The XQuery 1.0 and XPath 2.0 Full-text (XQFT) language has been developed by the W3C to extend XQuery and XPath with full-text search capabilities. XQFT allows users to specify a mix of structured and complex full-text predicates, and also allows users to score/rank such queries. The power and flexibility of XQFT gives rise to two interesting questions. First, is it possible to efficiently integrate a full-function XML query language with sophisticated full-text search? Second, is it possible to score and rank arbitrary XQuery and XQFT queries? In this demonstration, we present evidence that it is indeed possible to achieve the above goals. We demonstrate the Quark open-source data management system and show how we can seamlessly and efficiently integrate structured and unstructured search over XML data. In particular, we demonstrate (a) techniques for efficiently evaluating keyword search over virtual XML views, and (b) a framework for scoring both structured and full-text predicates.

Current systems that publish relational data as nested (XML) views are passive in the sense that they can only respond to user-initiated queries over the nested views. In this article, we propose an active system whereby users can place triggers on (unmaterialized) nested views of relational data. In this architecture, we present scalable and efficient techniques for processing triggers over nested views by leveraging existing support for SQL triggers over flat relations in commercial relational databases. We have implemented our proposed techniques in the context of the Quark XML middleware system. Our performance results indicate that our proposed techniques are a feasible approach to supporting triggers over nested views of relational data.

XML has emerged as a dominant standard for information exchange on the Internet. However, a large fraction of data continues to be stored in relational databases. At a high level, there are two approaches to supporting triggers over XML views. The first is to materialize the entire view and store it in an XML database with support for XML triggers. However, this approach suffers from the overhead of replicating and incrementally maintaining the materialized XML on every relational update affecting the view, even though users may only be interested in relatively rare events. In this paper, we propose the alternative approach of translating XML triggers into SQL triggers. There are some challenges involved in this approach, however, because triggers can be specified over complex XML views with nested predicates, while SQL triggers can only be specified over flat tables. Consequently, even identifying the parts of an XML view that could have changed due to a (possibly deeply nested) SQL update is a non-trivial task, as is the problem of computing the old and new values of an updated fragment of the view. We address the above challenges and propose a system architecture and an algorithm for supporting triggers over XML views of relational data. We implement and evaluate our system; the performance results indicate our techniques are a feasible approach to supporting triggers over XML views of relational data.

We consider the problem of efficiently producing ranked results for keyword search queries over hyperlinked XML documents. Evaluating keyword search queries over hierarchical XML documents, as opposed to (conceptually) flat HTML documents, introduces many new challenges. First, XML keyword search queries do not always return entire documents, but can return deeply nested XML elements that contain the desired keywords. Second, the nested structure of XML implies that the notion of ranking is no longer at the granularity of a document, but at the granularity of an XML element. Finally, the notion of keyword proximity is more complex in the hierarchical XML data model. In this paper, we present the XRANK system that is designed to handle these novel features of XML keyword search. Our experimental results show that XRANK offers both space and performance benefits when compared with existing approaches. An interesting feature of XRANK is that it naturally generalizes a hyperlink based HTML search engine such as Google. XRANK can thus be used to query a mix of HTML and XML documents.