원문정보
초록
영어
To ensure software quality software testing is done .Test data generation is one of the more expensive parts of software testing. So, for reducing software cost and development time automation of test data generation is required. The object-oriented paradigm can be challenging for generating test data, due to some aspects of its features like abstract class, encapsulation, inheritance and visibility. To reach high code coverage of object oriented code search based test data generation technique is used. Proposed approach show that how efficiency and effectiveness of search based test data generation using static analysis. System takes input as different java class files then instances are generated for that classes then generate sequence of method call for whole code coverage then using genetic algorithm which is very useful and work efficiently when there is large search space it use to reach test target and finally generate test cases in Junit format which are helpful for all developers to perform unit testing.
목차
1. Introduction
2. Related Work
2.1. Code based test data generation methods
2.2. Analysis of Various Automatic Test Generation Tools for Object Oriented Code
3. Challenges to Testing Object-Oriented systems
3.1. Encapsulation/Data Abstraction
3.2. Inheritance
3.3. Polymorphism
3.4. Abstract Classes
4. Proposed System
4.1. Pre-processing for Instance Generation
4.2. Test Data Generation
5. Implementation and Results
5.1. Input to Proposed System
5.2. Selection of target method
5.3. Output
5.4. Experimental Subject of input
5.5. Comparison of other approach with genetic algorithm
6. Conclusions
References