#paper

1 messages ยท Page 19 of 1

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Zenologia
[PaperMC/Paper] New comment on issue: #11955 Custom model data commands not working properly on 1.21.4 build 97

GH wont allow me to attach but heres the contents:

This is the global configuration file for Paper.

As you can see, there's a lot to configure. Some options may impact gameplay, so use

with caution, and make sure you know what each option does before configuring.

If you need help with the configuration or have any questions related to Paper,

join us in our Discord or check the docs page.

The world configuration options have been moved inside

their respective world folder. The files are named paper-world.yml

Docs: https://docs.papermc.io/

Discord: https://discord.gg/papermc

Website: https://papermc.io/

_version: 29
anticheat:
obfuscation:
items:
all-models:
also-obfuscate: []
dont-obfuscate:
- minecraft:lodestone_tracker
sanitize-count: true
enable-item-obfuscation: false
model-overrides:
minecraft:elytra:
also-obfuscate: []
dont-obfuscate:
- minecraft:damage
...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Mcdikmen
[PaperMC/Paper] Issue opened: #11965 Facing very high ping in my server (Like 5000 - 800 MS).

Spark Profile

https://spark.lucko.me/ECksyrJVyu

Description of issue

Hello,

I am facing very high ping in my server (Like 5000 - 800 MS). I am using 1.20.6 version paper. Also, there is no any tangible warn or error in my console.

Can u help me?

Plugin and Datapack List

Image

Server config files

https://mclo.gs/R3nSaIk - bukkit.yml
https://mclo.gs/c9Cvb0f - spigot.yml
https://mclo.gs/FaMxDVg - paper-global.yml
https://mclo.gs/72aC9ls - paper-world-defaults.yml
https://mclo.gs/iGbepj8 - server.properties.yml

Paper version

This server is running Paper version 1.20.6-151-ver/1.20.6@a4f0f5c (2024-10-30T19:13:22Z) (Implementing API version 1.20.6-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.20.6-149-6e71f41 (MC: 1.20.6)

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
ItsPerona
[PaperMC/Paper] Issue opened: #11966 High consumption by "LockSupport.parkNanos()"

Spark Profile

https://spark.lucko.me/1So4QkjtYe

Description of issue

I was having problems on a server with Datapacks and Plugins, I started removing plugins and then datapacks to the point of having nothing.

In Spark I can see that "LockSupport.parkNanos()" and "Unsafe.park()" make my server sometimes send problems of:
"Can't keep up is the server overloaded running 11038ms" every 5min approx.

Plugin and Datapack List

N/A

Server config files

[mclo.gs](https://mclo.gs/JskKGsH)

Paper version

1.21.4-106-main@ab1b312

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Zenologia
[PaperMC/Paper] New comment on issue: #11955 Custom model data commands not working properly on 1.21.4 build 97

what i noticed is that the piece of paper has the same 9 component data on single player vs multiplayer, but the texture does not render in multiplayer. videos attached are named for the testing, and Ive included the sample resource pack that im testing with.

command ran on server:

/minecraft:give @p minecraft:paper[custom_model_data={strings:["bandana1"]}]

command ran in single player:

/give @p minecraft:paper[custom_model_data={strings:["bandana1"]}]

1.21.4 test rp.zip

rustic folioBOT
rustic folioBOT
#
Gleefre
[PaperMC/Paper] Issue opened: #11973 zombie-villager-infection-chance is broken

Expected behavior

When zombie-villager-infection-chance is set to 100.0, it should be impossible for a zombie to kill a villager.

Observed/Actual behavior

A villager was killed by a zombie.

Steps/models to reproduce

Set zombie-villager-infection-chance is set to 100.0 on a difficulty other than hard (easy or normal), spawn a villager and a zombie a few times, and in some cases the villager will be killed instead of being infected.

Plugin and Datapack List

> plugins
[15:49:17 INFO]: Server Plugins (2):
[15:49:17 INFO]: Bukkit Plugins:
[15:49:17 INFO]:  - floodgate, Geyser-Spigot

> datapack list
[15:49:14 INFO]: There are 3 data pack(s) enabled: [vanilla (built-in)], [file/bukkit (world)], [paper (built-in)]
[15:49:14 INFO]: There are no more data packs available

Paper version

> version
[15:49:07 INFO]: This server is running Paper version 1.21.4-106-main@ab1b312 (2025-01-12T23:38:01Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You ar...
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] New comment on issue: #11962 Stopping a Goal can create an error

Can replicate but the suggested change can't work directly otherwise getGoal will always return null since it will fetch from the wrong key (minecraft:wrapped)

I did a test changing it (stupid me didn't keep my changes), but basically I think what I did is within the PaperVanillaGoal class, when getting the key from the Goal, I just checked if it as instance of wrapped goal, grabbed the inner goal to get the key from that.

I'll look into maybe doing a PR soon, and doing a bunch of testing.

rustic folioBOT
rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] New comment on pull request: #11975 Readd dropped hunk for ender dragon fight data

what's with all these "dropped hunks" what is that, and why are they causing bugs?

Just code that was accidentally skipped when the patches were manually reapplied in the hard fork process

Edit: Kinda also was mentioned by paper devs that such dropped hunks were the reason for such a long experimental period

why not just apply the patches on the code pre hard fork then do a diff with the code + patches post hard fork? Wouldn't this easily show the dropped hunks?

rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #11976 Relative teleporting entities broken 1.21.4

Stack trace


entity.teleport(location, TeleportFlag.Relative.PITCH, TeleportFlag.Relative.YAW);


java.lang.IllegalArgumentException: duplicate element: VELOCITY_ROTATION
        at java.base/java.util.ImmutableCollections$Set12.<init>(ImmutableCollections.java:798) ~[?:?]
        at java.base/java.util.Set.of(Set.java:704) ~[?:?]
        at org.bukkit.craftbukkit.entity.CraftEntity.teleport(CraftEntity.java:246) ~[paper-1.21.4.jar:1.21.4-DEV-9f74858]
        at org.bukkit.entity.Entity.teleport(Entity.java:143) ~[paper-api-1.21.4-R0.1-SNAPSHOT.jar:?]
...

Compiled against paper api 1.21.1 if that makes a difference

Plugin and Datapack List

Custom plugin teleporting entities with relative flag

Actions to reproduce (if known)

entity.teleport(location, TeleportFlag.Relative.PITCH, TeleportFlag.Relative.YAW);

Paper version

[02:39:29 INFO]: This server is running Paper version 1.21.4-DEV-main@9f74858 (1970-01-01T00:00:00Z) (Implementing API version 1.2...

rustic folioBOT
rustic folioBOT
#
jpenilla
[PaperMC/Paper] New branch created: move-generated-src
rustic folioBOT
#
jpenilla
[PaperMC/Paper] branch deleted: move-generated-src
rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] Issue opened: #11978 Update DataComponentTypes.INSTRUMENT to support more features

Is your feature request related to a problem?

The problem is the current component for Instrument is lacking features.

Describe the solution you'd like.

If you look at the Instrument Component you will see it has a few features Paper is missing:

  • sound_event/sound definition: Use any sound key
  • use_duration: A non-negative float for how long the use duration is.
  • range: A non-negative float for the range of the sound.
  • description: Text component to use as the instruments description

Describe alternatives you've considered.

Using NMS or NBT to make this happen, but that isn't really a viable option for me.

Other

No response

rustic folioBOT
#
lynxplay
[PaperMC/Paper] New comment on issue: #11955 Custom model data commands not working properly on 1.21.4 build 97

Even tested this with a server supplied resource pack, I cannot reproduce this.
Other triage members also were not able to reproduce this either.

There is not much we can do here if we cannot replicate this issue.
I guess ensure the pack sent from the server is up to date?

Given I cannot replicate this, I'll be closing the issue.
Obviously sorry for closing this as you obviously experience some issue, but the github bug tracker is not really the place to debug your personal setup.
Can ask around in paper-help on our discord, maybe someone wants to run through some more potential pitfalls.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
JHarris12345
[PaperMC/Paper] Issue opened: #11979 Fire aspect on boat issue

Expected behavior

Hitting a boat with a fire aspect weapon should not destroy it if the VehcileDestroyEvent is cancelled

Observed/Actual behavior

If you cancel the VehicleDestroyEvent and destroy a boat with a fire aspect weapon, the event will get called but the boat will get destroyed regardless of cancellation

Steps/models to reproduce

  1. Place a boat
  2. Get a fire aspect sword
  3. Always cancel the VehicleDestroyEvent
  4. Destroy the boat

This doesn't give off any EntityDamageEvent either

Plugin and Datapack List

N/A - tested on plain

Paper version

This server is running Paper version 1.21.1-132-ver/1.21.1@b48403b (2024-11-21T10:14:27Z) (Implementing API version 1.21.1-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.1-DEV-164e116 (MC: 1.21.1)

Other

No response

#
JHarris12345
[PaperMC/Paper] Issue opened: #11980 VehicleDestroyEvent fire aspect issue

Expected behavior

A player destroying a boat with a fire aspect weapon should still show the attacker as the player

Observed/Actual behavior

If you destroy a boat with a fire aspect weapon, the e.getAttacker() in the VehicleDestroyEvent is null. This makes it impossible to get the player from this event if they use fire aspect

Steps/models to reproduce

  1. Destroy boat with fire aspect weapon
  2. See how the e.getAttacker() is null

Plugin and Datapack List

N/A

Paper version

This server is running Paper version 1.21.1-132-ver/1.21.1@b48403b (2024-11-21T10:14:27Z) (Implementing API version 1.21.1-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.1-DEV-164e116 (MC: 1.21.1)

Other

No response

#
JHarris12345
[PaperMC/Paper] Issue opened: #11981 PotionSplashEvent doesn't prevent putting out fires

Expected behavior

If you cancel the PotionSplashEvent, it should prevent putting out fires with water potions

Observed/Actual behavior

Fires are always put out regardless of cancellation

Steps/models to reproduce

  1. Throw water splash potion (or lingering) at a fire
  2. Cancel potion splash event
  3. Fire goes out

Plugin and Datapack List

N/A

Paper version

This server is running Paper version 1.21.1-132-ver/1.21.1@b48403b (2024-11-21T10:14:27Z) (Implementing API version 1.21.1-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.1-DEV-164e116 (MC: 1.21.1)

Other

No response

rustic folioBOT
#
Doc94
[PaperMC/Paper] New comment on issue: #11980 VehicleDestroyEvent fire aspect issue

Fire Aspect call a ignite effect for well apply the fire.. this call EntityCombustEvent/EntityCombustByEntityEvent where the author is send (for the ByEntity) later when damage the entity the "damager" its not passed then when comes to the place where VehicleDestroyEvent is called the damager its not the player....
i dont find a nice way to make the call of damage for fire also know the damager... (or can add a extra field in entity for save the damager in this case and clear when the fire is clear... but still not clean)

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
MrPowerGamerBR
[PaperMC/Paper] New comment on pull request: #11774 Add PlayerUseBowWithoutProjectileEvent

Using this event to set an arrow fails when the player looks at a block instead of air (it still gets called).
I don't find an obvious reason for that.

I'm toying around with the same idea, and what I've noticed is that, if you hold the bow on the off hand, the bow does pull back.

If you listen to PlayerInteractEvent, you'll notice that the hand is called first, then the off hand second, but this DOES NOT happen when pulling the bow in the air, so maybe the server is overriding the use state when the off hand interact is called.

However, for some reason, cancelling the off hand event sadly does not make the main hand bow work... Cancelling the off hand does make the off hand bow not work tho.

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
NoyException
[PaperMC/Paper] Issue opened: #11982 Support using translatable component in ItemStack displayName and lore

Expected behavior

The displayName of ItemStack should be translated according to the player's locale.

Observed/Actual behavior

It only displays the key. (For example, block.minecraft.diamond_block instead of Diamond Block)

Steps/models to reproduce

    @Override
    public void onEnable() {
        GlobalTranslator globalTranslator = GlobalTranslator.get();

        TranslationRegistry translationRegistry = TranslationRegistry.create(Key.key("translationtest", "main"));

        translationRegistry.register("abc.myKey", Locale.US, new MessageFormat("This is a test."));
        translationRegistry.register("abc.myKey", Locale.GERMAN, new MessageFormat("Dies ist ein Test."));

        globalTranslator.addSource(translationRegistry);

        TranslatableComponent translatable = Component.translatable("abc.myKey");

        getCommand("ttest").setExecutor((sender, command, label, args) -> {
            sender.sendMessage(translatable);

            Player play...
#
Machine-Maker
[PaperMC/Paper] New comment on issue: #11982 Support using translatable component in ItemStack displayName and lore

As far as I can tell, the use of that plugin does not properly handle creative mode. The problem with doing this for items has always been creative mode. When you desync itemstack state between the client and the server, creative mode will override the server's state. Which will essentially remove the "translatable" part and just make the item rendered in whatever client's language for everyone on the server.

#
NoyException
[PaperMC/Paper] New comment on issue: #11982 Support using translatable component in ItemStack displayName and lore

As far as I can tell, the use of that plugin does not properly handle creative mode. The problem with doing this for items has always been creative mode. When you desync itemstack state between the client and the server, creative mode will override the server's state. Which will essentially remove the "translatable" part and just make the item rendered in whatever client's language for everyone on the server.

EDIT: I also just realized that the plugin page acknowledges this shortcoming and how it's not easily solvable

This is indeed a sad thing. But does that mean we can't accept an imperfect solution and would rather leave it as it is now

rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #11982 Support using translatable component in ItemStack displayName and lore

That โ€œimperfect solutionโ€ is literally โ€œintroduce a mechanism that will
mean that the base game will break your items in a well known use caseโ€.

