Query Rewriting using Semistructured Views

We consider the problem of rewriting semistructured queries using semistructured views. Given a client semistructured positive non-recursive query and a set of semistructured views, the algorithm nds a rewriting query that is equivalent to the client query and accesses only the views. Our solution is based on appropriately generalizing well understood techniques such as containment mappings and the chase. Furthermore, we develop an equivalence testing algorithm for checking the equivalence of the client query against the composition of the rewriting query with the views. We prove that our algorithm is sound and complete. Finally we investigate a specialized polynomial algorithm for composing the rewritten query with the views, based on which we can show that the semistructured query rewriting problem is NP-complete; i.e., it is no harder than the rewriting problem for conjunctive queries. Research partially sponsored by NSF, under Award Number IRI-9712239, and the Wright Laboratory, Aeronautical Systems Center, Air Force Material Command, USAF, under Grant Number F33615-93-1-1339. Research partially supported by NSF grant IRI{96{31952, ARO grant DAAH04{95{1{0192, Air Force contract F33615{93{1{1339 and the L. Voudouri Foundation.