An axiomatic approach to deciding query safety in deductive databases

  • Authors:

  • Michael Kifer;Raghu Ramakrishnan;Avi Silberschatz

  • Affiliations:

  • Department of Computer Science, SUNY at Stony Brook, Stony Brook, NY;Department of Computer Science, University of Wisconsin-Madison, Madison, WI;Department of Computer Science, University of Texas at Austin, Austin, TX

  • Venue:
  • Proceedings of the seventh ACM SIGACT-SIGMOD-SIGART symposium on Principles of database systems
  • Year:
  • 1988

Quantified Score

Hi-index 0.00

Visualization

Abstract

A database query is safe if its result consists of a finite set of tuples. If a query is expressed using a set of pure Horn Clauses, the problem of determining query safety is, in general, undecidable. In this paper we consider a slightly stronger notion of safety, called supersafety, for Horn databases in which function symbols are replaced by the abstraction of infinite relations with finiteness constraints [Ramarkrishman et. al 87] We show that the supersafety problem is not only decidable, but also axiomatizable, and the axiomatization yields an effective decision procedure. Although there are safe queries which are not supersafe, we demonstrate that the latter represent quite a large and nontrivial portion of the safe of all safe queries