S. Abiteboul and V. Vianu, Datalog Extensions for Database Queries and Updates, J. Computer and System Sciences 43 (1991), 62--124.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of view, this update query should delete only one tuple from emp non deterministically. From deductive databases point of view, this query should delete all tuples from emp. In the past decade, various approaches for the inclusion of update capabilities in rule based languages have been proposed [1, 2, 4, 7, 9, 10, 13, 21, 23, 24, 25, 27, 30]. However, most of them are logic programming based that support nondeterministic tuple at a time updates and fail to provide a natural account for setoriented (or bulk) updates in deductive databases syntactically and semantically. Some database oriented approaches either lack intuitive ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. J. Computer and System Sciences, 43(1):62-124, 1991.

....that the sequence of stages induced by such a formula reaches a xed point. However, if it does, this xed point is taken as the semantics of the formula. Otherwise, i.e. if the sequence does not become stationary, the result is de ned as being empty. It has been shown by Abiteboul and Vianu [AV91a] that partial xed point logic provides a precise characterisation of Pspace on nite ordered structures. Thus, showing that there are properties of nite ordered structures de nable in PFP but not in IFP would yield a separation of polynomial time and space. However, on unordered structures, ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43:62-124, 1991.

.... languages over nite structures cannot express many important relations, for example transitive closure, graph connectivity, acyclicity [3, 13, 15] The standard way to increase their expressive power is to extend rst order logic by various recursion like operators, such as xpoint operators [3, 9, 2, 19, 30]. Is it possible to increase the expressive power by using recursive datastructures instead of recursion like operators For example, one could use lists or trees, probably the most natural recursive datastructures. In this paper we show that the use of lists and trees does not increase the ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43:62-124, 1991.

....evaluated with polynomial space. Proposition 3.52. For every formula p PFP, the set of finite models of p is in PSPACE, in short: PFP C PSPACE. On ordered structures, one can use similar techniques as in previous capturing results, to simulate polynomial space bounded computation by PFP formulae [1, 90]. Theorem 3.53 (Abiteboul Vianu, Vardi) On ordered finite structures, PFP captures PSPACE. Proof. It remains to prove that every class ] of finite ordered structures that is recognizable in PSPACE, can be defined by a PFP formula. Let M be a polynomially space bounded deterministic Turing ....

....The multi dimensional u calculus is 67 Remark. Instead of evaluating a formula p L at single nodes v of G, we can also evaluate it at k tuples of nodes: G, L P iff G k, L, p. Example 5.23. Bisimulation is definable in L (in the sense of the remark just made) Let : X. A (Pb,1 Pb,2) [1](2)X A [2] 1)X) bGB For every transition system G we have that G, Vl, V2 iff Vl and v2 are bisimilar in G. Further we have that G,v V if and only if in G there exists point w, which is reachable from v (by a path of length 1) and bisimilar to v. Exercise 5.24. Prove that L is invariant ....

A. ABITEBOUL AND V. VIANU, Datalog extensions for database queries and updates, Journal of Computer and System Sciences, 43 (1991), pp. 62-124.

....before giving up the computation of a model. With a great deal of research focusing on overcoming these limitations, several proposals have been put forward that give up declarative semantics and fall back on procedural semantics e.g. those based on the inflationary fixpoint computation procedure [3,2,21]. This paper is motivated by the opposite conviction that, because of the important advantages offered by a model theoretic semantics, we should strive to preserve the declarative style of stable model semantics while trying to remove its limitations. This approach has already produced significant ....

....[1] In this case, for each D in D , P(D) can be partitioned into just two strata: D and P. 3 Semantics and complexity of DATALOG TM queries We assume that the reader is familiar with the basic notions of complexity classes [18,27] and of query language complexity evaluation (see for example [2, 3, 6, 8, 17, 19 21, 36, 41]) Definition 1. A (bound) DATALOG TM query Q is a pair (v G, P) where G is a ground literal, called query goal, P is a DATALOG program and v is a symbol denoting the type of semantics adopted for the query: for nondeterministic semantics, 3 for possible semantics, V for certain semantics ....

