Commonly used procedures for database retrievals and computational functions arc frequently required for the specification of data-intensive business programs. Existing program specification techniques use mostly procedural approaches which are tedious and highly repetitive for specifying validations. Tremendous efforts are still spent for re-specifying same procedures and computational functions in the process of specifying business programs.In this paper, we propose a program specification technique called the data derivation (DD) model to address the abovementioned problems. In this technique, we further enhance the reusability idea provided in fou* generation languages for the specification of business programs. We also adapt a declarative approach for the specification of validations to remove the repeated efforts inherited from procedural approaches.The data derivation (DD) model specifies a program through representing its data flows and their relationships with output operations attached to data flows. A DD model is formed by a composition of instances of generic components drawn from the set of formally defined generic components, with the incorporation of specification of output operations and constmints into the composition. It consist-s of an ER-like diagmm, called a data derivation (DD) diagram, and a specification of pammetera, attributes, output opcmtions and constraints involvd. The control structure of the DD model is defined by a set of rules. Whh the formats of its screens and reports defined separately, a DD model is executable and is sufficient for specifying progmms. The pseudocode specification of P is as shown in Figure 1. Note that the group of statements marked with "*V" specifies a pmcedurc which is the same with the procedure specified by the group of statements markedwith "*P. The former is for progmm validation and the latter is for program execution, The two groups of statements marked with "%Al" and "%A2" respectively .SpCCifY the aggmgatd function SUM in the respective 100PS, but the predefine SUM fisnction cannot be used in both cases. For specifying the abovementioncd program, both the 4GLs and tie 95 I
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.