The Research And Realization Of XML Data Storing Method Based On Relation Database

XML becomes the main form of data transimission in the current network because it can be extended and self-descriptive. However, its capability of data processing is very limited. After the realization of XML data storing to the relation database, we can use the database storage management technology in the relation database to handle XML data. Therefore, integrating XML into relation database storage method is very valuable.This thesis is based on acknowledge of XML and relation database, and researches depply the storing method of XML data in relation database. But there are still some problems, the storage method which is based on DTD can not adapt to the trend that XML documents are defined by XML Schema, and on the same time, the search method and the mode conversion rule of Bohannon storage method are not the best. This thesis proposes a full-scale mapping method which is based on the cost of search towards the former nature of XML DTD and the subsequent nature of XML Schema. Firstly, it adds the deep handling before the seeking of initial mode, and then it gets the initial mode through DOM tree analysis tool. It does the different treatment considering the difference between complex nodes and simple nodes when having the mode space. After that, this thesis imports the cost model which is based on Hash linking algorithm and finds the best mode from the mode space by improved Mountain-Climb Algorithm. Thus it improves the performance of the mapping method.At last, this thesis verifies the improvement of the mapping method through experiments. The results show that the improved mapping method has better performance compared with the classical method of Bohannon mapping.