Active integrity constraints: From theory to implementation

Abstract

The problem of database consistency relative to a set of integrity constraints has been extensively studied since the 1980s, and is still recognized as one of the most important and complex in the field. In recent years, with the proliferation of knowledge repositories (not only databases) in practical applications, there has also been an effort to develop implementations of consistency maintenance algorithms that have a solid theoretical basis. The framework of active integrity constraints (AICs) is one example of such an effort, providing theoretical grounds for rule-based algorithms for ensuring database consistency. An AIC consists of an integrity constraint together with a specification of actions that may be taken to repair a database that does not satisfy it. Both denotational and operational semantics have been proposed for AICs. In this paper, we describe repAIrC, a prototype implementation of the algorithms previously proposed targetting SQL databases, i.e., the most prolific type of databases. Using repAIrC, we can both validate an SQL database with respect to a given set of AICs and compute possible repairs in case the database is inconsistent; the tool is able to work with the different kinds of repairs that have been considered, and achieves optimal asymptotic complexity in their computation. It also implements strategies for parallelizing the search for repairs, which in many cases can make untractable problems become easily solvable.