If you want to put your items at risk of being modified by the client using
a plugin, thatโ€™s up to you, but this is just not something we can do inside
of paper by default, and Iโ€™m not sure it would ever make sense to slap this
behind a config due to how oddball destructive it is for a relatively niche
thing, mojang already has a solution for this.

On Fri, 17 Jan 2025 at 06:27, Noy @.***> wrote:

As far as I can tell, the use of that plugin does not properly handle
creative mode. The problem with doing this for items has always been
creative mode. When you desync itemstack state between the client and the
server, creative mode will override the server's state. Which will
essentially remove the "translatable" part and just make the item rendered
in whatever client's language for everyone on the server.

...

#
NoyException
[PaperMC/Paper] New comment on issue: #11982 Support using translatable component in ItemStack displayName and lore

That โ€œimperfect solutionโ€ is literally โ€œintroduce a mechanism that will
mean that the base game will break your items in a well known use caseโ€.

If you want to put your items at risk of being modified by the client using
a plugin, thatโ€™s up to you, but this is just not something we can do inside
of paper by default, and Iโ€™m not sure it would ever make sense to slap this
behind a config due to how oddball destructive it is for a relatively niche
thing, mojang already has a solution for this.
โ€ฆ

I see. That is very reasonable.

But since you mentioned that "mojang already has a solution for this", does that mean that the issue will be officially fixed in future releases?

rustic folioBOT
#
kangarko
[PaperMC/Paper] New comment on issue: #11768 Missing DataComponentValueConverter for MiniMessage -> Gson conversions

To add, this is still happening on the latest paper (1.21.4-114-main@cd9d6d6 (2025-01-15T11:21:28Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)):

[13:36:47 WARN]: java.lang.IllegalArgumentException: There is no data holder converter registered to convert from a class net.kyori.adventure.nbt.api.BinaryTagHolderImpl instance to a interface net.kyori.adventure.text.serializer.gson.GsonDataComponentValue (on field minecraft:banner_patterns)
[13:36:47 WARN]: 	at net.kyori.adventure.text.event.DataComponentValueConverterRegistry.convert(DataComponentValueConverterRegistry.java:90)
[13:36:47 WARN]: 	at net.kyori.adventure.text.event.HoverEvent$ShowItem.dataComponentsAs(HoverEvent.java:623)
[13:36:47 WARN]: 	at net.kyori.adventure.text.serializer.gson.ShowItemSerializer.write(ShowItemSerializer.java:151)
[13:36:47 WARN]: 	at net.kyori.adventure.text.serializer.gson.ShowItemSerializer.write(ShowItemSerializer.java:50)
[13:36:47 WARN]: 	at com.google.gson.TypeAdapter$1.write(TypeAda...
rustic folioBOT
rustic folioBOT
#
Samsuik
[PaperMC/Paper] Issue opened: #11983 Pushing a block outside the world border creates a glitchy moving block

Expected behavior

The piston should be able to push the block through the world border, or the block should be stopped as if it was pushed against an immovable block.

Observed/Actual behavior

After a block is pushed through the world border it turns into a glitchy moving block. I am unable to interact with the block and after reconnecting the block turns invisible but I can still collide with it.

If you increase the world border size the block will be able to move.

https://youtu.be/aEOddMXy0OA

Steps/models to reproduce

Push a block through the world border using a normal piston

Plugin and Datapack List

n/a

Paper version

This server is running Paper version 1.21.4-114-main@cd9d6d6 (2025-01-15T11:21:28Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are 2 version(s) behind
Download the new version at: https://papermc.io/downloads/paper

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Mitality
[PaperMC/Paper] Issue opened: #11984 EntityDamageByBlockEvent.getDamager() returning null for lava and possibly more damage sources

Expected behavior

Up to version 1.21.1, EntityDamageByBlockEvent.getDamager() returns the block responsible for the damage dealt as a valid Block object for damage caused by lava. I therefore expected it to work that way in newer versions too, as it's most certainly the expected behavior of this method in this case.

Observed/Actual behavior

On 1.21.3+, EntityDamageByBlockEvent.getDamager() just returns null for damage caused by lava.

Steps/models to reproduce

Here's what I did:

  • Create 2 localhost servers, one with PaperMC 1.21.1 and the other one with PaperMC 1.21.3 (newest builds).
    (no changes to those servers except accepting the eula)

  • Create a minimalistic plugin to log EntityDamageByBlockEvent.getDamager().

  • Join both servers, grab a bucket of lava and take a bath.

  • Compare console output:
    1.21.1: CraftBlock{...}
    1.21.3: null

Plugin and Datapack List

Just the minimalistic logging plugin:

public final class GetDamagerTest extend...
rustic folioBOT
#
alexfry756
[PaperMC/Paper] Issue opened: #11985 The trial Spawner requires 2 spawn eggs mob each.

Expected behavior

The trial Spawner requires 2 spawn eggs mob each. I played in the single player world and found that it takes 1 mob egg to activate the spawner. Also, during the test, I deleted the CMI plugin to test my hypothesis and everything remained the same, 2 eggs are still needed.. If possible, please fix it, I will be very grateful.

Observed/Actual behavior

A trial spawner must spend 1 Mob Summoning egg to activate instead of two mob summoning eggs.

Steps/models to reproduce

Plugin and Datapack List

AuthMe, Citizens, CMI, CMILib, DeluxeMenus, Interactions, PL-Hide, PlaceholderAPI, WorldEdit, WorldGuard

Paper version

This server is running Paper version 1.21.4-114-main@cd9d6d6 (2025-01-15T11:21:28Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are 3 version(s) behind
Download the new version at: https://papermc.io/downloads/paper
Previous version: 1.21.4-113-8e80d4e (MC: 1.21.4)

Other

No response

rustic folioBOT
rustic folioBOT
#
andfont
[PaperMC/Paper] Issue opened: #11986 Incorrect package name for class fatal error

Expected behavior

When I want to read the class package it does not give the information correctly and it is impossible to know the version, example:

public static String getBukkitVersion() {
String[] v = Bukkit.getServer().getClass().getPackage().getName().replace(".", ",").split(",")[3].split("-")[0].split("_");
return v[0].replace("v", "") + "." + v[1];
}

example error in plugin:
09:11:53 ERROR]: Error occurred while enabling PsABF v2.0.92 (Is it up to date?)
java.lang.ArrayIndexOutOfBoundsException: Index 3 out of bounds for length 3
at PsABF.jar/me.PsABF.pb.Main.getBukkitVersion(Main.java:58) ~[PsABF.jar:?]
at PsABF.jar/me.PsABF.pb.Main.onEnable(Main.java:66) ~[PsABF.jar:?]

Observed/Actual behavior

public static String getBukkitVersion() {
String[] v = Bukkit.getServer().getClass().getPackage().getName().replace(".", ",").split(",")[3].split("-")[0].split("_");
return v[0].replace("v", "") + "." + v[1];
}

exa...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #11988 setNoDamageTicks is broken

Expected behavior

setNoDamageTicks should make a given player invulnerable for the given amount of ticks

Observed/Actual behavior

They still take damage (in my case from explosions)

Steps/models to reproduce

player.setNoDamageTicks
world.createExplosion next to the player
they die from explosion damage

Plugin and Datapack List

N/A

Paper version

This server is running Paper version 1.21.4-DEV-main@a35cfe9 (1970-01-01T00:00:00Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are 1 version(s) behind
Download the new version at: https://papermc.io/downloads/paper
Previous version: 1.21.4-DEV-cd9d6d6 (MC: 1.21.4)

Other

No response

#
lynxplay
[PaperMC/Paper] New comment on issue: #11988 setNoDamageTicks is broken

So, this is a rather weird mechanic.

Previously, minecraft players had a spawn invulnerability which would behave like you described above (negate all damage).
setNoDamageTicks on players specifically would set the entities invul timer and the spawn invuln timer.
That is now gone in 1.21.4, changed by mojang. Players will take damage the second they load into the world.
The player setter hence cannot set that, and now players behave like any other entity in this regard, which is
"you are only invulnerable to damage sources that deal less damage than the damage that initially caused your invul", which in case of setNoDamageTicks is 0, so everything just damages.

This can obviously be fixed, but it feels pretty hacky and isn't a mechanic we really would wanna maintain I don't think?
Especially when better, event driven mechanics approaches exist to prevent damage on players.

rustic folioBOT
#
Doc94
[PaperMC/Paper] New comment on pull request: #11840 Add ExplodeEvent and Improvement in Spigot Explode events

With your PR that fire explosion doesn't spread fire on the ground which is an unintended side effect

i see the issue... not sure when i broke maybe the change from getType.isAir to isEmpty... i make a change for allow the empty things for the fire and looks is working. if its not a good "fix" can just revert the thing for the empty check

also i remove the shuffle in the KEEP behaviour because just make changes not related to the vanilla behaviour...

rustic folioBOT
#
lynxplay
[PaperMC/Paper] New comment on issue: #11988 setNoDamageTicks is broken

Yea we could but there is already a method for this and it would mess with the API getters on the last damage.
So the API was relying on internal behaviour that has been removed, at which point the usual way to go would be deprecation and then removal of the method, but in this case only the player-speciifc side effect was removed, which makes this a rather ugly case as we don't have a clear migration path for this.

In my honest opinion, this should just be implemented via EntityDamageEvent cancellation (that code would work backwards as well, no one is loosing anything there), but I'll leave it open for others.

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
SoSeDiK
[PaperMC/Paper] Pull request review comment: #11960 Fix inconsistencies between offline/online spawn position getter

What i think is confusing is the fact that an online player extends an offline player in the api.

Yes, that was my concern. Technically, you can have OfflinePlayer object representing an online Player instance (#getOfflinePlayer too returns online player if present), so you may call the method on an offline player expecting not to load chunks, but they will load due to the object being an online player.

New devs might stumble upon this quirk and will have to account for it. Deprecation would require to explicitly specify the wanted behavior, but I don't have a strong feeling against leaving this as is.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
MrPowerGamerBR
[PaperMC/Paper] New comment on pull request: #11774 Add PlayerUseBowWithoutProjectileEvent

For the sake of avoiding other server owners and developers banging their wall against the head trying to figure out how to do this: I wanted to create a plugin that uses the bow behavior to with a custom bow that does not require arrows (context: created a cue stick model for a bow, and I created a custom plugin to play billiards (more bow force = more ball velocity) without needing any weird hacks like spawning fake arrows in their inventory)

To do this, I needed to create this patch (similar to
this PR's patch but calls before any of the other arrow related events)

Changing the server to say that "hey there's an arrow for your bow!" DOES work if you are aiming at the sky, but does NOT work if you are looking at a block, because the client knows that there isn't an arrow in the inventory, so it ...

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Spottedleaf
[PaperMC/Paper] New branch created: test/save-on-flush
rustic folioBOT
#
niho2
[PaperMC/Paper] New comment on pull request: #11994 Consider off-hand when adding items to player inventory

~Fixes #11921~ (Closing as won't fix)

After review, it was determined that the proposed solution would break the API contract specified in the Inventory.getStorageContents() javadocs. The off-hand slot is intentionally excluded from storage contents, and addItem() should only work with storage contents as specified.

Instead of modifying addItem(), we should consider adding a new method addItemAnySlot() (similar to the existing removeItemAnySlot()).

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
NonSwag
[PaperMC/Paper] Issue opened: #11996 Setting the background color of a text display to null doesn't reset it

Expected behavior

I would expect that setting the background color of a text display to null would reset it to its default value
like in the picture
Image

Observed/Actual behavior

The background becomes a solid grayish white

Image

Steps/models to reproduce

  1. Spawn a text display
  2. set the background color to null

Plugin and Datapack List

irrelevant

Paper version

[18:18:02 INFO]: Checking version, please wait...
[18:18:02 INFO]: This server is running Paper version 1.21.4-118-main@73f7736 (2025-01-18T13:37:35Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.4-117-a35cfe9 (MC: 1.21.4)

Other

I skipped over the implementation, and the default is not -1 but 1073741824 now
using this code proves it

if (tagOptions.get...
rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #11996 Setting the background color of a text display to null doesn't reset it

Setting it to null would imply that there is no specific override of this,
so, apply parent derived style?

On Tue, 21 Jan 2025 at 17:27, David @.***> wrote:

Expected behavior

I would expect that setting the background color of a text display to null
would reset it to its default value
like in the picture
image.png (view on web)
https://github.com/user-attachments/assets/f3314dfa-3786-40a1-9d69-801a2eb66ffc
Observed/Actual behavior

The background becomes a solid grayish white

image.png (view on web)
https://github.com/user-attachments/assets/5c25a3d5-255c-45ec-a7b1-3b99285dc247
Steps/models to reproduce

  1. Spawn a text display
  2. set the background color to null

Plugin and Datapack List

irrelevant
Paper version

[18:18:02 INFO]: Checking version, please wait...
[18:18:02 INFO]: This server is running Paper version
@.*** (2025-01-18T13:37:35Z) (Implementing API version
1.21.4-R0...

rustic folioBOT
#
Jyguy
[PaperMC/Paper] Issue opened: #11997 Taking damage after hitting an entity slows sprinting speed

Expected behavior

Taking damage after hitting an entity should not alter sprinting speed. This is the behaviour seen on the vanilla server.

Observed/Actual behavior

The FOV and sprinting speed are reduced when taking damage after hitting an entity during uninterrupted sprinting.

