We present an algorithm to solve XPath decision problems under regular tree type constraints and show its use to statically type-check XPath queries. To this end, we prove the decidability of a logic with converse for finite ordered trees whose time complexity is a simple exponential of the size of a formula. The logic corresponds to the alternation free modal µ-calculus without greatest fixpoint, restricted to finite trees, and where formulas are cycle-free. Our proof method is based on two auxiliary results. First, XML regular tree types and XPath expressions have a linear translation to cycle-free formulas. Second, the least and greatest fixpoints are equivalent for finite trees, hence the logic is closed under negation. Building on these results, we describe a practical, effective system for solving the satisfiability of a formula. The system has been experimented with some decision problems such as XPath emptiness, containment, overlap, and coverage, with or without type constraints. The benefit of the approach is that our system can be effectively used in static analyzers for programming languages

We show that problems arising in static analysis of XML specifications and transformations can be dealt with using techniques similar to those developed for static analysis of programs. Many properties of interest in the XML context are related to navigation, and can be formulated in temporal logics for trees. We choose a logic that admits a simple single-exponential translation into unranked tree automata, in the spirit of the classical LTL-to-Büchi automata translation. Automata arising from this translation have a number of additional properties; in particular, they are convenient for reasoning about unary node-selecting queries, which are important in the XML context. We give two applications of such reasoning: one deals with a classical XML problem of reasoning about navigation in the presence of schemas, and the other relates to verifying security properties of XML views.

We present a tool for helping XML schema designers to obtain a high quality level for their specifications. The tool allows one to analyze relations between classes of XML documents and formally prove them. For instance, the tool can be used to check forward and backward compatibilities of recommendations. When such a relation does not hold, the tool allows one to identify the reasons and reports detailed counter-examples that exemplify the problem. For this purpose, the tool relies on recent advances in logic-based automated theorem proving techniques that allow for efficient reasoning on very large sets of XML documents. We believe this tool can be of great value for standardization bodies that define specifications using various XML type definition languages (such as W3C specifications), and are concerned with quality assurance for their normative recommendations.

Nous présentons une nouvelle perspective sur le h-index, ainsi qu’une contribution potentielle à celui-ci. 1.

XML is used for different purposes. We are interested in data-centric applications of XML where it is used to handle structured data in loosely coupled, distributed systems. In many such scenarios, it is important that the XML data complies to structural and integrity constraints, in particular to value-based constraints. The constraints should remain invariant under operations manipulating the data. In this paper, we present a technique to maintain the invariants. For a core procedural manipulation language, we show how to automatically derive the weakest precondition of procedures for the constraints. We present a formalization of a data-centric XML abstraction and an assertion language that enables weakest precondition generation. Our framework allows to automatically simplify and reduce the generated preconditions such that checking becomes independent of the size of the constraint definitions. This is achieved by isolating the aspects that are manipulated by the considered procedure from the aspects that remain unchanged. As the overall invariant can be assumed for the input data, only the isolated aspects have to be checked in prestates of procedures. 1.

Linear Temporal Logic (LTL) is one of the main tools used extensively in the field of verification, where many efficient model-checking algorithms have been developed over the years. Recently a tree analog of LTL has been proposed, where a modification of the LTL-to-Automata translation was introduced. This made it possible to use automata techniques for reasoning about properties of XML queries and specifications such as containment of XPath expressions, optimisation of XPath expressions under schema constraints, and security properties of XML views. However, as the size of the constructed automata grows, there is a need for optimisation techniques to reduce the size of constructed automata. The goal of this report is to construct an optimised automaton and to implement the translation from XPath to automata and use it on XML reasoning tasks. i Acknowledgements I would like to thank my supervisor, Professor. Leonid Libkin for introducing me to the world of automata and their connection with database systems. I am also grateful to Tony Tan for his support and guidance throughout the project and for his patience with me!

Abstract: We present an algorithm to solve XPath decision problems under regular tree type constraints and show its use to statically type-check XPath queries. To this end, we prove the decidability of a logic with converse for finite ordered trees whose time complexity is a simple exponential of the size of a formula. The logic corresponds to the alternation free modal µ-calculus without greatest fixpoint, restricted to finite trees, and where formulas are cycle-free. Our proof method is based on two auxiliary results. First, XML regular tree types and XPath expressions have a linear translation to cycle-free formulas. Second, the least and greatest fixpoints are equivalent for finite trees, hence the logic is closed under negation. Building on these results, we describe a practical, effective system for solving the satisfiability of a formula. The system has been experimented with some decision problems such as XPath emptiness, containment, overlap, and coverage, with or without type constraints. The benefit of the approach is that our system can be effectively used in static analyzers for programming languages