Data flow analysis within the ITOC information system design recovery tool

Most contemporary fourth-generation languages (4GL) are tightly coupled with the database server, and other subsystems, that are provided by the vendor. As a result organizations that wish to change database vendors are typically forced to rewrite their applications using the new vendor's 4GL. The anticipated cost of this redevelopment can deter an organization from changing vendors, hence denying it the benefits that would otherwise result, e.g., the exploitation of more sophisticated database server technology. If tools existed that could reduce the rewriting effort, the large upfront cost of migrating the organization's applications would also be reduced, which could make the transition economically feasible. The ITOC project is part of a large collaborative research initiative between the Centre for Software Maintenance at the University of Queensland and Oracle Corporation. The objective of this project is to design and implement a tool that automatically recovers both the application structure and the static schema definition of 4GL information system applications. These recovered system components are transformed into constructs that populate Oracle's Designer 2000 CASE repository. An essential component of the ITOC process is to determine the relationships between different columns in the database and between references to those columns and fields that appear within the user interface. This in turn requires analysis of data flow between variables in the 4GL programs. While data flow analysis has been applied in many applications, for example, code optimization and program slicing, this paper presents the results of using data flow analysis in the construction of a novel design recovery tool for 4GL-based information.