help-development

[ivory sleet]

Ugh I mean it could theoretically work across multiple nodes

Tho that being said I was just referring to the algorithmic stampedlock and java’s implementation of it

[orchid trout]

is it ok to do UserManager.updateBoard(this); in a user object?

[edgy crystal]

how do i add craftbukkit in 1.20.1?

[tender shard]

do you want to use NMS classes?

https://blog.jeff-media.com/nms-use-mojang-mappings-for-your-spigot-plugins/

[edgy crystal]

yes

[tender shard]

?nms

[undone axle]

https://blog.jeff-media.com/nms-use-mojang-mappings-for-your-spigot-plugins/

[edgy crystal]

thanks

[ivory sleet]

Might be a bit annoying if you are in an enterprise where unit tests and proper decoupling is nice to have

But else Id say rather fine

[echo basalt]

eh static

[sterile breach]

Yes. For example, in theory, if in async on an arraylist I remove a value but at the same time I get it on another thread. It's impossible to predict.

[echo basalt]

I think this is the only time I've coupled a user and usermanaged objects

[edgy crystal]

Malformed \uxxxx encoding.

[ivory sleet]

Arrays are different bro

[edgy crystal]

this comes if i reload my dependency

[tender shard]

?

?paste your pom

[undone axle]

https://paste.md-5.net/

[ivory sleet]

usually you wanna learn how the list is used if you do it in a multithreaded environment

[echo basalt]

multithreaded lists

might as well use a set

[ivory sleet]

cuz just wrapping it with a lock is ass

[edgy crystal]

https://paste.md-5.net/elemiyixim.xml

[sterile breach]

or a hashmap concurent will be the same. If it does a get after removing the value but the get "request" arrives before the delete request

[tender shard]

seems fine except that you didnt properly add the dependency nor did you add the specialsource plugin

[ivory sleet]

Wdym?

I think you should learn abit about concurrency

Right now you’re asking things that could very much be answered if you knew abit more about the topic

[orchid trout]

so i shouldnt have this in my user object

[tender shard]

why are you casting a player to HumanEntity

Player extends HumanEntity

[echo basalt]

I got no clue why you haven't made a getPlayer method yet

[orchid trout]

player is OfflinePlayer

ill rename it

[tender shard]

oh ok

[echo basalt]

I'd honestly treat the player class like a data class

and have facade classes for handling logic

or well

[ivory sleet]

That’s how you should have it

[echo basalt]

treat it like a wrapper :)

Still need to javadoc this

anything fancier should just be done elsewhere

[ivory sleet]

You have one DTO and one that facades the DTO, that way you decouple functionality from what users can do from their actual data they contain

[orchid trout]

im moving all the methods from user to usermanager

[sterile breach]

let's imagine

we have a concurent hashmap containing 1, dog
we create thread x and thread y
on thread x we try to delete 1 (and dog)
a few ms later on thread y we try to recover the value of 1.
and the "request" from thread x arrives a few microseconds after thread y. Is this impossible?

[sterile breach]

yeah I'll go and find out

[echo basalt]

It's possible in the context of networking but not in concurrency afaik

[ivory sleet]

Recover the value?

That’d only be possible with a callback

[sterile breach]

yes

[ivory sleet]

I mean its possible but why

No one designs it that way

It makes more sense if you have a thread safe map where
Key=A
Value=1

and u wanna increment value of A on both thread Alpha and Beta

[echo basalt]

he's basically asking this

Map<String, Integer> map = new ConcurrentHashMap<>();

map.put("joe", 123);

Thread one = new Thread(() -> map.remove("joe")).start();
Thread two = new Thread(() -> {
  int number = map.getOrDefault("joe", -1);
  assert number == 123;
}).start();

In my understanding the assertion always fails
It depends on who fires first

[ivory sleet]

using map.put(map.get) is problematic

so instead we’d use map.compute()

[sterile breach]

no, but several players doing things at the same time could lead to this. but is it's very imporable

[echo basalt]

That's literally why ConcurrentHashMap exists

[ivory sleet]

Delor thats where ur design discipline comes in

[ivory sleet]

Yes but that’s a rather improper way of simulating RC

[sterile breach]

prevents that?

[ivory sleet]

Read what I wrote earlier delor

[sterile breach]

okay

[ivory sleet]

Nah ConcurrentHashMap is just thread safe

[echo basalt]

^

[tranquil prairie]

Why does this code cause Caused by: com.comphenix.protocol.reflect.FieldAccessException: Field index 0 is out of bounds for length 0

PacketContainer destroyPacket = new PacketContainer(PacketType.Play.Server.ENTITY_DESTROY);
WrapperPlayServerEntityDestroy destroy = new WrapperPlayServerEntityDestroy(destroyPacket);
destroy.setEntityIds(new int[]{entityId});
destory.sendPacket(onlinePlayer)

[echo basalt]

Some versions use ints, others use intlists

I'd advise against the packet wrappers for this exact reason

[tender shard]

WrapperPlayServerEntityDestroy?`Where is that from?

[sterile breach]

I'm having trouble understanding what you mean here when you talk about x,y

[echo basalt]

And instead just writeSafely all 3 types

[tender shard]

isn't that wrapper thingy library dead since like 8 years

[ivory sleet]

and it means that regardless of how many threads u mutate the map on at the same time, the map will not freak out

[tranquil prairie]

nah theres a new one just not made by comphenix

[ivory sleet]

Actually tell me more about what you’re trying to do

[tender shard]

i'd just use the NMS package, then you see from the constructor which fields / types there are

[tranquil prairie]

I dont like using nms because of the problems is causes on different versions

[tender shard]

it can take both an IntList or an int array

[echo basalt]

That's the constructor

protocollib wraps the field

not the constructor

[tender shard]

well but apparently the wrapper is causing some issues

[echo basalt]

internally it uses an intlist

[tender shard]

the actual field is an IntList

[echo basalt]

here's a shitty simple way of doing it

[tranquil prairie]

so will just getIntegers().write(0, entityId) be good?

[ivory sleet]

Anyway @sterile breach regardless if you’re on just ONE jvm, or multiple JVMs and Redis you wanna ensure consistency, which means avoid indeterminacy

[echo basalt]

public static PacketContainer createEntityDestroyPacket(int entityId) {
  PacketContainer packet = new PacketContainer(Server.ENTITY_DESTROY);

  packet.getIntegers().writeSafely(0, entityId);
  packet.getIntegerArrays().writeSafely(0, new int[]{entityId});
  packet.getIntLists().writeSafely(0, List.of(entityId));

  return packet;
}

works across all versions

[tranquil prairie]

thank u

[echo basalt]

maybe use collections.singletonlist if you care about java8

[tender shard]

no clue, I wouldn't use protocollib i nthe first place

[tranquil prairie]

why not

[tender shard]

because it's more annoying than using NMS

[tranquil prairie]

practically every server has it

[tender shard]

and more error prone

getIntegers()... well what if it's a short in the next version? with NMS you can tell from the constructor directly what it takes as arguments

[echo basalt]

or you can just look at the nms packet class

[tender shard]

and then? then I do something like

if(version == 1.20) {
  myPacket.getIntegers()...
} else if(version == 1.19) {
  myPacket.getShorts()...
}

[echo basalt]

no

[tender shard]

I would rather just use normal NMS modules

[echo basalt]

writeSafely exists for a reason

[ivory sleet]

Yeah cuz practically most lazy plugins just use that

I’d also avoid protocollib as it compromises usability for a specific version to general practicality

[tranquil prairie]

how is it lazy it just saves time

[ivory sleet]

But thats cuz I haven’t written multi versioned plugins in ages

[tender shard]

if protocollib would have proper packet wrappers that worked in all versions, that'd be nice. But right now, protocollib just adds an additional layer without really adding any benefits

at least when you simply want to send packets, that is

[ivory sleet]

Myea it used to

In newer versions not so much

[tranquil prairie]

plus i like not having to run build tools on every version i want to support

[tender shard]

ProtocolLib:

public static PacketContainer createEntityDestroyPacket(int entityId) {
  PacketContainer packet = new PacketContainer(Server.ENTITY_DESTROY);

  packet.getIntegers().writeSafely(0, entityId);
  packet.getIntegerArrays().writeSafely(0, new int[]{entityId});
  packet.getIntLists().writeSafely(0, List.of(entityId));

  return packet;
}

NMS:

ClientboundRemoveEntitiesPacket packet = new ClientboundRemoveEntitiesPacket(entityId);

idk but I like the NMS way more

[tranquil prairie]

doesnt nms change every version tho

[tender shard]

that's why you use mojang maps

[tranquil prairie]

protocollib example supports all versions

[ivory sleet]

Slightly so

Abstraction, implementation inheritance also does

[tender shard]

unless the packet format changes, then your code compiles and then just throws errors at runtime

and as said, with mojang maps, the code basically doesn't change at all

[tranquil prairie]

ic

[tender shard]

updating my stuff from 1.19.4 to 1.20 was just changing "1.19.4" to "1.20" in my pom

protocollib is definitely handy to listen to incoming and outgoing packets but I'd never use it just to send packets

[ivory sleet]

I had to derive one interface when porting from 1.19 to 1.20

That was simple enough

[sterile breach]

I don't think it's useful for my current case because it seems imporable but i will do an example

[ivory sleet]

Well there’s only one reason u’d use CHM

If multiple writes can happen at the same time

or Ig if u wanna have some degree of atomicity

[tranquil prairie]

sorry for ugly code cleaning it up in a sec

[teal kindle]

Does the exact recipe choice support items with attribute modifiers?

