Research On Reading Techniques For Object-oriented Code

As a mainstream paradigm, object-oriented programming (OOP) is different from structured programming as the object is the basic unit in OOP and encapsulates the relevant behavior and data. It not only increases software flexibility, but also requires more during the inspection. As an important QA method, inspection is used to find defect in different artifacts, the source code for instance. During preparation, readers will work independently to find defects and the results will be discussed in more detail in subsequent team meetings. The reading approach is so called reading technique; therefore, an appropriate technique will benefit the reading effectiveness, and even the effectiveness of the whole inspection. However, it is difficult to make choice as several factors should be considered, such as language characteristics and the applicability of the reading techniques. What’s more, it should have other considerations for those novices. Of deep concern is that as yet there isn’t one specialized and concrete solution to deal with the situation mentioned above. Hence further investigation has realistic value.In order to further locate the problems in this specific area, we used the SLR method to summarize useful experience. On the basis of these relevant literatures, our team made a deeper analysis of the current challenges and solutions, identified that the combination of the reading techniques would make sense and the Checklist-Based as well as the Use Case-Based Reading Technique are applicable. The former provides clear guidance to the review contents, while the latter one captures the dynamic views of the programs with use cases. Finally, our team got the Extended Checklist-Based Reading Technique (E-CBR) to make a try of the combination. The checklist is a customized one with different levels for 00 and the dynamic nature of UBR is made full used of to guide reading. Besides, other measures are added to save cost. Moreover, our team carried out a case study to evaluate the new technique. The results of the preliminary investigation show that the new technique has a certain value in reading object-oriented code and it increases our confidence in further research.