Steps/models to reproduce

  1. Hold down your sprint key (or, enable toggle sprint)
  2. Begin running by holding W
  3. Attack an entity
  4. Take any form of damage (e.g. fire, fall)
  5. Observe that FOV and running speed are reduced

Demonstration video: https://youtu.be/MytXWu4hQGc

Plugin and Datapack List

None

Paper version

This server is running Paper version 1.21.4-118-main@73f7736 (2025-01-18T13:37:35Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version

Other

Enabling paper world config misc.disable-sprint-interruption-on-attack is a workaround to this issue but has other side-effects such as every consecutive hit being registered as a sprin...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Doc94
[PaperMC/Paper] New comment on pull request: #11840 Add ExplodeEvent and Improvement in Spigot Explode events

That change is completely irrelevant, i have already explained the cause of that issue in the original message. It's not only affect your new event but regular ones as well. So i will open a new issue about that (the fix you commited is a workaround). And you broke spigot further with your equivalent commit.

Yeah for the not explosion is handled by the keep maybe and how im now calm the event in that place the filter take all... Currently not sure why even need filter that... That block is in the list maybe just send all blocks and if user wanna they can filter that... Based in the first commit.
Or not sure if you have another suggest for this thing

#
Lulu13022002
[PaperMC/Paper] Issue opened: #11999 Fire explosion are reduced

Expected behavior

I expect the fireball to spread fire on the obsidian at the impact like in vanilla.

https://github.com/user-attachments/assets/2540734b-ad1d-4cae-b9fd-7baa33296dee

Observed/Actual behavior

The fire is not spread on the platform.

https://github.com/user-attachments/assets/04ea3d22-afd1-4073-bce1-29e817dba174

Steps/models to reproduce

  • Create a new world
  • Build an obsidian platform of 3x3 in the air
  • Summon a fireball (/summon minecraft:fireball)
  • Throw the fireball on the obsidian platform
  • Notice if the fire spread on the ground or not

Plugin and Datapack List

None

Paper version

30fdfb1aa173a8362e3a5bf19a4e309341f3ee19

Other

I have discovered this issue while reviewing https://github.com/PaperMC/Paper/pull/11840 which copied the old implementation.

rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12000 Attempted Double World add on ItemEntity['Goat Horn'

Stack trace

[11:59:03] [Server thread/ERROR]: Attempted Double World add on ItemEntity['Goat Horn'/298549, uuid='aeab8c83-5fdc-43a8-beee-40cae4aee3c5', l='ServerLevel[world]', x=XXXX53.26, y=16.13, z=XXXX17.03, cpos=[XX03, XX32], tl=0, v=true]
java.lang.Throwable: null
	at net.minecraft.server.level.ServerLevel.addEntity(ServerLevel.java:1510) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.server.level.ServerLevel.addFreshEntity(ServerLevel.java:1454) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.server.level.ServerLevel.addFreshEntity(ServerLevel.java:1449) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.world.entity.Entity.spawnAtLocation(Entity.java:2803) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.world.entity.animal.goat.Goat.dropHorn(Goat.java:362) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.world.entity.ai.behavior.RamTarget.tick(RamTarget.java:97) ~[paper-1.21.4.jar:1.21.4-DEV-a35cfe9]
	at net.minecraft.world.e...
#
lynxplay
[PaperMC/Paper] New branch created: bugfix/double-goat-horn-addition
#
kennytv
[PaperMC/Paper] branch deleted: bugfix/double-goat-horn-addition
rustic folioBOT
rustic folioBOT
#
Doc94
[PaperMC/Paper] New comment on pull request: #11840 Add ExplodeEvent and Improvement in Spigot Explode events

You can't remove the filter for the old events without updating the javadocs which is kinda a breaking change so not sure but the new one can hold empty blocks fine. Cancelling the events still spread fire it seems mainly because you removed the filter.

well already can consider a "breaking" change the thing about how now spigot fire the events for the keep behaviour and paper not (not in that events but yes in the new) but i make the rollback with the patch for the bukkit events.

the fire thing is maybe because just check the cancel for the keep behaviour and not in the main condition.. i move that for cover all calls.

rustic folioBOT
#
Spottedleaf
[PaperMC/Paper] branch deleted: test/save-on-flush
rustic folioBOT
rustic folioBOT
#
gragorther
[PaperMC/Paper] Issue opened: #12003 Minecraft server crashes and fills my logs

Stack trace

https://mclo.gs/dqO62Wz

Plugin and Datapack List

AdvancedTeleport, BetterChairs, BetterMessages, MiniMOTD, PingDisplay, Plan, ProtocolLib, SeeMore, ViaBackwards, ViaVersion

Actions to reproduce (if known)

No response

Paper version

This server is running Paper version 1.21.4-121-main@88bbead (2025-01-22T20:05:15Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.4-118-73f7736 (MC: 1.21.4)

Other

I use https://github.com/itzg/docker-minecraft-server and here is my docker compose file:

services:
  mc:
    image: itzg/minecraft-server
    container_name: paper
    labels:
      - traefik.enable=true
      - traefik.http.routers.http-analytics.entryPoints=http
      - traefik.http.routers.http-analytics.rule=Host(`analytics.gregtech.eu`)
      - traefik.http.middlewares.https_redirect.redirectscheme.scheme=https
      - traefik.http.middlewares.https_redirect.redirectscheme.perman...
rustic folioBOT
#
gragorther
[PaperMC/Paper] New comment on issue: #12003 Minecraft server crashes and fills my logs

Also could it perhaps have something to do with this warning I got from LibertyBans, which I just tried to install:

paper  | [16:45:13 INFO]: [LibertyBans] Enabling LibertyBans v1.1.0-RC2
paper  | [16:45:13 WARN]: [LibertyBans] *******************************************
paper  | We have detected bugs on your server which threaten your server's stability.
paper  | LibertyBans will continue to operate unaffected, but we strongly suggest you fix these bugs.
paper  | 
paper  | These bugs are (most likely) due to other plugins' mistakes. Each of the following plugins has shaded a library but did not relocate it. You should report each bug to the plugin author.
paper  | 
paper  | Plugin Name | Library Name | Class Detected
paper  | ----------------------------------------------
paper  | Plugin 'Plan 5.6 build 2959' | Jakarta-Inject | jakarta.inject.Provider
paper  | 
paper  | Note for advanced users: Understanding the consequences, you can minimize this warning by setting the system ...
#
electronicboy
[PaperMC/Paper] New comment on issue: #12003 Minecraft server crashes and fills my logs

Looks like a corrupted chunk, and the thing is stuck in a tizzy trying to read it

at 118,161 right?

Yes

Also could it perhaps have something to do with this warning I got from LibertyBans, which I just tried to install:

paper  | [16:45:13 INFO]: [LibertyBans] Enabling LibertyBans v1.1.0-RC2
paper  | [16:45:13 WARN]: [LibertyBans] *******************************************
paper  | We have detected bugs on your server which threaten your server's stability.
paper  | LibertyBans will continue to operate unaffected, but we strongly suggest you fix these bugs.
paper  | 
paper  | These bugs are (most likely) due to other plugins' mistakes. Each of the following plugins has shaded a library but did not relocate it. You should report each bug to the plugin author.
paper  | 
paper  | Plugin Name | Library Name | Class Detected
paper  | ----------------------------------------------
paper  | Plugin 'Plan 5.6 build 2959' | Jakarta-Inject | jakarta....
rustic folioBOT
rustic folioBOT
#
alexfry756
[PaperMC/Paper] Issue opened: #12004 Possibility to prohibit custom heads on the server. Please help!

Is your feature request related to a problem?

I, like several server authors, am the author of a fantasy RPG server and it has custom heads with unique characteristics, when a head with a custom texture must be held in a certain hand. But as soon as you place a head on any block, all its characteristics are lost.

Describe the solution you'd like.

Good day, dear developers. I would like to offer you a function for switching the installation of player heads on blocks. This is due to the fact that there is no worthy alternative to plugins in prohibiting the placement of heads on blocks. Server administrators could optionally switch whether it is possible to place heads on their servers.

Of course, there are servers in which builders place player heads with custom textures, but there are also servers where player heads are used as runes and custom items with unique attributes like in RPG games. But here's the problem, there are no normal plugins for 1.21.4 that could work w...

rustic folioBOT
#
NonSwag
[PaperMC/Paper] Issue opened: #12005 Player#sendEquipmentChange produces invalid packet with empty map (causes disconnect)

Expected behavior

Method call should silently fail

Observed/Actual behavior

ClientboundSetEquipmentPacket is sent anyway

Steps/models to reproduce

invoke Player#sendEquipmentChange with an empty map as slots argument

Plugin and Datapack List

irrelevant

Paper version

[21:30:36 INFO]: Checking version, please wait...
[21:30:36 INFO]: This server is running Paper version 1.21.4-121-main@88bbead (2025-01-22T20:05:15Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.4-118-73f7736 (MC: 1.21.4)

Other

ClientboundSetEquipmentPacket does not allow for empty slots

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Stis780
[PaperMC/Paper] Issue opened: #12006 InventoryMoveEvent (possible dupe with plugins)

Expected behavior

Amount for first slot in inventory in InventoryMoveEvent is bugged by Hoppers patch, I would expect the actual amount.

If InventoryMoveEvent is cancelled I wouldn't expect patch code to go rollback inventory or amounts since those could be already removed by other plugins or code... introducing dupes.

Observed/Actual behavior

I think all versions with this patch are affected:

https://github.com/PaperMC/Paper/blob/88bbead13b07914c01866e35df16b72110fa501e/paper-server/patches/features/0028-Optimize-Hoppers.patch

When server is using non-standard move amount for hoppers (ie 2,3,64... whatever but not default 1) and Plugins are modifying amounts or items in InventoryMoveEvent, items could get duped.

This would only work for items in first slot because of bugged patch in question.

After InventoryMoveEvent is fired, amount on slot 1 is rolled back even if InventoryMoveEvent is cancelled. Which shouldn't be the case?

Since InventoryMoveEvent was cancelled...

rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12009 PlayerInteractEvent RIGHT_CLICK_BLOCK fired when interacting with entities

Expected behavior

right clicking an entity with blocks behind them shouldn't fire playerinteractevent for the block

Observed/Actual behavior

It fires playerinteractevent for the block behind the entity

Steps/models to reproduce

right click a mob with blocks behind them

Plugin and Datapack List

N/A

Paper version

This server is running Paper version 1.21.4-DEV-main@88bbead (1970-01-01T00:00:00Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are 1 version(s) behind
Download the new version at: https://papermc.io/downloads/paper
Previous version: 1.21.4-121-88bbead (MC: 1.21.4)

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #12006 InventoryMoveEvent (possible dupe with plugins)

Generally, the type of interactions that would cause issues with the changes done here, outside of knowing the source stack size, have generally not been an issue given how precarious modifying state in any of these sorts of events is. Removing the clone was pretty important in terms of performance boosting here due to how expensive cloning the items is, especially with how hot this tight loop can get. This is, however, not a huge deal anymore given that item stacks now use immutable data components and the container for that supports a COW, and so, I think that this side-effect is actually mitigatable now

rustic folioBOT
#
Stis780
[PaperMC/Paper] New comment on issue: #12006 InventoryMoveEvent (possible dupe with plugins)

Amount being wrong for item in first slot is known issue, but that's not all.

Source inventory itemstacks simply CAN NOT be modified safely in IMIE, even if item is set in IMIE, IMIE is cancelled or something else.

In theory one could set transfer amount to 64, trying to work around this, then manually implement hopper code to reduce transfer amount do to 1 or whatever.

However this would always be required to run with 1 tick delay which would put redstone stuff out of sync.

@Warriorrrr I can put something together real quick if needed but any code that would try to modify items in source inventory during IMIE would fail if it touches first slot. Doesn't matter if calledSetItem is true, calledGetItem is true or whole IMIE is cancelled.

One would expect that if IMIE is cancelled, I can do whatever I want with source or target inventory, during IMIE.

Maybe it would also make sense that changing Item using setItem() on IMIE would change items that are being transferred? IMIE#ge...

#
electronicboy
[PaperMC/Paper] New comment on issue: #12006 InventoryMoveEvent (possible dupe with plugins)

One would expect that if IMIE is cancelled, I can do whatever I want with source or target inventory, during IMIE.

That has pretty much ever been promised from the API, cancelling an event will often run a form of "revert back to the state we saw before this event", as there's no nice way to properly manage what should occur when an event is cancelled outside of reverting the change, and there is generally no state tracking to see if people mess with stuff considered out of scope.

Even the original implementation of this event will wipe out any change you did to the source item inside of the inventory when cancelled. Hopper stuff is also generally a really precarious place due to how tight the logic is there, it makes doing much pretty harsh, especially given the performance complexities

rustic folioBOT
#
Stis780
[PaperMC/Paper] New comment on issue: #12006 InventoryMoveEvent (possible dupe with plugins)

I'm not sure if generic tick event for block entities wouldn't get called too often? In terms of performance?

Cancelling and doing own logic won't work (and ig this isn't promised by api, so ok api would roll it back).

Trying to use setItem() to implement own logic won't work (this should... could work?).

https://github.com/PaperMC/Paper/blob/88bbead13b07914c01866e35df16b72110fa501e/paper-server/patches/features/0028-Optimize-Hoppers.patch#L195

setItem() method on IMIE is not working as intended tho? I mean it's using getItem() from IMIE to add new items to target inventory when item is modified in IMIE right?

Why would it not also consider item from IMIE when removing an item from source inventory?

