help-development

[thorny python]

idk if it consume too many memory tho :v

[lost matrix]

WorldGuard actually just did this for the longest time without any major problems XD

[thorny python]

oops

ya indeed easy solution

[lost matrix]

Calculating if a vector points in a BoundingBox is quite cheap.
You can easily do several hundred checks in a tick without noticing it.

[thorny python]

hm

I have to check the problem size first, thanks

[midnight shore]

Is it possible to check it a player is in a box region?

[lost matrix]

Yes

[midnight shore]

How?

[lost matrix]

His location is just a vector. Check if this vector is in the BoundingBox

[thorny python]

x > x1 and x < x2 and z > z1 and z < z2

[midnight shore]

Ok

[thorny python]

:v

[lost matrix]

Something like

  public boolean isWithin(final Player player, final Location cornerA, final Location cornerB) {
    return this.isWithin(player.getLocation(), BoundingBox.of(cornerA, cornerB));
  }

  public boolean isWithin(final Player player, final BoundingBox box) {
    return this.isWithin(player.getLocation(), box);
  }

  public boolean isWithin(final Location location, final BoundingBox box) {
    return box.contains(location.toVector());
  }

[charred blaze]

how can i use " in messages?

nopermplayer.sendMessage(ChatColor.translateAlternateColorCodes('&', "Unknown command. Type "/help" for help."));

this gives me error

[hollow bluff]

Send the error then

[charred blaze]

Multiple markers at this line
- help cannot be resolved to a variable
- Syntax error on token "" for help."", delete
this token

[lost matrix]

Try to escape them "

\"

[charred blaze]

can you give me example

how to

[glossy venture]

"Unknown command. Type \"/help\" for help."

you need to escape quotes in strings

otherwise java will think its not a string anymore

and treat it like code

[charred blaze]

thanks.

[midnight shore]

Ty

[next stratus]

hey, I'm trying to import https://github.com/yannicklamprecht/WorldBorderAPI into my plugin but none of the repos work anyone got some advice?

[tardy delta]

https://github.com/yannicklamprecht/WorldBorderAPI/blob/main/how-to-use.md#how-to-add-it-with-maven search maven

[lilac dagger]

what class name do you think is appropriate for the start/stop methods?

[next stratus]

I've tried importing it via that but it can't seem to find it

[tardy delta]

depends on where do you use it for, i'd call it Service

so no errors in your pom?

[lilac dagger]

hmm, a service is actually what i was looking for thanks

[tardy delta]

hmm ._.

[simple summit]

What's the most efficient way to test if a world object instance is still loaded? There isn't an obvious World::isLoaded method.

[lilac dagger]

Bukkit.getWorld

uuid

[tardy delta]

try getting a location of it?

[spiral light]

Bukkit.getWorld() != null

[glossy venture]

maybe get a spawn chunk and check if its loaded

[tardy delta]

nah that will throw an ex probably

[glossy venture]

or does that work

[lilac dagger]

yes

[simple summit]

I was hoping for a way to avoid the hashmap lookup - and to be able to tell the difference between a world ref that hasn't been unloaded and one that just happens to be for a world that was unloaded and loaded again

[lost matrix]

If you keep a reference to a world then it will be kept loaded

[simple summit]

no, i can call Bukkit::unloadWorld and just because I hold onto that world object doesn't keep it loaded

i mean, yes the object still exists in memory - but that's not what i mean

[lost matrix]

It will be kept in memory

[ivory sleet]

It’s a bit annoying in your case assuming you rely on world objects themselves, as they’re not instantly removed when unloaded, rather they just get queued for collection whenever there are no hard/phantom references left. The lookup method is probably your safest bet.

[lost matrix]

Spigot just dereferences it

[simple summit]

ok - here's the situation. Plugin to create an arena world. I want to reset that world from time to time and I don't want the performance overhead of calling Bukkit::getWorld(String) shitloads

[lost matrix]

Generally speaking you should only reference worlds in volatile scopes

[ivory sleet]

Wat

the lookup isn’t gonna be the issue here

That’s a regular HashMap call at most

[simple summit]

hash lookups are slow vs just using an object ref

and if i'm calling a function for every block in a chunk - it adds up

[ivory sleet]

You’re just doing premature optimizations now

[lost matrix]

What? We are speaking about nanoseconds here. What are you saying?

[ivory sleet]

HashMaps are incredibly fast

Sure an object reference is faster

But they have different use cases

[simple summit]

yes they are, but 3ns is slower than 0

anyway - not the real point

[ivory sleet]

I’m out

[tardy delta]

lol

[simple summit]

question was about a way to test if a word has been "unloaded" or not - and the answer is no

not directly

[ivory sleet]

Yes

Server::getWorld

[simple summit]

have to go through the Bukkit world map

[ivory sleet]

I mean the answer is yes still

It’s doable

[simple summit]

ya, that's fine - i'll deal with it, just didn't want to miss anything "better"

[ivory sleet]

And you should be thankful it’s a hashmap and not some shitty linked list

Which would be even worse

[simple summit]

actaully a LL is faster than a hasmap for small object counts

[ivory sleet]

Not rly

[simple summit]

there's parts in the standard libraries that do full scans if under 5k entries

[tardy delta]

a what

[ivory sleet]

LinkedList scale extremely bad

[simple summit]

i forget what i was looking at when i saw that TBH

[ivory sleet]

Yeah especially when searching

[lost matrix]

And with small you mean < 3 or 4

[simple summit]

actaully, hashmaps seem to cache too

[ivory sleet]

Yes

[simple summit]

so the first call is "slow" vs successive calls for the same key

[ivory sleet]

It’s backed by an array

[simple summit]

anyway, have the info i came for - thanks

[ivory sleet]

Well should be said, a K type with bad hashCode implementation leaves you at O(n)

but since it’s String/UUID

It’s not gonna be a bad K type

Hence why it’s fast and reliable (:

[lost matrix]

I think unloading/loading might not be the best way to restore a world.
Depending on the size it would probably produce less lag to just use FAWE api and a schematic for that.

[simple summit]

async world edit?

[ivory sleet]

Smiles workload distributor guide 😌

[lost matrix]

😄

Or if its for your own project then a Docker setup is also a consideration.

[simple summit]

and no, it takes milliseconds to "restore" several region files with an unload load

wtf is FAWE?

[tardy delta]

fast async worldedit

[lost matrix]

async world edit?

[tardy delta]

it sucks

[ivory sleet]

Yeah well Ross are you deleting the regions files themselves or what?

[simple summit]

yes

[ivory sleet]

Ah makes sense

[simple summit]

i delete then and restore them from a zip

[lost matrix]

loading a world takes several ticks and lags out the server. There is no way to do that async.

[ivory sleet]

Fair, well anyhow conclusively getWorld is ya best bet if you wanna check whether a world is loaded or not

[simple summit]

it causes a long tick in the 80-150ms range, it's not ideal but doing it once an hour or so is good enough

[lost matrix]

Or you do it chunk by chunk and have +1ms for 50 ticks every hour

[ivory sleet]

Also 1 nano second is equal to 0.001 milli second and is equal to 0.000001 second

[simple summit]

nope

milli micro nano

[lost matrix]

uh oh

[ivory sleet]

Oh yeah

[simple summit]

you missed 1000x

[lost matrix]

Happens

[ivory sleet]

Then it’s even more negligible!

[simple summit]

that it does

milli micro nano peco fempto

[lost matrix]

pico

[ivory sleet]

Ross anyhow it’s good you prove our point that the map call is negligible (:

[simple summit]

but it's really not lol

[ivory sleet]

Literally is

Have you benchedmarked?

[simple summit]

and it's not the map so much as the string function calls

[ivory sleet]

wat

[simple summit]

look, you can't argue to me that a map call is faster than just using an object reference

[ivory sleet]

String::hashCode

[simple summit]

and to lower

[ivory sleet]

well, for IdentityHashMap then they may actually be equally fast

[simple summit]

but the world map is strings to worlds

anyway

[ivory sleet]

So?

[lost matrix]

You can do it several hundreds of thousands times in a single tick and wouldnt notice a difference.

[simple summit]

I'm not saying that the map is too slow for me to use

[ivory sleet]

So whats the issue then for getWorld?

