INFO: Roadmap for Programming XML with the Pull-Model Parser in the .NET Framework (313816)

SUMMARY

This article provides a roadmap for programming XML with the pull-model parser of the .NET Framework. To help with learning a Microsoft product or technology, roadmap articles provide links to useful information, including online documentation, Microsoft Knowledge Base articles, and white papers.

Overview

The .NET Framework provides two ways to parse XML data:

• Pull-model parser (XmlReader and related classes)
• DOM-model parser (XmlDocument and related classes)

The pull model provides forward-only, read-only, noncached access to XML data. The DOM model provides in-memory, random, read/write access to XML data. This article covers the pull-model parser of the .NET Framework. This implementation is compliant with the World Wide Web Consortium (W3C) XML 1.0 and the Namespaces recommendations.

The pull model is implemented with classes based on the XmlReader class. XmlReader provides read access to XML data, and also enables you to skip data that you do not want. Note that the pull model has several advantages over the Simple API for XML (SAX) model, as explained at the following MSDN Web site: SAX is not implemented in .NET Framework, although it can be built by using the XmlReader class.

XML in the .NET Framework is implemented under the System.Xml namespace. The following classes are related with the pull model:

• The XmlReader class is the abstract base class from which all other reader classes are derived. You can also implement your own reader class with XmlReader as the base.
• The XmlTextReader class represents a forward-only, noncached reader to access XML data. It lets you read data from different resources such as files, URLs, and streams.
• The XmlNodeReader class represents a parser over an XmlNode of a DOM document; this is typically used to read a specific section of a DOM tree.
• The XmlValidatingReader class represents a validating reader. It lets you validate XML data against DTD, XDR, and XSD schemas. Note that this is the only class that provides the validation feature.
• The XmlResolver class is used to resolve external resources, such as entities and schemas named by a URI.

System.Xml also provides several other classes, such as XmlNameTable, XmlDeclaration, XmlSpace, and XmlWhitespace, that are compliant with World Wide Web Consortiuem (W3C) specifications. For more information, see the SDK documentation.

Key Concepts and QuickStarts

The best place to start is with the following SDK documention link: Follow the links in the "See Also" section for samples of reading elements and attributes, validating against schemas, and skipping content that you do not want.

QuickStart tutorials are a quick way to understand what the .NET Framework technology offers leading-edge developers. Following are links to samples that are relevant to XmlReader class: back to the top

Knowledge Base Articles

Microsoft Knowledge Base How To articles provide step-by-step instructions for accomplishing specific tasks.

Typical steps in pull-model parsing include instantiating one of the XmlReader-based classes, based on the application's requirements, and reading the data after setting up the required properties.

For example, if performance is the key factor and no validation is required, XmlTextReader is a good choice. With the use of MoveToContent and Skip methods, it enables reading only the required data with improved performance. The following Knowledge Base articles provide code samples for programming XmlReader-related classes. You can also run XSL transformations (XSLT), query and modify XML data by using DOM, and serialize or deserialize XML data by using the System.Xml classes in .NET Framework. For more information, see the following Knowledge Base article:

Guidelines

Your choice of parser may depend on your application's requirements. Note that the DOM-model parser loads the whole XML document into memory whereas the pull model loads one node at a time. The pull model consumes less memory but does not provide random access to data.

The DOM model is suited for applications that require random, read/write access to the data where memory consumption is not a factor. The pull model is suited for applications that require speed and memory conservation. Under many circumstances, the required solution may be a hybrid of these two models. For example, if part of a very large XML document needs to be manipulated, it may be efficient to use the pull model to read it, and then construct a DOM with only the data needed for additional modification.

Troubleshooting

If you experience problems and want assistance, the best place to start is at the Microsoft Product Support Services (PSS) Web site: You have several options to find answers to your questions. You can search the Microsoft Knowledge Base, post your questions or share your experiences on Microsoft newsgroups, obtain the latest service packs and news about a specific product, or contact Microsoft Product Support.