Since the mid 1970s, software testing has been dominated by two majorparadigms, known as black box testing and white box testing. Strategies forblack box testing are based on the specification of the software componentunder test. Test suites developed using the black box paradigm reveal errorsagainst the functional requirements of software components. Strategies forwhite box testing are based on the internals of software components. Whitebox test suites are derived from the source code of the software componentunder test. Because programs can be represented in terms of graphs, solidcoverage criteria can be defined for white box testing strategies. In this paper,we start with a brief overview of the principles underlying black boxtesting and white box testing. Hereafter, we propose a process for testingsoftware components in which both paradigms are used at appropriatepoints in the software life cycle. We also touch upon the applicability ofboth paradigms in testing object-oriented programs.
展开▼