I found that some plugins, such as FastAsyncWorldEdit (on 1.12.2, not sure about newer) can cause chunks to unload inconsistently, which breaks chunk-based reset system and sometimes, even makes the world not reset at all, due to failed chunk unloads

this update fixes that problem by detecting if world has been edited improperly by such plugin and then falling back to world-based reset. If you're using FastAsyncWorldEdit and want more performant world resets, it is highly recommended to remove it, as it causes problems after a single block edit in any world, with the problems appearing in all worlds.

Additional improvements:

- moved all of disk operations to separate thread, the most intensive disk operations already were on separate thread, but now I've moved all small ones.

- Fixed a bug where the script was counting script reload as server startup

- Added custom events! on world reset start and on world reset complete, more on them will be explained below.



due to much increased amount of separate thread operations, there's a lot of waiting for other thread to finish, which makes worlds reset a little slower, but at better performance & absolute reliability, you can expect worlds reset to start taking up to 0.2 seconds more, but that is just waiting, no lag.

with increased reset times, comes a need to know when exactly the world has finished resetting, due to which I've added custom events!

you can tell when world starts to reset in any of your scripts with use of these events:
then to know when the reset has finished, you use this event:
the on complete of world reset event happens when the world has absolutely finished resetting and is available to be used for anything you wish

It is highly recommended to update to this version if you're using an older version of FastWorldReset as these changes make this script much more reliable and makes it perform far better on HDD disks.