Enumeration Problem in Evolutionary Testing

Evolutionary test is just the rise in recent years, automatic test data generation technology has been successfully applied in many practical engineering . However , in the white-box testing , when the enumeration program , the use of traditional fitness function to calculate the appropriate value formula terrain will lose oriented , resulting in degradation of the evolutionary search into a random search , which we call the enumeration . There is no study of the problem , similar enumeration flag variable is the research focus of evolutionary testing for the flag variable , our group has proposed a unified fitness function calculation rule . This paper analyzes the similarities enumeration flag variable , and extended to the Uniform Rules of the flag variable enumeration , and their difference is given a new iteration rule , the final enumeration the problem of appropriate value calculation rules . The experiments show that the rules in the case of failure of traditional methods to guide the evolutionary search to generate the required test data . Implanted programs in the appropriate value of the flag variable uniform rules given by the research group extended to the appropriate value of the enumeration rules implantation , the implanted code redundancy problem and code into the location selection problems, which affect implanted consideration and evaluation of the fitness function . This paper analyzes the problems and solutions , and program code into the appropriate value of the enumeration calculation rules . Finally, given the Design and Implementation of the algorithm used in the experimental test data generation .