Re-tooling Code Structure Based Analysis with Model-Driven Program Slicing for Software Maintenance

Static code analysis is a methodology of detecting errors in program code based on the programmer's reviewing the code in areas within the program text where errors are likely to be found and since the process considers all syntactic program paths; there is the need for a model-based approach with slicing. This paper presented a model of high-level abstraction of code structure analysis for a large component based software system. The work leveraged on the most important advantage of static code structure analysis in re-tooling software maintenance for a developing economy. A program slicing technique was defined and deployed to partition the source text to manageable fragments to aid in the analysis and statecharts were deployed as visual formalism for viewing the dynamic slices. The resulting model was a high-tech static analysis process aimed at determining and confirming the expected behaviour of a software system using slices of the source text presented in the statecharts.