coppro wrote:
While I don't have the code in front of me, couldn't you just add a virtual method called bSwordOnHalfTurn or something similar to determine whether the hit is processed?
Well, isn't that just the same than a check at the beginning of ProcessSwordHit? The difficulty isn't where to put the main code that bans a sword from hitting - it's deciding how to mark swords that *should* still activate on half-turns. Still, it's possibly a good idea to create a seperate bSwordOnHalfTurn method just to keep the checks self-contained and readable, especially given what checks should be made aren't quite set in stone.
The problem, I think, is that although hasted roles should definitely cause damage on half-turns (and in the patch I wrote, I classed that as the player and his Mimics), there are many places where a sword should still cause damage on a half-turn, simply because it's just appeared and didn't cause damage on the full turn. If the player deploys a Decoy, or a character with a sword appears right on top of a bomb, these really need to be handled.
So the exception is "
when a sword moves to a place it wasn't before"
. There's two ways this can happen: either something with a sword has moved during the half-turn, or a new sword has appeared (either a new role appeared or a sword was unsheathed).
The first (movement) possibly needs to be caught somehow, but I don't know how to do it. More to the point, I can't think of a current way for anything but the player and his mimics to move on a half-turn. Even characters will wait until the next full turn to do move commands. And even if one could without being hasted itself, we'd need a way of checking that they moved during the half-turn. I don't think things like sword orientation or prevX/Y work in this instance, since they don't get updated on half-turns if the monster doesn't move. So perhaps another boolean? Or maybe cross that bridge when we actually come to it....
The second was the one I was trying to focus on: the appearance of a sword where one wasn't before. This means retaining data of what state the sword was at the start of the half-turn or the end of the preceding full-turn. It also means catching things like characters using Appear or AppearAt and placing doubles. I'm not sure I caught all the possibilities though, hence some of the problems.
One solution would be that mimics (even preplaced mimics that don't move) are hasted while Beethro are. That kind of makes sense, right? (Or not...) Meh! I don't know. I didn't even reflect over the fact that this behaviour was a quirk when I saw it. But you are right of course, although I do want to point out that I don't think anybody has "raced to exploit something instead of reporting". And I'm not even sure that a change would actually break any rooms in any holds. I know it will effect some.
