I did a cargo run in that directory. That “works”, but I’m not sure that was the intention. Had to play around a minute to see that I needed to hit “Enter” to continue etc.

Hmmm… noticed the Architect constructors in the out_harness directory.

Kind of looks like some of the things I’m tripping over are possibly due to inconsistencies in keeping the files synced up(???)

example:
drunkards.rs in the output_harness uses self.map.tiles.iter() for instance, while drunkards in the drunkard dir uses mb.map.tiles.iter() (since it doesn’t have a map in the DrunkardsWalkArchitect anymore.)

Thanks! I’ve filed the architect constructors under a separate issue in the book bug tracker, that one seems to be me forgetting to merge some code. I’ll make sure that’s in there.

I agree that the harness could use a better introduction. The intent is to show how useful it can be to write code against a separate tool that shows how algorithms work before moving the code into a useful build. I think I got part way there, but didn’t really manage to show the working as well as I should have. I’ll see if I can rework that a bit for a smoother reader experience.

(As an aside, I highly recommend making little test harness programs when you try something new. It really helps to distill what’s important and makes testing a lot easier)