Define black box and white box testing. What are the advantages of each approach? Why are both necessary? b) Consider a program component Binary_Search (list, searched_string) which search a string in an array of maximum 100000 elements. i) What are test cases you would like to test this procedure based on equivalence classes? ii) What would be additional test cases based on boundary values. c) Path coverage testing requires that every possible path through the code be tested at least once. Decision point coverage testing requires that each outcome of each decision point be tested at least once. Why does decision point coverage usually require fewer test cases than full path coverage? What kinds of errors might this testing miss?