[tender shard]

yes

unless you're using a shapeless recipe

[upper hazel]

how to make sure that a certain item does not fall out of the inventory game when player dies bukkit api

[echo basalt]

mans using keywords

might as well just ask chatgpt

[ivory sleet]

Lol

[echo basalt]

what the actual fuck

[upper hazel]

how cumbersome...

I understand this manually work

[teal kindle]

I've changed my recipes to shaped to avoid issues, but I'm still having one. I have two recipes: one that crafts a custom sword (call it X) with attribute modifiers and another that "upgrades" that sword into another custom sword (Y). The first recipe works great, but the second one does not, and after doing some testing, I'm determined that the item meta of X when crafted is not equal to the one the game uses to create the recipe for Y since the UUIDs of the attribute modifiers are different. This makes sense since UUID's are supposed to be different to avoid collisions, but it causes my recipe for Y to always fail. Is there a way around this?

[tender shard]

oh well yeah the UUID has to be the same - "EXACT" choice

only thing you can do is to use a normal recipe and then listen to PrepareCraftItemEvent or what's it called and change the result manually

ofc then it won't show up in the recipe book properly

[teal kindle]

Wouldn't that become performance intensive with a lot of recipes?

[tender shard]

no

[orchid trout]

i made all my database and manager classes static instead of instance based and some of the times i try to use them it says Class cannot be resolved

[echo basalt]

ej

wh

[orchid trout]

ifds

oh

same package

[ivory sleet]

What about…

no static

(:

[orchid trout]

why can i only use them if the same package

at

[ivory sleet]

What does the error say

Crack@a

Crackma*

[echo basalt]

bam

don't fuck around with the concept of code quality, it will bite you in the ass

[ivory sleet]

(And when you go against it you do look like a clown more or less) [not to frighten u crackma but others]

[glossy venture]

why do people put private constructors in utility classes, like whats the issue

[echo basalt]

uhh

[glossy venture]

if someone constructs it

[echo basalt]

There's no point in constructing it and it isn't in its design

[ivory sleet]

Because you show intent of that its a utility class outside the class itself

[echo basalt]

So you make it impossible to construct further reinforcing that point

[glossy venture]

if theres no point in constructing it it shouldnt be

by a developer

[ivory sleet]

This this this this

[echo basalt]

There's no guarantee that some idiot will understand that

[glossy venture]

theres no consequence to constructing it though

[ivory sleet]

Its like sure u can have a manager class, but why name it with the suffix Manager, no real consequence there

You can have public fields I mean there is no real harm in that

[glossy venture]

ig, if someone cant understand that its a utility class by looking at static methods tho will they realize a private constructor is a pattern

[glossy venture]

true

[ivory sleet]

The more expressive and clearer a class is just by looking at it and its properties the more maintainable and readable a system does become

[young knoll]

What else am I supposed to end it with :c

It's a class that manages islands, what more do you want from me

[glossy venture]

javadocs are more expressive than a private constructor though

[ivory sleet]

I hate javadocs

They encourage you to write long elaborate comments

Put the effort into the actual code instead

[glossy venture]

tf

[ivory sleet]

javadocs are good to define rules of a function

but way too many misuse them Id assert

[glossy venture]

i understand being against comments in code but you shouldnt have to analyze the code of a method to understand how it works

[ivory sleet]

Nope

[glossy venture]

at least not for public api

[ivory sleet]

You should understand very intention of a method merely by reading its name

[glossy venture]

its much easier looking up an implementation detail in a javadoc than estimating the whole runtime of the method in your mind which is prone to errors

[ivory sleet]

AuthorizationHandler#login

[glossy venture]

you cant pack every detail of a method into its name

[ivory sleet]

Implementation detail in javadoc?

Lmao

No bud

Look at the implementation to learn the implementation

[glossy venture]

why

[ivory sleet]

Yes thats true but if you ever get to where that becomes a big issue, chances are your method is the size of goliath

[ivory sleet]

Because javadocs are not the primary focus, they are more prone to lying to you and to not be updated

[echo basalt]

gotta love writing 10x more boilerplate to save 5 lines everywhere

[glossy venture]