With that and counts on first slot fixed (that temp manipulation removed) there would no need for another event? Filters, different transfer amounts, etc, could simply be implemented in IMIE? Unless I'm missing something here.

#
electronicboy
[PaperMC/Paper] New comment on issue: #12006 InventoryMoveEvent (possible dupe with plugins)

IMIE basically works in this weird area of being before the item is placed into the target inventory, but it does a whole bunch of mutations to achieve that; changing the stack in the event would change the item which is tried to be added into the target inventory.

This event is generally a pretty hot area that nobody really wants to touch due to how precarious the implementation is, only made even more challenging due to the optimisation patch

rustic folioBOT
#
ColinBashful
[PaperMC/Paper] Issue opened: #12012 More Events for Recipe related stuff

Is your feature request related to a problem?

I'm currently developing an API allowing recipes with custom items as input, but there isn't any way to do the following:

  • Start smelting an item if a condition is met
  • Modify the result for the Crafter Block (not only if a valid recipe is given)

Describe the solution you'd like.

For the furnace:
An event triggered every tick, right before the furnace checks if it can smelt the given item.
This event should allow modifying the total cooking time required and provide a method to explicitly indicate whether the item can be smelted.

For the Crafter:
An event that's similar to PrepareItemCraftEvent.

Describe alternatives you've considered.

I tried mixins, but they don't work when only using PaperMC

Other

No response

rustic folioBOT
rustic folioBOT
#
NonSwag
[PaperMC/Paper] Issue opened: #12014 Sittable does not extend any entity class

Expected behavior

As documented, Sittable applies to Animals and since all implementing classes of Sittable share the Animals class it should extend Animals

Observed/Actual behavior

Sittable does not extend any entity class

Steps/models to reproduce

Plugin and Datapack List

Paper version

This server is running Paper version 1.21.4-126-main@fb5b173 (2025-01-25T20:51:32Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.4-121-88bbead (MC: 1.21.4)

Other

this makes it not possible to use sittable in generic contexts where you expect an instance of entity
Image

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Doc94
[PaperMC/Paper] New comment on pull request: #12017 Adding PredicateChoice to Paper API (updated version)

I just noticed there is already a PredicateRecipeChoice for potions, with the addition of the PredicateChoice this could be a bit confusing. Maybe it should be renamed to something more clear like a PotionRecipeChoice/PotionPredicateChoice?

well rename sounds a little breaking i think, maybe this new PredicateChoice can be named ItemPredicateChoice (? or add an extension of RecipeChoice for Potion and move things to that for make clear the diff... but not really sure.

rustic folioBOT
#
lynxplay
[PaperMC/Paper] New comment on pull request: #12019 Bring Entity#teleportAsync up to parity with Entity#teleport

Hi and welcome to paper :wave:

I don't see the usage of this? Why would you teleportAsync to an entity?
The entire point of that method is that you might teleport into a chunk that isn't loaded yet.
An entity instance existing means the respective chunk said entity is in also exists.
Just on a type-basis I don't see a usecase for these, but maybe I am just missing something. Could you elaborate on that a bit?

#
zJanny
[PaperMC/Paper] New comment on pull request: #12019 Bring Entity#teleportAsync up to parity with Entity#teleport

Hi and welcome to paper ๐Ÿ‘‹

I don't see the usage of this? Why would you teleportAsync to an entity? The entire point of that method is that you might teleport into a chunk that isn't loaded yet. An entity instance existing means the respective chunk said entity is in also exists. Just on a type-basis I don't see a usecase for these, but maybe I am just missing something. Could you elaborate on that a bit?

As far as i know, folia needs the teleportAsync method, so this would allow someone to create a Paper plugin with Folia support that uses this functionallity.

This would be one example i can think of, because i just encountered this issue with GrimAC

#
lynxplay
[PaperMC/Paper] New comment on pull request: #12019 Bring Entity#teleportAsync up to parity with Entity#teleport

Yea, I think I'll close this then.

Thank you anyway for the PR <3
For documentation purposes as to why the PR was closed:

Teleporting to an entity instance means the target entity exists.
Unless a plugin is doing terrible things, it is also loaded, as such the teleport target location is loaded.
As visible in the PR, this is then pretty much just syntax sugar for CompletableFuture.completedFuture(entity.teleport(targetEntity)), which is just not enough code saved to really be worth yet another method around teleportation, which is already one of the most overloaded bits of logic we have :sweat_smile:

Thank you again! Hope my line of reasoning here was understandable.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
LGDTimtou
[PaperMC/Paper] Issue opened: #12021 "Asynchronous Chunk getEntities call" when tab completing on: "tag @e remove" in console

Stack trace

> tag @e remove
[22:57:58 ERROR]: Thread Server console handler failed main thread check: Chunk getEntities call
java.lang.Throwable: null
        at org.spigotmc.AsyncCatcher.catchOp(AsyncCatcher.java:11) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.server.level.ServerLevel.getEntities(ServerLevel.java:2443) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.server.level.ServerLevel.getEntities(ServerLevel.java:1405) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.selector.EntitySelector.addEntities(EntitySelector.java:172) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.selector.EntitySelector.findEntities(EntitySelector.java:157) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.EntityArgument.getOptionalEntities(EntityArgument.java:69) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.EntityAr...
rustic folioBOT
rustic folioBOT
#
xism4
[PaperMC/Paper] New comment on issue: #12021 "Asynchronous Chunk getEntities call" when tab completing on: "tag @e remove" in console

Stack trace

> tag @e remove
[22:57:58 ERROR]: Thread Server console handler failed main thread check: Chunk getEntities call
java.lang.Throwable: null
        at org.spigotmc.AsyncCatcher.catchOp(AsyncCatcher.java:11) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.server.level.ServerLevel.getEntities(ServerLevel.java:2443) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.server.level.ServerLevel.getEntities(ServerLevel.java:1405) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.selector.EntitySelector.addEntities(EntitySelector.java:172) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.selector.EntitySelector.findEntities(EntitySelector.java:157) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.commands.arguments.EntityArgument.getOptionalEntities(EntityArgument.java:69) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
        at net.minecraft.comm...
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
myToasterDied
[PaperMC/Paper] Issue opened: #12023 TrialSpawner API is very lacking

Is your feature request related to a problem?

The trial spawner tilestate lacks features - the most important one being getting and setting the cooldown_ends_at value.

Providing access to this value would be highly beneficial, as it would allow users to get the remaining time before the trial spawner becomes available again, as well as modify or remove the cooldown as needed.

Describe the solution you'd like.

I propose these 2 functions:

getCooldownEndTime() returns a long that is the value, if there is no value it would return 0 or null.

setCooldownEndTime(Long) sets this value for the tilestate.

Describe alternatives you've considered.

.

Other

Image

Implementing this feature shouldn't be too complex, considering there is already a function in setting and getting the target_cooldown_length value. Of course, that's just my perspective. It's not me acutally doing the...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
NonSwag
[PaperMC/Paper] Issue opened: #12026 Expand Armadillo API

Is your feature request related to a problem?

Currently, it is not possible to define an armadillo's state (idle, rolling scared and unrolling)

Describe the solution you'd like.

A method to set the state

Describe alternatives you've considered.

NMS

Other

No response

rustic folioBOT
#
LoliColleen
[PaperMC/Paper] Issue opened: #12027 Placing blocks on offhand reaches paacket limit more easily

Expected behavior

The maximum speed of placing blocks with the main hand and the off hand should be the same

Observed/Actual behavior

It is easier to make blocks become "ghost blocks" when quickly placing blocks through an off hand. The main reason for blocks to become "ghost blocks" is that packets are sent up to the limit set by paper.

In ServerGamePacketListenerImpl code, can be noticed that:

// Spigot start - limit place/interactions
    private int limitedPackets;
    private long lastLimitedPacket = -1;
    private static int getSpamThreshold() { return io.papermc.paper.configuration.GlobalConfiguration.get().spamLimiter.incomingPacketThreshold; } // Paper - Configurable threshold

    private boolean checkLimit(long timestamp) {
        if (this.lastLimitedPacket != -1 && timestamp - this.lastLimitedPacket < getSpamThreshold() && this.limitedPackets++ >= 8) { // Paper - Configurable threshold; raise packet limit to 8
            return false;
        }

      ...
rustic folioBOT
#
TroDerp
[PaperMC/Paper] New comment on issue: #10655 hide_attributes flag not working

Going to close this as WAI. Mojang really should add a way to hide attributes without having to copy over the default attributes for an item type. We have no interesting in maintaining a hacky fix where we implicitly do that copying over if someone sets the hide flag, and we don't want to keep upstream's broken fix that just deletes all implicit attribute modifiers for items.

Should reconsider fixing this bug for the future MC version (whenever it comes), due to 25w04a changing how tooltips are hidden again and separating "show_in_tooltip" from the attribute component.

rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12029 Updated paper and my bacon got fried

Stack trace

https://pastebin.com/F1Rh90Xy

Server hard crashed right after updating from 30046e041049e4a942ba5367dba1b1aaaf1ba91e to 5a34bf0425f39d7b206a2c7de3bae860740a9da1

Plugin and Datapack List

N/A

Actions to reproduce (if known)

No response

Paper version

This server is running Paper version 1.21.4-DEV-main@5a34bf0 (1970-01-01T00:00:00Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version
Previous version: 1.21.4-DEV-88bbead (MC: 1.21.4)

Other

No response

#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12030 Shut down after hard crash fails at savePlayers

Expected behavior

the error shouldn't halt execution for the whole server jar, worlds should attempt to save

Observed/Actual behavior

The whole server halts execution indefinitely as the shutdown process fails

[17:29:57] [Paper Watchdog Thread/INFO]: [ProtocolLib] Disabling ProtocolLib v5.4.0-SNAPSHOT-739
[17:29:57] [Paper Watchdog Thread/INFO]: Saving players
[17:29:57] [Paper Watchdog Thread/WARN]: Asynchronous Save Players! Blocking thread until it returns 
java.lang.IllegalStateException: null
	at io.papermc.paper.util.MCUtil.ensureMain(MCUtil.java:116) ~[paper-1.21.4.jar:1.21.4-DEV-5a34bf0]
	at net.minecraft.server.players.PlayerList.saveAll(PlayerList.java:1073) ~[paper-1.21.4.jar:1.21.4-DEV-5a34bf0]
	at net.minecraft.server.players.PlayerList.saveAll(PlayerList.java:1069) ~[paper-1.21.4.jar:1.21.4-DEV-5a34bf0]
	at net.minecraft.server.MinecraftServer.stopServer(MinecraftServer.java:1016) ~[paper-1.21.4.jar:1.21.4-DEV-5a34bf0]
	at net.minecraft.server.dedicated...
rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] New comment on issue: #12029 Updated paper and my bacon got fried

Seems like a player probably right clicked while flying with elytra at the edge of a chunk, resulting in a useitempacket "to" field being within an unloaded chunk, I feel like we should be checking the chunk is loaded first though?

diff --git a/net/minecraft/server/network/ServerGamePacketListenerImpl.java b/net/minecraft/server/network/ServerGamePacketListenerImpl.java
index 5738709..9655001 100644
--- a/net/minecraft/server/network/ServerGamePacketListenerImpl.java
+++ b/net/minecraft/server/network/ServerGamePacketListenerImpl.java
@@ -1997,6 +1997,15 @@ public class ServerGamePacketListenerImpl
                 float f8 = f3 * f5;
                 double d3 = this.player.blockInteractionRange();
                 Vec3 to = from.add((double) f7 * d3, (double) f6 * d3, (double) f8 * d3);
+
+                // Paper start - Don't allow using items on unloaded chunks
+                if (this.player.level().getChunkIfLoadedImmediately(Mth.floor(to.x()) >> 4, Mth.floor(to.z(...
rustic folioBOT
rustic folioBOT
#
KioProject123
[PaperMC/Paper] Issue opened: #12036 Exact recipe for the stonecutter is incorrectly displayed

Expected behavior

Unmatched recipes are not displayed or cannot be used

Observed/Actual behavior

Recipe results are displayed incorrectly
Image
Image

Steps/models to reproduce

1ใ€Add an exact recipe for stonecutter

        ItemStack source = new ItemStack(Material.STONE);
        source.editMeta(itemMeta -> {
            itemMeta.itemName(Component.text("a special stone"));
        });
        Bukkit.addRecipe(new StonecuttingRecipe(new NamespacedKey("a", "a"), new ItemStack(Material.DIAMOND), new RecipeChoice.ExactChoice(source)));
        Bukkit.addRecipe(new StonecuttingRecipe(new NamespacedKey("b", "b"), new ItemStack(Material.WHITE_WOOL), Material.STONE));

2ใ€Give yourself a plain stone (/give xxx stone)
3ใ€View on Stonecutter

Plugin and Datapack List

Only test plugin

Paper versio...

