Automatic test data generation to cover exception-dependent paths using symbolic execution

• Grant number: 14/08713-9
• Support Opportunities: Regular Research Grants
• Duration: August 01, 2014 - July 31, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Marcelo Medeiros Eler
• Grantee: Marcelo Medeiros Eler
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Software testing is one of the key activity to software quality assurance. Several testing techniques were devised to help testers to create and to evaluate test cases. Symbolic execution approaches have been used as an effective technique to automatic generate test data to cover test requirements derived from structural testing control flow criteria. During symbolic execution, each execution path of the program under test is represented by a constraint sequence which must be satisfied so that specific path is executed. Concrete solutions are found for the constraint sequences found by constraint solvers, and those solutions are used as data test to execute each analyzed path. Most approaches only generates test data to execute exception free paths, which are paths that do not depend on exceptions to be executed. The automatic generation of test data for exception dependent paths, which are paths that are executed only when a given exception is activated, is not handled by current approaches.The automatic generation of test data for exception dependent paths has not been properly addressed by current approaches because the constraints that needs to be satisfied to launch an exception are not explicitly written in the source code. There are no constraints to sign an exception activation, unless it is an exception defined by the developer. Otherwise, only the source code analysis under the point of view of each exception can reveal which test data can lead to an exception activation. Therefore, this grant proposal aims at investigating the test data generation of exception dependent paths and to propose techniques to define constraints able to assure that the target exceptions are properly launched and to avoid that unwanted exceptions are thrown. (AU)

Scientific publications

(5)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

SILVA, DAVI BERNARDO; ELER, MARCELO MEDEIROS; DURELLI, VINICIUS H. S.; ENDO, ANDRE TAKESHI. Characterizing mobile apps from a source and test code viewpoint. INFORMATION AND SOFTWARE TECHNOLOGY, v. 101, p. 32-50, SEP 2018. (14/08713-9)

ELER, MARCELO M.; ENDO, ANDRE T.; DURELLI, VINICIUS H. S.. An empirical study to quantify the characteristics of Java programs that may influence symbolic execution from a unit testing perspective. JOURNAL OF SYSTEMS AND SOFTWARE, v. 121, p. 281-297, NOV 2016. (14/08713-9)

SILVA, DAVI BERNARDO; ENDO, ANDRE TAKESHI; ELER, MARCELO MEDEIROS; DURELLI, VINICIUS H. S.; IEEE. An analysis of automated tests for mobile Android applications. PROCEEDINGS OF THE 2016 XLII LATIN AMERICAN COMPUTING CONFERENCE (CLEI), v. N/A, p. 9-pg., 2016-01-01. (14/08713-9)

DE CASTRO, CAMILA FARIA; OLIVEIRA, DECIO DE SOUZA, JR.; ELER, MARCELO MEDEIROS; IEEE. Identifying Characteristics of Java Methods that May Influence Branch Coverage: An Exploratory Study on Open Source Projects. PROCEEDINGS OF THE 2016 35TH INTERNATIONAL CONFERENCE OF THE CHILEAN COMPUTER SCIENCE SOCIETY (SCCC), v. N/A, p. 8-pg., 2016-01-01. (14/08713-9)

ELER, MARCELO MEDEIROS; ENDO, ANDRE TAKESHI; DURELLI, VINICIUS H. S.; ELER, DANILO MEDEIROS; IEEE. On How Characteristics that Hinder Test Data Generation Using Symbolic Execution Combine: An Analysis of the SF100 Benchmark. PROCEEDINGS OF THE 2016 35TH INTERNATIONAL CONFERENCE OF THE CHILEAN COMPUTER SCIENCE SOCIETY (SCCC), v. N/A, p. 6-pg., 2016-01-01. (14/08713-9, 13/03452-0)