something like

@throws IllegalArgumentException If the argument isn't a valid email
``` or something can be useful

[ivory sleet]

Look at java.lang.Runnable

Shits been there since java 3 or sth

And its usecase has been significantly extended

[ivory sleet]

Yes but thats defining the rules of the function

[glossy venture]

yeah

[ivory sleet]

And what I mean bu that is the declaring the contract of the function

Thats food

And good

[slender elbow]

api contract != implementation details

[glossy venture]

no shit

its still a javadoc

[slender elbow]

which is not implementation behaviour

[glossy venture]

so

[mortal hare]

wtf happened to intellij recently

[slender elbow]

so, like conclube said, if you want to learn the implementation, you go and read the implementation

[ivory sleet]

orby, let me ask you, what color does ur ide paint ur javadoc comments in?

[mortal hare]

i started to freeze loading my projects

for no reason

[glossy venture]

gray

[mortal hare]

and i need to close it via task manager

[ivory sleet]

Yeh exactly its the “dont look at me grey” “forget about me grey” “oh im here but u cant see me grey” or dark green for that part

[glossy venture]

javadocs are displayed differently if its not your source file in intellij

[tender shard]

i like writing javadocs 🥲

[glossy venture]

also its on the method when u hover over it

[ivory sleet]

The colors used by ides are made to make the comments and javadoc a secondary focus, because when u look at code u wanna primarily (probably) read the code and not the comments

Yea they are orby, but usually u only get the first lines or sth no?

[glossy venture]

its probably because as a developer you dont want the javadocs you wrote for other people in your face

[ivory sleet]

I mean I have my javadoc and comment colors set to bright orange

[glossy venture]

i like the gray

only reason i put comments in code

[ivory sleet]

Well if I see a comment in a code it better be telling me something that wasn’t properly possible to communicate with through the code itself

[glossy venture]

dont look at my code lmao

[ivory sleet]

I shall not then :>

[slender elbow]

actually changed by jdk 19 👀

[tender shard]

how about this

[mortal hare]

dude i've restarted this project

4 times

[glossy venture]

good

[mortal hare]

and it still bootloops

wtf

[ivory sleet]

Yeah sure

[ivory sleet]

Shii my bad

[ivory sleet]

Tho alex personally I try to make my comments talk about the code they’re commented over/under, not some other code in some other module

[glossy venture]

long but prob needed because its substituted

the logic isnt in the method body

[mortal hare]

comments shouldnt be that long jesus

[tender shard]

wdym? it's a JsonSerializer / JsonDeserializer for gson

[glossy venture]

you cant really derive shit from the method body

so i had to put long comments

[ivory sleet]

That was general criticism

Not particularly thrown at u

Yeah orby what I was saying are ofc just conventions, or well I’d say they are but knowing when to avoid those are just as crucial

Sometimes you have to talk about other modules in a comment, or talk about implementation, given its justifiability

[glossy venture]

yeah ig

javadocs prob arent needed most of the time

unless its an interface

like bukkit api

[ivory sleet]

Myea

[tender shard]

well javadocs makes a promise, the implementation doesn't

if javadocs tell me something returns a mutable list, I know it's a mutable list. If I only look at the implementation and see there that the list is mutable, well - that could change in the next version

[glossy venture]

true

[rare rover]

how would one detect when someone right/left clicks a client side block aka packet

[ivory sleet]

I think java did a horrible design mistake regarding this mutability design speaking of which, they just violate liskov’s substitution principle straight of making immutable lists a private implementation and then just putting that into javadocs instead, well at least we’re getting sequenced collections

[hazy parrot]

had same issue with spring yesterday

are you running project in wsl ?

[honest adder]

Hello !
I made a minecraft plugin in 1.16.5
The problem is that when I want to apply color to message in my config.yml, the default color remains (white):
Main : https://paste.helpch.at/daweteyuti.scss
FlySpeed : https://paste.helpch.at/uyohulaweb.typescript
WalkSpeed : https://paste.helpch.at/erataxudug.kotlin
Color : https://paste.helpch.at/netigadiva.typescript
Config.yml : https://paste.helpch.at/cewukukani.yaml

[placid moss]

console doesnt have spigot colors

[honest adder]

Yes I am talking about the messages in the config.yml

[placid moss]

are you sure the config is being loaded in right (the messages loaded inside ofthe code match with the message in the config)

[honest adder]

Yes as u can see

[tender shard]

print out the return value of getConfig().saveToString(), does it include the color codes?

[honest adder]

I create a value with these functions and I print the result?

[tender shard]

just do System.out.println(getConfig().saveToString()) in onEnable

[honest adder]

Ok one sec

[tender shard]

as expected, your config does not include any color codes

[honest adder]

https://starb.in/XCq7sO.md

Arfff

[tender shard]

is it possible that only your included config.yml includes the color codes but the actual saved config.yml does not?

[honest adder]

I did not understand, the color codes are present in the config.yml I use a Color class to translate and in my FlySpeed ​​​​class you can see how I use all that to use & for the colors

[kindred pasture]

Hey guys I need advice on directional object. So I disabled vanilla crop growth and have my own system and when a pumpkin or melon grows I manually convert a PUMPKIN_STEM into a ATTATCHED_PUMPKIN_STEM. The issue is I need to set the direction data on the block with Directional#setFacingDirection(face); but BlockData#setBlockData() does NOT take a Directional object. So now Im stuck.

This is a snippet of the code. https://pastecode.io/s/3npr0pj5 I tried playing with casting but it doesn't seemt work?

[tender shard]

in which config.yml?

[kindred pasture]

I know setting the type to ATTATCHED automatically connects it to the fruit but I later check the direction of the stem and just setting the type seems to make it not an instance of Directional which is weird

[tender shard]

the one included in your .jar, or the one in your plugin's data folder?

[honest adder]

Config.yml : https://paste.helpch.at/cewukukani.yaml

[remote swallow]

Pass it the directional instance anyway

[kindred pasture]

It doesnt take it, i can't its a error

[remote swallow]

Is there a seDriection something

?kd-s

?jd-s

[undone axle]

https://hub.spigotmc.org/javadocs/spigot

[kindred pasture]

on BlockData? sadly no

Only on Directional

but .setBlockData doesnt take Directional whatsoever so

I also tried casting the directional back to BlockData to pass it in

But unfortunately that makes it not an instanceof directional which I need

[honest adder]

My config.yml are in my file of my plugin -> but included in jar (u can use my code to see if u want)

[remote swallow]

Are you using the right one, is a subclass of BlockData which means you can pass it to setBlockData

[young knoll]

You are likley using the other one

[tender shard]

org.bukkit.material.Directional != org.bukkit.block.data.Directional

you have to use the block.data.Directional import

[kindred pasture]

ohhhhhhhhh

theres two??

sheeesh

thank u sm!

[ivory sleet]

Bukkit troll :,)

[kindred pasture]

rip

[honest adder]

For my problem I rechecked but yet the code seems right to me I don't see where it's wrong

I had made a slightly simpler plugin and the code is identical in terms of managing the config / colors è_é

[tender shard]

i don't really understand. it looks like you have an old config.yml in the plugin's folder that does not include color codes

and the color codes are only inside your default config in the .jar

[honest adder]

Wait I think I understood what you were telling me I will try something thank you

[00:19:05] [Server thread/INFO]: player_only: '&cCette commande doit être exécutée par un joueur !'

Thanks i think its work i will try IG

Its work thanks !

[tender shard]

np

[cinder abyss]

Hello, in BlockPistonExtendEvent, e.getBlocks contain blocks before being pushed or after ? (mushroom blocks after being updated or before ?)

I think it's before, because it's will

thanks the doc

[tender shard]

the event is also cancellable, so it must be before moving

[cinder abyss]

this is smart

it's perfect then if it's before

[cinder abyss]

if I change a block's data, it will changes after pushing?

[tender shard]

i guess so

[cinder abyss]

okay

just setting the mushroom block to his state before pushing

will work on both retract and extend events

almost perfect```java
@EventHandler
public void onMushroomBlockPush(BlockPistonExtendEvent e) {
for(Block block : e.getBlocks()) cancelMushroomStateChange(block);
}

public void cancelMushroomStateChange(Block mushroom){
    if(!CUSTOM_BLOCKS.contains(mushroom.getType())) return;
    MultipleFacing facing = (MultipleFacing) mushroom.getBlockData();
    mushroom.setBlockData(facing, false);
}```

filtering e.getBlocks() will be better

not working

maybe I need to delay

3 ticks

lol

just need to get piston relative and add 1 🤓

[tranquil prairie]

Hey how do I "reload" a player? Something I am doing only works if I teleport a player into another world and then teleport them back as that reloads the player for everyone. How do I have the same effect without teleporting them?

[cinder abyss]

what are you doing ?

that reload the player

[tranquil prairie]

What?

I am doing something with armorstands and it needs to reload the player otherwise anyone else wont see it

[cinder abyss]

well, I don't see what can do that

[tranquil prairie]

well nick plugins have to

but also this thing im doing apparently needs it as it only works if i reload the player in a bad way

[cinder abyss]

what are you doing with armor stand?

exactly

[tranquil prairie]

just making a nametag

but it doesnt show unless player is reloaded

[cinder abyss]

hum

well idk sorry

I could help you with mushroom blocks actually but not with armorstands :/

[slender elbow]

you'd be surprised by the amount of events that happen after the action takes place and (sometimes poorly) attempt to revert it if it's cancelled

[tender shard]

Yeah ik but shouldnt be the case here i guess

Neither

== the type

Not the itemstack

getHelmet().getType() == Material.WHATEVER

[weak meteor]

Process 'command 'C:\Program Files\Java\jdk1.8.0_202\bin\java.exe'' finished with non-zero exit value 1

i already changed enviroment vars, project sdk

what else do i do

[remote swallow]

that wont work how you want

[tender shard]

please use proper method names and early returns and also why are you using the wrapper type instead of the primitive for the health double

[remote swallow]

if they dont have a the helmet, they next 3 ifs wont run

when they could have all but a helmet

[weak meteor]

why doesnt just run everything in 1 if statement

[drowsy helm]

Thats even worse

[tender shard]

i didnt understand the goal

[weak meteor]

well idk

[tender shard]

hurt them by 4 hearts if they do NOT have full netherite, but don't hurt them when they move?

i dont get it

[weak meteor]

maybe earle returns will work

early

[tender shard]

also note that all of the equipment methods are nullable so you can't just do getType() without checking if it's null

then please refactor your code to be more readable

[livid dove]

I'd have a container class that has a boolean set to true or false if they do an e event thar gives then full netherite on equip... and then check the boolean on move.

Checking their head, shoulders knees and toes on move is a heft waste of thread lol

[young knoll]

Good thing alex has an ArmorEquipEvent

:p

[tender shard]

    @EventHandler
    public void onMove(PlayerMoveEvent event) {
        if(!hasFullNetherite(event.getPlayer())) {
            event.getPlayer().setMaxHealth(event.getPlayer().getMaxHealth() - 8);
        }
    }

    private static final Map<EquipmentSlot, Material> ARMOR_SLOTS = new HashMap<EquipmentSlot, Material>() {
        {
            put(EquipmentSlot.HEAD, Material.NETHERITE_HELMET);
            put(EquipmentSlot.CHEST, Material.NETHERITE_CHESTPLATE);
