Implementing Source Code Metrics for Software quality analysis

Developing a high quality software product in an economical way is one of the fundamental goals of any software engineering activity. As computers are being used in almost every conceivable area in the contemporary world, quality of software becomes a key factor in the strategic success of a business and human security in general. Finding determinants of software quality and mapping them into quantitative measures is a crucial factor in sustainable success of an end product. Software metrics as means of quality analysis has attracted a lot of attention among researchers and practitioners in last one decade. Mapping of program characteristics into these metric values indicate structural complexity and behavior of an information system. In this case study, the five software metricslines of code (LOC), cyclomatic complexity (MVG), Halstead volume (HV), number of modules (NOM) and lines of comment (COM) have been utilized to analyze a set of three java based sorting programs. Three software measurement tools have been applied on them to judge their performance with respect to the metrics mentioned therein. Also a derived metric maintainability index has been calculated from the base metrics to indicate relative maintainability of the source code. Comparative analysis of the chosen tools have also been undertaken to reveal how they differ in delivering results for the same programs. Further, some other quality factors which can be derived from the constituent metrics are mentioned in a later sub-section.