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12037 The blockState must be placed to call this method when calling createBlockState()

Expected behavior

If I want to make an enumset of blocks that are instanceof Container, I'd do something like this on startup (using the paper createblockstate api)

 for (Material value : Material.values()) {
            if (value.isBlock() && value.createBlockData().createBlockState() instanceof Container) {
                CONTAINER_MATERIALS.add(value);
            }
        }

Observed/Actual behavior

using the paper createblockstate api lets me create a blockstate without placing it in the world, but it fails for one material: MOVING_PISTON:

Caused by: java.lang.IllegalStateException: The blockState must be placed to call this method
        at com.google.common.base.Preconditions.checkState(Preconditions.java:513) ~[guava-33.3.1-jre.jar:?]
        at org.bukkit.craftbukkit.block.CraftBlockState.requirePlaced(CraftBlockState.java:332) ~[paper-1.21.4.jar:1.21.4-DEV-88bbead]
        at org.bukkit.craftbukkit.block.CraftBlockState.getWorld(CraftBlockS...
#
metabrixkt
[PaperMC/Paper] Issue opened: #12038 First executions of delayed and repeating tasks on `FoliaAsyncScheduler` run in a single timer thread, making tasks block each other

Expected behavior

  1. Bukkit.getAsyncScheduler().runDelayed(...) and Bukkit.getAsyncScheduler().runAtFixedRate(...) schedule the task using FoliaAsyncScheduler#timerThread: it doesn't matter if it's a delayed or a repeating task
  2. FoliaAsyncScheduler.AsyncScheduledTask#run() gets run for the task
  3. All runs of the task are executed on FoliaAsyncScheduler#executors

Observed/Actual behavior

  1. Bukkit.getAsyncScheduler().runDelayed(...) and Bukkit.getAsyncScheduler().runAtFixedRate(...) schedule the task using FoliaAsyncScheduler#timerThread: it doesn't matter if it's a delayed or a repeating task
  2. FoliaAsyncScheduler.AsyncScheduledTask#run() gets run for the task
  3. Each run is meant to be dispatched into the pool (FoliaAsyncScheduler#executors) but the first one isn't. The work is dispatched into the pool only when AsyncScheduledTask#state is STATE_ON_TIMER, which is instead always STATE_SCHEDULED_EXECUTOR for first runs of all task:
    1. `As...
rustic folioBOT
rustic folioBOT
#
jpenilla
[PaperMC/Paper] New branch created: player-jspecify
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
luddebjorken
[PaperMC/Paper] Issue opened: #12043 Villagers attempting to pathfind through walls

Expected behavior

  • Villagers pathfind correctly to beds and workstations according to their schedule, going through doors to get into houses.

Observed/Actual behavior

  • Villagers attempt to access bed/workstation through the wall instead of walking inside the building, if the bed/workstation is close enough to a wall

Steps/models to reproduce

  • Construct a small box/house with an opening on one end
  • Place bed inside the box adjacent to one of it's walls opposite the opening
  • Spawn villager outside the wall on the side with the bed
  • Set time to night
  • Observe villager pathing

Plugin and Datapack List

None

Paper version

[23:50:10 INFO]: Checking version, please wait...
[23:50:10 INFO]: This server is running Paper version 1.21.4-131-main@1004374 (2025-01-28T21:51:56Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)

Other

Might not replicate 100%. Making walls >4 blocks thick sometimes fixes the issue.

Tested same world on vanilla server with no...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] Issue opened: #12044 WorldCreator - option to overriding setting initial spawn

Is your feature request related to a problem?

The main issue with creating a new world via WorldCreator, is it can take upwards of 10 seconds (or longer on some machines I'm assuming)
After throwing in like 20 debug timings, I came to the conclusion that it's setting up the initial spawn area that is causing the massive delay.

In my test:

[14:13:20 INFO]: 'set biome provider' took: 2 ms
[14:13:20 INFO]: 'create dimension key' took: 0 ms
[14:13:20 INFO]: 'create ServerLevel' took: 39 ms
[14:13:20 INFO]: 'setup world border' took: 2 ms
[14:13:20 INFO]: 'initialized serverLevelData' took: 0 ms
[14:13:27 INFO]: 'set initial spawn' took: 6855 ms       #  <-------- eeek, nearly 7 seconds
[14:13:27 INFO]: 'init world' took: 0 ms
[14:13:27 INFO]: Preparing start region for dimension minecraft:world_custom_biomes
[14:13:27 INFO]: Preparing spawn area: 100%

It took almost 7 seconds to setup the initial spawn. This is due to Minecraft having to search around to find a valid ...

rustic folioBOT
#
Machine-Maker
[PaperMC/Paper] New comment on pull request: #12017 Adding PredicateChoice to Paper API (updated version)

I just noticed there is already a PredicateRecipeChoice for potions, with the addition of the PredicateChoice this could be a bit confusing. Maybe it should be renamed to something more clear like a PotionRecipeChoice/PotionPredicateChoice?

Since PredicateRecipeChoice isn't public, delete it, and make PotionMix#createPredicateChoice return an instance of the new type. But also deprecate that method, and suggest to use the new method you create.

Here's a great example of what I talked about in my review, suggesting to keep implementation types out of the "api" because we can now just delete PredicateRecipeChoice entirely.

rustic folioBOT
#
Machine-Maker
[PaperMC/Paper] New comment on issue: #12036 Exact recipe for the stonecutter is incorrectly displayed

Yeah, this is not fixable. In fact we might want to not allow exact ingredients as inputs for stonecutter recipes. Stonecutter recipes are unique, in that the client and server have to agree on the number and output of each recipe for a given input because of the button arrangement. If there is any desync, the menu behaves strangely.

There are no exact ingredients as far as the client is concerned, so in this example the client thinks there are 9 valid recipes for the input stone. But the server thinks there are only 8 valid ones, because it checks the exact ingredient. This leads to this desync, and the behavior observed here.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
KioProject123
[PaperMC/Paper] New comment on issue: #12036 Exact recipe for the stonecutter is incorrectly displayed

Yeah, this is not fixable. In fact we might want to not allow exact ingredients as inputs for stonecutter recipes. Stonecutter recipes are unique, in that the client and server have to agree on the number and output of each recipe for a given input because of the button arrangement. If there is any desync, the menu behaves strangely.

There are no exact ingredients as far as the client is concerned, so in this example the client thinks there are 9 valid recipes for the input stone. But the server thinks there are only 8 valid ones, because it checks the exact ingredient. This leads to this desync, and the behavior observed here.

Please do not remove the exact recipes for the stonecutter. The plugin can send fake recipe update packets when players use the stonecutter, ensuring that the client displays the recipes synchronized with the server. I'm not sure if this fix can be ported to Paper, as the solution seems too hacky.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] Issue opened: #12048 ChunkGenerator - Implement a way to get noise terrain height in surface generation

Is your feature request related to a problem?

My problem is I can't get the vanilla terrain heigh in a generator.

Describe the solution you'd like.

I hope I can explain this well.
Let's say I have created my own ChunkGenerator

I have implemented the generateSurface method

I have also implemented shouldGenerateNoise ... telling Minecraft that it will generate the base noise terrain.

My plan is to just generate the surface.
Issue is there is (from what I can find) no way to get the vanilla noise terrain heigh (basically what vanilla has already generated in the chunk by the time it gets to my surface...

rustic folioBOT
#
Machine-Maker
[PaperMC/Paper] Pull request review comment: #12017 Adding PredicateChoice to Paper API (updated version) 
    public PredicateChoiceImpl {
        Preconditions.checkArgument(stackPredicate != null, "The item predicate cannot be null");
        Preconditions.checkArgument(exampleStack != null, "The example stack cannot be null");
        Preconditions.checkArgument(!exampleStack.isEmpty(), "Cannot have empty/air example stack");
        exampleStack = exampleStack.clone()
    }

can use a fancy record compact constructor here since its just preconditions

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
notPhyx
[PaperMC/Paper] Issue opened: #12050 TabCompleteEvent doesnt Work

Expected behavior

Triggers when sender uses Tab Completion

Observed/Actual behavior

So i tried to cancel out some Tab Completions, but the event doesnt even trigger, i added some debugging, it didnt trigger aswell, the Listener is registered.

Steps/models to reproduce

@EventHandler
public void onTabComplete(TabCompleteEvent event){
    Bukkit.getLogger().info("executed");
    if (event.getSender() instanceof Player) {
        if (event.getSender().hasPermission("transient.bypass.completions")) return;
        List<String> completions = event.getCompletions();

        completions.removeIf(completion ->
                hiddenCommands.stream().anyMatch(completion::startsWith));

        event.setCompletions(completions);
    }
}

Plugin and Datapack List

Image

Paper version

This server is running Paper version 1.21.4-110-main@c44f8...

rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #12050 TabCompleteEvent doesnt Work

The initial question would be 'what are you trying to tab complete', tab completion is mostly handled on the client these days, with the client only asking the server to complete stuff it's told to ask the server about (i.e. the legacy bukkit command system); stuff like the root command is sent to the client entirely (and, chat completions are 100% on the client)

rustic folioBOT
#
KoekjesBoyy
[PaperMC/Paper] Issue opened: #12051 Chunks loading slow on server

Expected behavior

Chunks are expected to load fast. i have switched from spigot to paper for the use of velocity and for some reason it loads extremely slow.

Observed/Actual behavior

I have a 32GB ram 3.30 ghz 4 cores processor. where i run the server on. The chunks load very slow you will see it if you fly too quickly and i get stuck in the not loaded chunks and wait for them to load.

Steps/models to reproduce

All i did was switch from spigot to paper and whenever i play on paper the chunk load slow whenever i play on spigot it is back to fast chunk loading
Before you all say stay on spigot i have to stay on paper for the plugins and velocity.

Plugin and Datapack List

Plugin list:
Axiompaper, Buildmode(my own plugin), Essentials, essentialschat, essentialsspawn, Gsit, Luckperms, MV-core, MV-inventories, Timber, Vault, Worldedit.

Paper version

This server is running Paper version 1.21.4-136-main@e0711af (2025-01-31T12:30:12Z) (Implementing API version 1....

#
electronicboy
[PaperMC/Paper] New comment on issue: #12051 Chunks loading slow on server

This is a support topic and requires too much potential back and forth to
be flooding people from the issue tracker. Paper artificially limits the
rate at which stuff is done to prevent DoSโ€™ing clients, and you've got the
concurrent sends and loads set to 1 chunk;

On Sat, 1 Feb 2025 at 13:16, Luca @.***> wrote:

Please use our Discord https://discord.com/invite/papermc for support
with performance issues. The issue tracker isn't really a good place for
diagnosing performance problems.

This isnt performance im pretty sure its a bug i tried resolving this
issue already there was no solution

โ€”
Reply to this email directly, view it on GitHub
https://github.com/PaperMC/Paper/issues/12051#issuecomment-2628949916,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AAJMAZBOLW7XJW2FT5AP5TD2NTCK7AVCNFSM6AAAAABWJI5PD6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMRYHE2DSOJRGY
.
You are receiving this because you are su...

rustic folioBOT
rustic folioBOT
#
kompot7
[PaperMC/Paper] Issue opened: #12052 Minecraft freezes with a crossbow

Expected behavior

Hello, I'm seeing this problem with 1.19.4 players When players on version 1.19.4 load a crossbow (only if it's enchanted with Impaler IV, I don't know about lvl below), then minecraft freezes. I'm not sure if it's because of you, but try to do something.

Observed/Actual behavior

Occurs when the player loads a crossbow with the Impaler IV enchantment.

Steps/models to reproduce

Occurs when the player loads a crossbow with the Impaler IV enchantment.

Plugin and Datapack List

no

Paper version

Paper 1.21.4-82-main@08fc33c

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
electronicboy
[PaperMC/Paper] New branch created: fix/add-lz4-region-comp-support
#
NicoNekoDev
[PaperMC/Paper] Issue opened: #12054 LZ4 compression not supported

Expected behavior

LZ4 compression to work (added in snapshot 24w04a)

Observed/Actual behavior

Error: [EnumValueSerializer] Invalid enum constant provided, expected one of [GZIP, ZLIB, NONE], but got LZ4

Steps/models to reproduce

Edit server.properties > region-file-compression=lz4
Edit config/paper-global.yml > unsupported-settings.compression-format: ZSTD

Plugin and Datapack List

No plugin, no datapack, fresh server.

Paper version

This server is running Paper version 1.21.4-136-main@e0711af (2025-01-31T12:30:12Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)

Other

The LZ4 does not exists in the unsupported settings in paper config, which is why it doesn't work.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Y2Kwastaken
[PaperMC/Paper] Pull request opened: #12055 Fix MenuType builder locations clashing with existing blocks.

Was notified on discord about a bug relating to "clashing" of sorts when on top of a chest block, and using the no location construction. This occurs because when standing on top of a non full block e.g. in this specific case a chest. The player's BlockPos is still in the same block of the chest thus cuasing the no location construction to open the chest the player is on top of.

I Addressed this by setting the BlockPos to BlockPos.ZERO instead of the player's location the goal is to avoid this clash. My thought here is that BlockPos.ZERO could possibly also cause issues if a specific chest block were to exist at (0, 0, 0) within a world, so I don't really think this is the "best" decision. I'll be revising this PR to attempt to take better shots at addressing this quirky behavior.

I was experimenting with the MenuType API and ran into an issue where when standing on a non-full block (such as a chest or brewing stand), the InventoryView will be that of the container. So any ...

rustic folioBOT
#
KioProject123
[PaperMC/Paper] New comment on pull request: #12055 Fix MenuType builder locations clashing with existing blocks.

Using the Zero coordinates may cause unnecessary chunk loading.

By binding a location in an unloaded chunk to this builder it is likely that the given chunk the location is will load. That means that when, building this view it may come with the costs associated with chunk loading.

Perhaps you can set the block position outside the world height boundaries, which ensures that a virtual inventory is always returned, such as player.blockPosition().atY(player.level().getMinY() - 1).

rustic folioBOT
rustic folioBOT
#
Y2Kwastaken
[PaperMC/Paper] New comment on pull request: #12055 Fix MenuType builder locations clashing with existing blocks.

