PICT Case Study

There are three core principles in developing PICT which are speed in generating test, ease of use and extensibility of core engine. PICT take input of plain-text file that specified by tester as test factor (parameters) and test factor values (values of parameter) (Czerwonka, 2006). Two phases need to be passed through in order to generate test cases in PICT, preparation and generation. Firstly, all information needed in generation phase need to be computed in preparation phase. This information includes parameter-interaction structure that defined interaction of values to be covered. From parameter-interaction, set of slots was produced. Slot is a list of possible combination that can be made from available parameter. Each slot can …show more content…
UnLike AETG, pairs in PROW can either be visited once, more than once or not visited. This result occurred because PROW support constarint in their algorithm.
CTWeb is a combinational testing tool for web application. The relationship between CTWeb and PROW is, CTWeb implemented PROW algorithm along with other algorithms in their development. Test parameter and parameter value was insert into CTWeb in two ways, manually or upload the value file. CTWeb also support constraints and weight where the value can be defined by CTWeb user.
Another additional features of CTWeb is its ability to set base test suite where a list of test case was used as base for PROW algorithm. Having all information needed, CTWeb execute PROW algorithm for the second times to reduce pairs obtained from the first execution. Then, the result will be sorted according to the weight of each pairs. Considered general PROW algorithm while ignoring the pre and post PROW algorithm. The complexity of PROW algorithm can be calculated as O(n) for while statement in line 2 since it iterate until specified value is meet. Then, in for loop the maximum number is when no more remaining pair is found, n. Since the second for loop also have same maximum number, Big-O notation for this for loop is O(n2). Thus, the lower bound and final result of Big-O notation is
…show more content…
lei, 2007), IPOG-D (Y. lei, 2007), IPOG-F(M. Forbes, 2008), IPOG-F2 (M. Forbes, 2008) and PaintBall. IPOG is the main algorithm for ACTS as it is the best in term of test set size and time required to produce test set. IPOG-D and PaintBall were better algorithm for larger system while IPOG-F and IPOG-F2 produce smaller test set but it required more time. Other than that, ACTS support extension of test set. When executing a test, ACTS can produced test set from scratch or can extend available test set. A test set may needed to be extend in case of new parameter was added or new interaction strength was required. Another feature of ACTS is, mixed strength generation. This features support multiple interaction strength within the same parameter quantity. This was done by grouping the parameter together before new parameter strength was executed. For instance, a system having seven parameter, P1, P2, P3, P4, P5, P6 and P7. Parameter P1, P2, P3 and P4 need to be test in strength of 3-way but P2, P3 and P5 need to be tested with interaction strength of two. ACTS produce combination for both strength and eliminate the duplication before display the final test

Related Documents