Yanjun Lou and Z. Meral Ozsoyoglu. LLO: an object-oriented deductive language with methods and method inheritance. In Proceedings of the 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....relates to previous work on inheritance in the context of (rule based) objectoriented languages as follows. Some of these approaches base inheritance on syntactic criteria, either on the class hierarchy alone [Bre87] on unification of terms denoting objects [AN86, CCCR 90] or on signatures [LO91] In [BM92] an additional labeling concept of rules affects inheritance. Other approaches consider inheritance as a model theoretic default mechanism [BL91, LV92, KLW90] The techniques we study are in some sense between these two directions. We use syntactical criteria to be able to provide ....

Y. Lou and Z. Meral Ozsoyoglu. LLO: An object oriented deductive language with methods and method inheritance. In Proc. SIGMOD , 1991.

....of the language is defined in terms of an inflationary fixpoint operator, extended by the mechanism of old invention. We remark that the language supports object creation, but neither forms of object deletion nor state update functionalities are available. LLO LLO (Logic Laguage for Objects) [44] is a declarative query language for an object oriented data model, including also methods. A method is defined as a clause, whose head is the method interface and whose body is the method implementation. A query is seen as a message and the unification algorithm takes care of message handling. ....

....of stete chenges without side effects. In other proposeis, multi heeded cleuses ere used for similer purposes. However, the notion of updeting object stete does not fit well in object oriented extensions of logic progremming. In eddition, elso epproeches developed in the detebese field, like e.g. [37, 39, 44], do not consider stete evolution. The epproeches deriving from the logic progremming field, moreover, do not consider e scheme level, thet is the notion of cless. Moreover, even though inheritence is supported in some ceses [21, 27] the form of inheritence considered is emong objects, thet is e ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An Object-Oriented Deductive Language with Methods and Methods Inheritance. In Proc. of the ACM SIGMOD Lnt'l Conf. on Management of Data, pages 198 207, 1991.

....without side effects. In other proposals, multi headed clauses are used for similar purposes. However, the notion of updating object state does not fit well in object oriented extensions of logic programming. In addition, also approaches developed in the database field, like e.g. Fre94,GLR90,LO91] do not consider state evolution. Many of the approaches [AK89,GLR90,CCCR 89] moreover, do not consider the behavioral component of objects, that is, methods. We think that this is an important issue because it overcomes the dichotomy between data and operations of the relational model. ....

....do not consider the behavioral component of objects, that is, methods. We think that this is an important issue because it overcomes the dichotomy between data and operations of the relational model. Few proposals moreover, deal with behavioral inheritance and overriding. In addition to [ALUW93,LO91,McC88] these topics have been addressed in [BJ95,DT95,JL95] All these proposals extend F logic [KL90] or F logic variations) with behavioral inheritance. In F logic, indeed, only structural inheritance is directly captured. For behavioral inheritance, the non monotonic aspects introduced by ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An Object-Oriented Deductive Language with Methods and Methods Inheritance. In Proc. of the ACM SIGMOD Int'l Conf. on Management of Data, pages 198--207, 1991.

.... research prototypes have been implemented, e.g. SALAD (see Section 2) NAIL [18] KIWI [22] or ALGRES [4] Also first attempts of combining Datalog with concepts from object oriented databases exist resulting in various deductive object oriented database system prototypes like COMPLEX [12] LLO [16], LOL [3] CLOG [13] or Quixote [34] 2 LDL LDL (Logical Data Language) was designed at MCC as purely declarative logic based language. It provides the following powerful extensions of pure Datalog (a complete presentation gives [19] C simple data types including string, integer, and real ....

Y. Lou, Z.M. Ozsoyoglu. LLO: an ObjectOriented Deductive Language with Methods and Methods Inheritance. Proc. ACM Intl. Conf. Management of Data, 1991.