I would refrain from using positions outside the world boundaries maybe Mojang decides in the future to disconnect the client for block changes outside the world Also, last time I did virtual inventories myself the block had to be within the reach of the player plus a few additional blocks to even open

I understand your concern with outside the world positions, however it will be tested thoroughly as an option. Also if you were unable to reach the inventory within a a couple blocks you incorrectly set up your menu api. Distance only really matters for physical menus that are stored within the world. Virtual menus are virtual and thus disconnected from the world, even if they create a fake tile the tile is not actually in the world. I have also other things I wish to attempt to mitigate these issues, such as using the players upper body position instead, which is slightly more favorable if I can avoid these glitches with that.

rustic folioBOT
rustic folioBOT
#
Gn132
[PaperMC/Paper] Issue opened: #12056 Fishing rod bobber despawn upon entering nether portal. AKA enderpearl teleport chamber papermc bug.

Expected behavior

Vanilla and also expected behaviour:
Fishing rod bobber is still loaded upon entering nether portal, unless player uses fishing rod again. When used again it despawns.

Observed/Actual behavior

Fishing rod bobber is left in loaded by enderpearl chunk despawns upon entering nether portal.
disable-unloaded-chunk-enderpearl-exploit is set to false.

Steps/models to reproduce

  1. Creating enderpearl chunk loader (water column with bubbles, 6 blocks deep into ground), throwing enderpearl inside.

  2. Create ender pearl teleport chamber OR command block inverted with redstone torch OR dispenser with single item inverted with redstone torch (Screenshot 1)
    Image

  3. Use fishing rod on pressure plate, leaving bobber on it.

  4. Enter portal with fishing rod in hand.

  5. Actual behavior: Bobber despawns, activating dropper/command block/teleport chamber upon entering portal.
    Expected...

rustic folioBOT
#
JoField08
[PaperMC/Paper] Issue opened: #12057 Zombified Piglins spawn in places they shouldn't - PortalbasedPiglin Farm

Expected behavior

Zombified Piglins should spawn in the nether portals

Observed/Actual behavior

Zombified Piglins spawn near to nether portals, outside the portals on random blocks.
Most of the time under the portals.

Steps/models to reproduce

  1. Build a regular Zombified Piglins Farm and start it
  2. Wait 5-10 minutes
    -> Piglins will spawn next to portals oder under them
    https://i.imgur.com/eCeAgGl.png

Plugin and Datapack List

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

Paper version

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

Other

Other servers are also affected by this issue.
There is also an issue on Folia https://github.com/PaperMC/Folia/issues/320

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Malfrador
[PaperMC/Paper] New comment on issue: #12050 TabCompleteEvent doesnt Work

That event is only called for bukkit commands. It is not called for commands using the new brigadier command API, as tab completion for those is generally handled on the client. You can specify permissions while registering the command there, and the client will respect those for tab completion.

I am unable to reproduce the event not working on the latest build. Its called just fine for all bukkit commands, so I am going to close this issue. For further support, I would recommend using the discord server.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Andre601
[PaperMC/Paper] New discussion: #12059 Change Version format for the API

Is your feature request related to a problem?

Only "problem" is readability of the version, as it has a lot of fluff that doesn't really seem to fit or make sense to have.

Describe the solution you'd like.

Given the hard-forking of Paper, it would make sense to consider changing the version structure used for the API.

As of right now is the version the same format as used by Spigot, which is <mc_version>-R0.1-SNAPSHOT, which is a lot to read for very little info or use.
Why is there R0.1? Does it have any use? Are releases really always snapshots?

I would propose to have version changes made by using the following, if technically achievable:

  • <mc_version> for proper releases matching the MC version.
  • <mc_version>-DEV for the latest available build of the API for the given MC version.
  • <mc_version>-<commit> (or <mc_version>-<build_number>?) for releases of a specific commit (or build?), if a dev wants to pin the version to a specific one (for wha...
Discussion Category: Ideas
#
electronicboy
[PaperMC/Paper] New comment on discussion: #12059 Change Version format for the API

The version string is what is deployed to the maven repo, and that's a 1:1 sorta deal, to have multiple things would require multiple deployments, and that's generally not how stuff like nexus, etc, expects versioning to be handled. Ideally, we'd have release builds, but, every mechanism I've seen of that has generally flopped.

SNAPSHOT needs to stay as that's how stuff remembers that it's a snapshot build and to deal with snapshot versioning caveats; R0.1 is an artifact of the days when bukkit used to do some form of releasing

Discussion Category: Ideas
rustic folioBOT
rustic folioBOT
#
CPieter
[PaperMC/Paper] New comment on pull request: #12017 Adding PredicateChoice to Paper API (updated version)

Since PredicateRecipeChoice isn't public, delete it, and make PotionMix#createPredicateChoice return an instance of the new type. But also deprecate that method, and suggest to use the new method you create.

Here's a great example of what I talked about in my review, suggesting to keep implementation types out of the "api" because we can now just delete PredicateRecipeChoice entirely.

Most issues with the PR should now be addressed, but I'm not quite sure how to properly replace the PotionMix predicate choice. The new PredicateChoice requires an example stack, but this might not really make sense for use in the PotionMix. I'm hesitant to make the example stack nullable since not having one can break recipes. Maybe the PotionMix update should be left to a seperate PR?

rustic folioBOT
rustic folioBOT
#
JoField08
[PaperMC/Paper] New comment on issue: #12057 Zombified Piglins spawn in places they shouldn't - PortalbasedPiglin Farm

Try this map

  • make sure the difficulty is set to hard
  • start the farm
  • you have to hit the piglins with a sword
    --> after a few minutes piglins will spawn on the floor

I tested this to make sure only paper is responsible for this issue. There were no other plugins on the server.

This server is running Paper version 1.21.4-138-main@5395ae3 (2025-02-02T21:42:11Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are running the latest version

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
electronicboy
[PaperMC/Paper] Pull request review comment: #12061 Blazingly Fast multi block changes

also, actually, I'm not sure we can support going this deep for block modifications, at best we could probably go towards LevelChunkSection#setBlockState to run the small amount of logic needed to keep some counters in check;

The big concern we had around this kinda thing in the past was how to expose it in a manner suitable for plugins wanting to use this stuff, as well as, how do you refine how much work should be done to allow plugins to minise the side-effects induced; I don't think that a plugin wants to use this API to paste a schematic for their game only for it to not function as intended due to stuff like not updating the tickables state, etc

I'm also not really sure where the logic is, but, the game already has the means to send updates to clients without having to manually send them, and that logic supports dealing with batching, instead of flooding clients with individual packets

#
erenkarakal
[PaperMC/Paper] Issue opened: #12062 Failed to decompress chunk data

Stack trace

[20:33:08] [Paper Common Worker #0/ERROR]: [ca.spottedleaf.moonrise.patches.chunk_system.io.MoonriseRegionFileIO] Failed to decompress chunk data for task: Task for world: 'world_the_end' at (-135,4) type: CHUNK_DATA, hash: 901536466
net.minecraft.nbt.ReportedNbtException: Loading NBT data
	at net.minecraft.nbt.CompoundTag.readNamedTagData(CompoundTag.java:538) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag$1.loadCompound(CompoundTag.java:57) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag$1.load(CompoundTag.java:42) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag$1.load(CompoundTag.java:35) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag.readNamedTagData(CompoundTag.java:532) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag$1.loadCompound(CompoundTag.java:57) ~[paper-1.21.4.jar:1.21.4-126-fb5b173]
	at net.minecraft.nbt.CompoundTag$1.load(...
#
Malfrador
[PaperMC/Paper] Pull request changes requested: #12061 Blazingly Fast multi block changes

I really like the idea of this API, IMO having something like this (maybe with some disclaimers regarding ticking entities) would be very nice to have. Settings blocks with the API is currently indeed really slow.

Few things that probably aren't ideal. Might be worth looking at WE/FAWE code here too to see how they deal with this, especially in regard to making sure things behave as expected after edits.

#
Malfrador
[PaperMC/Paper] New comment on issue: #12062 Failed to decompress chunk data

I ran out of disk space while the server was running.

The chunk seems to be entirely missing from the chunk data. -135/3 and -135/5 are there, -135/4 just isn't or is completely invalid NBT. So I guess when you ran out of disk space, it failed to save entirely. There is not really much that can be done here, the game in general doesn't handle running out of disk space very well.

Given that this apparently isn't corruption of some sort and just caused by external factors, I am going to close this issue, this is not something Paper can really fix. Blindly regenerating chunks like that might be unintended behaviour for other cases of actual corruption.

rustic folioBOT
#
Malfrador
[PaperMC/Paper] New comment on issue: #12057 Zombified Piglins spawn in places they shouldn't - PortalbasedPiglin Farm

This seems to be working as intended.
I tested it both on the latest Paper build and in vanilla, using the provided map with difficulty hard. Both in Vanilla and Paper, a zombiefied piglin is spawned as reinforcement after some time, as evident by them having a minecraft:reinforcement_callee_charge attribute modifier. Those can spawn in any free space around the leader and occasionally even spawn in the chamber below the farm.

An easy way to test this is by speeding up time with /tick rate (though Paper is able to sustain a much higher rate here than the vanilla server).

This is a Java Edition-only feature, so maybe you tested against Bedrock? Additionally, this was recently (1.21.2) changed to actually spawn zombified piglins and not normal zombies.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
SirYwell
[PaperMC/Paper] New comment on pull request: #12061 Blazingly Fast multi block changes

I'd generally love to have a nice batch-editing API in Paper, but I'm not so sure about this design. Maybe it's worth to go a step back and look at use-cases to determine what's needed.
With a method that just takes a list of locations + the block data to set, you might be able to place blocks fast, but how do you know where you want to place which block? As soon as you need to read existing blocks from the world, this suddenly isn't really useful anymore. So I'd recommend to figure out what you actually want to cover with this API, and whether that's really a common use-case.

Also thread-safety is a bit more tricky, pretty sure this could lead to errors if the underlying palette is resized while a chunk packet for the chunk to modify is built.

#
SirYwell
[PaperMC/Paper] New comment on pull request: #12061 Blazingly Fast multi block changes

If you already have a location -> block data mapping, yes. But from my experience, this is rarely what people want to do. As soon as you need to access world state to determine whether to place a block at a position, this doesn't work anymore. But that's a pretty common use-case (i.e. replacing specific blocks).
One common use-case where this solution is sufficient would be placing schematics. However, in such scenario, an explicit mapping of location -> block data in any way is pretty inefficient.

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
TT-Candy
[PaperMC/Paper] Issue opened: #12065 The minecart entity's movement behavior does not match the vanilla version.

Expected behavior

As show in the video๏ผŒthe minecart moves along with the machineโ€˜s operation and does not detach from its intended position.

https://github.com/user-attachments/assets/8b4dcdad-c237-46ad-98c8-48792eee006f

Observed/Actual behavior

As demonstrated in the video๏ผŒthe minecart falls out of the machine during operation.

https://github.com/user-attachments/assets/a1988697-d5c3-4632-84e4-2600c12184f0

Steps/models to reproduce

Due to technical limitations๏ผŒ Iโ€™m unable to create a simplified reproduction setup and have instead provided the complete machine within a world save file for testing purposes.

็Ÿฟ่ฝฆๆต‹่ฏ•.zip

Plugin and Datapack List

While the scenario is somewhat complex๏ผŒI can confirm this issue is unrelated to plugins or datapacks.Further details are provided below.

Paper version

This issue has been identified through additional investigation as being caused by Paper-based serve...

#
TT-Candy
[PaperMC/Paper] Issue opened: #12066 The minecart entity's movement behavior does not match the vanilla version.

Expected behavior

As show in the video๏ผŒthe minecart moves along with the machineโ€˜s operation and does not detach from its intended position.

https://github.com/user-attachments/assets/8b4dcdad-c237-46ad-98c8-48792eee006f

Observed/Actual behavior

As demonstrated in the video๏ผŒthe minecart falls out of the machine during operation.

https://github.com/user-attachments/assets/a1988697-d5c3-4632-84e4-2600c12184f0

Steps/models to reproduce

Due to technical limitations๏ผŒ Iโ€™m unable to create a simplified reproduction setup and have instead provided the complete machine within a world save file for testing purposes.

็Ÿฟ่ฝฆๆต‹่ฏ•.zip

Plugin and Datapack List

While the scenario is somewhat complex๏ผŒI can confirm this issue is unrelated to plugins or datapacks.Further details are provided below.

Paper version

This issue has been identified through additional investigation as being caused by Paper-based serve...

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on pull request: #12061 Blazingly Fast multi block changes

I don't really get what a whole bunch of this PR is trying to do, it looks like a bunch of logic all slapped in random locations trying to address issues which aren't really addressed with the proposed fixes; I also really do not like that the method proposal here looks like it's the 1st class mechanism of setting multiple blocks into the world.

Providing a level of low level access to the world would be nice and is something we have discussed with some parties before, but, I would much rather see this proposed as a discussion somewhere where we can focus on creating a concrete API plan before we start worrying about the implementation details within

#
electronicboy
[PaperMC/Paper] New comment on pull request: #12061 Blazingly Fast multi block changes

I mean code wise, I get the intent of the PR, but, the logic looks a bit misdirected in some areas or doesn't take advantage of the proper underling mechanisms in the server which already solve issues that this API faces

I think we should have a proper discussion over what such API should look like, as most of the gains in terms of setting blocks is going to be avoiding the extra overhead of the CB layer which isn't always needed, but also in terms of batching. Having a raw means to toss in a Map isn't all that I could see desired from such an API, and so, we should probably have a proper discussion over if such a method is 'enough' or if there are other concerns for consumers of this API

rustic folioBOT
rustic folioBOT
#
cyningstan
[PaperMC/Paper] Issue opened: #12067 Fisherman won't buy pufferfish

Expected behavior

A fisherman of sufficient rank (my test was a master) who has pufferfish in their list of trades, should accept pufferfish and exchange them for emeralds.

