Hands-on Rust: can_place logic is "inside out"

Title: Hands-On Rust (Chapter 11: prefab)

Just played a couple of amulet-less games. With a bit of debugging, I believe that your can_place logic is inside-out. Current code starts with assumption that you can’t place and then set’s it to true if any of the positions work:

        let mut can_place = false;// (4)
        dimensions.for_each(|pt| {// (5)
            let idx = mb.map.point2d_to_index(pt);
            let distance = dijkstra_map.map[idx];
            if distance < 2000.0 && distance > 20.0 && mb.amulet_start != pt {// (6)
                can_place = true;
            }
        });

I think you need to start with a more positive attitude (you can_place unless you find something invalid):

        let mut can_place = true;
        dimensions.for_each(|pt| {
            let idx = mb.map.point2d_to_index(pt);
            let distance = dijkstra_map.map[idx];
            if distance >= 2000.0 || distance < 20.0 || pt == mb.amulet_start {
                can_place = false;
            }
        });

I’m also seeing some other weirdness:

• player stepping off into darkness,
• inaccessible floor tiles (I can see them diagonally)
• monsters on top of one another (pretty sure I’ve seen that)

I’m assuming the others are typos on my part and I have a good bit of debugging ahead of me. Will let you know what I find.

BTW, normally (i.e. in other environments I’ve coded in), every time something “weird” happened, I’d set up a unit test to catch it and then fix it (ensuring it’s never happen again). Would be interesting how unit testing i handled where there’s so much randomness involved.