Fixes#1774.
This commit is designed to fix the clientside door desync issue. Based
on the order of events that I've been able to see, the way that door
opening works is like this:
1. Client sends a door open request.
2. Server echoes request back to client.
3. Both server and client simulate door opening.
4. The client that requests the initial door open sends a tile square to
the server for some reason.
In TShock, under all circumstances, we send a tile square back to the
client that sends one in, unless you have the
`tshock.ignore.sendtilesquare` permission. This adds a deviation: it
does not network data back if the event is just a door change. Doing
this is safe from the perspective of actual gameplay. A previous
iteration of this commit synchronized data to other clients, but that
seemed superfluous.
This does not really solve the underlying problem or answer the question
as to why sending a tile square back to the client seems to throw it
off, but it does. I was not able to replicate the desync issue anymore
with this branch. I expect that it will be safe to keep, because the
improved logic will only happen if the tile square had no effective
changes in addition to the door changes.
Run /sync if your doors disappear. This will resync your local client
with the server state. For more information, please see the associated
changelog entry.
Per packet debug logs are redundant for people with the packet monitor
plugin. If you need packet monitoring, please install the packet monitor
plugin.
This change applies @AxeelAnder's suggeted patch from #1845 to attempt
to resolve spawn point issues. If you remove your bed spawn point it
should send you back to the map spawn point with this.
This is in response to discussion #1859. Due to the nature of this
already terrible handler, I refuse to add any other types of checks for
whether or not this item is banned and lava. If you want to ban lava,
ban a lava bucket. Don't ban an infinite lava bucket, because finite
lava is still infinite lava in this topsy turvy game where the client
rules the world.
It looks like the only time the client informs the server of the spawn location is via the PlayerSpawn packet. In SSC, understandably, we don't trust any updates from the client if we can avoid it. In this scenario, we have code in PlayerSpawn to check for beds when they attempt to spawn, but no code to handle negative spawn locations. The client sends -1,-1 coordinates when the player has no spawn. We should trust this.