Observed/Actual behavior

The fisherman does not accept the pufferfish; clicking on the pufferfish in the trade list (while carrying >=4 pufferfish) has no effect; pufferfish can be dragged into the appropriate UI slot but no emeralds are offered. The trade is not frozen.

Steps/models to reproduce

The behaviour was observed in PaperMC 1.21.4-138. I also tried the same things in vanilla single-player 1.21.4 to see if this was a vanilla issue (apparently it isn't, see below).

  1. Click on the pufferfish in the list of trades. You should see your pufferfish appear in the lefthand UI slot, and the appropriate number of emeralds in the right; grabbing the emeralds makes the sale. This happens in vanilla single-player. In PaperMC, you hear the click sound, but the pufferfish do not appear in the lef...
#
Warriorrrr
[PaperMC/Paper] Pull request opened: #12068 Add Vault API

Supersedes #11159

Adds new API for vault blocks

Changes made compared to the previous PR:

  • Added some javadocs.
  • Renamed the display loot table method & changed the getRewardedPlayers return type to align with spigot.
  • Removed itemsToEject methods, these items are only populated when a player is opening a vault, but these can already be changed with the block dispense loot event, and also can't think of any use case for changing items in the middle of ejecting.
  • Removed rewarded players methods with player instances.
  • Added API for state update time.
  • Made the VaultBlockEntity#serverData field public to be able to directly update the server data, without having to access the actual tile entity.
rustic folioBOT
#
cyningstan
[PaperMC/Paper] New comment on issue: #12067 Fisherman won't buy pufferfish

The only possible relevant differences I can think of between my game and @Warriorrrr's are (a) I'm in survival where it looks like Warriorrrr is in a creative flatworld, and (b) my fisherman was trained up in a previous version (the world has existed since 1.19 and this Fisherman has been there since 1.20 at the latest). They did buy pufferfish in 1.21.3 though. If nobody can reproduce this then I'm happy to hoard pufferfish and see if the issue disappears when 1.21.5 drops :-)

rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] New comment on issue: #11287 Settings worlds to not auto-save still makes them save

I've had this issue before, and forgot to report it.
Looking thru the code, I think I see the issue,
In MinecraftServer#stopServer:

<img width="484" alt="Image" src="https://github.com/user-attachments/assets/6e731dc4-3f26-4e32-bc71-f93454cc9626" />

It appears CraftWorld uses the vanilla logic for no-auto-saving.
Minecraft appears to just overrides and forcefully decides to save against your wishes when the server stops.
I'm assuming Mojang did this on purpose, using the no-save option to just stop auto-saving every 5 minutes, but still continuing to save when the server stops.

I guess in a sense the World.setAutoSave is kinda of correct, as it's not AUTO saving (every 5 min), but maybe another method should be added to prevent saving altogether.

#
electronicboy
[PaperMC/Paper] New comment on issue: #11287 Settings worlds to not auto-save still makes them save

autosaving is not full saving, the entire point of that mechanism is to disable the periodic saving in order to allow an external program to backup the world, it prevents various things like chunks from being able to unload, and doesn't even prevent everything from saving.

The big issue with this sorta thing is that the entire data system is leaky as there was never any real considerations of mojang to having separate levels being as separate as CB and people like to think they are, but, also, there's the entire headaches over what disabling saving should and should not affect (and the fact that this is an entire set of discussions and work for such a niche usecase in which we don't have any form of mechanism for assuring that these kind of things would keep working to the degree that people would expect for read-only worlds, etc, etc)

rustic folioBOT
rustic folioBOT
#
Warriorrrr
[PaperMC/Paper] New comment on pull request: #12068 Add Vault API

May not be the scope of this PR, but perhaps these could also be addressed here?

* [VaultUnlockEvent #11687](https://github.com/PaperMC/Paper/discussions/11687)

* [VaultStateChangeEvent #11679](https://github.com/PaperMC/Paper/discussions/11679)

Or would this rather be a separate PR?

My current scope is just the vault block, I'll change the PR name

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Warriorrrr
[PaperMC/Paper] Pull request opened: #12069 Add vault change state event

Closes #11679

Adds an event for when a vault block changes its state, allowing plugins to get and/or modify the resulting vault state (though doing something like forcefully setting the vault state to ACTIVE will cause the event to be thrown every second if there's no player nearby).

I'm not completely sure about having a getPlayer, because it requires us to pass the player instance through the unlock method and may not be completely accurate for the case where the vault becomes active. Without it, the diff would be less complex but the event would also be less useful.

rustic folioBOT
#
Devlrxxh
[PaperMC/Paper] New discussion: #12070 Efficient way for large Block Placement

Is your feature request related to a problem?

As of now everyone knows bukkit doesn't directly offer any way to do large sized batch-editing. And I have issues trying to do massive block placements since there isn't any way to do it directly except using reflection and NMS.

Describe the solution you'd like.

A way to be able to do batch-editing while it not costing any performance issues, its fine if there are stuff not updating for example ticking tile entities. so a method something like setBlocks(Map<Location, BlockData> blocks); I have made a pr about my solution about this before but its need some discussion

Describe alternatives you've considered.

Using NMS and reflection https://github.com/Devlrxxh/BlockChanger

Other

No response

Discussion Category: Ideas
#
electronicboy
[PaperMC/Paper] New comment on discussion: #12070 Efficient way for large Block Placement

It's generally a discussion over how such a mechanism should look, not a "should we add this". We already know that there has been demand for such a thing over the years, it's the API design issue and considering how much/what control should be given over stuff, because "it's fast but it corrupts game state" is not something we can be overly excited for when it comes to API, especially if we're going to be considering it a first class citizen of the API

Discussion Category: Ideas
#
electronicboy
[PaperMC/Paper] New comment on discussion: #12070 Efficient way for large Block Placement

Majority of the slowdown is in all the extra API fluff that the CB layer adds; maybe we can add a means to skip certain operations and delay them until layer, i.e. letting the light system deal with relighting chunks in the future rather than "right now"; but, busting world state by default without any form of control is just not something that I can personally support

Discussion Category: Ideas
#
electronicboy
[PaperMC/Paper] New comment on discussion: #12070 Efficient way for large Block Placement

Part of the reason why I said that this should be a discussion is so that we can nail down the details of what this API should look like; corrupting world state is something I'm not fond of, especially not as the default option;

The idea would be to nail down what the overarching API should look like (It CANNOT be a singular method looking like a first class mechanism of modifying the world), and should likely offer some control over what operations are skipped, and, where viable, use already built in/optimised paths for handing stuff; i.e. delaying relighting of chunks until post work so we can just relight the entire chunk if over X blocks, using the servers already existing mechanism for sending out block updates rather than trying to handle that ourselves, etc

Discussion Category: Ideas
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
FireInstall
[PaperMC/Paper] Issue opened: #12071 Warning, you've not updated in a while!

Expected behavior

I just did run the runServer task added by the gradle plugin xyz.jpenilla.run-paper, version 2.3.1. And since I didn't started the testserver for a few days the plugin correctly fetched the latest paper version, at time of writing it's Build 138.
However, at startup checked itself and told me I would be out of date, witch ofcause isn't true as it was updated mere seconds ago!

Observed/Actual behavior

> Task :runServer
Fetching Paper builds for version 1.21.4...
Latest build for 1.21.4 is 138.
Downloading Paper 1.21.4 build 138...
Done downloading Paper, took 19.812s.
Verified SHA256 hash of downloaded jar.
Starting Paper...

Applying patches
Starting org.bukkit.craftbukkit.Main
*** Warning, you've not updated in a while! ***
*** Please download a new build as per instructions from https://papermc.io/downloads/paper ***
2025-02-06T00:48:16.572772100Z ServerMain WARN Advanced terminal features are not available in this environment
[01:48:16 INFO...
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
Pman3
[PaperMC/Paper] Issue opened: #12072 NewChunks patch

Is your feature request related to a problem?

There are certain exploits like Trouser-Streak's NewerNewChunks module which lets players find old chunks and allows for chunk trailing bases.

Describe the solution you'd like.

Paper's anti-xray seems to prevent the Pallete exploit in the Overworld, and makes it much more difficult for chunk trailing to be utilized. In the nether, however, the liquid flow allows for Liquid exploit module under NewerNewChunks to be used to follow chunk trails. Would there be any chance to implement such a patch, or is there already one which exists in paper or in a 3rd party plugin? If liquid and the blocks could somehow be obfuscated, then I believe this could be patched.

Describe alternatives you've considered.

Paper's xray seems to somewhat fix the pallete exploit in the Overworld, but it doesn't work when players use chunk trailing exploits in the nether.

Other

Link to Trouser Streak (required meteor to run) - https://github.co...

rustic folioBOT
#
Warriorrrr
[PaperMC/Paper] New comment on issue: #12072 NewChunks patch

If your world is entirely pre-generated, you can use chunky to tick liquids and other blocks for some time to prevent that from being used, but allowing adding non-still liquids to anti xray doesn't seem very useful.

I don't think something like this is really something that can be prevented in a clean way, and also not enough of a priority to warrant an open issue.

rustic folioBOT
rustic folioBOT
#
Warriorrrr
[PaperMC/Paper] New comment on issue: #11677 Map data is continually sent once a second even with no changes

Replacing the true with !vanillaMaps could work to reduce the amount of packets being sent while idle, since if there's no plugin renderers we can assume that dirtyDecorations will be accurate.

Moving with a map will still cause packets to be sent because your player arrow has to be updated, but it's still beneficial.

diff --git a/net/minecraft/world/level/saveddata/maps/MapItemSavedData.java b/net/minecraft/world/level/saveddata/maps/MapItemSavedData.java
--- a/net/minecraft/world/level/saveddata/maps/MapItemSavedData.java	(revision e2399547e7f434d9bd7542a70f01d3be8d14b23c)
+++ b/net/minecraft/world/level/saveddata/maps/MapItemSavedData.java	(date 1738863833572)
@@ -632,7 +632,7 @@
             }
 
             Collection<MapDecoration> collection;
-            if ((true || this.dirtyDecorations) && this.tick++ % 5 == 0) { // CraftBukkit - custom maps don't update this yet // TODO fix this
+            if ((!vanillaMaps || this.dirtyDecorations) && this.tick++ % 5 == 0)...
rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #12072 NewChunks patch

The only real fix for this that wouldn't be a hack would be to have a fully stable world generation output, i.e. not relying on things like the world ticking in order to properly handle putting the game into its 'final form', i.e. such as the ticking of fluids.

The only other option would be to ensure that a generated chunk has been properly randomly ticked to completion, noting that you also need chunk neighbours to be there to end up in the correct state properly. All options around this are stupidly invasive, especially if you can't pre-gen. Maybe there is some cat and mouse you can pull off with specific implementations of these tools, I recall scepticising some things a plugin could do a good while ago, but, that's generally 100% out of scope for us.

From a 'real world' standpoint, I would generally just focus on pregenning a fair area around spawn and hoping that puts off enough folk, but I can't see there being much of a tenable solution here.

rustic folioBOT
#
NoyException
[PaperMC/Paper] Issue opened: #12073 Add EntitySpawnByEntityEvent

Is your feature request related to a problem?

I'm currently making a roguelike plugin. I need to find out whether all the enemies are killed to accomplish a combat, so that's why I need to take those entities spawned by enemies into consideration (They should also be considered as enemies in this combat). So I wonder how can I find the entity who causes a spawn event.

Describe the solution you'd like.

Add EntitySpawnByEntityEvent which inherits EntitySpawnEvent

Describe alternatives you've considered.

  • Add a method getWhoSpawn() in EntitySpawnEvent
  • Add a method getWhoSpawn() in Entity

Other

No response

#
NoyException
[PaperMC/Paper] New comment on issue: #12073 Add EntitySpawnByEntityEvent

entities spawned by enemies, such as? Vanilla doesn't really have much of a "spawned by" concept, and I ponder the capability of exposing such a concept without having any form of proper cause API would just make such a concept be relatively broken in general, especially when it comes to determining what means it was "spawned by X"

For example, a Jokey can spawn a vehicle, an egg can spawn a chicken, a small slime can be spawned by a dying big slime, an endermite can be spawned by a ender pearl, etc

#
NoyException
[PaperMC/Paper] New comment on issue: #12073 Add EntitySpawnByEntityEvent

All of those generally sound okay, bar the slime one, as that falls into a really weird area as that's not considered a 'spawn' as such, that's effectively a transformation (and is already exposed in a specific event)

You're right, there's already a SlimeSplitEvent. But what about the others? I thought, maybe we can refactor the SpawnReason, make it an abstract class instead of a enum to carry more information?

rustic folioBOT
#
Ojiie55
[PaperMC/Paper] Issue opened: #12074 Unsupported settings tripwire hook dupe

Is your feature request related to a problem?

In 1.21.4 the tripwire hook dupe was working in vanilla but now in papermc its not working and i need a option in the config to maybe enable it so i can get emeralds with villager when trading

Describe the solution you'd like.

Add a unsupported setting for it

Describe alternatives you've considered.

String dupe but got patched in 1.21.4 vanilla anyways

Other

No response

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
TT-Candy
[PaperMC/Paper] Issue opened: #12075 The Wither's mob AI behavior is inconsistent with the vanilla Minecraft.

