ÁùºÏ²ÊÖ±²¥¿ª½±

Boundary value testing is a subtype of equivalence class testing and a common testing method for creating test cases. Boundary value testing is a testing method facilitated by our testing tool TPT.

During test modelling it is worthwhile to create supplementary test cases covering the values at the (inner and outer) boundaries of any ordered ranges you specified. Those test cases can be created by way of a boundary value analysis.

Common Errors at Boundaries

Testing the boundaries of ordered ranges oftentimes identifies common human-made programming errors. 

A common error at boundaries has to do with having used ¡°>¡± rather than ¡°>=¡°. 

Moreover, when an event is specified to take place after x seconds, it is not always clear whether the event should commence starting at the 0th cycle or the 1st cycle, for instance.

Another error-prone operation involves the result of a calculation being outside of the range of digits that can be possibly represented (integer overflow).

Boundary Values¡¯ Special Role

Boundary values are assigned a special role which is why it helps to specify the boundary behaviour of your model as much as possible.

Moreover, boundary values are an essential test exit criterion that is oftentimes required in test plans.

In TPT you can learn more about equivalence classes and boundary value testing via the ¡°Help¡± button and then choosing ¡°Equivalence Classes¡±.