XML schema evolution management using XML-based transformation rules

Abstract

XML is rapidly emerging as the most widely adopted technology for information representation and exchange in Web-based information systems. Main features of XML are related to the use of tags for defining nested document structures and to the introduction of document types called DTD in describing the structure of documents. Due to the lack of definition features and constraints with DTD, XML Schema has emerged as a powerful type building mechanism. Similar to a schema evolution in conventional databases, XML schema is subject to change in document structure which calls for an efficient schema evolution management strategies. Different from the evolution management in conventional databases, evolutions in XML schema are characterized by its hierarchical nature of document and various type changes in the hierarchy. The size of document data to be shared by a large number of users across the Internet can hardly justify the version control of schema evolution management widely adopted in conventional databases. For this and more, an efficient strategy is required to solve the XML-specific problems. In this thesis, we propose a strategy for XML schema evolution management based on schema mutation in that an up-to-date schema is maintained and documents are transformed according to the new schema. For this we have defined a set of XML-schema evolution rules each of which implements an automatic document transformation action. The document transformation rules are defined based on a set of well-classified XML schema type changes. The rule actions convert the old XML documents into the new documents according to the change in XML schema format.