Testing interval trees

After implementing interval trees, I have started to think about testing them. It would be great to have some test script. To do this, I would need to think about different scenarios, that can happen – different ways of inserting and removing data can lead to different rotations, for example. Of course, I did not want to spend too much time on testing.

Therefore I have written small test program, that inserts random events into interval tree and list, does several searches, deletes some events and  does some searches again. I have found several bugs this way.

But how can I be sure, that all interesting situations were tested ? I have used great gcov and lcov tools. Almost 90% of code was covered by tests, and although the remaining code is not so important, I would like to add more tests in the future. The valgrind did not show me any memory leaks. This is my tip for your  better sleep: test your code.