....types. Further, in object oriented systems one defines behavioral semantics in terms of operations, or methods, associated with objects. Our approach is to associate constraints to objects. Constraints can be user defined and dispersed over object types as methods associated with object types [7, 10]. Because constraints are tightly coupled with the object, it is simple to maintain object consistency within an object when the object is updated [2] But, because of this tight coupling, it is hard to maintain overall database consistency if the effects of the update of an object are propagated ....

....this difficulty, there are three well known approaches: naive approach, limited approach, and expert approach. In a naive approach, all constraints are defined in one place (e.g. a root node or the metaclass ) and all constraints are considered for enforcement. The second approach, used in [5, 6, 7, 8, 10], allows constraints to be defined and associated with an object. Only those constraints (so called intraobject constraints) associated with an object or inherited from a supertype object are considered when an update or a query is presented to the object. However, no propagation of update ....

[Article contains additional citation context not shown here]

Yanjun Lou and Z. Meral Ozsoyoglu. LLO: an object-oriented deductive language with methods and method inheritance. In James Clifford and Roger King, editors, Proc. ACM SIGMOD Intl. Conf. on Management of Data, pages 198--207, Denver, CO, 1991.

....it has gained a lot of attention, e.g. 5, 19, 33] There are a number of important issues not yet quite understood. These include for instance typing issues and object creation. The problem in object creation is the naming of the newly created object. In IQL [5] LOGRES [19] LOCO [37] or LLO [38], object identities come from a unique infinite sort, and one new identifier is taken from that sort when necessary. In other formalisms (e.g. O logic of [39] or F logic of [33] new objects are denoted using Skolem functions so the programmer directly controls their names. A number of ....

Y. Lou and Z. M. Ozsoyoglu. LLO: an object-oriented deductive language with methods and method inheritance. In sigmod, pages 198--207, 1991.

....schema and data. The authors describe two kinds of inheritance: monotonic inheritance and inheritance with overriding. The semantics of overriding is static. Unlike most of the work in this area, the authors investigate updates. Lou and Ozsoyoglu extend Datalog to include object oriented features [79]. The resulting language is LLO. In LLO, the schema and data are separate. Methods are defined by clauses, and inheritance is achieved through type and unification mechanisms. The unification mechanism searches the class inheritance hierarchy bottom up to find the method with best matching type. ....

....LDL. However, this limits the expressiveness of the language. In Gulog, an answer to a query is a substitution as in logic programming. However, in Gulog, the substitutions are typed and objects rather than values are substituted. Our approach is most similar to that in IQL [6] ILOG [56] and LLO [79] where the answer is a collection of object identifiers. This is different from F logic [65] and LIFE [9] where there is basically no distinction between types and instantiations of types. In [14] answers to DTL CHAPTER 3 : GULOG 46 queries are a set of homogeneous elements. That is every answer ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An object-oriented deductive language with methods and method inheritance. In Proc. of the ACM SIGMOD International Conference on the Management of Data, pages 198--207, 1991.

....is extended by functional reduction, called narrowing [6] But as aforementioned our interest is on bottom up algorithms. The main contributions of our work can be summarized as follows: Computed methods of a DOOD can be considered as algorithms executed when needed, as suggested in [1, 11]. Their evaluation is based on fast functional reduction, improving efficiency but fully preserving the declarative minimal model semantics of a program. Moreover, termination of some logical programs is achieved which typically are unsafe otherwise. An earlier approach discusses reduction rules ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An object oriented deductive language with methods and method inheritance. In J. Clifford and R. King, editors, Proc. ACM SIGMOD Intl. Conference on Management of Data, 1991.

....functional and set valued methods. Many of the logics that have been described in the literature do not address overriding (Abiteboul, 1989) Beeri, 1990) Chen, 1989) Kifer, 1989) Maier, 1986) or multiple inheritance (Abiteboul, 1990a) Beeri, 1990) Chen, 1989) Kifer, 1989) (Lou, 1991), Maier, 1986) Three proposals that do seriously consider these issues are Kifer et al. s F logic (Kifer, 1990) McCabe s L O (McCabe, 1992) and Abiteboul et al. s datalog meth (Abiteboul, 1993) F logic is a lot richer than Gulog. In particular, in F logic, it is possible to query the schema. ....

