White-box techniques

Unlike black-box techniques, white-box testing techniques focus on the structures and behavior within the test object. This is why white-box techniques are also known as structural or structure-based testing techniques.

Here too, test cases, test conditions and test data are derived from the test basis. As well as requirements (for example, to define the expected result), the test basis also includes information about the code base, the system architecture, design details, and other aspects of the structure of the component(s) you are testing.

Running a test analyzes the inner state of the test object (i.e., the point of observation, or PoO) lies within the test object). In exceptional cases, you can intervene in the object’s operation—for example, if you can’t provoke the desired operating error using the component’s own interface, you will have to perform a “hard” set of the operation. In other words the point of control (PoC) is within the test object. Test cases can be derived from the structure of the application’s code or specifications (see figure 5-2). The objective of white-box techniques is to verify a specified degree of structural coverage (for example, 80% of the statements in the test object need to be covered by test cases). You can then increase the degree of coverage by systematically deriving further tests.

Fig. 5-2PoC and PoO for black-box and white-box techniques