            put(EquipmentSlot.LEGS, Material.NETHERITE_LEGGINGS);
            put(EquipmentSlot.FEET, Material.NETHERITE_BOOTS);
        }
    };

    private static boolean hasFullNetherite(Player player) {
        for(Map.Entry<EquipmentSlot, Material> entry : ARMOR_SLOTS.entrySet()) {
            EquipmentSlot slot = entry.getKey();
            Material requiredMaterial = entry.getValue();
            
            ItemStack equippedItem = player.getEquipment().getItem(slot);
            
            if(equippedItem == null) return false; // No item, this cannot be netherite
            Material actualItem = equippedItem.getType();
            
            if(actualItem != requiredMaterial) return false; // This is not the required item
        }

        return true;
    }

[livid dove]

hello darkness my old friend

[tender shard]

this is how I'd check for full netherite items

alternatively one could ofc hard code it

but I like this more

here's the ArmorEquipEvent https://github.com/JEFF-Media-GbR/ArmorEquipEvent

[livid dove]

The sacred jedi texts

Just to be lazy I'd of stored just the slots as a set and then checked to see if those slots on the player had NETHERITE in the material string

[remote swallow]

ill put a netherite axe on my chest

[tender shard]

was about to say

[young knoll]

I’ll put a netherite block on my head with /hat

[tender shard]

also it's in the move event so I'd rather use something that doesn't require string comparing all the time

[livid dove]

I mean that's utterly fair lol

[tender shard]

the player's gonna die very fast

what a stupid idea in the first place lol

[livid dove]

Hey u just wasted time coding it

Lol

[young knoll]

Just don’t get hit 5head

[livid dove]

I think I may of interfaced myself into a corner

[young knoll]

The solution is more interfaces

[remote swallow]

then add more

[livid dove]

What's the dea with implementing an interface into an abstract, and then impmenting a child of the interface into a concrete of the abstract? Is that legal? Lol
I
Don't wanna get arrested by the Oracle police

[echo basalt]

try

[livid dove]

It can be done, I'm just curious if it's the right play.

E.g making an block log abstract and then the concretes implementing child's of it like blockchange

Feels sorta dirty :S

Probably the heaviest plugin I've ever done and wanted folks takes

I know I know, its not quite -8 health on not full netherite. But still

[unkempt peak]

Anyone know how I can create a BlockState from a material in 1.19.2 mojang mappings?

[tender shard]

BLockStates are API, has nothing to do with mojang maps

Material#createBlockData

oh wait, blockstates

erm wait

[remote swallow]

cant you get it from data

[tender shard]

Material.DIRT.createBlockData().createBlockState()

but I think this was added in 1.20.1

[unkempt peak]

Yeah i need net.minecraft.world.level.block.state.BlockState in 1.19.2 from a Material

[remote swallow]

june

[remote swallow]

theres probably some Craft class to convert it

but idk if the state exists without a block

[tender shard]

what exactly do you need this for?

[unkempt peak]

So, I'm trying to rewrite the NMS Chunk method from https://www.spigotmc.org/threads/methods-for-changing-a-massive-amount-of-blocks-up-to-14m-blocks-s.395868/ in 1.19.2 mojang mappings

I need to get a BlockState from a material where the block doesn't exist

[remote swallow]

does that not want data?

[unkempt peak]

This is what it wants BlockState net.minecraft.world.level.chunk.LevelChunk.setBlockState(BlockPos blockposition, BlockState iblockdata, boolean flag)

Unless I have the wrong method, but that's the only thing I could find

[tender shard]

Material bukkitMaterial = Material.DIRT; // only works for Materials where isBlock() returns true
net.minecraft.world.level.block.state.BlockState nmsBlockState = CraftMagicNumbers.getBlock(bukkitMaterial).defaultBlockState();

[unkempt peak]

Thanks

[tender shard]

np

btw is there a reason why you're on 1.19.2 instead of 1.19.4 or 1.20.1? o0

1.19.2 is a weird version

[unkempt peak]

That's just the version I started this project on, I'll upgrade it after I finish the main features

[tender shard]

oh ok. just wondered because I saw many people using specifically 1.19.2 in the last days

[remote swallow]

id recommened updating before you do a load of nms

[tender shard]

tbh, updating with mojang maps is as easy as changing one property in your pom

[remote swallow]

well yeah, its not too hard to update just annoying

[unkempt peak]

Well I'm going to want it to work with 1.19 anyways so the order doesn't really matter

[remote swallow]

1.19.2 nms wont work on 1.19.3 or 1.19.4

[tender shard]

btw does gradle support using sth like "properties" in the versions of dependencies (I mean, ofc it does) but does IntelliJ also understand that properly?

[unkempt peak]

Yeah ik

[remote swallow]

yeah

[tender shard]

oki

is there any tutorial on how to write a simple gradle plugin? I wanna write a plugin to run the specialsource .jar

[remote swallow]

https://docs.gradle.org/current/userguide/custom_plugins.html

[young knoll]

Holy shit he’s doing it

Special source on gradle

[tender shard]

I've already read taht but I don't get it, am I really supposed to write my plugin inside the build file using groovy?

[remote swallow]

no you can do it in java classes

[tender shard]

but then why do they say "You can include the source for the plugin directly in the build script"

i don't really understand that

it's the part at "Packaging a plugin -> Build script"

[remote swallow]

bc you arent required to make it a seperate project

[remote swallow]

kotliny syntax

you can make a gradle plugin through intellij project wizard and go from there

[tender shard]

either this tutorial is completely incomplete, or I am stupid - they don't mention anything about how to do it in java, do they?

[remote swallow]

down at the bottom

no code examples though

[tender shard]

I also don't get why they keep referring to some obscure "buildSrc" directory

what's that supposed to be

[remote swallow]

no clue

[tender shard]

typical gradle docs ugh

[remote swallow]

create a gradle plugin project and see what you have

[tender shard]

yeah well that's what I#m trying to do

but the docs don't really give any useful info

[remote swallow]

https://github.com/freefair/gradle-plugins/blob/master/lombok-plugin/src/main/java/io/freefair/gradle/plugins/lombok/LombokPlugin.java

[tender shard]

I'll call it SpecialSauce

[young knoll]

Lost in the sauce

[tender shard]

what the

[remote swallow]

case

[tender shard]

yeah I know

but why

why does that have to be the same

and if it has to be the same, why do I have two available input fields

if it has to be the same anyway, there should be only one input field

[remote swallow]

you would think case but no

[tender shard]

like literally, why can I enter both a name and an artifactid if both must be the same

this is like asking "do you want fries with that" and when I say "yes", they say "sorry we're out of fries"

yeah bro then don't ask

lol

[young knoll]

“Do you want fries with that”
“Yes”
“Well too bad motherfucker”

[remote swallow]

https://tenor.com/view/too-damn-bad-sarcasm-holes-gif-8841847

[tender shard]

am I supposed to set my groupid etc in settings.gradle or in the publication section in build.gradle?

I can't find anything about that in the docs

IJ's default gradle template uses a weird mixture where the project name is defined in settings.gradle while groupid and version are defined in the build file as normal property

[remote swallow]

Name is always in settings

[surreal grail]

Hey
I followed the wiki on the SpigotMC wiki site to use the BungeeCord configuration System but somehow my config doesn't load. (https://www.spigotmc.org/wiki/using-the-bungee-configuration-system/ )
The config file is getting created with the values inside but somehow when I try to access the values by using configuration.getString("") it does return an empty String.

Am I doing anything wrong by loading/accessing the config?

        try {
            Configuration configuration = ConfigurationProvider.getProvider(YamlConfiguration.class).load(new File(this.getDataFolder(), "config.yml"));
            ProxyServer.getInstance().getLogger().info(configuration.getString("database.host"));
        } catch (IOException e) {
            throw new RuntimeException(e);
        }

[tender shard]

are you sure you're properly creating the file?

[surreal grail]

yeah like I said the file exists

[tender shard]

IIRC, bungee doesnt have the concept of paths

you'll have to do sth like getConfigurationSection("database").getString("host")

[surreal grail]

alright I will try it like this now

ProxyServer.getInstance().getLogger().info(configuration.getSection("database").getString("host")

It still prints out an empty String

[tender shard]

loop over the keys and see what it prints out

configuration.getKeys().stream().forEach(System.out::println);

[surreal grail]

[16.08 03:19:35.380] INFORMATION: [Proxy-1] [03:19:34 INFO]: database.host
[16.08 03:19:35.380] INFORMATION: [Proxy-1] [03:19:34 INFO]: database.port
[16.08 03:19:35.380] INFORMATION: [Proxy-1] [03:19:34 INFO]: database.database
[16.08 03:19:35.380] INFORMATION: [Proxy-1] [03:19:34 INFO]: database.user
[16.08 03:19:35.380] INFORMATION: [Proxy-1] [03:19:34 INFO]: database.password

well atleast it shows the keys

[tender shard]

hmm

show your config pls

[surreal grail]

## Database
database.host: "aa"
database.port: 3306
database.database: "aa"
database.user: "aa"
database.password: "aa"

I removed some values cuz I don't wanna leak my db ip

[tender shard]

yeah I think the problem is that you're using dots in the key names. bungeecord tries to separate that into getSection("database").getString("host")

but there isn't actually a section called database

try it like this

database:
  host: localhost
  port: 3306
  etc: blabla

see if that works

[surreal grail]

Okay

[tender shard]

your code can stay the same, just change the yaml

[surreal grail]

[16.08 03:24:24.861] INFORMATION: [Proxy-1] [03:24:24 INFO]: localhost

It did work now

Thank you so much 🙏

[tender shard]

np!

mc-dev plugin at its worst again - that's not even valid yaml

[echo basalt]

takes like 2ms to fix

also intellij is somewhat at fault as it likes to indent

and plugins that indent on top cause issues

(Example: copilot)

[shy finch]

I want to make some more PRs to spigot but the workflow is hellishly slow... Use buildtools to get latest, open in intellij, create and run task to apply the patches (and configure the task so it will actually run with the buildtools path), make changes, create and run task to create patches (configure the task so it will actually run with the buildtools path), test, then go back to applying patches if things don't work.

When all is done, creating the patches for me makes all these whitespace changes so I basically have to manually copy-paste the patches I need for the PR, then discard all changes, then paste my changes...

Surely I am missing something (probably a lot). Does anybody have a video or anything that might help me improve the workflow?

I'm looking to improve the SculkCatalyst API if that makes anybody want to lend a hand 😛

^ that workflow is livable for my PRs that were like <10 lines of code, but this one will require a bit more

[tender shard]

yeah that is indeed very annoying.

howeeeever, opening the buildtools directory isn't the correct way in the first place

?contribute

[undone axle]

You can find information about contributing to Spigot at the following links:
https://www.spigotmc.org/wiki/cla/
https://www.spigotmc.org/wiki/guide-contributing-to-spigot/
https://hub.spigotmc.org/stash/projects/SPIGOT/repos/craftbukkit/browse/README.md
https://hub.spigotmc.org/stash/projects/SPIGOT/repos/craftbukkit/browse/CONTRIBUTING.md

[tender shard]

second link

and for the whitespace and stuff, don't use auto-formatting in your IDE (at least not if you haven't loaded the codestyle file) and use the included maven profile for the codestyle

other than that, if somebody knows a better workflow, I'd also be interested in getting pinged about that lol

[young knoll]

I don’t actually use buildtools

Other than that that’s about it

[shy finch]

Well when I tried directly from the jira repo, I had a really hard time building a .jar to run

