XML-Schemaevolution: Änderung eines XML-Schemas mit automatisierter Adaption assoziierter XML-Dokumente

The eXtensible Markup Language (XML) is a well-established and standardized format for exchanging and storing structured and semi-structured information. If the structure of an XML document complies with the standard of the W3C (World Wide Web Consortium), the XML document can be seen as well-defined. The XML Schema, abbreviated XSD (XML Schema Definition), is one schema language of the W3C for specifying requirements for the structure and content of XML documents. An XML document is called valid, if it fulfills all restrictions and conditions of an associated XML Schema. The modification or evolution of an XML Schema could have different reasons, for example the error correction, the clearance of obscurities, the further development of applications or in general the change of requirements for exchanged information. The resulting problem of modifying an XML Schema is that existing XML documents, which were valid against the former XML Schema, could consequently lose their validity and have to be adapted as well (co-evolution). The above mentioned validity problem and its solution are the main topics of this dissertation. The following approaches and mechanisms were developed and used for the solution of the validity problem. The conceptual model EMX (Entity Model for XML-Schema) is a simplified representation of an XML Schema. The resulting three-layer architecture consists of a model, a schema and a document layer. Since a unique mapping between EMX and XSD exists, modifications are applied on the conceptual model. Modifications are formally described by the domain-specific transformation language ELaX (Evolution Language for XML-Schema). The applied operations are logged. The ruled-based algorithm ROfEL (Rule-based Optimizer for ELaX) then reduces the number of logged ELaX operations by identifying and removing unnecessary, redundant and invalid operations. This reduction is an essential prerequisite for the following automatic derivation of transformation steps. The classification of operations is the basis of an impact analysis of schema modifications to the document layer. If an adaption is necessary, or it’s possible because of the ambiguous mapping of the schema and the document layer, XML document transformation steps are automatically derived. These steps are used for the adaption of the DOM (Document Object Model) of XML documents which are associated with the modified XML Schema. The research prototype CodeX (Conceptual design and evolution of XML schemas) serves as demonstrator, which can be used for the evaluation of the above mentioned approaches and mechanisms presented in this dissertation.