?>
<diffgr:diffgram 
         xmlns:msdata="urn:schemas-microsoft-com:xml-msdata"
         xmlns:diffgr="urn:schemas-microsoft-com:xml-diffgram-v1"
         xmlns:xsd="http://www.w3.org/2001/XMLSchema">

   <DataInstance>
   </DataInstance>

  <diffgr:before>
  </diffgr:before>

  <diffgr:errors>
  </diffgr:errors>
</diffgr:diffgram>

The DiffGram format consists of the following blocks of data:

  • <DataInstance>
    The name of this element, DataInstance, is used for explanation purposes in this documentation. A DataInstance element represents a DataSet or a row of a DataTable. Instead of DataInstance, the element would contain the name of the DataSet or DataTable. This block of the DiffGram format contains the current data, whether it has been modified or not. An element, or row, that has been modified is identified with the diffgr:hasChanges annotation.
  • <diffgr:before>
    This block of the DiffGram format contains the original version of a row. Elements in this block are matched to elements in the DataInstance block using the diffgr:id annotation.
  • <diffgr:errors>
    This block of the DiffGram format contains error information for a particular row in the DataInstance block. Elements in this block are matched to elements in the DataInstance block using the diffgr:id annotation.

DiffGram Annotations

DiffGrams use several annotations to relate elements from the different DiffGram blocks that represent different row versions or error information in the DataSet.

The following table describes the DiffGram annotations that are defined in the DiffGram namespace urn:schemas-microsoft-com:xml-diffgram-v1.

Annotation Description
id Used to pair the elements in the <diffgr:before> and <diffgr:errors> blocks to elements in the <DataInstance> block. Values with the diffgr:id annotation are in the form [TableName][RowIdentifier]. For example: <Customers diffgr:id="Customers1">.
parentId Identifies which element from the <DataInstance> block is the parent element of the current element. Values with the diffgr:parentId annotation are in the form [TableName][RowIdentifier]. For example: <Orders diffgr:parentId="Customers1">.
hasChanges Identifies a row in the <DataInstance> block as modified. The hasChanges annotation can have one of the following three values:
inserted
Identifies an Added row.
modified
Identifies a Modified row that contains an Original row version in the <diffgr:before> block. Note that Deleted rows will have an Original row version in the <diffgr:before> block, but there will be no annotated element in the <DataInstance> block.
descent
Identifies an element where one or more children from a parent-child relationship have been modified.
hasErrors Identifies a row in the <DataInstance> block with a RowError. The error element is placed in the <diffgr:errors> block.
Error Contains the text of the RowError for a particular element in the <diffgr:errors> block.

The DataSet includes additional annotations when reading or writing its contents as a DiffGram. The following table describes these additional annotations, which are defined in the namespace urn:schemas-microsoft-com:xml-msdata.

Annotation Description
RowOrder Preserves the row order of the original data and identifies the index of a row in a particular DataTable.
Hidden Identifies a column as having a ColumnMapping property set to MappingType.Hidden. The attribute is written in the format msdata:hidden[ColumnName]="value". For example: <Customers diffgr:id="Customers1" msdata:hiddenContactTitle="Owner">.

Note that hidden columns are only written as a DiffGram attribute if they contain data. Otherwise, they are ignored.

Sample DiffGram

An example of the DiffGram format is shown below. This example shows the result of an update to a row in a table before the changes have been committed. The row with a CustomerID of "ALFKI" has been modified, but not updated. As a result, there is a Current row with a diffgr:id of "Customers1" in the <DataInstance> block, and an Original row with a diffgr:id of "Customers1" in the <diffgr:before> block. The row with a CustomerID of "ANATR" includes a RowError, so it is annotated with diffgr:hasErrors="true" and there is a related element in the <diffgr:errors> block.

<diffgr:diffgram xmlns:msdata="urn:schemas-microsoft-com:xml-msdata" xmlns:diffgr="urn:schemas-microsoft-com:xml-diffgram-v1">
  <CustomerDataSet>
    <Customers diffgr:id="Customers1" msdata:rowOrder="0" diffgr:hasChanges="modified">
      <CustomerID>ALFKI</CustomerID>
      <CompanyName>New Company</CompanyName>
    </Customers>
    <Customers diffgr:id="Customers2" msdata:rowOrder="1" diffgram:hasErrors="true">
      <CustomerID>ANATR</CustomerID>
      <CompanyName>Ana Trujillo Emparedados y helados</CompanyName>
    </Customers>
    <Customers diffgr:id="Customers3" msdata:rowOrder="2">
      <CustomerID>ANTON</CustomerID>
      <CompanyName>Antonio Moreno Taquera</CompanyName>
    </Customers>
    <Customers diffgr:id="Customers4" msdata:rowOrder="3">
      <CustomerID>AROUT</CustomerID>
      <CompanyName>Around the Horn</CompanyName>
    </Customers>
  </CustomerDataSet>
  <diffgr:before>
    <Customers diffgr:id="Customers1" msdata:rowOrder="0">
      <CustomerID>ALFKI</CustomerID>
      <CompanyName>Alfreds Futterkiste</CompanyName>
    </Customers>
  </diffgr:before>
  <diffgr:errors>
    <Customers diffgr:id="Customers2" diffgr:Error="An optimistic concurrency violation has occurred for this row."/>
  </diffgr:errors>
</diffgr:diffgram>

See Also

XML and the DataSet | Loading a DataSet from XML | Writing a DataSet as XML Data | Creating and Using DataSets