Some aspects of the implementation of a relational data base sublanguage

Abstract

A relational model of data has been proposed by Codd [1] for protecting users of formatted data systems from the potentially disruptive changes in data representation caused by growth in the data base and changes in traffic. The adoption of this relational model of data has permitted the development of a universal data sublanguage, based on an applied predicate calculus, called ALPHA. Although ALPHA has been intended to be a sublanguage of the languages used by all terminal users of a shared, formatted data base, only [2] the semantics of it has been fully described by Codd. However, before ALPHA can be implemented, it is necessary to specify the syntax of this data sublanguage. Accordingly, the work described here has involved the specification of the syntax of this data sublanguage and the development of a related PL/1 computer program, called the translator. The translator carries out the following functions: (i) accepts as input a modified form of source statements; these statements include both data sublanguage ALPHA statements and computational facility statements (ii) checks that conditions are met for the relational calculus; interactive debugging facilities are provided (iii) produces as output a set of coding tables suitable as input to an interpreter; these tables, a modifier version of those based on the work of Palmero, [3] are presented in a precise manner which clearly reflects the subsequent operations to be carried out. It is intended that this work will form a basis for the development of a data base system.