In particular, I'd like to hear from the devs whether that was brainstormed and discarded (and why), or if it simply never came up in the decision making.

The unlikely case that a third type of creep is to be added must be considered- must then players patch their way around three separate movement systems? Four? While some might argue this is slippery-slope reasoning, I feel that the unreasonableness of such an approach for ten different systems is proof enough of it's unreasonableness for two.

I would be extremely curious to hear the thought process behind this development choice- if it was overlooked at the time, considered and rejected, considered out of scope, etc.

   PowerCreep.prototype.moveTo = register.wrapFn(function(firstArg, secondArg, opts) {
        if(!this.room) {
            return C.ERR_BUSY;
        }
        return globals.Creep.prototype.moveTo.call(this, firstArg, secondArg, opts);
    });

It looks innocuous but there's a problem here; It goes against expectations (and general OOP). Creep.prototype.moveTo is expected to only be called on creeps. Therefore any overrides, any internal behavior of moveTo, expects this to be an instance of a Creep, or a class derived from one. A PowerCreep is neither, it's an entirely separate class abusing the scope of a Creep method. There are several methods and properties like this, that could be written once and reused, but the only common base is RoomObject which for obvious reasons is too abstract for this.

I understand why a PowerCreep can't simply be a child class of Creep, too many differences. But If you find yourself writing calls like this it's time to consider writing a shared based class.

This would allow for things like hits, carry, moveTo, my, owner, etc and any other methods or properties identical between the two to be written once and reused across as many types of actors as you want.

I'm a strong proponent for having a shared base class. Due to the way prototype inheritance works, adding such a prototype to the engine would not cause any breaking changes to existing player code.

On the other hand, I totally agree with what you're saying: creeps and power creeps should inherit from a common base class, not just for simplicity, but as a general rule of programming too. I'd definitely prefer my code to simply "work" with little to no modification once I reach the power creep stage.

The real problem is that this isn't a cut-and-dry issue. There are nuances that need to be considered at all levels. I can totally understand the player base wanting something, and the developers simply being unable to deliver. They're really stuck between a rock and a hard place here.

But that's my view as a gamedev myself, so I may be biased.

This wouldn't be the first time they made a breaking change- it wouldn't even be the first time they made a breaking change without pushing it to PTR first. Yeah, it's possible that inactive players will suffer for that, but realistically if you're not paying attention to the PTR and whatnot, you don't really have any right to get upset. You're actively not participating in the game- does it make a difference if you get wiped because one of your neighbors started eyeballing your territory vs because you didn't pay attention to patch notes?

User error is not a valid excuse for bad design either.

Personally, I disagree- this is absolutely a cut and dried issue. I'm all for recognizing the nuance in complex situations, but the truth is that the longer this is left unchanged the more problematic is becomes. Precedent exists for pushing breaking changes.

Besides, note- Both Warinternal and Semp are players that use prototypes heavily. @WarInternal I know for certain uses, for lack of a better phrase, a "Prototype-based architecture". The idea that these changes shouldn't be made for the sake of players who rely heavily on prototypes is completely swatted down by the fact that those very players are the ones asking for these changes.

EDIT: Actually, would this even be a significantly breaking change? Only players who had gone into the power creep move prototype would be having problems with the change breaking, and those players would explicitly be the ones trying to fix this problem.

Not only that, was removing preboosting not a breaking change? And that was a hell of a lot more widespread than people mucking about in the power creep movement prototype.

Come to think of it, you could just as easily just point both the power creep and regular creep move commands to the prototype they're inheriting from and mark them as depreciated. That both wouldn't break anything and allow would allow people to get their code fixed.

If we introduce a LivingEntity (choose any name) class between RoomObject and PowerCreep/Creep, the empty class by itself will change exactly nothing in the behavior. Existing properties and methods will still be inherited.

If we then move the method move to it, both classes would inherit the method as-is. The move methods on PowerCreep could be removed entirely, as instead of a hacky cross-prototype call, they'd inherit them from the LivingEntity class. And If you ever wanted to introduce a third type of movable actor you'd already have a decent base class to work with.

