Schema-based Xml Query Rewriting And Indexing Technology Research

XML (Extensible Markup Language) is a markup language introduced by W3C (the World Wide Web Consortium) in February 1998. Because of its unique technique superiority, it has been widely accepted as a new standard for representing and exchanging data on the Web. To construct various applications on XML, querying and getting data from XML data source accurately is an essential step. Based on what mentioned above, the dissertation does some theoretical and experimental research on abstracting, querying and indexing XML data.Firstly, this paper analyzes the central components for constructing a query system, proposes a query framework and discusses the general process from document parsing to query.Secondly, based on the tree model of DTD and XPath, the paper studies the optimization of the XPath query expression itself. According to the tree model of DTD, the paper puts forwards three rewriting methods.Thirdly, the paper proposes a structural index called DOBI. By extracting DTD and storing OB structure, DOBI can well resolve the most basic connection question between the ancestor and offspring on structural query. It can efficiently realize all kinds of query. After analyzing the efficiency of query, the paper proves DOBI can quickly establish the relations among elements, reduce the route visiting times, save I/O resources, realize the joins of XML documents effectively as well improve the efficiency of query.Fourthly, According to the XML XUpdate language proposed by Tatarinov etc, the paper proposes an updated algorithm based on DOBI index which can keep the validity of documents and make the high-efficient data upgrade possibly.Finally, the paper summarizes the whole research work and plans the expectation of the future work.