It may happen, that test results do not match our expectations. If the tested function is not trivial, we can use debugger to step through the function and examine variables. However, if we simply set a breakpoint in the function, it will be detected as invalid test termination and the test will end. This means that winIDEA will restore target state as it was before the test run. To avoid this and enable debugging, we can set the debug mode flag in testIDEA. It is shown as toggle button in the testIDEA toolbar:
Procedure for debugging is therefore the following:
The debug functionality described above is flexible, but requires several operations to start debugging. To make things easier in the common situation, when we want to debug a function from the beginning, a single click debug mode has been added to testIDEA. Button called Quick debug button is located on the testIDEA toolbar:
With a single click the target is initialized for execution of the test, which is currently selected in the test specification tree, then execution point is set to function start, and focus is set to winIDEA. Now everything is ready for debugging.
If tests contain test-points and stubs, we usually want them to get applied during debugging, so that target state matches state during normal test run. However, when we manually step through application code in winIDEA, testIDEA has no control and can not apply test points and stubs. To avoid this, we should not step in winIDEA, but do the following:
If stub or test-point will be encountered during the run, they will be applied.
Note: Do not modify breakpoints set by testIDEA for stubs and test-points!