Abiteboul, S., V. Vianu, V.: Datalog Extensions for Databases Queries and Updates. J Comput Syst Sci 43, 6124, 1991.

....offered by different variants of resolution. The update language proposed by [NT89] provides in addition explicit control by allowing sequential , conditional and iterative operators in rule bodies. A comprehensive study of various extensions of Datalog with fixpoint semantics can be found in [AV91] deterministic and nondeterministic extensions are studied w.r.t. their expressive power and complexity. Connections to procedural languages are given which also exhibit many interesting forms of programmed control. A different way to control evaluation is pointed out in RDL1 [dMS88] here ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43(1):62--124, 1991.

....no abstract, SQL independent semantics was given. Production rules have recently been reconsidered, since they seem well suited as a language for active rules. Therefore, referential actions which are specialized active rules can also be formalized by production rules, e.g. in the style of [AV91, PV95]. However, by axiomatizing referential actions as a logic program P and employing a declarative semantics, the resulting set of updates can be justified and explained in a more intuitive way using the rules of P . This is due to the fact that declarative semantics like the well founded or stable ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. Journal of Computer and System Sciences, 43(1):62--124, 1991.

....correspondence requires that we restrict our attention to structures which contain a distinguished relation that is always a total order of the universe. In particular, Immerman and Vardi independently showed: Theorem 1 ( 18, 26] On ordered structures LFP = P. Similarly, Abiteboul and Vianu [2] showed that the expressive power of PFP is equivalent to that of a while language introduced by Chandra and Harel [7] which Vardi [26] proved captures PSPACE on ordered structures. Theorem 2 ( 2] On ordered structures PFP = PSPACE. In the absence of order, these logics are too weak, i.e. ....

....showed: Theorem 1 ( 18, 26] On ordered structures LFP = P. Similarly, Abiteboul and Vianu [2] showed that the expressive power of PFP is equivalent to that of a while language introduced by Chandra and Harel [7] which Vardi [26] proved captures PSPACE on ordered structures. Theorem 2 ([2]) On ordered structures PFP = PSPACE. In the absence of order, these logics are too weak, i.e. everything that can be expressed in IFP is polynomial time computable, but the converse does not hold, and similarly for PFP and PSPACE. 2.2 Finite Variable Logics Finite variable logics turn out to ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. J. of Computer and System Sciences, 43:62-124, 1991.

....be axiomatised with a nite number of variables. Thus, for instance, LFP is the closure of rst order logic under an operation for forming the least xed points of positive formulas. Similarly, PFP closes rst order logic under an operation forming the iterative xed points of arbitrary formulas [2]. It was independently proved by Immerman and Vardi [14, 19] that LFP exactly characterises polynomial time complexity (the class PTIME) on structures equipped with a linear order, while Abiteboul and Vianu [2] show that PFP similarly characterises polynomial space complexity (the class PSPACE) on ....

.... order logic under an operation forming the iterative xed points of arbitrary formulas [2] It was independently proved by Immerman and Vardi [14, 19] that LFP exactly characterises polynomial time complexity (the class PTIME) on structures equipped with a linear order, while Abiteboul and Vianu [2] show that PFP similarly characterises polynomial space complexity (the class PSPACE) on ordered structures. By results of Kolaitis and Vardi [15] and Dawar et al. 8] we know the following fact: For any sentence of LFP (or PFP) there is a k and a set of sentences of L such that Mod( ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43:62-124, 1991.

