D.J. Russomanno
Prolog, query interface, open database connectivity
A Prolog-based query interface with optional data segmentation via modules is presented in this paper. The motivation of this work is to provide an implementation of a useful, but simple interface that relieves the Prolog programmer from having to recall the com- ponent order within a tuple and the complete deļ¬nitions of a set of underlying facts when forming declarative queries. The imple- mentation includes a schema level representation and accompanying query predicates, such that the order and number of columns within a Prolog query becomes unimportant, resulting in increased data independence. The interface supports three architectures: 1) the traditional Prolog environment, 2) an environment in which Prolog ground clauses have been extracted from heterogeneous data sources, and 3) as an abstraction layer in a loosely coupled architecture with interfaces to Open Database Connectivity (ODBC) compliant sources in which data is accessed as though it existed as ground clauses within the Prolog environment. As compared to traditional approaches to forming a Prolog query, the current implementation of the interface introduced an additional 17.6 microseconds to retrieve a tuple using a 2.0 GHz Intel Xeon processor with 1 GB of RAM.
Important Links:
Go Back
