Using white-box-test the implementation of a component is tested and not the specification. So the developer is allowed to review the sourcecode during generation of the necessary tests.
The goal is to meet specified criteria of sufficiency. A general correctness of the tested component is not guaranteed. Therefore this method should be combined with other testing methods.
The black-bos-tests are part of software testing. Using this method, developers create tests before they develop the corresponding component. The target to reach is to test the specification of the component and not the implementation.
The disadvantage of this method is that new functions are added to the components because there are new demands and so on. The software tests will still work but the new features are not tested. So there is the need to use additional testing methods.
The test-driven Development is a part of software testing and describes that the tests have to be created before the development of the components. So called grey-box-tests are used which are a combination of the advantages of white-box-tests [1] and black-box-tests:
1. The software tests are written by the developer of the component himself (white-box-test)
2. The test is written without knowledge about the implementation of the component
The advantage of this method is that the tests are not adapted to work with the component, the component has to work with the test. In face of this advantage you should also use additional methods of software testing.