....# r(X; Y; Z) q(V; X ) q(X; Y ) #s(X; Y; Z) r(X; Y; Z) q(X; Y ) q(Y; Y ) Now, q occurs negatively and, contrary to the case of Datalog programs, there are various possible semantics leading to di#erent queries. In this paper we consider the nonin#ationary semantics (NI semantics) cf. [3]) and the well founded semantics (WF semantics) 11] In both, # induces a sequence q 0 ;q 1 ; of subsets of D D (see below) Set q t : a; b) there is an n 0 s:t: a; b) # q n for all n n 0 (q t , the truth set, consists of those pairs that eventually are in all members of ....

S. Abiteboul, V. Vianu, Datalog extensions for database queries and updates, J. Comput. System Sci. 43(1) (1991) 62--124.

....no abstract, SQL independent semantics was given. Production rules have recently been reconsidered, since they seem well suited as a language for active rules. Therefore, referential actions which are specialized active rules can also be formalized by production rules, e.g. in the style of [AV91, PV95]. However, by axiomatizing referential actions as a logic program P and employing a declarative semantics, the resulting set of updates can be justified and explained in a more intuitive way using the rules of P . This is due to the fact that declarative semantics like the well founded or stable ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. Journal of Computer and System Sciences, 43(1):62--124, 1991.

....a little overstated; the term was introduced in the context of undecidability. 2 [13, 16, 9] for some of its earlier formulations. The main motivation has been that common relational database queries are expressible in relational calculus algebra [13] Datalog : and various fixpoint logics [3, 32, 9, 10]. Most importantly, as shown in [27, 47] every PTIME query can be expressed using fixpoints of first order formulas on ordered structures; and, as shown in [3] it suffices to use Datalog : syntax under a variety of semantics to express various fixpoint logics. In addition, extensions have been ....

.... relational database queries are expressible in relational calculus algebra [13] Datalog : and various fixpoint logics [3, 32, 9, 10] Most importantly, as shown in [27, 47] every PTIME query can be expressed using fixpoints of first order formulas on ordered structures; and, as shown in [3], it suffices to use Datalog : syntax under a variety of semantics to express various fixpoint logics. In addition, extensions have been proposed to this framework to manipulate complex object databases, based on high order formulas over finite structures, e.g. 1, 2] The study of functional ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. J. Comput. System Sci., 43 (1991), pp. 62--124.

....8225 y Currently visiting INRIA Rocquencourt, France facilitated the integration of logic programming techniques in databases. The main motivation has been that common relational database queries are expressible in relational calculus and algebra [17] Datalog : and various fixpoint logics [4, 5, 29, 13, 14]. Most importantly, as shown in [23, 38] every PTIME query can be expressed using Datalog : on ordered structures; and, as shown in [4] it suffices to use Datalog : syntax under a variety of semantics to express various fixpoint logics. In addition, extensions have been proposed to this ....

.... been that common relational database queries are expressible in relational calculus and algebra [17] Datalog : and various fixpoint logics [4, 5, 29, 13, 14] Most importantly, as shown in [23, 38] every PTIME query can be expressed using Datalog : on ordered structures; and, as shown in [4], it suffices to use Datalog : syntax under a variety of semantics to express various fixpoint logics. In addition, extensions have been proposed to this framework to manipulate complex object databases, based on highorder formulas over finite structures, e.g. 2, 1] see [3] for a short ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. J. Comput. System Sci., 43 (1991), pp. 62--124.

....Y ) r(X; Y; Z) q(V; X) q(X; Y ) s(X; Y; Z) r(X; Y; Z) q(X; Y ) q(Y; Y ) Now, q occurs negatively and, contrary to the case of Datalog programs, there are various possible semantics leading to different queries. In this paper we consider the noninflationary semantics (NI semantics) cf. [AV91]) and the well founded semantics (WF semantics) VG93] In both, Pi induces a sequence q 0 ; q 1 ; of subsets of D Theta D (see below) Set q t : f(a; b) j there is an n 0 s.t. a; b) 2 q n for all n n 0 g (q t , the truth set, consists of those pairs that eventually are in all ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. Journal of Computer and System Sciences, 43(1):62--124, 1991.

