I was gonna say the same thing, but the tests are not rigorous enough so it's really the creator's fault.

In the C# version of this the tests are all using the same reference to the array so if your function modifies the array inline it will not behave as intended.

edit: To people saying this is intended, what if you write code which abuses this fact? It's probably possible to pass all tests with enough trial and error.

this is a case where you can purposefully cause side effects upon the test data between tests. not good. sloppily written tests.