Lou, Y. and Z. M. Ozsoyoglu, LLO: An object-oriented deductive language with methods and method inheritance. In Proc. of the ACM SIGMOD International Conference on the Management of Data, pages 198--207, 1991.

....many researchers in several different areas. Besides the area of logic programming, where it has been proposed as a framework to specify polymorphic types [23] and set abstractions [2] HiLog has also been used as a declarative query language for deductive [6, 16] and object oriented databases [11, 12]. Despite its popularity, the viability of HiLog as a logic programming language depends on whether it can be implemented efficiently. In particular, the performance of a HiLog implementation must be able to compete with that of modern Prolog systems. Before we can expect Prolog programmers to use ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An object-oriented deductive language with methods and method inheritance. In Proceedings of the ACM SIGMOD International Conference on the Management of Data, pages 198--207, Denver, Colorado, May 1991.

....(and predecessors) and by IQL (and predecessors) ffl ROCK ROLL provides a full object oriented model of behavioural knowledge, in which there is a clear distinction between the interface to and the implementation of a type. Few proposals tackle this problem, of which ConceptBase [26] and LLO [33] are well known exemplars. This is a necessary feature for a system to be classified as behaviourally object oriented, and has origins in the notion of abstract data types. In a DOOD context, it implies devising ways of distinguishing within the database theory the declaration that instances of a ....

Y. Lou and Z.M. Ozsoyoglu. LLO: An Object-Oriented Deductive Language with Methods and Method Inheritance. In James Clifford and Roger King, editors, Proc. ACM SIGMOD International Conference on the Management of Data, pp. 198--207. ACM Press (1991).

....in such a case the computation moves to another object (we will also say that a context switch has happened) Our approach to deductive object databases relates to previous work on logical object databases as follows. Some of these approaches do not consider state evolution of deductive objects [12, 19, 27, 28, 24, 2, 15, 21, 23]. Others consider state evolution [5, 16] but objects have the granularity of terms. By contrast we consider an object with larger granularity, that is a theory, i.e. a set of logical clauses. Moreover many of the considered approaches do not consider the behavioral component of objects (i.e. the ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An Object-Oriented Deductive Language with Methods and Methods Inheritance. In Proc. Int'l ACM Conf. on Management of Data, pages 198--207, 1991.

....language is implemented by mapping it onto an existing deductive database system [BM92] The above proposals significantly play down the role of an independently defined data model as the basis from which languages can be defined with different operational semantics. Other such proposals are LLO [LO91] and ORLOG [JL93] Both are inspired by F logic into postulating a higher order syntax, which provides undeniably elegant solutions in certain cases, but is difficult for occasional users to fully grasp. The higher order approach also incurs the penalty of a less than intuitive declarative ....

Y. Lou and Z.M. Ozsoyoglu. LLO: An ObjectOriented Deductive Language with Methods and Method Inheritance. In James Clifford and Roger King, editors, Proc. ACM SIGMOD Conf., pages 198--207, 1991.

....F logic is complex, and still incomplete, it provides the most complete description available of the object oriented concepts it addresses. Space does not permit a more extensive survey of theoretical work in this area, but other notable contributions have been made by Beeri [6] Lou and Ozsoyoglu [24] and Meseguer and Qian [26] 3 Practice So much for the discussion of the logical foundations of deductive object oriented database systems. What can database users do with them now Well, many object oriented concepts together with some deductive concepts are provided in the extended ....

Y. Lou and Z. M. Ozsoyoglu. LLO: An object-oriented deductive language with methods and method inheritance. In Proc. ACM SIGMOD Int. Conf. on Management of Data, pages 198--207, Denver, CO, 1991.

