Thesis
A fault model for pointcuts and advice in AspectJ programs
Washington State University
Master of Science (MS), Washington State University
2006
Handle:
https://hdl.handle.net/2376/568
Abstract
This thesis presents a fault model for pointcuts and advice, the two main constructs of the AspectJ programming language. The fault model provides a fault/failure analysis of how a fault, in a pointcut or a piece of advice in a program, can cause a data state in the program to become corrupted, and how that erroneous data state can propagate to the final state of the program, thereby manifesting a failure. The fault model also includes a catalog of fault types that are believed to represent faults likely to be introduced in programs by programmers writing AspectJ code. Each type of fault is described in terms of how it appears syntactically in source code as well as in how it can cause an infection of program state. The fault types are identified from a careful analysis of the syntax and the semantics of the pointcut and advice constructs. The fault model can help testers and programmers identify places in a program where faults are most likely to appear, and identify what kinds of faults to look out for when using a certain language feature. The fault model is claimed to be a good foundation for fault seeding, mutation testing, program inspections, and evaluation of testing strategies for AspectJ programs. Examples are given that demonstrate the model's suitability for these purposes. It is also believed that the fault model can be used to derive test adequacy criteria and devise testing strategies.
Metrics
3 File views/ downloads
21 Record Views
Details
- Title
- A fault model for pointcuts and advice in AspectJ programs
- Creators
- Jon Swane Baekken
- Contributors
- Roger T. Alexander (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; Pullman, Wash. :
- Identifiers
- 99900525193601842
- Language
- English
- Resource Type
- Thesis