[simple summit]

just that if there was some World::isLoaded i'd use it lol

[ivory sleet]

No that wouldn’t work

[lost matrix]

Btw detecting when a world is unloaded is quite the hassle

[ivory sleet]

Unless you create some super shady proxy adapter design

[midnight shore]

Hi again, how can i set the display name (above the head) of an EntityPlayer? (I'm trying to rename an NPC)

[lost matrix]

Change his GameProfile 😄

[lilac dagger]

would it be better to restore the chunk files in order to reset a world?

[tardy delta]

that goes brr

[lilac dagger]

i was thinking of creating an implementation for that

[ivory sleet]

Probably Freestyler

Altho it sounds quite dangerous

Unless you’re wiping all of them

[simple summit]

the nasty part is not corrupting shit

[lilac dagger]

yeah, all of them

[simple summit]

and not confusing clients in render range

[midnight shore]

Sorry, could you send me an example? I'm new to nms and i don't really know how to change this

[lilac dagger]

it's for a minigame

[ivory sleet]

That probably depends on what exactly you’re dealing with

[lilac dagger]

so the players won't be on the map

[ivory sleet]

If it’s about data management corruption is probably gonna be a pivot

[simple summit]

it's about taking stuff out of a zip and putting it in the world

for me anyway

[ivory sleet]

In your case yes

Yeah

[lost matrix]

Maybe this works

  public CompletableFuture<Void> unloadWorld(final String worldName, final boolean save) {
    Preconditions.checkArgument(Bukkit.getWorld(worldName) != null);
    Bukkit.unloadWorld(worldName, save);
    return CompletableFuture.runAsync(() -> {
      while (Bukkit.getWorld(worldName) != null) {
        try {
          Thread.sleep(100);
        } catch (final InterruptedException e) {
          e.printStackTrace();
        }
      }
    });
  }

[dire marsh]

what the fuck

that looks cursed

[simple summit]

or just listen for the WorldUnloadEvent event and wait a tick

[tardy delta]

final 😶

[ivory sleet]

I am not sure but maybe a PhantomReference/WeakReference of type World would be fine where you register the reference to a reference queue and then later callback

[lost matrix]

But it sounds like he keeps references to the World objects which could make the WeakRef useless.

[ivory sleet]

Yeah, then it’s probably a rotten idea to do it like that

[lost matrix]

Hm the method above might not work as intended. I think you actually need to check the file handles on the world and if they are released.

[midnight shore]

please

[ivory sleet]

Yeah, should probably also use a custom thread pool to not slow the common pool down 😅

[lost matrix]

true

[lost matrix]

NMS is not part of the API. Its undocumented and if you use it you are on your own most of the time.
I would just look into other peoples plugins and see how they did it.

Or look at the protocol and reverse engineer how vanilla does it.

[simple summit]

Humm, could the session.lock file be watched... seems like the sort of thing it should exist for actually

[tardy delta]

there's probably a better way to rename an npc

just respawn it with another name?

[ivory sleet]

Probably, you can actually register a FileWatcher

[grim ice]

isnt having everything as final and changing it to public only if u need it public better?

[tardy delta]

public variables ew

[ivory sleet]

Or sth like that idr the name exactly

[blazing scarab]

No

That breaks encapsulation and turns objects into passive data bags

[tardy delta]

then normal people just create a getter

[ivory sleet]

System.out

[midnight shore]

oh ok

[blazing scarab]

I dont have any questions to java since arraycopy method not following naming conventions, breaking SRP and taking objects exist

[ivory sleet]

Lol

Ye

[lost matrix]

Might work like this. I dont think there is a better way to check if a File is locked.

  public CompletableFuture<Void> unloadWorld(final String worldName, final boolean save) {
    Preconditions.checkArgument(Bukkit.getWorld(worldName) != null);
    Bukkit.unloadWorld(worldName, save);
    final File worldFile = Objects.requireNonNull(Bukkit.getWorld(worldName)).getWorldFolder();
    return CompletableFuture.runAsync(() -> {
      while (this.isLocked(worldFile)) {
        this.busyWait(100);
      }
    }, Executors.newSingleThreadExecutor());
  }

  private void busyWait(final int millis) {
    try {
      Thread.sleep(millis);
    } catch (final InterruptedException e) {
      e.printStackTrace();
    }
  }

  private boolean isLocked(final File file) {
    if (file.isDirectory()) {
      final File[] children = file.listFiles();
      if (children == null) {
        return false;
      }
      for (final File child : children) {
        if (this.isLocked(child)) {
          return false;
        }
      }
      return true;
    } else {
      return file.canWrite();
    }
  }

[ivory sleet]

Well you could ask yourself who wrote some of the java classes

[blazing scarab]

Also pretty sure "out" is stateless, which is, eh, acceptable

[ivory sleet]

Is it?

[grim ice]

oh right nvm

i mean

[tardy delta]

lol

[grim ice]

the opposite of final

[ivory sleet]

I mean generally speaking, any method that returns void must have a side effect in principle

[grim ice]

thats what i meant

[blazing scarab]

I think is uses some native implementation which is not shown to us

[ivory sleet]

Ah fair

[blazing scarab]

State doesnt means it has no side-effects, as said before, good objects should represents real life entity, the state object itself encapsulates is just real life entity coordinates

[ivory sleet]

Well, anything that is stateful generally has side effects, whilst stateless things do not.

[blazing scarab]

Oh wait

Yeah ur right

[lost matrix]

Tell me that a record has side effects

[blazing scarab]

I messed up definitions

[ivory sleet]

Technically at instantiation, but otherwise records shouldn’t… hopefully

[blazing scarab]

Passive data bags!!! Sometimes useful ngl

[ivory sleet]

Well, you probably know more about me in regards to the entire elegance of object orientation 🥲

[lost matrix]

Shouldnt side effects be avoided everywhere? Stateful or not? Mabye we have a different understanding about what a side effect is...

[ivory sleet]

I think of side effect as a change to the system

Whatever that change might be

[lost matrix]

For me its an implicit change to the system

So a setter has no side effects for me. But a setter that also increments a counter does.

[ivory sleet]

Ah, well I from my point of view, a setter changes the state of the system, else it would be no point in calling it as it returns void

[simple summit]

CPP has a better construct for this discussion than java - there you define a method as const if it does not change any member state

aka is purely a calculation - or IIRC purely external side effects

[lost matrix]

I think java has annotations for that. Contracts and what not.

[ivory sleet]

Yeah purely external side effect is the term

[blazing scarab]

aka procedure

[simple summit]

I don't think java (stock) has a way to say this method doesn't change member state - but there's probably an annotation for it in some library... and hey the language keeps evolving

[ivory sleet]

And even Haskell has purely external side effects… pretty much unavoidable

[lilac dagger]

final does that @simple summit

[simple summit]

no it doesnt

[lilac dagger]

along with an immutable view of whatever you want

if you want the return itself to not change

[simple summit]

final on a method means you can't override it

not that the method can't change member state

[tardy delta]

on class that you cant extend it

[lilac dagger]

and reset it too

you can't file = newFile;

[ivory sleet]

Ross believe they meant for variables

In which it’s true unless you use sth like unsafe or reflection

[full elbow]

anyone know how to disable elytra?

[simple summit]

Anyway in the spirit of this discussion - avoid public static (non-final) things as a rule and use singletons VERY sparingly. Otherwise the best way to create robust extensible software is to follow a test-first practice (I prefer BDD myself). Testability of an implementation will really guild you to good encapsulation of concepts and motivate you to use more dependency injection instead of dependency creation (it's healthy to avoid newing complex things within your member initializers and constructors.

[ivory sleet]

I mean singletons won’t really destroy your testability

[simple summit]

    @EventHandler(ignoreCancelled = true)
    public void onEntityToggleGlideEvent(EntityToggleGlideEvent event) {
        if (!config.roofPreventElytra)
            return;

        if (!(event.getEntity() instanceof Player))
            return;

        final Player player = (Player)event.getEntity();
        if (player.getWorld().getEnvironment() == World.Environment.NETHER && player.getLocation().getBlockY() > 127) {
            event.setCancelled(true);
            player.setGliding(false);
            tellCantDoThatHere(player, "glide");
        }
    }

[ivory sleet]

I do both BDD and TDD

Sometimes even DDD

[simple summit]

singletons are cancer - they have their place but they are a pattern of last resort

[ivory sleet]

Nah

Sure it’s more of a procedural pattern to some extent

But really no

They are good

[simple summit]

it's the coupling they cause that is the problem

[ivory sleet]

Well

You can for one just pass the instance to the constructor

[simple summit]

you're much better off to have a singelton factory than 20 different singletons

that's all i mean

[ivory sleet]

Or if you need some sort of lazy init, a supplier that returns the singleton instance can be used

What

Where does factory come in play here xD

It’s not impossible to mock singletons either

[simple summit]

it allows you to break the coupling problem

[ivory sleet]

We have libraries and other stuff

There is no additional coupling problem if you design your system correctly

As said

You can still pass the instance through a constructor

Explicitly or with a Supplier

[simple summit]

lol, i don't disagree - it's just easier to fuck up your system design by overusing singletons than to use them with extreme caution

[ivory sleet]

Additionally, if you want to create abstractions, wrap the singleton with an adapter

Yeah Ross you’re right

I just don’t believe it couples your system

[simple summit]

I'm talking to a "new programmer" you're talking to me about an academic argument.

