ABSTRACT
Data flow testing is an unfortunate term because it suggests some connection with data flow diagrams; no connection exists. Data flow testing refers to forms of structural testing that focus on the points at which variables receive values and the points at which these values are used (or referenced). We will see that data flow testing serves as a “reality check” on path testing; indeed, many of the data flow testing proponents (and researchers) see this approach as a form of path testing. While dataflow and slice-based testing are cumbersome at the unit level; they are well suited for object-oriented code. We will look at two mainline forms of data flow testing: one provides a set of basic definitions and a unifying structure of test coverage metrics, while the other is based on a concept called a “program slice.” Both of these formalize intuitive behaviors (and analyses) of testers; and although they both start with a program graph, both move back in the direction of functional testing. Also, both of these methods are difficult to perform manually, and unfortunately, few commercial tools exist to make life easier for the data flow and slicing testers. On the positive side, both techniques are helpful for coding and debugging.
