Object Oriented Software Engineering Workshop 1
Software Metrics
Objectives
To generate control flow graphs and understand how they are used to determine the complexity of program code.
Understand how software metrics can be used to measure and improve the quality of software.
You will achieve these objectives by completing a set of tasks.
Setup 1
Download vcs.zip from OOSE Moodle page and unzip it to a folder of your choice. Create a Java eclipse project and name it VehicleControlSystem. Copy oose.vcs, vehicle.types and img folder from unzipped content into the src and project directo- ries in Eclipse respectively. Your eclipse directory structure should be similar to Figure
To run the project, right click Controller.java and run as Java application. A class
Figure 1: Project directory structure for Vehicle Control System on Eclipse
Figure 2: Design Diagram for a Vehicle Control System
diagram of the system design is as shown in Figure 2. You can also access this diagram by clicking Design.ucls in oose.vcs package.
Setup 2
Download CKMetrics.jar from OOSE Moodle page to a folder of your choice. CKMetrics is a tool for determining Chidamber and Kemerer (CK) metrics suite for java class files and Jar APIs. CKMetrics operates by using Apache Byte Code Engineering Library (BCEL) to analyse class files that are contained in a jar file. Assume you have A.jar located in your local drive ../JarLocation. To determine the quality of A using CKMetrics:
Select Load File button from the toolbar. You can either browse and select the folder containing A or select A. For the former, CKMetrics will load all the jar files contained in the folder.
Then select Compute button from the toolbar. CKMetrics will compute the mean CK metrics value for each selected jar file.
Optionally, you can include jdk class files in the analysis, or only analyse publicly accessible classes, fields and methods by checking includeJdk and onlyPublic checkboxes respectively.
Figure 3 is a screenshot showing CKMetrics’s output. You can also select a .class file following above steps to analyse java classes.
Figure 3: Analysis output from CKMetrics
Finally, to build the jar file required to check the quality of Vehicle Control System
based on above steps:
Right click on the project in Eclipse.
Select export.
Export as runnable jar file.
Tasks
Draw the control flow graph for the method initialiseVehicle in Controller.java.
(1 Mark)
Assume the objective is to increase encapsulation, modularity and reduce the com- plexity of VehicleControlSystem. Identify 3 metrics that can be used to measure these quality factors in VehicleControlSystem. (1 Mark)
Identify the most important class in VehicleControlSystem that should be refac- tored to enhance the quality factors in Task 2. Justify your choice in a maximum of 200 words. (1 Mark)
Propose and implement four refactorings for the class identified in Task 3 to enhance quality factors mentioned in Task 2. In maximum of 200 words, discuss the extent suggested refactorings enhances quality factors mentioned in Task 2. (1 Mark)
Draw a new class diagram for VehicleControlSystem after implementing suggested refactorings in Task 4. In maximum of 200 words, discuss any structural difference in the design of VehicleControlSystem after implementation. (1 Mark)
Deliverables
Turn in a pdf document electronically via Moodle for Workshop 1. You should clearly state at the top of the document your name and registration number. The document should also contain your solution to tasks 1- 5 with relevant screen- shots of CKMetrics and class diagrams.
A source folder for Vehicle Control System containing the implementation of your refactoring tasks.
A runnable jar file of the Vehicle Control System after implementing your pro- posed refactoring tasks.
Assessment
Submissions is due by 16.30 on 5th February 2019. You should submit your solution on Moodle in the appropriate upload slot for the laboratory.
Tutors and demonstrators will be in your allocated Lab to offer assistance. Endeavour to attend your labs to discuss issues rather than sending emails. Its likely that you will receive no response by email or response may be slow.
As per the Code of Assessment policy regarding late submissions, submissions will be accepted for up to 5 working days beyond due date. Any late submissions will be marked as if submitted on time, but reduced by 1 mark for each additional day. Submissions received more than 5 working days after the due date will receive an H (band value of 0).