As for the common monkey-patching that we do that is 1) never been officially supported I thought and 2) surprise, still not broken. If you monkey-patch move on the Creep method, it will still inherit from the super class, only now it's not breaking PC because they have their own copy of move, from the LivingEntity class, and we have the option of patching that if we want to affect both.

If anybody can think of something it would break, I'm all ears.

Edit:

As for the if (!this.room) check in PC move, a prototype call can still be made but it needs to stay within it's own chain. If PC inherited from LivingEntity, something like:

PowerCreep.prototype.moveTo = register.wrapFn(function(firstArg, secondArg, opts) {
        if(!this.room) {
            return C.ERR_BUSY;
        }
        return globals.LivingEntity.prototype.moveTo.call(this, firstArg, secondArg, opts);
    });

would be acceptable, if LivingEntity were the parent class to PowerCreep.

@SemperRabbit has a working branch of the code, and I can't see a single thing that would break if done properly (since seamless inheritance like this is literally the point of objects)

@artch can you name a single thing that would actually break with this change?

• grabbing Creep.move with var x = Creep.prototype.move.
• grabbing Creep.moveTo with var x = Creep.prototype.moveTo.
• grabbing Creep.moveByPath with var x = Creep.prototype.moveByPath.
• calling c.moveTo() with any arguments
• replacing the Creep move prototype with Creep.prototype.move = x; and having creep movement use the new function
• replacing the Creep moveByPath prototype with Creep.prototype.moveByPath = x; and having creep movement use the new function
• replacing the Creep moveTo prototype with Creep.prototype.moveTo = x; and having creep movement use the new function

The only thing which would break, as far as I can tell, are:

• overwriting Creep.prototype.moveTo = x; and then expecting somePowerCreep.moveTo() to use the new creep moveTo function
• overwriting Creep.prototype.moveByPath = x; and then expecting somePowerCreep.moveByPath() to use the new Creep moveByPath function

Both of these things are breaking only to people who've written PowerCreep code in the time since they've been out, and both have the easily fixable solution of overwriting Moveable.moveTo / Moveable.moveByPath. There's no way this can affect idle users, even if they extensively use the prototype system.

All you need to do is publish an announcement far enough in advance that this is the plan and anyone relying on it can prepare for it by doing this:

PowerCreep.prototype.move = Creep.prototype.move;
PowerCreep.prototype.moveTo = Creep.prototype.moveTo;
PowerCreep.prototype.moveByPath = Creep.prototype.moveByPath;
PowerCreep.prototype.transfer = Creep.prototype.transfer;
PowerCreep.prototype.withdraw = Creep.prototype.withdraw;
PowerCreep.prototype.drop = Creep.prototype.drop;
PowerCreep.prototype.pickup = Creep.prototype.pickup;
PowerCreep.prototype.say = Creep.prototype.say;

For anyone who does this or isn't relying on the buggy behaviour there are no breaking changes!

Note: I would strongly suggest that all of the offending methods above should be fixed, not just the move ones, so the class should probably be called BaseCreep or something rather than Movable.

I assumed it would break more stuff. I was surprised that Creep.prototype.move still worked the same.

I really think this is a good idea. And it fixes the bug of changes to Creep.move affecting PowerCreep.move.

@artch, idk if you have any code, but you and @o4kapuk are more than welcome to spawn in too. I'd love to get feedback on what the devs think. btw, figuring out the data() call was more tricky than I thought (see the commits from today lol)

Again, everyone can take a look at my branch here: https://github.com/semperrabbit/screeps_engine/tree/BaseCreep

How is pull handled here?

Use braces on all multi line if statements. That seems to be the style in the rest of the repo.

Consider breaking this up into two PRs. The first one creates the new base-creep type in the existing creep file renaming classes but otherwise not moving code around. The second one moves it all to the new file. This would be far easier to review as the unchanged functions would be nearly diff free. The second one would have tons of diffs but very little new code. The easier the PR is to review the more likely the devs will accept it.

We're still having memory problems with high room counts. I suspect the indexes are to blame. FIND_MOVEABLES is a fine idea but I would discourage a new index just concat the FIND_POWER_CREEPS and FIND_CREEPS calls. For completeness add a FIND_MY_MOVEABLES as well.