| Type of Document |
Master's Thesis |
| Author |
Mungara, Mahesh Babu
|
| Author's Email Address |
mmungara@vt.edu |
| URN |
etd-10252003-144535 |
| Title |
A Method for Systematically Generating Tests from Object-Oriented Class Interfaces |
| Degree |
Master of Science |
| Department |
Computer Science |
| Advisory Committee |
| Advisor Name |
Title |
| Edwards, Stephen H. |
Committee Chair |
| Nance, Richard E. |
Committee Member |
| Rosson, Mary Beth |
Committee Member |
|
| Keywords |
- all-DU-pairs
- class Interfaces
- all-nodes
- Specification-based Testing
|
| Date of Defense |
2003-09-15 |
| Availability |
unrestricted |
Abstract
This thesis describes the development and evaluation of a manual black-box testing method inspired by Zweben's test adequacy criteria, which apply white-box analogues of all-DU-pairs and all-nodes to a flow graph generated from the black-box specification. The approach described herein generates tests from a matrix representation of a class interface based on the flow graph concept. In this process, separate matrices for all-DU-pairs and all-nodes guide the generation of the required tests. The primary goal of the research is not to optimize the number of tests generated but to describe the process in a user-friendly manner so that practitioners can utilize it directly, quickly, and efficiently for real-world testing purposes.
The approach has been evaluated to assess its effectiveness at detecting bugs. Both strategies - all-DU-pairs and all-nodes - were compared against three other testing methods: the commercial white-box testing tool Jtest, Orthogonal Array Testing Strategy (OATS), and test cases generated at random. The five approaches were applied across a sample of eleven java classes selected from java.util.*. Experimental results indicate that the two versions resulting from this research performed on par with or better than their respective equivalent approaches. The all-DU-pairs method performed better than all other approaches except for the random approach, with which it compared equally. Experimental evaluation results thus indicate that an automated approach based on the manual method is worth exploring.
|
| Files |
| Filename |
Size |
Approximate Download Time
(Hours:Minutes:Seconds)
|
| 28.8 Modem |
56K Modem |
ISDN (64 Kb) |
ISDN (128 Kb) |
Higher-speed Access |
| |
appdx.pdf |
189.97 Kb |
00:00:52 |
00:00:27 |
00:00:23 |
00:00:11 |
00:00:01 |
| |
Thesis.pdf |
518.37 Kb |
00:02:23 |
00:01:14 |
00:01:04 |
00:00:32 |
00:00:02 |
|