Like I had missing dependencies

[tender shard]

did you run applyPatches with the path to your buidltools "decompile" directory?

[shy finch]

I don't quite remember 😅

Let me pull it up

[tender shard]

i have to add, I never actually did any changes to spigot itself, only bukkit and craftbukkit

[shy finch]

well yeah for this I would need to modify bukkit, craftbukkit, and nms

[tender shard]

in 99% of cases that's all you'll need though

yeah NMS is part of craftbukkit's patches

[wet breach]

The patches probably should be done from spigot repo

[tender shard]

this reminds me, I have to change my jira email address

[shy finch]

and craftbukkit and bukkit are 2 separate repos, so how can I add an event in bukkit, call it in craftbukkit, then test the .jar result?

Because from jira I have a bukkit and a craftbukkit repo

2 different ones

[wet breach]

Do it from spigot repo

[young knoll]

You just build craftbukkit

It includes Bukkit in it

[wet breach]

Spigot literally clones cb and bukkit after they are done and has its own patches

[tender shard]

1. you add it to bukkit, then you mvn install it
2. you add it to craftbukkit, then you mvn package it and use the resulting .jar

[young knoll]

• make changes
• maven install bukkit
• maven package cb

[wet breach]

So it should be done from there otherwise you risk spigot changing something unexpectedly

[young knoll]

Don’t make changes in the spigot repo

Then you end up with the spigot patches in your commits

[wet breach]

You can exclude them from commits

[echo basalt]

while we're in this topic

[young knoll]

I guess you could just not apply the spigot patches

[echo basalt]

I suggest making an event for when a command is forcibly run through dispatchCommand 🙏

[young knoll]

But that’s beside the point

[tender shard]

anyone know how to change the output directory for (and ONLY for) the final .jar built by gradle?

[echo basalt]

great question

I've been using gradle for the past like 6-7 years and I still don't know any of it

[shy finch]

I'm too small brain for this... so lets say I am starting fresh and have fresh forks in my hub.spigotmc, I should be:

1. Clone Bukkit to a folder and open in Intellij
2. Clone CraftBukkit to a folder and open in Intellij
3. ...? add something or other

[remote swallow]

destinationDurectory = file()

[shy finch]

wait a gotchu one sec

[young knoll]

Then do this @shy finch

[shy finch]

or if you use gradle.kts

tasks.named<ShadowJar>("shadowJar") {
    destinationDirectory.set(file("../some_folder"))

[remote swallow]

That requires shadow

[tender shard]

I'm not using shadowJar

[shy finch]

I tend to use shadowJar even when I don't need really need to... less to learn

[wet breach]

This probably wouldnt be hard

[echo basalt]

It isn't

It's just that I've had to use Unsafe to override the command map with a proxied one

[wet breach]

Like one or two lines lol

[echo basalt]

because there's no event and everything is locked down

private final field with a protected constructor or whatever

[wet breach]

Well there is but not for the method specifically because it wasnt originally intended to stop a plugin from doing something

[young knoll]

private final fields don’t require unsafe

[echo basalt]

p sure this was for logging purposes

doesn't need to be cancellable

But literally any plugin that forces the console to do something won't fire console command events

[wet breach]

It should fire the servercommand event

[echo basalt]

it doesn't

Because that event fires when reading console input

instead of on the command map

[wet breach]

Because the command is suppose to be ran as that command sender. Must of been an oversight

[wet breach]

I would probably make a jira ticket for it

[echo basalt]

cant bother

[remote swallow]

Give me an event name and ill pr it this or next week

[echo basalt]

ConsoleCommandEvent

if it doesn't exist

[remote swallow]

Isnt that server

[wet breach]

Dispatch event and fix servercommand event not firing

[tender shard]

typical gradle moment

[wet breach]

The last one should be classed as an uncaught bug

[echo basalt]

just fire events on the command map ffs

the event fires

just not when forced

[young knoll]

I should probably update my gradle wrappers

I think I’m still on 7.4

[echo basalt]

anyways it's like 3am and I'm really not feeling like working

even though I gotta deliver a project tonight

lmfao thanks bukkit

[tender shard]

good luck, chances are high that everything breaks

[tender shard]

that's also how I learnt to handle InterruptExc

i never really understood it

[echo basalt]

thanks aikar!

[tender shard]

does anyone know what this means?

FAILURE: Build failed with an exception.

* What went wrong:
A problem occurred configuring root project 'specialauce'.
> Failed to notify project evaluation listener.
   > org.gradle.api.tasks.bundling.Jar.setClassifier(Ljava/lang/String;)V

* Try:
> Run with --stacktrace option to get the stack trace.
> Run with --info or --debug option to get more log output.
> Run with --scan to get full insights.
> Get more help at https://help.gradle.org.

BUILD FAILED in 422ms

[kind hatch]

What is specialauce?

[tender shard]

that is my project

[young knoll]

He’s lost in the sauce

[wary topaz]

Run with --scan to get full insights.

[tender shard]

same thing

[echo basalt]

I wonder how many lines of code luckperms has

And I also wonder if there's any OS plugin that has crossed the 100k limit yet

[tender shard]

probably some noob plugin that creates a few items using the same 1000 lines duplicated a hundred times

[echo basalt]

I mean something qualit

[shy finch]

PS C:\Users\colli\Desktop\Code\craftbukkit> mvn install bukkit
[INFO] Scanning for projects...
[INFO] 
[INFO] -----------------------< org.bukkit:craftbukkit >-----------------------
[INFO] Building CraftBukkit 1.20.1-R0.1-SNAPSHOT
[INFO]   from pom.xml
[INFO] --------------------------------[ jar ]---------------------------------
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time:  0.073 s
[INFO] Finished at: 2023-08-15T22:16:27-04:00
[INFO] ------------------------------------------------------------------------
[ERROR] Unknown lifecycle phase "bukkit". You must specify a valid lifecycle phase or a goal in the format <plugin-prefix>:<goal> or <plugin-group-id>:<plugin-artifact-id>[:<plugin-version>]:<goal>. Availab
le lifecycle phases are: pre-clean, clean, post-clean, validate, initialize, generate-sources, process-sources, generate-resources, process-resources, compile, process-classes, generate-test-sources, proces
s-test-sources, generate-test-resources, process-test-resources, test-compile, process-test-classes, test, prepare-package, package, pre-integration-test, integration-test, post-integration-test, verify, in
stall, deploy, pre-site, site, post-site, site-deploy. -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/LifecyclePhaseNotFoundException
PS C:\Users\colli\Desktop\Code\craftbukkit> 

not sure where I am supposed to be running mvn install bukkit

[kind hatch]

I think the largest project I have done is ~42K lines. Even then I know for a fact that some of it could be reduced down.

[shy finch]

WeaponMechanics is sitting at about 65k lines of code

[echo basalt]

biggest I've done is about 45k

we rewrote it in minestom and it's like 20k

[tender shard]

Just run „mvn install“

Without bukkit

[young knoll]

or mvn install -P development

To run codestyle checks

[shy finch]

it really wants bukkit:

(result of running mvn install)

[tender shard]

Weird name, id expect the dev profile to NOT run code checks

[tender shard]

Run „mvn install“ on the bukkit project

After that you run „mvn package“ on craftbukkit

[shy finch]

I assume that is after I run applyPatches

on craftbukkit

because aint no chance in hell this is compiling

[tender shard]

yes sure, as explained in the second ?contribute link

[shy finch]

alright thanks, appreciate it. This definitely improves the workflow

[tender shard]

1. you mvn install the bukkit project
2. then you open craftbukkit and run applypatches.sh with the decompile directory from buildtools as parameter
3. now craftbukkit should compile just fine using mvn package

how you do changes to spigot though, I have no clue. but you shouldn't really need that

[shy finch]

and from there create the server... now don't hate me, but if I create a test plugin for this server, is the fastest way to depend on custom server to add a system gradle lib?

Or does everybody have a fancier way

[wet breach]

Such a small number

[tender shard]

you can do mvn install on craftbukkit instead of mvn package - then you can depend on the craftbukkit dependency normally (ofc you must specify mavenLocal() as repo)

it's important that you get rid of the normal spigot repository in your plugin's build file - otherwise it will get overwritten if a new snapshot is available in the repo

oh wait, craftbukkit is not available in any repo

[shy finch]

oh, that makes more sense... I guess I wouldn't actually nee to run mvn install on craftbukkit though since why would I need the plugin to depend on craft bukkit?

[tender shard]

yeah you usually don't need that

you should be able to just depend on bukkit, I was a bit stupid

but be sure to remove the spigot repo

in your plugin

otherwise it might update to the latest snapshot and use that instead of your own custom version

[shy finch]

I think I am going to make a video on this at some point... mostly for my forgetful ass 😆

[tender shard]

i should probably turn that into a blog post

[shy finch]

epic collab time

[tender shard]

I have a habit of being upset at overly complicated or outdated tutorials and then write a blog post once I figured it out myself lol

[shy finch]

we can just edit the spigot wiki right?

[tender shard]

most articles, yes

not all

[shy finch]

it looks like we can

[tender shard]

yeah that article is unlocked

but I prefer to write on my blog

?nms

[undone axle]

https://blog.jeff-media.com/nms-use-mojang-mappings-for-your-spigot-plugins/

[tender shard]

the wiki is kinda weird

I'm still waiting for someone to rewrite the buildtools article

(or to get the buildtools GUI merged)

[shy finch]

wiki always feels incomplete/outdated

but I don't want to be the one to fix that 😆

[tender shard]

the buildtools article is a horrible mess

and it's annoying to edit the wiki because it uses BBcode instead of normal wiki formatting D:

i should ask md to install a mediawiki

[shy finch]

markdown purist over here

[tender shard]

markdown is fine too but I'm used to wikipedia

[kind hatch]

That's what I'm waiting for. I was gonna write a bit for it.

[tender shard]

tbh, I don't think anyone will miss the current article, feel free to edit it already lol

unfortunately very few people contributed to the discussion about the buildtools article that I started

basically only 2 people and one was complaining about not being able to download buildtools lmao

[young knoll]

We didn’t even get much contribution on the experimental discussion thread

Sadge times

[kind hatch]

Yea, I gave my thoughts on it. Was hoping for some feedback on it.

Could also just split up the one page like this if people don't want to maintain multiple pages.

• BuildTools (GUI)
• BuildTools (CLI)

[tender shard]

experimental discussion thread?

[young knoll]

About the changes in the experimental branch

[kind hatch]

https://www.spigotmc.org/threads/looking-for-testers-feedback-on-spigot-pull-request-converting-enums-to-classes.604316/

[tender shard]

I don't think that's needed. the article should imho look like this:

more is literally not needed

[echo basalt]

what version of jdbc has spigot been running?

[tender shard]

that's from latest CraftBukkit

there's also mysql-connector-j 8.0.33

[kind hatch]

That's where I kinda disagree. Going to a wiki page and it only containing like 8 lines of text would give off the impression of a poor wiki article.
I'm not saying to write an entire manual for BuildTools, but keep the simplicity of your screenshot while still providing useful information.

Those errors that are currently listed on the wiki are something that should stay. The links to the Discord/IRC should be made more noticeable, the general description could be updated and the removal of the same command but different version would all be meaningful changes.

[tender shard]