[ivory sleet]

At least not from my experience

Well, all things considered, new programmers usually don’t get the concept of tight coupling

[simple summit]

my point is you're not wrong and I don't disagree - just that new players will screw themselves with it

[ivory sleet]

Indeed

[simple summit]

anyway, back to code 😉

[ivory sleet]

[vocal cloud]

[blazing scarab]

i dont think you need singletones in well-designed systems

mstbe that lazy init but i never had any usecases for such solution

[brave sparrow]

There's definitely valid uses for singletons

[blazing scarab]

for example?

[buoyant viper]

programs 😎

[ivory sleet]

I have used it for serializers mostly

Don’t know if it was well designed, but it was definitely the best choice of design in my case

[hasty prawn]

I still stand by using a Singleton for the plugin instance

[ivory sleet]

That’s arguably fine

[simple summit]

I agree that singletons are not "necessary" - they are however "beneficial in some situations". Static state is always a coupling disaster waiting to happen.

[ivory sleet]

Altho JavaPlugin::getPlugin

[hasty prawn]

that's a lot of typing

[simple summit]

they must be used with care

[blazing scarab]

Fine if they dont encapsule any state, just logic

[simple summit]

and not be the default "i'm lazy so i always use singletons" lol

a static utility class is not a singleton

[ivory sleet]

Yea

[blazing scarab]

I use singletones as well for persistent daya types

[ivory sleet]

Ah fair

I mean all singleton says is that we should at compile time ensure only one instance exists, and it should have a global access point

[simple summit]

there's nothing bad about static helper methods - the contamination comes from mutable static state

[blazing scarab]

We know right

[ivory sleet]

I mean I consider static state doable sometimes, but mostly it fucks you up in multithreaded environments

[simple summit]

I'm clarifying for all readers - not just those active in this conversation 😉

[full elbow]

pain

[grim ice]

yeah

thats also why some people dont like singletons cuz like u have to type out ur class and method name every time

[brave sparrow]

that's arguably a singleton though: there's one static instance of the plugin that you can get statically at runtime and use to modify state

[grim ice]

but cant u just

import static

[brave sparrow]

that's not really the main complaint people have with singletons lol

[full elbow]

wait what is the best website to code for beginner?

[chrome beacon]

Any website

Do what you want to

[neon minnow]

Hi guys can anyone help me I am learning more but, basically im trying to convert java to kotlin and i have this problem:
https://cdn.discordapp.com/attachments/912851861923569675/941370925704683550/unknown.png

Thank you

if i use Collection<Player> players.add(p) will go red, and just say unresolved reference

https://cdn.discordapp.com/attachments/912851861923569675/941371342085820416/unknown.png

[young knoll]

Pass Bukkit.getOnlinePlayers to the constructor of an array list

[neon minnow]

Can you show me how that would look please ?

I will try

[young knoll]

new ArrayList<>(Bukkit.getOnlinePlayers())

[neon minnow]

dont know if u saw the img (discord link) but i have problem with the for loop

add is unresolved reference.

players.add(p)

[young knoll]

You don’t need a loop if you do that

[wet breach]

you could use addAll() if you are just going to add all the players

[neon minnow]

https://media.discordapp.net/attachments/912851861923569675/941375106607898704/unknown.png ?

[young knoll]

I don’t see a new

[wet breach]

ArrayList<Player> players = new ArrayList<>();
players.addAll(Bukkit.getOnlinePlayers());

one way of doing it if you want your own array for all online players

[young knoll]

You can just use the arraylist constructor

[neon minnow]

Thanks

[wet breach]

Think you could even do new ArrayList<>(Bukkit.getOnlinePlayers);

[young knoll]

Mhm

[young knoll]

That’s what I said to do

[wet breach]

ArrayList<Player> players = new ArrayList<>(Bukkit.getOnlinePlayers);

[young knoll]

No idea how to translate that to Kotlin though

I don’t use Kotlin

[wet breach]

I didn't see that, but maybe the other way I showed might be easier to translate to kotlin?

[young knoll]

No idea

[wet breach]

like you I don't use Kotlin either

[sullen dome]

is there an easy way to create spawn egg recipes without creating one for every spawn egg?

as they are all different items

[wet breach]

probably could use an enum

place all the spawn egg items in said enum then you could easily apply the recipe regardless of which egg it is

[young knoll]

Yeah you could loop over a list of the spawn egg items and do it that way

[wet breach]

well with an enum wouldn't need to do any looping, just check which it is

[young knoll]

I think they want 1 recipe for each item

But don’t want to specify 50 recipes or whatever

[wet breach]

yeah, place all the eggs in a single enum

then it wouldn't matter which egg it is, just need to check that it is an egg of sorts

[mortal hare]

does anyone have issues with intellij

it spins up to 99% CPU usage

for no apparent reason

after a while

[wet breach]

https://github.com/frostalf/ServerTutorial/blob/master/ServerTutorial/src/main/java/pw/hwk/tutorial/enums/Permissions.java#L15

this is what I am talking about, you can put more then a single value in one enum

Now what I don't know is if you could register a recipe just using one egg, might have to create your own methods for checking?

but with an enum, could easily still do it I suppose, technically would be registering multiple recipes but just more easily done instead of typing out every registration XD

[young knoll]

It should

I would use a set but yeah

[midnight shore]

Hi, i'm doing a trail with blocks and this blocks disappear after a few seconds. I tried getting the player location but that block is just in the middle of the player, so i tried subtracting to the location, but weirdly the block appeared up the player and not down

[young knoll]

Yes

Also that isn’t a valid config path you are using

Config paths can’t end with .

[midnight shore]

this is my code

[young knoll]

Set

HashSet to be specific

[wet breach]

that middle number is the y value, which is up or down

positive values make it go up

subtracting or negative values make it go down

[midnight shore]

but i'm subtracting, not adding

[young knoll]

Subtracting 1 should be the block below the player

[lost matrix]

I need to write a text file with common responses...
You should not read a config while the server is running. Rather read it once the server starts or a player connects (if its player specific data)
and store the data in properly named variables. Configs are there for persistence. They do not exist to be abused as data objects.

[young knoll]

I mean

It’s basically just a map

[wet breach]

which event are you using? Playermove?

[midnight shore]

yes

[lost matrix]

Do you just add a Map<String, Object> to your classes or do you use proper fields?
Accessing those objects is error prone.

[young knoll]

I always use map<String, Object>

Fields are for nerds

[lost matrix]

I would use Block#getRelative() instead

[wet breach]

instead of using the getTo for the location, try getting the players current location instead. using event.getPlayer().getLocation()

[midnight shore]

okay

i'll try ty

[wet breach]

I only say try that, because some things with the events

they haven't quite happened yet

from appearance it shouldn't make a difference but you never know lol

[lost matrix]

What are you trying to do here?

[wet breach]

make a trail of blocks below the player

as they move

for some reason instead of the block going down using the subtract method it made it go up more?

[lost matrix]

add(0, 1, 0) gives the block above

[young knoll]

They are using subtract

[lost matrix]

Thats why i would use getRelative. Its more concise.

Block below = event.getTo().getBlock().getRelative(BlockFace.DOWN);

[wet breach]

That may work too

[lost matrix]

But i would heavily filter this event

[brave sparrow]

That’s not going to solve it if the subtract is returning the block above

That means the getTo is messed up somehow

Try setting that block with no offset and see what happens

[wet breach]

which is my theory

[young knoll]

Also I would use sendBlockChange

Don’t want people getting a ton of gold

[lost matrix]

  public void onMove(final PlayerMoveEvent event) {
    final Location to = event.getTo();
    final Location from = event.getFrom();
    final int toX = to.getBlockX();
    final int toY = to.getBlockY();
    final int toZ = to.getBlockZ();
    final int fromX = from.getBlockX();
    final int fromY = from.getBlockY();
    final int fromZ = from.getBlockZ();
    final boolean blockPassed = toX != fromX || toY != fromY || toZ != fromZ;
    if (!blockPassed) {
      return;
    }
    // What happens if player goes from one block to another
  }

[lost matrix]

That too

[brave sparrow]

We’re not there yet lol

They need to make it work before they can optimize it

[midnight shore]

which is? like i said before i'm new

[wet breach]

you have an error at the top. From should be getFrom

[sullen dome]

my idea was wrecked anyway. didnt know that most mobs dont even have heads lmao

[young knoll]

They do with minecraft-heads

[brave sparrow]

What happens if you set e.getTo().getBlock() instead of subtracting

Which block gets changed?

[lost matrix]

thx

[sullen dome]

yea but that would be a kind of complicated thing

i just wanted to make eggs craftable with their heads lol

[lost matrix]

Block below = event.getTo().getBlock().getRelative(BlockFace.DOWN);

This has to work

[brave sparrow]

That’s the same as subtract

[midnight shore]

the one in te middle of the player

[wet breach]

well it could be how the event works, I don't have my IDE opened to look at the implementation

[brave sparrow]

Very interesting

[brave sparrow]

Can you print out the coordinates of the getBlock() and the getBlock().subtract()?

Also you’re not jumping or anything right?

[midnight shore]

plain walk

[brave sparrow]

Ok so yeah print it out for us

Also what spigot version are you using

[midnight shore]

this works

[midnight shore]

1.17

[brave sparrow]

Ok

This is so weird

[lost matrix]

getTo gets the location at the players feet

[wet breach]

weird that subtracting is essentially doing adding though

[brave sparrow]

The issue is that subtract is adding

Lol

It makes no sense

[wet breach]

only scenario I can think of for that happening is somehow a negative is getting in there

[lost matrix]

yeah... did he subtract a negative number? 😄

[brave sparrow]

Nope

subtract(0, 1, 0)

It’s bizarre

[lost matrix]

What version is he using?

[wet breach]

1.17

[lost matrix]

Weird

[wet breach]

the only way I can see a positive doing that

is if they are subtracting from 0

[brave sparrow]

I just looked at the stash and there weren’t any changes

[wet breach]

that is if the player is already at 0

[brave sparrow]

What?

[midnight shore]

can someone come in vc?

please

[brave sparrow]

0-1=-1

Not 1

Lol

[wet breach]

I am aware of this, but if the server expects it to be positive at all times for y, this would cause some issues if it was absolutes

[neon minnow]

what is DANDELION_YELLOW for material.

cant find it

[young knoll]

YELLOW_DYE

I assume

[neon minnow]

ty

hm no

[midnight shore]

guys please

[young knoll]

Version?

[neon minnow]

1.8.8

[brave sparrow]

That would be a very specific bug lol

[young knoll]

Lol

[midnight shore]

something weird is happening

[neon minnow]

lol

[young knoll]

INK_SAC with meta

Don't remember which number is yellow, good luck

[neon minnow]

Unresolved reference: INK_SAC

[midnight shore]

the block gets placed and then it gets unplaced but for some reason also placed again

[brave sparrow]

INK_SACK

[wet breach]

indeed, but would only happen at y0 and it is possible that it has always been there just no one noticed since most people don't do much stuff at y 0 lmao

[young knoll]

I knew that

[brave sparrow]

The event fires every time they moved and you’re not checking to make sure they left the block

[wet breach]

I would have to look later at the server code though

[lost matrix]

It could be a problem with the objects that are being used. Do changes to the Location
retrieved by getTo reflect to the game state?

[wet breach]

but that is the only thing I can think of that would cause such a thing

I don't know of anything else that would

[lost matrix]

Mabye he just moves the player down

[midnight shore]

but i'm sure i'm leaving the block

[brave sparrow]

You are

Here’s what’s happening

They move

You set it to gold and cache the old state to be reset after x time

Then they move again over the same block

You set it to gold and cache the old state, which is gold because you already set it

[midnight shore]

oohhh i see the point

[brave sparrow]

Then you have it reset at x+1

Now after x it goes back to stone or whatever

And after x+1 it’s gold

[wet breach]

this seems like a plausible thing too

since the location from the event is in relation to the player lol

[midnight shore]

how do i stop that from happening @brave sparrow ?

[brave sparrow]

@midnight shore

[midnight shore]

oh

my bad

ty

[lost matrix]

This still doesnt fix the issue with revisiting

[brave sparrow]

True

You’d need some sort of Set of blocks you’ve changed and haven’t reverted yet

[midnight shore]

my goodness all of this just for a dumb trail of gold blocks

[lost matrix]

Hm im cooking up something with a PriorityQueue...

Ok i stopped writing this. Way too spoon feedy. But this is actually not a trivial problem.

[grim ice]

Aw

wanted to see it

[midnight shore]

i'm trying using a hasmap

[lost matrix]

Some fragments:

  private final Map<Block, RevertableBlock> revertableBlockMap = new HashMap<>();
  private final PriorityQueue<RevertableBlock> revertableQueue = new PriorityQueue<>();

  ...
  @RequiredArgsConstructor
  private static class RevertableBlock implements Comparable<RevertableBlock> {
    private long lastVisit;
    private final BlockData originalData;

    @Override
    public int compareTo(@NotNull final SpigotSandbox.RevertableBlock other) {
      return (int) (other.lastVisit - this.lastVisit);
    }
  }

Then polling every tick until current time is reached. Revisiting -> edit timestamp then re-insertion into PrioQueue.

[midnight shore]

😖it doesn't work

[tardy delta]

i see weird stuff there

[midnight shore]

?

i got it working i think

just adding a check to see if the block is gold

[tardy delta]

im not really following what you are doing

[mortal hare]

what is a priority queue

and why it is not for example Deque used here

[midnight shore]

Done! it is working just pertect

[ivory sleet]

PriorityQueue is basically that every element has a priority associated to it dovidas

[mortal hare]

is it something similiar to tree

with levels

also

due to how its sorted

isnt it a slow

queue

// Java program to demonstrate the
// working of PriorityQueue
import java.util.*;
  
class PriorityQueueDemo {
    
      // Main Method
    public static void main(String args[])
    {
        // Creating empty priority queue
        PriorityQueue<Integer> pQueue = new PriorityQueue<Integer>();
  
        // Adding items to the pQueue using add()
        pQueue.add(10);
        pQueue.add(20);
        pQueue.add(15);
  
        // Printing the top element of PriorityQueue
        System.out.println(pQueue.peek());
  
        // Printing the top element and removing it
        // from the PriorityQueue container
        System.out.println(pQueue.poll());
  
        // Printing the top element again
        System.out.println(pQueue.peek());
    }

i found this example

output:

10
10
15

this means that it pulls values sorted by lowest values

[young knoll]

Surly you just need a simple FIFO queue?

[mortal hare]

nah, im just researching

I would've used Deque if i needed it

[tardy delta]

An unbounded priority queue based on a priority heap. The elements of the priority queue are ordered according to their natural ordering, or by a Comparator provided at queue construction time, depending on which constructor is used.```

[mortal hare]

i've just never seen this class in my life

an element with maximum ASCII value will have the highest priority.

maximum ASCII value is 255

[tardy delta]

theres a constructor where you can pass in a comparator

[mortal hare]

then why the it is printing lowest numbers instead

in the example

[tardy delta]

aah food on my keyboard

[mortal hare]

rip

[tardy delta]

why am i even eating here

[mortal hare]

its so weird

that highest priority values in PriorityQueue are not at the head of the Queue

[tardy delta]

i never used a queue 😳

is there some way to avoid this shit tho?

[mortal hare]

put the variable inside anonymous func

or create a new final variable

[tardy delta]

Variable 'user' is accessed from within inner class, needs to be final or effectively final

smh this sucks

[eternal night]

I mean, you would want to join the completable future if you need its value on the current thread

[tardy delta]

im in the async pre login event

[eternal night]

Well then join the completable future ?

[tardy delta]

lets wonder what that does

[eternal night]

it holds the current thread until the completable future is done

[tardy delta]

wew thats good to know

[quiet ice]

Or just use AtomicReference

[eternal night]

dunno how creative an atomic ref is if you need the value afterwards

[quiet ice]

I think you could use WeakRef too however pretty sure that that will get GC'ed, but under some circumstances it might be a better option than a full blown atomic ref

[ivory sleet]

Really depends on what FourteenBrush wants, if it’s to use the value instantly in coordination linearly to the snippet they are showing us a mere ::join might be sufficient, but yeah if he wants to store it in some reference non blocking definitely AtomicReference.

[tardy delta]

i want to load the user from the database and cache it after that

to log him in when the actually joins

i think thats better to do it in the playerjoin instead of asyncprelogin

[young knoll]

I load them in the asyncprelogin

Maybe I shouldn't since they can still be denied access

[tardy delta]

hmm

[ivory sleet]

FourteenBrush no

A dos attack could make the server crash if you used PJE

[lavish hemlock]

Can't you check if the event gets cancelled?

[tardy delta]

heh?

[young knoll]

Yes I could do that

[ivory sleet]

Assume a lot players are spam joining

[tardy delta]

ah that way

[ivory sleet]

(And leaving)

Yeah

[young knoll]

I assume disallowing the login event does not fire the quit event

[lavish hemlock]

Probably not

[ivory sleet]

Yeah probably not

[lavish hemlock]

Since quit likely happens when a player has joined already

[young knoll]

Hmm, does it fire any event?

[tardy delta]

but whats an alternative then conclure?

[ivory sleet]

iirc tho, disallowing the APPLE event won’t cancel execution of PLE or sth weird like that

I mean just load em in APPLE

[tardy delta]

now were talking about apples

[young knoll]

I load data in the APPLE

[ivory sleet]

Yaa

[young knoll]

But If they get disallowed in that event, the data will stick around

Since the data only gets unloaded in the PQE

[tardy delta]

this sounds weird going from dos attack to apples 🤡

[lavish hemlock]

https://www.youtube.com/watch?v=xS-YgkrQ-Q4

[ivory sleet]

var dto = repository.getOrMake(event.getUniqueId())

storageInterface.loadUser(dto).join();

or sth

[tardy delta]

what would dto even be?

im loading them based on an uuid

[ivory sleet]

the user object from the cache

[tardy delta]

if they join they arent in cache yet

[ivory sleet]

Basically you have a cache object which is capable of creating objects on its own

Well the method would create a new user and put it into the map for the UUID given

[wet breach]

not sure why you are wanting to do it on the prelogin, I would recommend login event for getting DB stuff, but it depends what exactly the data is and what it is for. If the stuff in the DB isn't to be used right away but is just needed for some things, then use the join event. If you kind of need it right away then best to do it at the login event. In either case make sure your DB calls are done on another thread.

[ivory sleet]

and then return it

APPLE does not use the server thread frostalf

So assuming he is declaring a callback for APPLE, it should be fine

[wet breach]

don't know what apple is

[young knoll]

AsyncPlayerPreLoginEvent

[ivory sleet]

^

[wet breach]

Ah ok

[young knoll]

    public void loadUser(UUID uuid) {
        User user = loadUserData(uuid);

        this.userMap.put(uuid, user);

        Bukkit.getScheduler().runTask(this.plugin,
                () -> Bukkit.getPluginManager().callEvent(new UserLoadedEvent(user)));
    }

That's what I do

[tardy delta]

an user object stores a lot of data which comes from the database so why would i create an user object when i dont know the values? and then load it?

[wet breach]

in either case prelogin is a bit iffy for DB calls though, because a lot can happen where the player doesn't make it to the login event 😛

[ivory sleet]

Well, your data mode shouldn’t be dependent on the storage implementation

[neon minnow]

How to set an item in inventory that is enchanted? thanks

[ivory sleet]

It’s just architectural layering at its finest to some extent

Let me give you a class of mine

[young knoll]

In my case I need to load the data that early, because I may disallow connection based on that data

I just need to make sure I somehow unload it if the player never makes it to the Join event and thus the Quit event doesn't get called

[tardy delta]

like this is my user class to load
https://github.com/FourteenBrush/MagmaBuildNetworkReloaded/blob/master/src/main/java/me/fourteendoggo/MagmaBuildNetworkReloaded/user/User.java

it depends on data or it wouldnt make sense to exist

[wet breach]

this is what expiring caches are good for 😛

[ivory sleet]

FourteenBrush I see, it’s hard to explain but let me try

[tardy delta]

xd

[young knoll]

Good idea

Wonder if an existing impl of that exists that expires every X seconds unless a condition is met

[wet breach]

Guava has one

[lavish hemlock]

There's also Caffeine

Which is based off Guava afaik

[wet breach]

in is actually quite nice using it because you don't have to worry about it in the end if it isn't needed

the queue will take care of it once it expires 🙂

it expires when it isn't being used 😛

or you could have a pre-determined time to expire everything

[lavish hemlock]

If you just wanna check if the value is still be used

I'm pretty sure most caches support weak keys/values

[wet breach]

yeah in guava it has a method to get it again if it expired in the cache

[young knoll]

Yeah I want to expire if Bukkit.getPlayer(user.getUUID) == null

[wet breach]

you could easily make a runnable to do that

check the cache for any non-valid things and then force expire them

[tardy delta]

i did it like this
https://github.com/FourteenBrush/MagmaBuildNetworkReloaded/blob/5281b7ca63b2446c82dc599264cb396042605691/src/main/java/me/fourteendoggo/MagmaBuildNetworkReloaded/storage/cache/UserRepository.java#L56

[wet breach]

but since the cache items will expire anyways at some point

[tardy delta]

to keep the cache up to date

[wet breach]

you could just simply let the cache mechanism handle it

[young knoll]

Ah crap some users may not actually be a player

That may be an issue

[tardy delta]

conclure do you mean that final fields in my case would be bad wo i better use setters to set stuff when loading the user

i better wont interrupt you typing the bible

[lost matrix]

Guava caches expire content when the cache is accessed. Thats also possible.

Unless you specify a scheduled executor

[ivory sleet]

So usually we have certain classes which are serving the purpose of transferring data between two points (could be two methods, two classes or what not).

We then have a repository, or well a lookup map.

Thing is, in this situation we want to isolate the data transfer object itself from the repository. This is done by not letting the data transfer object know about the repository. However it is fine to let the repository know about data transfer objects, although make it minimal. For instance we could do sth like:

class UserRepo{
  final Map<UUID,User> map;
  final Function<UUID,User> factory;
  
  Optional getIfPresent(UUID id):
    yield ofNullable(map.get(id));

  User getOrCreate(UUID id):
    yield map.computeIfAbsent(id,factory);
}

class User {
  final UUID id;
  AtomicReference<UserSnapshot> ref;

  void set(UserSnapshot s):
    ref.set(s);

  UserSnapshot get():
    yield ref.get();

  void update(UnaryOperator op):
    ref.updateAndGet(op);
}
class UserSnapshot {
  final SomeData;
  final OtherData;

  UserSnapshot(
    SomeData someData,
    OtherData otherData
  )

  //one of these methods for OtherData also
  UserSnapshot setSomeData(SomeData d):
    yield new UserSnapshot(d,this.otherData)

  SomeData getSomeData():
    yield this.someData
}

In this case User objects are able to live without a direct data fetch from your database

However that has many benefits

For instance testability

And the fact that your user isn’t coupled to the persistence layer / storage

When you update userdata here, it’s done atomically using User::update

(UnaryOperator<T> is just a Function<T,T>)

Anyhow

Essentially you also have a database implementation maybe something like:

interface UserStorage {
  CompletableFuture<Void> load(User user);

  CF<Void> save(User user);
}```

CF - completablefuture just lazy

Now assuming you might have a class MySqlUserStorage implements UserStorage (to implement the proper load and save)

This way you can load any user object at any given point of time, which helps with a lot of things in terms of testability and scalability

[young knoll]

I think I just had a stack overflow reading all that

[modest garnet]

how do i get server tps as an int

[tardy delta]

my brain is hurting

[young knoll]

Is getTPS exposed yet?

[ivory sleet]

Maybe I should write a complete example FourteenBrush?

[tardy delta]

so i would save the actual data on the usersnapshot?

[ivory sleet]

Cache it is a better word

[young knoll]

(no)

[ivory sleet]

essentially yes tho

The UserSnapshot is an immutable data carrier which represents the user data at a given point of time, it’s very nice and thread safe

In this way we reduce the amount of mutations (by having as said, a snapshot which we delegate all the data responsibility to)

[modest garnet]

?

[young knoll]

NMS

Wonder if MD would allow TPS to be exposed

[tardy delta]

but when creating an user, it should have a snapshot or it wouldnt make sense to make a user if you cant request data on it?

[ivory sleet]

well, just instantiate a UserSnapshot and set it on the user?

[young knoll]

So a user may exist with no data

[ivory sleet]

Yeah I forgot the getter methods

My bad

[elfin atlas]

Does someone know how I can fix this error? (Protocollib)

com.comphenix.protocol.reflect.FieldAccessException: No field with type [I exists in class PacketPlayOutEntityDestroy.
        at com.comphenix.protocol.reflect.StructureModifier.writeInternal(StructureModifier.java:365) ~[ProtocolLib.jar:?]
        at com.comphenix.protocol.reflect.StructureModifier.write(StructureModifier.java:345) ~[ProtocolLib.jar:?]
        at me.katze.av.utils.VirtualEntity.destory(VirtualEntity.java:77) ~[av-0.1.jar:?]
        at me.katze.av.utils.VirtualEntityManager$1.run(VirtualEntityManager.java:47) ~[av-0.1.jar:?]
        at org.bukkit.craftbukkit.v1_17_R1.scheduler.CraftTask.run(CraftTask.java:101) ~[patched_1.17.1.jar:git-Paper-233]
        at org.bukkit.craftbukkit.v1_17_R1.scheduler.CraftScheduler.mainThreadHeartbeat(CraftScheduler.java:483) ~[patched_1.17.1.jar:git-Paper-233]
        at net.minecraft.server.MinecraftServer.b(MinecraftServer.java:1559) ~[patched_1.17.1.jar:git-Paper-233]
        at net.minecraft.server.dedicated.DedicatedServer.b(DedicatedServer.java:479) ~[patched_1.17.1.jar:git-Paper-233]
        at net.minecraft.server.MinecraftServer.a(MinecraftServer.java:1475) ~[patched_1.17.1.jar:git-Paper-233]
        at net.minecraft.server.MinecraftServer.x(MinecraftServer.java:1274) ~[patched_1.17.1.jar:git-Paper-233]
        at net.minecraft.server.MinecraftServer.lambda$spin$0(MinecraftServer.java:319) ~[patched_1.17.1.jar:git-Paper-233]

    public void destory(Player viewer) throws InvocationTargetException {
        PacketContainer packet = main.protocolManager.createPacket(PacketType.Play.Server.ENTITY_DESTROY);
        packet.getIntegerArrays().write(0, new int[] {this.entityId});
        main.protocolManager.sendServerPacket(viewer, packet);
    }

[young knoll]

So you mutate the users data

[ivory sleet]

In principle just the reference to the data snapshot

[young knoll]

So you make a new DataSnapshot when you update anything?

[ivory sleet]

Yep

[tardy delta]

but the Function<UUID,User> factory; might be something like User::new so when caching it for the first time, the snapshot will be null until the db completed to set it?

[young knoll]

That doesn't seem ideal if you are updating stuff often

[ivory sleet]

Myes

[ivory sleet]

How so?

[young knoll]

Lot of Object creation

[ivory sleet]

Jvm handles that relatively good

[young knoll]

Fair enough

[tardy delta]

i was just planning on doing stuff with getters and setters

[ivory sleet]

Like this scales a lot better generally than having a big dto with a lot of volatile fields, which means you need to implement atomically safe methods for all of the fields

[young knoll]

Technically my data is one field

[ivory sleet]

Ah

[young knoll]

Map<String Object>

[ivory sleet]

🥲

[young knoll]

It needed to be easy for addons to add new data to the user

[ivory sleet]

Maybe not the ideal case then, like a ConcurrentHashMap directly in the User could be fine

Ye

[lost matrix]

When is a CompletableFuture not lazy?

[ivory sleet]

Anyways FourteenBrush assuming you want to get some data then you could do sth like:
user.get().name();

And for changing the name:

user.get().name(newName);

And if you increment the points of a user:

user.update(oldSnapshot -> oldSnapshot.score(oldSnapshot.score()+1));
(Could probably write a nicer method to increment but everything here would be thread safe and atomically safe)

[young knoll]

I realized something annoying in java recently

parseInt cannot throw a NPE, but parseDouble can

[ivory sleet]

Good question, since I’m on phone I wrote CF as an acronym for CompletableFuture

[ivory sleet]

How come :0

[young knoll]

parseInt will just throw a NFE if the input string is null

[tardy delta]

oh god youre on phone

wrote that whole bible on phone

[ivory sleet]

Lol ye

[young knoll]

parseDouble will throw a NPE if the input string is null

[ivory sleet]

Interesting inconsistency

[tardy delta]

hmm ill take a closer look at it

[ivory sleet]

Yeah, ofc you could go with classic setters and getters, but it’s just that you’ll run into some pretty devastating issues when the plugin scales

[young knoll]

Just never add any new features

problem solved

[ivory sleet]

🥲

[young knoll]

This is why I still use Java 1

[ivory sleet]

Let’s remove JDK from JDK

[tardy delta]

is that even something jdk1

[young knoll]

No idea

[ornate heart]

When you do PlayerInventory#getContents does that include their armor?

[tardy delta]

well thanks for helping me somewhat further tho 🙉

[modest garnet]

how do you get a BukkitRunnable to stop repeating

[tardy delta]

.cancel()

[ivory sleet]

TechLewi you basically call cancel() inside run, or if you reference the BukkitTask object, invoke/call cancel() on that.

[young knoll]

I can find JDK 1.1 on oracles site

No 1.0

[lost matrix]

Or get its id and tell Bukkit to stop the task with that id

[modest garnet]

thanks, real quick any chance you know how to get the consoles response from a command ran by the plugin

[young knoll]

Oof

Some kind of listener on the logger?

[lost matrix]

Why would you want that?

[modest garnet]

because i dont want to dive into NMS yet and i need to get the server TPS, an alternative is to run the /tps command in console and get the response

[young knoll]

That is even more janky IMO

[modest garnet]

lol

[ivory sleet]

?scheduling yeah btw

[undone axle]

https://www.spigotmc.org/wiki/scheduler-programming/

[lost matrix]

Implement a custom CommandSender and let it dispatch a command using Bukkit.dispatchCommand()

[ivory sleet]

TechLewi not sure exactly

[young knoll]

I can PR a double[] getTps

[ivory sleet]

Like there are some ways of doing it

[young knoll]

But IDK if MD would accept it

[ivory sleet]

Though, none are perfect

[young knoll]

wait hold up

Nope, no getTPS in server.spigot either

[modest garnet]

well i can get the double of tps with
MinecraftServer.getServer().recentTPS however when converted to a string it gives me [D@73601f1d

[young knoll]

It's an array

TPS from last 1m, 5m, and 15m by default

[lost matrix]

@RequiredArgsConstructor
public class ScrapingCommandSender implements CommandSender {
  
  private final Consumer<String> messageConsumer;
  
  @Override
  public void sendMessage(@NotNull final String message) {
    this.messageConsumer.accept(message);
  }
  ... and a ton more

Then something like

    Bukkit.dispatchCommand(new ScrapingCommandSender(System.out::println), "/help");

[young knoll]

Huh

Less janky than expected

[lost matrix]

    Bukkit.dispatchCommand(new ScrapingCommandSender(msg -> doSomething(msg)), "/help");

[modest garnet]

hmm

[young knoll]

Although you still gotta parse the string output

And then worry about servers with different outputs

[lost matrix]

This can be used for example if you have a plugin that has a toplist in chat but no hologram support.
Just fill a hologram with the chat response ^^

Used this way back when i still used plugins from other people 😛

[modest garnet]

lol, i appreciate that however i think i just found a way by mistake lol

[shy wolf]

how to get the view in InventoryMoveItemEvent?

[young knoll]

there isn't one

That event is for hoppers

[shy wolf]

what

why

the

any way

than what is the event for move item for inv

[young knoll]

Click

InventoryClickEvent is generally the answer for anything to do with items in inventories

[shy wolf]

ok thanks

[terse ore]

How do I make an itemStack by an itemMeta?