....is selected using some conflict resolution strategy and the corresponding actions are executed. This process is repeated until there are no more triggered rules. In the database community, such a forward chaining or fixpoint semantics has been studied for a number of Datalog variants (see, e.g. AV91] thereby providing a logic based formalization of production rules: Let Datalog : denote the class of Datalog programs which allow negated atoms in rule bodies. The inflationary Datalog : semantics (I Datalog) turns the wellknown immediate consequence operator TP developed for (definite) ....

.... for all databases; moreover, confluence is no longer guaranteed if instead of the presented semantics, a nondeterministic semantics is used [AS91] On the other hand, nondeterminism can be a powerful programming paradigm which increases the (theoretical and practical) expressiveness of a language [AV91,GGSZ97] 1 For efficiency reasons however, the distinction between events and conditions may be crucial in practice. 2 Another noninflationary semantics called P Datalog is obtained by only keeping the newly derived tuples in each iteration; see also Section 4.5. 6 A problem with these ....

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. Journal of Computer and System Sciences, 43(1):62--124, 1991.

....quite similar to the original query language QL rst proved to be computationally complete by Chandra and Harel [16] it does not harm to refer to it by the same name. Using as before a representation of n ary relations on U by binary relations on U , we then have: Theorem 2 ( 16] see also [3, 4, 1]) Every computable generic query is expressible by a QL program. We feel this result is a nice a posteriori con rmation of Tarski s conviction that relation algebra (with pairing) is a formalism with all the expressive power one needs. We conclude this section with three remarks. First, given ....

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43(1):62-124, 1991.

....eciently on programs that solve polynomial time problems. With current research focusing on overcoming these limitations, several proposals have been put forward that give up declarative semantics and fall back on procedural semantics, e.g. based on the in actionary xpoint computation procedure [2, 1, 16]. This paper is largely inspired by the conviction that modeltheoretic semantics o ers important bene ts, and, therefore, we should strive to preserve it. Thus, we propose a novel approach whereby strati ed DATALOG programs are extended to include (1) an explicit non deterministic construct called ....

....possible DATALOG : queries, i.e. all possible query goals on all possible DATALOG : programs. 3 Semantics and Complexity of DATALOG : We assume that the reader is familiar with the basic notions of complexity classes [13, 21] and of query language complexity evaluation (see, for instance, [1, 2, 6, 12, 14, 15, 16, 27, 31]) The answer to a query Q on a database D over DBP is de ned as follows: 1. under the non deterministic semantics (Q ) a) true if there is a stable model M of P [ D for which G is true in M (b) false if either there exists no stable model or there is a stable model M of P [ D for which G ....

S. Abiteboul and V. Vianu. Datalog Extensions for Databases Queries and Updates. In Journal of Computer and System Science, 43, pages 62-124, 1991.

No context found.

S. Abiteboul and V. Vianu, Datalog Extensions for Database Queries and Updates, J. Computer and System Sciences 43 (1991), 62--124.

No context found.

S. Abiteboul and V. Vianu. Datalog Extensions for Database Queries and Updates. Journal of Computer and System Sciences, 43(1): 62-124, 1991.

No context found.

Abiteboul, S. and V. Vianu, V., Datalog Extensions for Databases Queries and Updates. Journal of Computer and System Sciences, 43, pages 62--124, 1991.

No context found.

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43:62--124, 1991.

No context found.

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Comp. and System Sc., 43(1):62--124, August 1991.

No context found.

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Computer and System Sciences, 43:62--124, 1991.

No context found.

S. Abiteboul and V. Vianu, Datalog extensions for database queries and updates, Journal of Computer and System Sciences, vol. 43 (1991), pp. 62--124.

No context found.

S. Abiteboul and V. Vianu. Datalog extensions for database queries and updates. Journal of Comp. and System Sc., 43(1):62--124, August 1991.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC