Application of Object-Oriented Metrics To C++ and Java
Ushā Kumārī, Sucheta Bhasin
Abstract
Ushā Kumārī, Sucheta Bhasin
Abstract
The increasing importance of software measurement has led to design and development of new software measures. As the development of object-oriented software is rising, more and more metrics are being defined for object-oriented programming languages. But there is a little understanding of comparative application of many of these measures to different programming languages. As a consequence, it is very difficult for software managers and developers to select suitable measures and programming languages for object-oriented systems. In this paper, we investigate 23 object-oriented metrics proposed by various researchers. The metrics are applied to same set of 15 programs coded in C++ and Java in an attempt to make comparative study of these two prominent object-oriented programming languages. The work done here reveals that these object-oriented metrics can be applied successfully to both Java and C++. But, Java proves to be more object-oriented than C++ as per intuition. However, counting rules defined for computation of these metrics should be framed more carefully.
OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The increasing importance of software measurement has led to design and development of new software measures. As the development of object-oriented software is rising, more and more metrics are being defined for object-oriented programming languages. But there is a little understanding of comparative application of many of these measures to different programming languages. As a consequence, it is very difficult for software managers and developers to select suitable measures and programming languages for object-oriented systems. In this paper, we investigate 23 object-oriented metrics proposed by various researchers. The metrics are applied to same set of 15 programs coded in C++ and Java in an attempt to make comparative study of these two prominent object-oriented programming languages. The work done here reveals that these object-oriented metrics can be applied successfully to both Java and C++. But, Java proves to be more object-oriented than C++ as per intuition. However, counting rules defined for computation of these metrics should be framed more carefully.
Key concepts: Computer science, Java, Programming language, Object-oriented programming, Software development, Software metric, Generics in Java, Real time Java