[jagged thicket]

@earnest gyro

can u see this

[young knoll]

ItemStack#getItemMeta

[lost matrix]

The ItemMeta does not contain the Material. So you cant just convert an ItemMeta to an ItemStack.

[pearl bloom]

Hello, I've been googling now for absolute ages. If someone could help me out with an issue, I'd be soo grateful.

[young knoll]

Oh BY an itemMeta

Yeah that isn't really possible

[ivory sleet]

?ask

[undone axle]

If you have a question, please just ask it. Don't look for staff or topic experts. Don't ask to ask or ask if people are awake or available. Just ask the question to the channel straight out, and wait patiently for a reply. Make sure you use the right channel regarding the topic of your question. Create a thread in case the channel is already in use!

[pearl bloom]

NamespacedKey randomKey = new NamespacedKey(this,"ieyyyysssds");
        ShapedRecipe randomRecipe = new ShapedRecipe(randomKey, Items.getJoint());
        randomRecipe.shape(" P "," D ","LLL");
        randomRecipe.setIngredient('P', Material.PAPER);
        randomRecipe.setIngredient('D', Material.DIAMOND);
        randomRecipe.setIngredient('L', Material.LEATHER);
        Bukkit.addRecipe(randomRecipe);    

What is possibly wrong here

[lost matrix]

What lets you think there is something wrong here?

[pearl bloom]

Since I can't craft it 😄 (Spigot 1.16.4 btw)

[lost matrix]

Show the whole code and a screenshot of the crafting table pls. This looks fine to me.

[terse ore]

?scheduling

[undone axle]

https://www.spigotmc.org/wiki/scheduler-programming/

[pearl bloom]

Idk what code part you want, the Items.getJoint()?

[lost matrix]

Why is there an iron ingot?

[young knoll]

Iron ingot != paper

[pearl bloom]

My bad 😄

I can spawn the getJoint by command, so there should be no issues there

[lost matrix]

Show the method where you register the recipe in

[pearl bloom]

I've literally just thrown it into onEnable rn 😄

[lost matrix]

Hm. Does your plugin come up as green when you type /pl ?

[pearl bloom]

Yeah, a lot of other stuff I've written in the same Core plugin works. It's just this part

[lost matrix]

Did you forget to compile the plugin and add it into your plugins folder?

[pearl bloom]

Nope

[lost matrix]

Add a debug message after each recipe and check if its displayed in the console pls.

[pearl bloom]

The furnace recipe works

[lost matrix]

Add a debug message after each recipe and check if its displayed in the console pls.

[pearl bloom]

It's in the onEnable() Ofcourse it will show 😄

[lost matrix]

*If no exception is thrown before that and you actually copied the compiled jar

[lost matrix]

And you are sure that you joint your local server?

[pearl bloom]

Yeah 😛

[lost matrix]

And no exception is being thrown at any time?

[blazing scarab]

Damn drug propaganda in minecraft

[pearl bloom]

None

So it's not just me being crazy xD

I've done thousands of recipes, now all of sudden it's not working. Super weird

[lost matrix]

Where did you get your server jar from?

[pearl bloom]

BuildTools

[pearl bloom]

You are more than welcome to join the server and test xD

[lost matrix]

I have no idea... Lets throw some random fixes at it.
Register the recipes 4 ticks later.

  @Override
  public void onEnable() {
    Bukkit.getScheduler().runTaskLater(this, () -> {
      // Register recipes here
    }, 4L);

  }

[pearl bloom]

Alright, Ima just test to run the server with only my plugin

[lost matrix]

Ah important: Never reload.

[lost matrix]

You should always do that when developing your plugins. Have a local minimal setup with just your plugin and its dependencies.

[pearl bloom]

I am 😛

Now I'm removing the dependencies for my plugin

Still nada

Ill try to delay it

Delaying it didn't help either

[lost matrix]

listen for the PrepareItemCraftEvent and check if getRecipe returns null.

[pearl bloom]

Alright, booting up. Let's see what it says 😄

o.O

[lost matrix]

Let me guess. Now it works.

[pearl bloom]

No

[lost matrix]

So...

[pearl bloom]

But it realises the recipe is there o.O

[lost matrix]

Either the result is null or another listener sets the result to null

[young knoll]

Or the result is air or has a stack size of 0

Or something weird

[heavy temple]

?scheduler

[young knoll]

?scheduling

[undone axle]

https://www.spigotmc.org/wiki/scheduler-programming/

[heavy temple]

Oh thns

Thnks

[pearl bloom]

Sure, u want the ItemStack?

[tardy delta]

@ivory sleet would it make any sense to replace an entire usersnapshot instead of using the getters and setters?

[lost matrix]

Print out Recipe#getResult() in the craft event

[pearl bloom]

I actually already tried what you're suggesting. I changed to a vanilla item instead of a custom

[ivory sleet]

Imo yes fourteen brush

[lost matrix]

*If recipe is not null

[tardy delta]

ah on an user object creation

[pearl bloom]

Wow that was a tiny picture xD But the result is the actual Itemstack

[lost matrix]

Well then another listener is tempering with the PrepareItemCraftEvent

[pearl bloom]

Wdym?
Why would it be?

[lost matrix]

Set the result to null or something... no idea why-

Maybe recipes with permissions.

Are you op?

[pearl bloom]

Yeah

[lost matrix]

🤷

[pearl bloom]

I really appriciate the help atleast 😄

I've gotten a bit closer to the issue 😛

[quaint mantle]

Hi, how do i update plugin from 1.17.1 to 1.18? thanks

[tardy delta]

just change the version of your spigot dependency

[pearl bloom]

Anyone else wanna have a go why this doesn't work? :3

[spiral light]

did you test if the recipe exist and gets it shape and so on from the api after adding it

[quaint mantle]

How do i do that?

[tardy delta]

using maven?

[quaint mantle]

yeah

[pearl bloom]

I mean, the console reacts to the recipe from PrepareItemCraftEvent so

[tardy delta]

change the version in your pom and reload it

[quaint mantle]

ok thanks

[spiral light]

i would try setting up an runnabl the prints every tick the result slot

maybe its changing but not vissible or its changing but get deleted directly

[pearl bloom]

Interesting idea. I'll try it

[mortal hare]

do bukkit itemmeta objects use NBTTags internally?

[pearl bloom]

Nothing is literally changing ;-;

There is noo way... This is NOT working 😄

[blazing scarab]

Nah https://hub.spigotmc.org/stash/projects/SPIGOT/repos/craftbukkit/browse/src/main/java/org/bukkit/craftbukkit/inventory/CraftMetaItem.java

[mortal hare]

private NBTTagCompound blockData;

Yes it does

its a wrapper

https://i.imgur.com/UHdFRqz.png

i made nbt serializer for ItemStacks

its more user friendly than using yaml tags

[blazing scarab]

No

[mortal hare]

There's literally constructor

pointing to NBTTagCompound object

[blazing scarab]

Well, it stores the nbt. However, modification method like setLore does not touch the tag

[blazing scarab]

Yes. It just reads the compound tag, not actually wrapping it

[mortal hare]

    @Overridden
    int applyHash() {
        int hash = 3;
        hash = 61 * hash + (hasDisplayName() ? this.displayName.hashCode() : 0);
        hash = 61 * hash + (hasLocalizedName() ? this.locName.hashCode() : 0);
        hash = 61 * hash + ((lore != null) ? this.lore.hashCode() : 0);
        hash = 61 * hash + (hasCustomModelData() ? this.customModelData.hashCode() : 0);
        hash = 61 * hash + (hasBlockData() ? this.blockData.hashCode() : 0);
        hash = 61 * hash + (hasEnchants() ? this.enchantments.hashCode() : 0);
        hash = 61 * hash + (hasRepairCost() ? this.repairCost : 0);
        hash = 61 * hash + unhandledTags.hashCode();
        hash = 61 * hash + (!persistentDataContainer.isEmpty() ? persistentDataContainer.hashCode() : 0);
        hash = 61 * hash + hideFlag;
        hash = 61 * hash + (isUnbreakable() ? 1231 : 1237);
        hash = 61 * hash + (hasDamage() ? this.damage : 0);
        hash = 61 * hash + (hasAttributeModifiers() ? this.attributeModifiers.hashCode() : 0);
        hash = 61 * hash + version;
        return hash;
    }

wtf is this

it seems like metadata version flag or some sorts?