which other information would be needed?

all those "currently listed errors" are things that only happened 7 years ago

and ofc the article will be longer than 8 lines if every command line option will be explained

[kind hatch]

People still run into them for some reason. Not even a week ago, someone ran into this issue.

[tender shard]

yeah well okay, if this really happened to someone days ago, then it should stay - at the very bottom in a section called "Troubleshooting", but not in the middle of the article

(imho)

[kind hatch]

Yea and that makes sense.

[tender shard]

right now the article is like 91825 lines where 99% of people would only need to read 0.1%, but the important 0.1% are hidden somewhere in the middle

and that's why people rather use paper because they can just download it

[kind hatch]

That's it. Time to rewrite everything from scratch.

[tender shard]

most paper users are incapable of reading more than 30 lines of text and hence they are so overly confused by spigot, they just cannot run buildtools. that's why I think that the article should start with a huge DOWNLOAD BUILDTOOLS HERE link, and then in the text below, ofc we can explain common issues and command-line options etc, but we definitely need the huge "DOWNLOAD IT HERE" banner at the top

[eternal oxide]

wiki page needs a big "Download SPIGOT" button that does everything an idiot needs 🙂

[kind hatch]

I want to know how they do the buttons on the main wiki page.

[tender shard]

oh yeah there must be a Download button

[kind hatch]

Having a download button like that would help.

One for discord too

[tender shard]

we should have a spigot.jar downloadable that only contains buildtools, then runs itself and builds latest spigot, and then replaces itself with the finished .jar

so people think they actually downloaded spigot

[kind hatch]

You know, you might be onto something since a lot of people try to run the BuildTools jar thinking it's spigot. 🤔

[echo basalt]

p sure that's what paper does

[eternal oxide]

or just a RUNME.jar

[tender shard]

no clue how paper works

but I'm pretty sure that spigot went down to 20% of users not because paper is "so much faster" but because they just offer a "Download here" button, while spigot doesnt

[eternal oxide]

agreed

most users are just looking for a download button

[tender shard]

yeah. we must talk to md_5 about this

[eternal oxide]

we need to keep buildtoosl and all that for us devs, but there needs to be a simple one button automated system for end users

[tender shard]

although having BT gui merged is already a huge step into the right direction

[kind hatch]

It would also help if the Jenkins page, (which you are redirected to if you click the download button on the navbar) mentioned in all caps.

THIS IS NOT A PRECOMPILED SPIGOT JAR

Every opportunity you have to warn the user, you should take it.

[tender shard]

okay guys I got an idea

[eternal oxide]

end users doesn;t even want to go to jenkins. they want an easy to find "Download" button

[kind hatch]

Well that's where it takes you.

It just redirects you to Jenkins

All because of the legal bullshit

[eternal oxide]

Honestly even I've never used that menu. Spigot site is very unintuitive

I alwasy search download buildtools if I need the link

[tender shard]

Spigot-Server.jar - a GUI, similar to the current buildtools gui

1. General (you can select the version) and change server.properties stuff there like port, motd
2. a huge START button - it checks if there's already a spigot-1.X.X.jar in the current folder - if not, it runs buildtools in the background
3. a console GUI, similar to the already existing one, but basically... all in one. A whole GUI. so you just download Spigot-GUI.jar, run it, select version, and that's it

[kind hatch]

That sounds a lot like those Bukkit GUIs that existed back in the day.

[tender shard]

i haven't seen any, but this one could be official

[eternal oxide]

I'd definitely vote for an idiot proof download button

[tender shard]

it doesnt even have to have many features, all it must offer is to be able to select a version and have a start button and be linked as "Download" on the homepage

[tender shard]

yeah exactly

[kind hatch]

Hold up, here's a thought.

[tender shard]

even buildtools gui isn't idiot proof enough

[echo basalt]

Can't we just make a

launcher with a builtin version config

that just runs buildtools

in a cache folder

[tender shard]

that's what I suggested

[kind hatch]

Nvm, it wouldn't be compiled on the user's computer.

[echo basalt]

Legit just make it fetch buildtools from spigot's jenkins

put it in a cache folder

run it

[tender shard]

yep that's basically what I had in mind

maybe with an additional tiny config GUI for port, motd, max players, etc

a button to open the plugins folder in explorer

[echo basalt]

that's server.properties

[tender shard]

yeah

but i na GUI

[echo basalt]

I don't even mean a gui

[tender shard]

yeah a button would be enough

[kind hatch]

You are straight up describing the Bukkit GUIs.

[echo basalt]

Just a standalone jar that runs automagically

[kind hatch]

Give me a sec, I'll see if I still have a screenshot or the original programs.

[echo basalt]

with maybe a tiny gui with a countdown and a server version

[eternal oxide]

We definitely need something super simple that an idiot can use. When the forks hard fork we are going to have issues.

[echo basalt]

and a headless mode

[tender shard]

ok shadow next project. but this time pls lemme do a gridbag layout 😄

[echo basalt]

git like 6 projects in my queue atm

[tender shard]

yeah like the current buildtools gui

if you do java -jar ... it runs in CLI mode, if you just double click it, it opens theGUI

[echo basalt]

I got no clue why I'm in the contributors list I just asked to be on it

[kind hatch]

Soo, I do have the programs, but I'll have to see if they run through WINE.

[tender shard]

send it to me

oh

Shadow

send the exe

I'm on win rn

[kind hatch]

Which one you want?

[tender shard]

v2

ofc

[kind hatch]

It's been a long time since I used those. (cough timestamps)

[tender shard]

nothing happens 🥲

[kind hatch]

Could try the other one.

Could also try running it in compat mode. (If that's even still a thing)

[tender shard]

that also just quits

no dialog, no error msg, nothing

[kind hatch]

That's odd.

I wonder if they needed a certain .NET framework.

[tender shard]

I have another idea

[kind hatch]

I'm still gonna try wine

[tender shard]

Buildtools gui could generate a tiny bat and sh file to start spigot

Called Start-Spigot-Windows.bat and Start-Spigot-Linux.sh

It could also use javafinder to get the correct java

[worldly ingot]

Separate PR for that ideally

[unkempt peak]

Does anyone know what can cause an out of bounds exception when using net.minecraft.world.level.chunk.LevelChunk.setBlockState? (mojang mapping)

[worldly ingot]

Under a new flag that's off by default too

[worldly ingot]

Probably a call to Chunk#setBlock() either < (0, 0) or > (15, 15)

Without plugin intervention, highly unlikely to occur

[tender shard]

yeah sure, once the GUI gets merged

[unkempt peak]

Oh it says in the stacktrace the exception is from net.minecraft.world.level.chunk.IChunkAccess.b

[kind hatch]

I don't even remember how this worked on windows anymore. 😢

[tender shard]

paste the whole stacktrace

shadow have you ever used XAMPP?

[unkempt peak]

?paste

[undone axle]

https://paste.md-5.net/

[unkempt peak]

https://paste.md-5.net/udofapatoz.css

[kind hatch]

I think once, but it confused the hell out of me.

[unkempt peak]

placeBlocksByChunk: https://paste.md-5.net/juvuzakusa.cpp

[tender shard]

the Spigot Bootstrapper could be a bit like this.
1 button to start spigot (or build it if not found yet)
1 button to edit spigot.yml
1 button to edit server.properties
(etc etc)
1 button to open plugins folder
yeah that's it

and at the bottom, the server logs including prompt for commands

and ofc a schnitzel should show with a 1/100 chance

very important

anybody knows why this isn't working? It cannot find "SpecialSaucePlugin" which is not surprising, but that's exactly how the baeldung tutorial suggested to do it :/ https://www.baeldung.com/gradle-create-plugin

[tender shard]

Can somebody explain to me the difference between the following two things?
Option one, works fine

buildscript {
    repositories {
        mavenLocal() // plugin published to maven local
    }
    dependencies {
        classpath 'com.jeff-media:specialsauce:1.0-SNAPSHOT'
    }
}

plugins {
    id 'java'
}

apply plugin: 'com.jeff-media.specialsauce'

Option two, does not work, see error below

buildscript {
    repositories {
        mavenLocal() // plugin published to maven local
    }
    dependencies {
        classpath 'com.jeff-media:specialsauce:1.0-SNAPSHOT'
    }
}

plugins {
    id 'java'
    id 'com.jeff-media.specialsauce' version '1.0-SNAPSHOT'
}

Error:

Build file 'C:\Users\mfnal\IdeaProjects\testproject\build.gradle' line: 12

Plugin [id: 'com.jeff-media.specialsauce', version: '1.0-SNAPSHOT'] was not found in any of the following sources:

* Try:
> Run with --stacktrace option to get the stack trace.
> Run with --info or --debug option to get more log output.
> Run with --scan to get full insights.
> Get more help at https://help.gradle.org.
BUILD FAILED in 959ms

[slender elbow]

you define the plugins repository in the settings.gradle(.kts), not in the main build script, uh something like

pluginManagement {
  repositories {
    gradlePluginPortal()
    mavenLocal()
  }
}

and get rid of the whole buildsxript configuration

[tender shard]

but then why does "apply plugin" work? ALso neither the gradle docs, nor the above mentioned baeldung tutorial mention anything about that

https://docs.gradle.org/current/userguide/custom_plugins.html#sec:custom_plugins_standalone_project the gradle docs itselfs don't mention ANY repository having to be specified, not even maven local

I am very confused

[slender elbow]

uhm yes it does?

[tender shard]

hm? where?

uuugh emily why are you not verified lol

[slender elbow]

"using your plugin in another project"

like 7 lines below lol

[tender shard]

on which page are you now?

[slender elbow]

that same page you linked

[tender shard]

I don't see any "using your plugin in another project" part

[slender elbow]

okay maybe more then 7 lines lmao but further down

idk I just scrolled down

lemme

[tender shard]

oh there. yeah well but I have published it to maven local already

and also added it to the normal dependency as classpath dependency

I wonder what the difference is between plugins { } and apply plugin:

why does one work without having to specify a separate pluginManagement repository, while apply plugins does not require that

erm other way around

[slender elbow]

the buildscript and apply plugin thing are kinda being phased out

[tender shard]

are you sure? I'm using 8.2.1 (latest) and not getting any warnings

[slender elbow]

yeah but they encourage/prefer the other way for new stuffs

[tender shard]

I still don't understand why one (plugins { } but not apply plugin) requires me to declare a plugin repository - how does it work for other plugins? e.g. why do I not need to declare a repository for lombok or the java plugin?

[carmine mica]

there's an implicit repository, the gradle plugins portal

[slender elbow]

they're in the gradle plugin portal which is ad- yeah

(edited)

[tender shard]

wtf

[slender elbow]

okay least favourite papermc representative

sorry that was mean

I love you

[tender shard]

they add their website automatically but not mavenLocal()?

that is... funny

but ok, thanks. I am still confused about why it's so different

[carmine mica]

adding maving local sounds like a really bad idea

end up accidentally using some modified local version without realizing

[tender shard]

how else would I test my own plugin before publishing it anywhere

[carmine mica]

well you can add it yourself for testing

[slender elbow]

that's what snapshot repos are for, aren't they? :p

[carmine mica]

oh I mean adding it by default, sorry

not just adding it whenever. adding it is useful for testing like you said

but adding maven local implicitly I don't think is good

[tender shard]

my plugin is already using a snapshot version, and the gradle docs don't mention anything about using the gradlePluginPortal by default

