Incremental querying in the concurrent CLP language IFD-Constraint Pandora
Jimmy H. M. Lee, Ho-fung Leung · 1996
Lee and Leung propose the IQ-scheme [11], a framework for efficient incremental query processing in concurrent constraint logic programming.They show that the performance of the IQ-scheme is in general better than the scheme proposed by Van Hentenryck and Le Provost [22].IFD-Constraint Pandora is an instance of the Constraint Pandora class of languages that supports finite domain and interval constraints.In this paper we show how the IQscheme can be incorporated in [FD-Constraint Pandora.We first describe the IFD-Constraint Pandora language, its operational semantics and constraint satisfaction mechanism.Then we present an implementation scheme for the [Qscheme in the IFD-Constraint Pandora language, which is based on the concept of constraint effective point.Finally, we present some preliminary results of applying the lQ-scheme to an IFD-Constraint Pandora program that solves the mapcoloring problem. INTRODUCTIONAccording to the incremental query processing [19] or incremental execution [3,4,22] framework in logic programming, users enter queries by increments.After an increment is entered, Prdog displays an answer, which is the result of, not the increment but, the query accumulated so far.In this framework, variables with the same identifiers in different increments denote the same variable.Thus increments are continuations of the same query.The aim is to minimize re-execution by making use of information gathered in the execution of the previous query increments.Van Hentenryck and Le Provost [22] propose a scheme (VHLP-scheme hereafter) for incremental query processing in the CLP framework [10].In the VHLP-scheme, the increments are all constraints instead of goals: user is allowed to add new constraints to the query or delete old constraints from it, but the goals are not changed.Van Hentem'yck and Le Provost argue that the VHLP-scheme, which is based on "Permission to make digital/hard copy of all or part of this material without fee is granted provided that copies are not made or distributed for profit or commercial advantage, the ACM copyright/server notice, the title of the publication and its date appear, and notice is given that copying is by permission of the Association for Computing Machinery, Inc.(ACM).To copy otherwise, to republish, to post on