• Journal of KIISE:Software and Applications
• /
• v.37 no.10
• /
• pp.768-772
• /
• 2010

Concolic testing generates test data by combining concrete program execution and symbolic execution to achieve high test coverage. CREST is a representative open-source test tool implementing concolic testing. Currently, however, CREST aims at exploring all possible execution paths. In case of testing a specific branch or block, thus, it can be ineffective. This paper suggests a goal-oriented concolic testing that generates test data to execute a given branch or block.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.123-132
• /
• 2012

This paper presents a goal-oriented test data generation technique that specializes concolic testing. The proposed technique, referred to as GCT (Goal-oriented Concolic Testing) produces test inputs which execute a specific target. Concolic testing can be seen as the brute force approach to search the space of all possible paths until a required test input is found. In contrast, GCT restricts the number of program paths that are explored by using data flow information to identify statements that should be executed beforehand in order for the target to be executed. We conducted experiments to evaluate the performance of GCT with programs with flag variables to show its effectiveness.

• Journal of Internet Computing and Services
• /
• v.9 no.2
• /
• pp.89-105
• /
• 2008

So far, most of research on automated test data generation(ATDG) deals with programs without pointers. Recently, few works hove been done on ATDG in the presence of pointers, but they ore path-oriented techniques which require the specification of on entire program path to be tested or a program to be executed. This paper presents a new technique for generating test data even without specifying a program path completely. The presented technique is a static technique which transforms the test data generation problem into a SAT(SATisfiability) problem and makes advantage of SAT solvers for ATDG. For the ends, we transform a program under test into Alloy which is the first-order relational logic and then produce test data via Alloy analyzer.