Generating Test Data for Branch Coverage

  • Authors:
  • Mary Lou Soffa;Aditya P. Mathur;Neelam Gupta

  • Affiliations:
  • -;-;-

  • Venue:
  • ASE '00 Proceedings of the 15th IEEE international conference on Automated software engineering
  • Year:
  • 2000

Quantified Score

Hi-index 0.02

Visualization

Abstract

Branch coverage is an important criteria used during the structural testing of programs. In this paper, we present a new program execution based approach to generate input data that exercises a selected branch in a program. The test data generation is initiated with an arbitrarily chosen input from the input domain of the program. A new input is derived from the initial in-put in an attempt to force execution through any of the paths through the selected branch. The method dynamically switches among the paths that reach the branch by refining the input. Using a numerical iterative technique that attempts to generate an input to exercise the branch, it dynamically selects a path that offers less resistance. We have implemented the technique and present experimental results of its performance for some programs. Our results show that our method is feasible and practical.