anyway, thanks for your answers, I am still confused about how I would have been supposed to find this out myself

e.g. the difference between apply plugins and plugins { } automatically using the plugin portal but not maven local - where is that documented?

[slender elbow]

idk I just scrolled down lol

[tender shard]

I have now read the full page and didn't find anything about it

[carmine mica]

its here https://docs.gradle.org/current/userguide/plugins.html

https://docs.gradle.org/current/userguide/plugins.html#sec:custom_plugin_repositories

that page also discusses the difference you were looking for

between the plugins {} block and the legacy method

[tender shard]

thanks! I have tried to find the difference but somehow all it says is that "apply plugins" is "legacy"

[carmine mica]

oh well maybe it doesn't describe the functional difference, but it does say that it shouldn't be used

idek what the functional difference is, I have never used buildscript and I only ever see it on things using much older gradle versions

[tender shard]

yeah me neither, this was the first time today that I really had to use it (I've only seen it on 10 year old scripts) lol

I still find it highly weird that mavenCentral() has to explicitly be added as repository but on the other hand, gradle has no problem in using an arbitrary website for plugins

that makes little sense to me

but mavenLocal() is not trustworthy? idk ugh

mavenLocal() is the only thing where oneself has control over

[carmine mica]

idk if its because its not trustworthy, I think it can just lead to more unexpected things, like using modified local versions of stuff when you totally don't wnat to

[tender shard]

mavenLocal() should always be the first place to lookat imho

yeah well if one doesn't want modified versions in their maven local, one shouldn't install stuff there

it's literally the only repository where one has full control over

[carmine mica]

but maybe I installed it there 2 months ago

I don't dig around in there and delete stuff

[tender shard]

me neiter but I also don't install random stuff there 😄

[carmine mica]

I want to know that what I clone from a repo will use exactly the right things

mavenLocal doesn't adhere to that

[tender shard]

but you don't even know from wich repository you got something

if you have 5 repos declared in your repositories { } section, you can't even know where dependency XY came from

[carmine mica]

you can specify it if you want

but you can also get a report that says it iirc

[tender shard]

yeah but usually you don't do that

same as you usually don't overwrite stuff in maven local with random things

[eternal night]

Imagine you are looking for MM

and here he is

[tender shard]

what is MM

[eternal night]

not merging PRs

😭

Machine

I just am too lazy to spell out MachineMaker and I am already back off to studying so now I am poking him to merge A PR

[tender shard]

no clue what MachineMaker is tbh

[eternal night]

[slender elbow]

rip

[eternal night]

[tender shard]

Okay so tl;dr i have to declare mavenLocal in pluginRepositories to use plugins using plugins {} from maven local, right?

[carmine mica]

correct

[tender shard]

Thx!

[carmine mica]

in pluginManagement.repositories

[eternal night]

merci

[tender shard]

Erm yeah thats what i meant

I am used to maven lol

Oh since so many gradle ppl are here rn

Is there a good comprehensive book about gradle?

Like, an actual book that somebody can recommend

[slender elbow]

the docs are amaze...... a big maze, lol

[tender shard]

Haha yeah

[carmine mica]

I think its hard to write a book that wouldn't be outdated dramatically in a year

[tender shard]

True

[carmine mica]

gradle's rate of change is very very quick, its one of the bigger issues people have I think

[tender shard]

yeah that is indeed very annoying

[carmine mica]

you can still just use the older version, nothing requiring you to update really

[eternal night]

but then my ~/.gradle folder is filled with so many versions

who has the disk space

[tender shard]

I learnt java completely from books, I haven't studied anything remotely similar to CS or stuff, but I managed to learn java and maven and stuff fine with normal books, kinda annoying that it's not possible with gradle lol

everytime I look for docs they are outdated or in a very weird order (e.g. I have a question at step 3 and it gets answered in step 27 of the docs)

so I always have to ask people who already know the answer and then I wonder "how did they manage to learn this without asking someone who already knew this"

[slender elbow]

exhibit A: what you just experienced lol

[tender shard]

gradle is a rabbit hole and I sometimes think it's not possible to understand "most of it", you either have to know everything or you know nothing

[slender elbow]

also stackoverflow questions from a decade ago with a valid answer for gradle 2 or whatever that still show up at the top of search results

that's my biggest problem with SO in general, eck

[tender shard]

the IntelliJ docs are equally weird btw

(I'm talking about writing IJ plugins)

yeah anyway, thank you all, I'm a big step closer to my goal of writing a very simple gradle plugin now lol

and it's only 6 am

[eternal night]

true, I woke up 30 mins ago

[tender shard]

oh btw I have another question about gradle.

in maven, groupId, artifactId and version are essential parts of a pom. in gradle, it seems like only the "artifactId" is essential while groupId and version just seem to be "normal properties" that aren't actually needed to define.

this surprises me especially since subprojects (which seem to be like maven modules) inherit the parent stuff mostly, but in gradle the artifact is declared in the "global" gradle.settings file, which makes little sense to me because usually, groupId and version stay the same in "child modules/projects" and only the artifactId changes

I wonder what the reason is to separate the groupId/version from the artifactId into two different files

maybe I should have studied CS instead of law

[valid beacon]

Good afternoon! Are there any programmers who can take my order for 5000₽?

[eternal oxide]

?services

[undone axle]

If you wish to request or offer development/art/building/administration services, please do so at https://www.spigotmc.org/forums/services-recruitment-v2.54/

[shy finch]

Hmm so working on a PR for sculk catalysts, but it seems a class is not included in the craftbukkit repo after I applied patches

I guess I did a step wrong with buildtools?

(this is a problem since I need to modify the code in this class, I need to add an event)

[sterile breach]

Imagine: we have two servers

The server work like this
Get a value un the cache and delete it

Server 1: we get the value of "1" (is a dog)

We give a diamond to the player
(In the same time an other player di a request)

Server 2 grt the value of 1 and give a diamond

Server1 delete the value.

It doesn't make sense, but it's an example. The 2 players got a diamond for something that should have given only 1.

[buoyant viper]

did u drag over the file from the work/decompile-XXXX dir?

u need to manually drag in the files u want to edit if they arent there and ./applyPatches.sh to them p sure

[shy finch]

that sounds sketchy but ok

I guess only classes that already have patches are included?

[buoyant viper]

yes

[ivory sleet]

Can u fix ur spelling errors, ive no clue what u mean

[sterile breach]

Yes

[ivory sleet]

But it does sound like you need to sync the players inventories as well

[sterile breach]

We have 2 servers. Conecte a la db.

We have a code that works this way. When a player makes a command, it retrieves the value of key 1 (which will be dog).
Then it deletes it.
So it gives:

Db.get
Dg.delete

Except that if on weaner 1 a player x does this command and starts it
Db.get but just after on another server a player also does this command and the first db.delete has not yet arrived, the second player's db.get will return something even though it should no longer exist. Do you get it?

[sterile breach]

Not for what I want to do, but yes, at the moment I was wondering how to synchronize them between servers.

[ivory sleet]

Yes in that case you need to synchronize the start of that command or whatever operation/process you’re initiating

[sterile breach]

Yes, but if it's on 2 different servers?

[ivory sleet]

This

By synchronize I don’t mean java synchronized or java concurrent hashmap

I mean synchronize across servers

You have a proxy server right?

[sterile breach]

Yes, I was thinking of a synchronized boolean, and when it's set to True we wait for it to be false before making the request, but even then, if the 2 seeveyrs send the boolean to the other server at the same time, it's set to True. It won't work

[sterile breach]

Yes

[ivory sleet]

Anything I say has nothing to do with java

because java synchronization and thread safety is only something you care about when you’re on that jvm itself, outside we talk more generic

[ivory sleet]

And that is a terrible design I should say also

[ivory sleet]

I would use the proxy server as some sort of manager and permission master

Basically each server send a redis message, it goes through the proxy and the proxy tells whether that message should be ignored/denied or not

That way there is a central service that ensures operations may be synchronized across and deduplicates if necessary

[sterile breach]

Oh yes, so everything runs in a single thread on the bungee server and there's no risk of desynchronization

[ivory sleet]

Na

[sterile breach]

And concurrence

[ivory sleet]

On the proxy you want it multi threaded

But at some point you do need to have some synchronization mechanism which I’d use a lock for

Since you’re essentially trying to avoid RC where an operation that should only be executed once could if you’re not careful enough be executed twice

and speaking of all this its not like u drop thread safety on server 1 and 2

That shit is still there

[sterile breach]

How?

[ivory sleet]

By coding your stuff thread safe and implement atomicity etc if needed

[sterile breach]

so you'd advise me to run everything through the proxy, making a kind of lock and queue system?

[shell robin]

public class BossOfTheForce {

    private Map<UUID, BossBar> bossBars = new HashMap<>();

    private WitherSkeletonBossbar witherSkeletonBossbar;


    public BossOfTheForce(Location loc, AetheriaBosses plugin) {
        WitherSkeleton witherSkeleton = (WitherSkeleton)
                loc.getWorld().spawnEntity(loc, EntityType.WITHER_SKELETON);

        witherSkeleton.setCustomName(ChatColor.DARK_RED + "Boss Of The Force");
        witherSkeleton.setCustomNameVisible(true);
        witherSkeleton.getEquipment().setItemInMainHand(ForceAbilityItem.createForceSword());
        Attributable skeletonAt = witherSkeleton;
        AttributeInstance attribute = skeletonAt.getAttribute(Attribute.GENERIC_MAX_HEALTH);
        attribute.setBaseValue(1500);
        witherSkeleton.setHealth(1500);
        witherSkeleton.setGlowing(true);

        witherSkeletonBossbar = new WitherSkeletonBossbar(plugin);
        Player targetPlayer = (Player) witherSkeleton.getTarget();
        if (targetPlayer != null) {
            witherSkeletonBossbar.createBossBar(witherSkeleton, targetPlayer);
        }
        new BukkitRunnable() {
            @Override
            public void run() {
                if (witherSkeleton.isDead()) {
                    return;
                }
                if (witherSkeleton.getTarget() == null) {
                    for (Entity entity : witherSkeleton.getNearbyEntities(10, 10, 10)) {
                        if (entity instanceof Player) {
                            witherSkeleton.setTarget((Player) entity);
                        }
                    }
                }

                witherSkeleton.addPotionEffect(new PotionEffect(PotionEffectType.SPEED, Integer.MAX_VALUE, 1));
                witherSkeleton.getLocation().getWorld().spawnParticle(Particle.CRIT, witherSkeleton.getLocation(), 10);
            }
        }.runTaskTimer(plugin, 100L, 100L);
    }
    private void updateBossBar(WitherSkeleton witherSkeleton) {
        witherSkeletonBossbar.updateBossBar(witherSkeleton);
    }


    private void removeBossBar(WitherSkeleton witherSkeleton) {
        witherSkeletonBossbar.removeBossBar(witherSkeleton);
    }
}```
```java
public class WitherSkeletonBossbar {

