it seems there are some border cases with selection - say you have multiple paths of the same length to attacking-positions
as I understand it you should select the one from those where end up in the first attacking-positions in reading order!
I made a mistake here first as well - I reused an A* algorithm I keep around for AoC (it was handy in every single year) and I did not really concentrate while reading so I thought it a good idea to use the simple manhatten-distance for a heuristic.
Luckily the example given showed me the mistake rather quickly.
Overall I'm actually not sure if I got it really right but my algorithm worked for my input on both parts.
If you have your input around I can test it and give you the answer (I'm actually curious if mines would be right)
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.