The second one is not an issue. The first one is fixed.

Issue with function name fixed. Now it is the same as in the tests.

You didn't really get where is laying your current problem. ;) Your code isn't failing on the efficiency side, it's failing on the path reconstruction side (I've been there too... ;o )

Finding what you're doing wrong at this point might be a bit uneasy, but that's the "only" real difficulty of the kata (once you know how to get rid of the performances problem)

good luck! ;)

while I don't see your current solution, it's probably have to do with how you check if the node you're looking at has been visited already