....in the language spaces they generate; in the second we confront them with Ullman s criticisms [52] to argue that under his assumptions their responses may not be adequate. Finally, we indicate how adequate responses are possible. We wish to remark that there is a wealth of research activity (e.g. [12, 43], among others) whose fruits would fit under one of the labels we used to discuss certain approaches to DOODs. The fact that we have not taken them as examples should not be taken as an implicit judgment of comparative merit. Other proposals are more difficult to classify within a general ....

Yanjun Lou and Z. Meral Ozsoyoglu. LLO: An Object-Oriented Deductive Language with Methods and Method Inheritance. In James Clifford and Roger King, editors, Proceedings of the 1991 ACM SIGMOD International Conference on the Management of Data, pages 198--207, Denver, Co, May 1991. ACM Press. SIGMOD Record, 20(2), June 1991.

....range of the queries expressible in these languages and what needs to be done to expand this range; also, very few of these fifty languages extend beyond relational model 1 . Aside from these, a number of non visual (declarative) formal languages were also proposed [CKW89, KL89, Cru92, Cru89, LO91, KKS92, OW89, BNST91, Kup90, Hul87, INV91] and recently one of these [KL89] is extended into a visual language[Cru92] see Section 8 for a discussion on the limitations of the few visual languages developed for object oriented databases. We believe that there is a strong need to synthesize the ....

.... and show that these compare favorably with the corresponding fragments of other set based languages discussed in [Vad] Note that grouping plays an important role in set based models, and different languages incorporate it differently: BNST91] uses an explicit grouping construct, AG88, CKW89, LO91] use data functions or its variations, O 2 [BCD90] XSQL[KKS92] use ad hoc constructs ALL , STBE[OMO89] uses window hierarchy, etc. Our VQL does not use any special construct for grouping, because restricted universal quantifier enables direct encoding of grouping. Practical front: VQL ....

[Article contains additional citation context not shown here]

Lou, Y. and Ozsoyoglu, Z.M. Llo: An object oriented deductive language with methods and method inheritence. In Proceedings of the 1991 ACM SIGMOD International conference on management of data, 1991. 109

....databases where the data has spatial properties, includes sequences and complex objects, and queries are of exploratory in nature. Although many query languages have been proposed for the object oriented model (e.g. IQL [AK 89] Hilog [CKW 89] F logic [KL 89] ORION [KKD 89] O 2 [BCD 90] LLO [LO 91] XSQL [KKS 92] NOODLE [MR 93] spatio temporal models [ITB 92, BVZ 93] there are very few visual query languages that handle both spatial and sequence data in a seamless manner. VISUAL uses a unique iconized object called spatial enforcement region to model spatial relationships. VISUAL is ....

Y. Lou and Z.M. Ozsoyoglu, "LLO: An Object-Oriented Deductive Language with methods and Method Inheritance", Proc. ACM SIGMOD Conf., 1991.

....volume, real time deadlines, spatial properties, includes sequences and complex objects, and queries are of exploratory in nature. Although many query languages have been proposed for the object oriented model (e.g. IQL [AK 89] Hilog [CKW 89] F logic [KL 89] ORION [KKD 89] O 2 [BCD 90] LLO [LO 91] XSQL [KKS 92] NOODLE [MR 93] spatio temporal models [ITB 92, BVZ 93] there are very few visual query languages that handle multimedia data with spatial and sequence properties in a seamless manner. VISUAL is nonprocedural, and uses the example element concept of Query by Example (QBE) Zl ....

Y. Lou and Z.M. Ozsoyoglu, "LLO: An Object-Oriented Deductive Language with methods and 33 Method Inheritance", Proc. ACM SIGMOD Conf., 1991.

No context found.

Yanjun Lou and Z. Meral Ozsoyoglu. LLO: an object-oriented deductive language with methods and method inheritance. In Proceedings of the 1991.

No context found.

Y. Lou and Z. Ozsoyoglu, "LLO: an object-oriented deductive language with methods and method inheritance," SIGMOD Records, vol. 20, pp. 198--207, June 1991. SIGMOD'91 Management of data. Proceedings of the 1991 ACM SIGMOD International Conference.

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