Can you please elaborate? Who added extra test case? Can you paste test cases here.

I believe he's talking about this test case on Leetcode: ["abc","ab"], this fails using the provided implementation. Since the ordering is invalid ("ab" should come before "abc"), they expect an empty string as the result instead of returning "abc" as the correct orderi...

Adding the following snippet to the inner for loop handles this test case:

// Handle cases like ["abc", "ab"] if (j == length - 1 && w1.length() > w2.length()) { return ""; }

This is not a valid input. From the problem statement:

It is given that the input is a valid dictionary and there exists an ordering among its alphabets.

I understand it is not a valid input,. Just to be on the safer side, for the sake of interviews, we can use the following to check for such a testcase: string w1 = words[i], w2 = words[i + 1]; for (int j = 0; j < max(w1.length(), w2.length()); j++) { if(j== w1 .length()...

I struggled with the same issue with the new, invalid test cases. Here's the Python code

        # populate the graph and inDegree
        for i in range(0, len(words) - 1):
            w1, w2 = words[i], words[i+1]
            for j in range(0, max(len...