This paper investigates the view update problem for XML views published from relational data. We consider XML views defined in terms of mappings directed by possibly recursive DTDs compressed into DAGs and stored in r...This paper investigates the view update problem for XML views published from relational data. We consider XML views defined in terms of mappings directed by possibly recursive DTDs compressed into DAGs and stored in relations. We provide new techniques to efficiently support XML view updates specified in terms of XPath expressions with recursion and complex filters. The interaction between XPath recursion and DAG compression of XML views makes the analysis of the XML view update problem rather intriguing. Furthermore, many issues are still open even for relational view updates, and need to be explored. In response to these, on the XML side we revise the notion of side effects and update semantics based on the semantics of XML views, and present efficient algorithms to translate XML updates to relational view updates. On the relational side, we propose a mild condition on SPJ views, and show that under this condition the analysis of deletions on relational views becomes PTIME while the insertion analysis is NP-complete. We develop an efficient algorithm to process relational view deletions, and a heuristic algorithm to handle view insertions. Finally, we present an experimental study to verify the effectiveness of our techniques.展开更多
In this paper we introduce the view specification language and consider the problem of answering queries posed on semistructured data views. The two main approaches ,query rewriting and view materialization,are outlin...In this paper we introduce the view specification language and consider the problem of answering queries posed on semistructured data views. The two main approaches ,query rewriting and view materialization,are outlined.We also discuss the views and its maintenance for XML data. Based on the above discussion a view architecture for integrating data from heterogeneous source or from the WWW is proposed.展开更多
On how to storage XML metadata in digital library and make optimization according to the statistics gener ated by the Database engine,this paper introduces a feasible solution which involves view materialization and c...On how to storage XML metadata in digital library and make optimization according to the statistics gener ated by the Database engine,this paper introduces a feasible solution which involves view materialization and colonygenetics algorism.展开更多
基金Wenfei Fan is supported in part by EPSRC under Grants No.GR/S63205/01,No.GR/T27433/01,and No.EP/E029213/1.
文摘This paper investigates the view update problem for XML views published from relational data. We consider XML views defined in terms of mappings directed by possibly recursive DTDs compressed into DAGs and stored in relations. We provide new techniques to efficiently support XML view updates specified in terms of XPath expressions with recursion and complex filters. The interaction between XPath recursion and DAG compression of XML views makes the analysis of the XML view update problem rather intriguing. Furthermore, many issues are still open even for relational view updates, and need to be explored. In response to these, on the XML side we revise the notion of side effects and update semantics based on the semantics of XML views, and present efficient algorithms to translate XML updates to relational view updates. On the relational side, we propose a mild condition on SPJ views, and show that under this condition the analysis of deletions on relational views becomes PTIME while the insertion analysis is NP-complete. We develop an efficient algorithm to process relational view deletions, and a heuristic algorithm to handle view insertions. Finally, we present an experimental study to verify the effectiveness of our techniques.
文摘In this paper we introduce the view specification language and consider the problem of answering queries posed on semistructured data views. The two main approaches ,query rewriting and view materialization,are outlined.We also discuss the views and its maintenance for XML data. Based on the above discussion a view architecture for integrating data from heterogeneous source or from the WWW is proposed.
文摘On how to storage XML metadata in digital library and make optimization according to the statistics gener ated by the Database engine,this paper introduces a feasible solution which involves view materialization and colonygenetics algorism.
