Fault Localization Automation using Code Coverage Data and Diff Utilities

University essay from Lunds universitet/Institutionen för datavetenskap

Abstract: Performing fault localization is one of the most time consuming and expensive aspects of software development today. In large software systems the manual fault localization techniques quickly become unmanageable and as such there is a growing need for automated solutions. In this master’s thesis we propose and evaluate a new method for fault localization based on comparing code coverage and diff utility data. The implemented tool is called FLAC which consists of both a Java application that performs the necessary functionality and a Qlik Sense application designed for visualization purposes. The FLAC prototype is restricted to JavaScripts but the method can be applied to all types of software development projects that employ version control and have the ability to gather code coverage data from test executions. Our method is evaluated by applying it to otherwise stable software that has been injected with faults, analyzing efficiency and performance information and also through interviews. Our results indicate that this method can significantly increase fault localization efficiency.

  AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)