    private AetheriaBosses plugin;
    private Map<UUID, BossBar> bossBars = new HashMap<>();

    public WitherSkeletonBossbar(AetheriaBosses plugin) {
        this.plugin = plugin;
    }

    public void createBossBar(WitherSkeleton witherSkeleton, Player targetPlayer) {
        UUID bossUUID = witherSkeleton.getUniqueId();
        BossBar bossBar = plugin.getServer().createBossBar(ChatColor.RED + "Wither Skeleton Boss", BarColor.RED, BarStyle.SOLID);
        bossBar.addPlayer(targetPlayer);
        bossBars.put(bossUUID, bossBar);
        updateBossBar(witherSkeleton);
    }

    public void updateBossBar(WitherSkeleton witherSkeleton) {
        UUID bossUUID = witherSkeleton.getUniqueId();
        double healthPercentage = witherSkeleton.getHealth() / witherSkeleton.getAttribute(Attribute.GENERIC_MAX_HEALTH).getValue();

        if (bossBars.containsKey(bossUUID)) {
            BossBar bossBar = bossBars.get(bossUUID);
            bossBar.setProgress(healthPercentage);
        }
    }

    public void removeBossBar(WitherSkeleton witherSkeleton) {
        UUID bossUUID = witherSkeleton.getUniqueId();
        if (bossBars.containsKey(bossUUID)) {
            BossBar bossBar = bossBars.get(bossUUID);
            bossBar.removeAll();
            bossBars.remove(bossUUID);
        }
    }
}``` I can't see the bossbar

There is no error at console

Please tag if you reply

[smoky anchor]

@shell robin I feel like the createBossBar is never getting called.
If you spawn the entity, it's not gonna target a player immediately.

Other potential problems:
You're not deleting the bossbars, not sure how bukkit handles those but there is a method to remove them.
Also, not sure if you can cast the target to player safely. Could target normal skeleton that shot it. (Tho this will never happen in your code)
Not sure why you're adding the potion effect every 100 ticks, either give it the effect once with -1 for the duration (infinite, works only on like 1.19.4+)
Your boss will always target the player that joined the server last (upon losing target), unless the getNearbyEntities has different implementation than I expect it to have.

Not sure if I'm missing something, but you're not updating the bossbar either

[harsh totem]

BanList.Type.NAME is deprecated. Shouldn't I use PROFILE?

[chrome beacon]

Yes use Profile

[harsh totem]

I put null as the date of the ban and my IDE doesn't let me do it.

ban.addBan(dead.getPlayerProfile(), ChatColor.RED + "You lost all your hearts!" + ChatColor.WHITE, null, null);```

it only works if I cast it to Date

[shell robin]

well, what should i do

i don't understand exactly where is the error

[smoky anchor]

Just noticed, the tasks never gets cancelled if even if the entity is dead.

[sullen marlin]

yes and? that's fine

[smoky anchor]

I would say create the bossbar with the entity, add targets once it has some

[shell robin]

these are possible but still prevent this bossbar from being seen?

[smoky anchor]

no, the problem is that it never gets created since the entity can't target you right after it spawns

[shell robin]

what if I wait for 5 seconds?

[smoky anchor]

what if the player hides/tps away

you can't "wait" in this case
check periodically until the skeleton has a target ig

[smoky anchor]

or this

[shell robin]

If I just set target != null it won't work :D right?

[smoky anchor]

I-
I have no idea what you're talking about rn
I gave you two solutions that should work and I think I explained (what I think is) the problem fairly well

[shell robin]

okay okay

i understand

thank you

i will try both of them

[grim oak]

I tried this and added it as a blank texture, but it seems to retexture it to this