Date of Award

12-2017

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Engineering and Sciences

First Advisor

Keith Brian Gallagher

Second Advisor

William H. Allen III

Third Advisor

Anthony Smith

Fourth Advisor

Phil Bernhard

Abstract

High Volume Automated Testing is a powerful family of software testing techniques which enable a variety of testing goals, including the discovery of hard-to-reproduce bugs, which can enable new levels of quality assurance when applied correctly. This thesis presents a software tool, Yeager, which may be used in conjunction with existing test code to execute tests similar to Long Sequence Regression Tests based on an inferred state-model of the system under test as provided by tester annotations of state transitions caused by individual test code snippets. The usefulness of the package is evaluated through the development and deployment of a HiVAT campaign on the open-source Monica Personal CRM system, which also aids in the explanation of the package’s use. The package does enable such testing in a fast and easy-to implement manner while also enabling testers to better understand the structure of the system under test. The Yeager package cannot enable HiVAT alone, it must be implemented as part of an existing automated testing suite. The package, contextualized with a chapter outlining the state of and history of HiVAT in general, provides a new way for testers in the field to implement these powerful techniques for only marginal additional effort.

Comments

Copyright held by author

Download

Share

COinS