if the path was 5 norths and 5 easts, your code would say it was a valid path.

I agree. I think the author doesn't know that a null string is not the same as an empty string. In Java, we have:
String aNullString = null;
String anEmptyString = "";
aNullString.equals(anEmptyString) returns false.

Is there a drunkenDoodling method for the Java version? I tried both drunkenDoodling(str) and drunken_doodling(str). Both times, the compiler says it cant't find the function.

This kata is beautiful now!

What I am trying to tell you is that the Java version of your kata doesn't follow the form of most katas.
Here is what I see for your Kata (there are no examples, even though it says "Let's see some cases:"):
Description:

We need a function prime_bef_aft() that may give the largest prime bellow a certain given value n, befPrime/bef_prime, and the smallest prime larger than this value, aftPrime/aft_prime. The result should be output in a list like the following:

If n is a prime number will give both primes, n will not be included in the result.

Let's see some cases:

Happy coding!!

Those are not in the problem description in the Java version. The are only in the test. Look at other postings. Other postings list examples in the problem descriptions. One of the criteria for judging if a kata is ready to go beyond beta is that it have examples.

one of the expected things in the problem description is examples of the expected results. For example, "prime_bef_aft(4) should return [3,5]"

Problemm description is missing examples

I assumed that when I left a comment that it would be tagged with where I was when I made the comment. I was in Java. The test code doesn't work

test code does not test for correct return type.

I am having the same problem in Java. The tests don't compile.