Play Home Contribute GitHub Join Us Discourse Staff Members GitHub Contribution Guides Team

Is there a reason why only certain methods are "plannable"?


I thought I posted this question awhile ago, but now I’m thinking I never got around to it:

When you have a programmable unit that uses plans, it seems that only certain methods (say, move, etc) can be used in that plan. Is there a reason why custom methods aren’t allowed? As it stands now, if you want a plan you’re required to use the “say” and “hear” exchange rather than just having a method call that does what you want.


All methods that work with plan() have to be coded just so to be able to continue the previous action when called again unless they are done, and it’s a very buggy and tricky process that I try to avoid whenever possible by making levels use chooseAction() instead.

I wrote up some more about the complications here:


I’ve experienced performance issues with chooseAction that I don’t get with plan. There’s also the issue that you mention with needing to wait for the previous action to finish before taking a new action. Plan has that built in, while chooseAction doesn’t.

I understand that this is probably a difficult thing to accomplish, but I would really like a good way to be able to make a method call and be able to specify “wait until this is actually done before going on to the next line”. I tried adding in my own while(test for completion){this.wait()} loops, but that gets treated as an infinite loop when the level loads, so I don’t think that’s really doing the trick.