anyone explain that first solution, or refer math theory about that, so confused.

DFS won't give you a fastest route to the specified field

Try to print the result here in CodeWars editor (before returning it, or whatever the case), and you'll see that it doesn't return True :P

ok was able to find the bug, I had a default initializer wich was reused between attempts. \o/ got it now !

No, that shouldn't be possible. You'd see the error here too.

Seems like python regular tests expect True / False, while random tests expect 1 / 0 (yep, weird). Due to loose types, they evaluate to same thing, so it just means you returned False, when True was expected.

Well, I hope, but I tried most of the test before/after the one that fail and they all pass.

I also notice that that in case of error like RecursionError: maximum recursion depth exceeded in comparison the kata swallow the error and return always False.

Is it possible?

All tests are correct, and there're no issues with CW's Python environment. Are you sure you're reading the logs and tests' output in the correct order?

......W..
..W.....W
....W....
WWW......
.W.WWWWW.
.WW....W.
.WW..W.W.
W....W.W.
WW.WW....

False should equal 1

Ok I have a wierd behaviour with python 3.6. On my local pc I have True for this but when I run it on codewars I get False and the result expect 1. WTF?