Study Of XQuery Processing Based-on Relational Database

Along with the increasing development of Internet-based business application, XML (eXtensible Markup Language) has emerged as the standard format for data expression and exchange, which leads to requirements involving XML data management. Relational Database System has been almost perfect in the domain of techniques, and it's dominant in business data management area. It's been pay wide attention and regards to how to realize XML data management taking advantage of Relational Database. The keys for wide applications of XML lie in query and searching in XML data sources. It's very important to Web data management. Within a large number of XML query languages, XQuery is one part of the standard drafts of W3C(World Wide Web Consortium). And it has been used in applications widely. It is becoming the SQL for XML. Accordingly, XQuery processing based on Relational Database is emerging one of the main contents of XML data management. It makes significant effect on the applications of XML. The content of XML data management includes storing XML documents into Relational Database using some of mapping approaches, and query processing of user's XQuery on primary XML documents. Different form SQL, XQuery aims at XML documents, but the data which XQuery aim at are stored in relational tables in database according to various mapping approaches. Thus makes an serious obstacle to query processing.In this thesis, the problem is discussed mainly from two aspects, which is processing for query language transformation of XQuery to SQL, and processing for adaptation to multiple mapping approaches. The processing course and realizing methods have been analyzed. Many current representative solutions have been cmpared. The feasible route to resolve the problem and to attain the requirements of application is got as the result.Based on above, an XQuery processing method is designed, which adopts XQuery processing solution based on virtual XML view. The query processing rules are induced after analyzing the bottom processing mechanism of realizing query language transformation function. What's more, an auto creating mechanism of virtual XML view which is build on bottom processing mechanism is proposed to solve the adaptability problem of multiple mapping approaches. There are two kinds of mapping approaches, for structure-mapping and model-mapping. As for the adaptability of structure-mapping approach, virtual XML view is constructed by virtue of XML structure tree while is realized with twelve parameters for the adaptability of model-mapping approach. And two algorithms are designed to complete the above works. The method put forward in this thesis aims to good adaptability and extensibility, and provides convenience to users' query on XML document that is stored in relational database using XQuery. At the end of this thesis, experiments of constructing mechanism of virtual XML view are done to support the proposed method.