This paper proposes a two-step approach to define rules for maintaining materialized XML views specified over relational databases. The first step concentrates on identifying all paths of the base schema that are relevant to a path of the view, with respect to an update. Hence, the corresponding views paths can be affected by the update. The second step creates rules that maintain all view paths that can be affected by the update. The paper then discusses how to automatically identify all paths in the base schema that are relevant to a view path with respect to a given update operation and how to create the appropriate maintenance rules.

This work addresses the problem of incremental maintenance of XML views defined on top of relational data. In order to incrementally maintain a XML view, event-condition-rules should be specified for the relational source. Such rules are responsible for correctly modifying the XML view content in order to reflect changes made to the base source.This work proposes an approach where incremental view maintenance rules are derived from view correspondence assertions, which specify relationships between the view schema and the base source schema. The adoption of correspondence assertions also facilitates the task of formally proving that the derived rules correctly maintain the view.

In this work we study the problem of how to incrementally maintain materialized XML views of relational data, based on the semantic mappings that model the relationship between the source and view schemas. The semantic mappings are specified by a set of correspondence assertions, which are simple to understand. The paper focuses on an algorithm to incrementally maintain materialized XML views of relational data.

No abstract