Basic Block Coverage for Unit Test Generation at the SBST 2022 Tool Competition

Conference Paper (2022)
Author(s)

P. Derakhshanfar (TU Delft - Software Engineering)

Xavier Devroey (University of Namur)

Research Group
Software Engineering
Copyright
© 2022 P. Derakhshanfar, Xavier Devroey
DOI related publication
https://doi.org/10.1145/3526072.3527528
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 P. Derakhshanfar, Xavier Devroey
Research Group
Software Engineering
Pages (from-to)
37-38
ISBN (print)
978-1-6654-6234-1
ISBN (electronic)
978-1-4503-9318-8
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Basic Block Coverage (BBC) is a secondary objective for search-based unit test generation techniques relying on the approach level and branch distance to drive the search process. Unlike the approach level and branch distance, which considers only information related to the coverage of explicit branches coming from conditional and loop statements, BBC also takes into account implicit branchings (e.g., a runtime exception thrown in a branchless method) denoted by the coverage level of relevant basic blocks in a control flow graph to drive the search process. Our implementation of BBC for unit test generation relies on the DynaMOSA algorithm and EvoSuite. This paper summarizes the results achieved by EvoSuite's DynaMOSA implementation with BBC as a secondary objective at the SBST 2022 unit testing tool competition.