[lavish hemlock]

1. The fuck is Overridden

2. That looks like a strangely-named hashCode

[burnt raven]

i am using lombok to simplify my development process and i the SLF4J logger in lombok is not outputting anything to the console, is there a way i can attach lomboks logger to the server console's one

[lavish hemlock]

Well the server logger uses Log4j, not SLF4J.

You'd probably need to import the SLF4J binding.

[burnt raven]

how?

also does this post relate?

https://www.spigotmc.org/threads/how-to-implement-slf4j-logger-to-plugin.462707/

[lavish hemlock]

https://mvnrepository.com/artifact/org.apache.logging.log4j/log4j-slf4j-impl

I assume that would work.

[burnt raven]

ill try it

https://gyazo.com/01a7b629f3edbbae8e7a990e61390bc5
@lavish hemlock

[lavish hemlock]

It's probably trying to call a method from a newer version of Log4j

(We don't support 1.8 anymore btw)

[burnt raven]

um

ok, but what version of the binding should i use then

[chrome beacon]

The one that 1.8 has

Which one that is idk

[burnt raven]

case in point

[lavish hemlock]

They use this version of Log4j

[blazing scarab]

lombok moment

[burnt raven]

so what version of the binding do i need

[tardy delta]

hey @ivory sleet would i do my caching stuff in the userrepo class too?

[blazing scarab]

How does your user repo looks like?

[burnt raven]

2.0?

[lavish hemlock]

Probably this one: https://mvnrepository.com/artifact/org.apache.logging.log4j/log4j-slf4j-impl/2.0-beta9

[ivory sleet]

basically yes

repo/lookup/caching manager/glorified map whatever you call it

[tardy delta]

ah okay

thanks 💗

[ivory sleet]

[young knoll]

I like Glorified Map

[burnt raven]

@lavish hemlock it doesnt error out but it says it failed to load class still

[lavish hemlock]

Hmm shit

[tardy delta]

i think my current cache interface isnt enough

[blazing scarab]

czn u give a link

[tardy delta]

the implementation is in the same package / cache
https://github.com/FourteenBrush/MagmaBuildNetworkReloaded/blob/master/src/main/java/me/fourteendoggo/MagmaBuildNetworkReloaded/storage/Cache.java

[burnt raven]

wait @lavish hemlock the scope was just set to test, it initialized now and i get this error:

java.lang.NoSuchMethodError: 'org.apache.logging.log4j.Logger org.apache.logging.log4j.spi.LoggerContext.getLogger(java.lang.String)'

but the method exists

org.apache.logging.log4j.spi.LoggerContext.getLogger(String name)

@ivory sleet

[modest garnet]

how can i get an bit of code to sleep for x milliseconds

[tardy delta]

run it on another thread and call Thread.sleep(millis) on it maybe?

[blazing scarab]

Whats the xy

[modest garnet]

hmm um well i have a bukkitRunnable which repeats every 5s but if tps is less than 15 then the runnable will cancel and the server will restart broadcasting at custom intervals?

[modest garnet]

??

[frozen socket]

Can anyone explain to me how to do queries or edit multiple regions at a time with worldguard/worldedit?

[tardy delta]

BukkitRunnable run task timer (1, 20 *5) probably

[blazing scarab]

what are you trying to do

[wheat compass]

I'm asking here because I can't really search the forums for it easily. (keyword function will talk about java functions instead of datapacks)

I'm making a simple plugin that has a custom command to run the vanilla /function command to run a datapack function
I know this isn't the best way to do things, but I'm low on time so I'm just adapting the datapack. Is it possible to do this? If so, it feels like it would be a single line of code solution.
Sorry if this is bothering

[solid lion]

I need some help if anyone wouldnt mind

so i rewrote all my code to one of my plugins and now when i toggle it, it always says [Plugin] Toggled "on" (it doesnt toggle off) also it doesnt do what its supposed to do (its and autosmelt plugin so it wont smelt the blocks just drops the ore naturally) all other commands work as intended. ALSO ive ran debug messages for each block and it is identifying the blocks correctly without error, just not smelting as intended

[young knoll]

Bukkit#getConsoleSender#dispatchCommand iirc

[tardy delta]

isnt /function vanilla so you cant override the behaviour?

[wheat compass]

I want vanilla behavior

[young knoll]

They don't want to

[wheat compass]

Just need some way to call it within a plug-in. Thanks

[young knoll]

Ah that isn't even it

[modest garnet]

its an autorestart plugin where when tps is lower than a certian amount it will restart the server, which you can set broadcast intervals about restart in a config, but i need to sleep the code or something or just get it to wait for next broadcast interval

[blazing scarab]

Use bukkit scheduler

[young knoll]

It's Bukkit#dispatchCommand(Bukkit#getConsoleSender, <command>)

[blazing scarab]

Sadly there's no datapack api present at the moment

[heavy temple]

Who speak spanish?

[modest garnet]

can i use a scheduler within a runnable?

[lavish hemlock]

This server is English-only.

[blazing scarab]

yes

[modest garnet]

oh ok

[heavy temple]

Oh raios

Xd

[tardy delta]

hola is that spanish?

[wheat compass]

Thank you

[heavy temple]

Yeah

[tardy delta]

damn i know spanish

por favor

idk what that means

[heavy temple]

Pero si puedes entender más cuando alguien te escribe más texto mucho mejor

[tardy delta]

ah yes yes english only

[modest garnet]

wait so i would create a scheduler within a runnable and another runnable in the scheduler

[tardy delta]

you dont create it

?scheduling

[undone axle]

https://www.spigotmc.org/wiki/scheduler-programming/

[heavy temple]

The meaning Is please

[tardy delta]

ah

[burnt raven]

how do i get the existing config for the logger so i can use it in another program without it looking like this:
https://gyazo.com/f510b34196a79e589ac4315ab5aa7efe

https://i.gyazo.com/f510b34196a79e589ac4315ab5aa7efe.png

[heavy temple]

I can understand a little that speak here but i have only basic english level

[chrome beacon]

Why don't you use your plugins logger?

[burnt raven]

well im using a utility i made that makes my plugin more optimized and i use lombok for logging in that util, and i need to port the plugin logger config into the lombok logger config so it doesnt look like your logging a logger

[modest garnet]

just read this but i dont see how to implement it into my current code can i send it in a paste bin

[heavy temple]

How I can connect the Twitch API that to send message when a streamer is on a live?

[hybrid spoke]

the api will tell you

[earnest gyro]

oh finally

i can see this channel

Hey guys, i'd like to ask something

Im trying to make a plugin which allows you to see your last 5 kills in sword lore.
The problem after 5 kills it stops registering kills in sword lore. (Thats because 5 line is allowed)
How can fix this? Where should i start?

[young knoll]

Delete the oldest one

[earnest gyro]

well i have think about that but i couldnt make it properly

Can you point me where should i look at for this?

[young knoll]

I would keep the last 5 names in the PDC and then modify that first, then use that to update the lore

That way you can remember which lore lines need to be changed

[earnest gyro]

i meant it as a link but thank you so much ur help

and your time

have a nice day/nightr

[velvet dragon]

[Server thread/ERROR]: Could not load 'plugins\mcPlugin.jar' in folder 'plugins'
org.bukkit.plugin.InvalidPluginException: No public constructor

Im getting this error and im not quite sure why.

import org.bukkit.Bukkit;
import org.bukkit.ChatColor;
import org.bukkit.plugin.java.JavaPlugin;

class Base extends JavaPlugin {

    public Base(){}
    
    @Override
    public void onEnable() {
        Bukkit.getLogger().info(ChatColor.AQUA + "onEnable" + this.getName());
    }

    @Override
    public void onDisable() {
        Bukkit.getLogger().info(ChatColor.AQUA + "onDisable" + this.getName());
    }
}```
This is all i have so far with my plugin.yml like this
```yml
name: JansFirstPlugIn
version: 0.1
author: MrUnkn0wn
main: Base

and this as my project strukture

nevermind cant send screenshots :x

forgive me if this is a stupid problem its my first time creating a plugin

[earnest gyro]

i guess "main: Base" this is the problem

[velvet dragon]

i thought so too but when i changed it it instead told me that it cant find the class

[earnest gyro]

you should use the whole class name

[velvet dragon]

so as far as i can tell it is finding the class atm

[earnest gyro]

then you should add the class name at the end