Expected behavior

The Wither does not lose its hostility towards Iron Golems when the chunk unloads.

Observed/Actual behavior

After unloading and reloading a chunk,the Wither momentarily loses its hostility,causing it to escape the Wither cage.

Steps/models to reproduce

Construct the relevant mechanism within a chunk.
Move away from the chunk to ensure it unloads
Return to reload the chunk and observe the Wither's behavior.

Image

Plugin and Datapack List

Issue persists in a clean Paper instance without plugins or datapacks.
This issue is unrelated to plugins or datapacks.

Paper version

All Paper version.

Other

No response

#
kyoshuske
[PaperMC/Paper] Issue opened: #12076 PaperPlugin does not support SpigotLibraryLoader

Expected behavior

SpigotLibraryLoader should download the required maven repo provided in paper-plugin.yml.

Observed/Actual behavior

SpigotLibraryLoader does not download anything.... but when i use plugin.yml instead of paper-plugin.yml it works as intended

Steps/models to reproduce

  1. add paper plugin to your server
  2. add required libraries to paper-plugin.yml
  3. start server

Plugin and Datapack List

only this plugin and nothing else

Paper version

Server: Paper 1.21.1-122-master@4430e96 (and yes i tried using the latest paper build)
Java: Eclipse Adoptium Temurin-21.0.3+9

Other

No response

rustic folioBOT
#
HaHaWTH
[PaperMC/Paper] Pull request opened: #12077 Separate tick count to ensure vanilla parity

Closes #12075.

This pull request introduces another entity tick counter totalTickCount for api usage, reset vanilla tickCount instead of restoring value from Spigot's Spigot.ticksLived tag to keep vanilla parity.

Analysis

Spigot's restore tickCount behavior(See original patch) causes potential issues with Wither's ai, see code below, and lead to wither loses its target when entities are reloaded:
image

Api Changes

Now Entity#getTicksLived gets value from totalTickCount to avoid api behavior breakage, Entity#setTicksLived sets both tickCount and totalTickCount.

Currently the totalTickCount is only used for apis, if the breakage is fine, the extra counter can be removed.

rustic folioBOT
#
ShaneBeee
[PaperMC/Paper] Draft pull request opened: #12078 Entity#teleport changes to use vanilla logic

This PR aims to create a new method for teleporting using vanilla logic.
Minecraft handles a lot of the teleport logic that Bukkit/CB has used over the years.
A lot of the CraftBukkit teleport code is so old, and limits what users can do with teleporting.
This chance allows the user to rely on Minecraft to handle the teleport logic rather than CraftBukkit limitations.

#A Few Notes:

  • I chose the method name teleportVanilla as a temporary placeholder. Would definitely like a better name.
  • I haven't added all the possible methods yet for Bukkit Entity, I wanted to get feedback first before writing all those out
  • I slightly changed ServerPlayer teleport logic to handle the player's passengers (Minecraft by default doesn't allow players having passengers, this is seen when attempting to use the /ride command... the API however let's us do this, so I wanted to make sure that worked.).

I have done a fair bit of testing, including:

  • Player with a stack of passengers (I ...
rustic folioBOT
#
Malfrador
[PaperMC/Paper] New comment on pull request: #12078 Entity#teleport changes to use vanilla logic

I personally would prefer if the Craftbukkit teleport logic was simply aligned with vanilla, instead of introducing an entirely new method. The goal should be to align API behavior with vanilla where it makes sense, and here there really is little reason to keep the limitations of the Craftbukkit method.

Of course this then should be done in a major update so people are aware of the change.

#
ShaneBeee
[PaperMC/Paper] New comment on pull request: #12078 Entity#teleport changes to use vanilla logic

I do agree with having this take over Bukkit logic over a new method.

The only reason I wrote a new method was because I attempted something along these lines a year or so ago, back on CraftBukkit.
It was noted by someone not to change Bukkit logic.

I personally agree the Bukkit logic is old and silly (why are we explicitly blocking Passengers from TPing when Minecraft allows it?)

I am happy to remove the new method and just plop this into the existing methods if that is what you guys would like me to do.

rustic folioBOT
#
ayu3422543
[PaperMC/Paper] New comment on issue: #3316 Is this really "NOT A BUG OR A CRASH"

plz help me Iam getting this problem

[09:03:34 ERROR]: --- DO NOT REPORT THIS TO PAPER - THIS IS NOT A BUG OR A CRASH - git-Paper-318 (MC: 1.20.2) ---
[09:03:34 ERROR]: The server has not responded for 10 seconds! Creating thread dump
[09:03:34 ERROR]: ------------------------------
[09:03:34 ERROR]: Server thread dump (Look for plugins here before reporting to Paper!):
[09:03:34 ERROR]: [ChunkTaskScheduler] Chunk wait task info below:
[09:03:34 ERROR]: [ChunkTaskScheduler] Chunk wait: [( -11,-18) in 'world']
[09:03:34 ERROR]: [ChunkTaskScheduler] Chunk holder: NewChunkHolder{world=world, chunkX=-11, chunkZ=-18, entityChunkFromDisk=false, lastChunkCompletion={chunk_class=net.minecraft.world.level.chunk.ProtoChunk,status=minecraft:initialize_light}, currentGenStatus=minecraft:initialize_light, requestedGenStatus=minecraft:full, generationTask=null, generationTaskStatus=null, priority=BLOCKING, priorityLocked=false, neighbourRequestedPriority=IDLE, effective_priority=BLOCKING, oldTi...

rustic folioBOT
rustic folioBOT
#
badvision
[PaperMC/Paper] Issue opened: #12079 Unable to open autocrafters

Expected behavior

Right-clicking an autocrafter should open an inventory screen letting you configure the crafter

Observed/Actual behavior

Nothing happens in the UI; on the server side there is an error (same as reported in SPIGOT-8005)

`[21:09:59] [Server thread/ERROR]: Failed to handle packet net.minecraft.network.protocol.game.PacketPlayInUseItem@524c55e7, suppressing error
java.lang.IllegalStateException: Title not set
at com.google.common.base.Preconditions.checkState(Preconditions.java:513) ~[guava-33.3.1-jre.jar:?]
at net.minecraft.world.inventory.Container.getTitle(Container.java:96) ~[spigot-1.21.4-R0.1-SNAPSHOT.jar:4432-Spigot-d421948-f152cad]
at org.bukkit.craftbukkit.v1_21_R3.inventory.CraftInventoryView.<init>(CraftInventoryView.java:30) ~[spigot-1.21.4-R0.1-SNAPSHOT.jar:4432-Spigot-d421948-f152cad]
at org.bukkit.craftbukkit.v1_21_R3.inventory.view.CraftCrafterView.<init>(CraftCrafterView.java:13) ~[spigot-1.21.4-R0.1-SNAPSHOT.jar:4432-Spigot-d421948-f15...

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
badvision
[PaperMC/Paper] New comment on issue: #12079 Unable to open autocrafters

Yeah, I only wish I'd asked for something that merited a good use of your time -- except of course maybe giving you a sense of personal satisfaction that your hard fork was successful. I just built a bamboo plank farm that uses two chained auto crafters and it does seem to be working fine. I ran into some weird issues with rails and will open a ticket for that if I can find a good way to explain the issue (and confirm it's not a bug already reported.)

Thanks again!

#
badvision
[PaperMC/Paper] Issue opened: #12081 Breaking redstone near rails changes rail layout

Expected behavior

After placing rails the way I want them, I expect that their layout/orientation should not be changed when breaking unpowered redstone next to them.

Observed/Actual behavior

What happens in this case is the rails get changed to an S-shape. I understand that switches can be used to change rails, but I don't expect that breaking unpowered redstone should trigger the logic for this. Also, I don't think that redstone state changes should affect rails that do not have any bends in them, but I am not 100% clear on the vanilla behavior either, so some of that could be a "me" problem.

Steps/models to reproduce

Create something like the following, then break the redstone.

Image

Plugin and Datapack List

No plugins needed to reproduce this issue.

Paper version

Paper version 1.21.4-138-main@5395ae3 (2025-02-02T21:42:11Z)

Other

No response

rustic folioBOT
#
lynxplay
[PaperMC/Paper] branch deleted: fix/add-lz4-region-comp-support
#
lynxplay
[PaperMC/Paper] Pull request review comment: #12068 Add Vault block API

Actually pretty unsure on this statement.
A set would be useful for insertion schematics, but the returned thing is unmodifiable/immutable anyway.
Consumers cannot do anything about the contents in the first place but calling contains which, that works on a collection just as well as on a set.
I don't see much benefit from this being a set, tho I'll chat it over with others.

rustic folioBOT
#
spring-dependency-management
[PaperMC/Paper] Issue opened: #12082 Effect.STEP_SOUND is actually the block break effect

Expected behavior

levelevent id 2001 should be PARTICLES_DESTROY_BLOCK, which spawns block break particles and makes the noise, but

Observed/Actual behavior

in the effect api its

/**
* Sound of a block breaking. Needs {@link org.bukkit.block.data.BlockData} as additional info.
*/
STEP_SOUND(2001, Type.SOUND, org.bukkit.block.data.BlockData.class, Material.class), // Paper - block data is more correct, but the impl of the mtehods will still work with Material

Steps/models to reproduce

N/A

Plugin and Datapack List

N/A

Paper version

This server is running Paper version 1.21.4-DEV-main@742968e (1970-01-01T00:00:00Z) (Implementing API version 1.21.4-R0.1-SNAPSHOT)
You are 1 version(s) behind
Download the new version at: https://papermc.io/downloads/paper

Other

No response

rustic folioBOT
#
lynxplay
[PaperMC/Paper] branch deleted: player-jspecify
rustic folioBOT
#
ElFrod0
[PaperMC/Paper] Issue opened: #12083 Server GUI partially freezing when using 'help' command

Expected behavior

Full output of help command whilst allowing us to use even more commands later on.

Observed/Actual behavior

After running help command, the output is not complete and we can't send any more commands via the Server GUI anymore. We can't even stop the server by closing the GUI.
In that case we have to stop the server via standard-ish console or just straight up kill the process.

Steps/models to reproduce

  1. Start server without --nogui flag.
  2. Wait for the server to load fully.
  3. Write help in the server's GUI and send.
  4. Observe the command's output is not complete. You cannot send any more commands after that as well.

Plugin and Datapack List

No plugins nor datapacks

Paper version

1.21.4-143-1a04e96 (MC: 1.21.4)

Other

No response

rustic folioBOT
#
NonSwag
[PaperMC/Paper] Issue opened: #12084 Inversed visual fire & lock fire ticks

Is your feature request related to a problem?

Currently, we have the ability to set an entity visually on fire without it being actually burning
but the way visual fire works, it is not possible to disable visual fire if the entity is really burning

Describe the solution you'd like.

Internally, the visualFire field could use a nullable Boolean object and default to null and make the entity burn if true
not burn if false
and send the actual fire state when null
I understand this will cause problems with existing plugins that utilize this method, other suggestions are more than welcome

Describe alternatives you've considered.

Other

Also, I would like to suggest adding a lockFireTicks method to be able to make entities burn or not start burning as long as the ticks are locked

rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
rustic folioBOT
#
KiKi-XC
[PaperMC/Paper] Issue opened: #12085 Request to add RCON event

Is your feature request related to a problem?

I would like to request the addition of an RCON login event to the Minecraft API. Currently, there is no event to listen for RCON login attempts, which makes it difficult to implement security measures such as detecting brute force attacks on the RCON interface.

Why is this needed?

Brute-force protection: This would allow server admins to track failed login attempts, lock out IPs after a number of failed attempts, and integrate with firewalls.
Logging and monitoring: Server admins could log RCON activity and detect unauthorized access attempts in real time.
Improved security: It would allow for more secure server management, without relying on external tools or scripts.
Adding this feature would help improve RCON security and better protect Minecraft servers.

Describe the solution you'd like.

Currently, there is no event to listen for RCON login attempts in Minecraft. Adding an RCON login event would provide plugin developer...

rustic folioBOT
#
electronicboy
[PaperMC/Paper] New comment on issue: #12085 Request to add RCON event

While I can support this as a low priority, I'm generally not sure why you'd be worrying about the security of something which is completely insecure in general when it would make more sense to use existing secure alternatives, i.e. one of the few ssh server plugins, or, create something designed to be secure which would serve as a direct alternative

rustic folioBOT
#
KiKi-XC
[PaperMC/Paper] New comment on issue: #12085 Request to add RCON event

While I can support this as a low priority, I'm generally not sure why you'd be worrying about the security of something which is completely insecure in general when it would make more sense to use existing secure alternatives, i.e. one of the few ssh server plugins, or, create something designed to be secure which would serve as a direct alternative

"Thank you for considering my idea and offering to prioritize it at a lower level. I understand your point, but my goal is to develop a more comprehensive rcon brute-force protection plugin. During today's practical implementation, I found that this event is missing, which is crucial for supporting my plugin's development.

In such a large community like the Bukkit plugin ecosystem, I couldn't find any plugins specifically addressing rcon security. This is why I want to create a more comprehensive and user-friendly plugin and contribute it to the community."

rustic folioBOT
#
masmc05
[PaperMC/Paper] New comment on issue: #12085 Request to add RCON event

The rcon system is really insecure and flawed, if you want security, as mentioned earlier, use something properly designed to be secure, like ssh protocol or a custom protocol as an alternative to rcon. Rcon and security are a bit of incompatibile concepts, and just an event would not be enough to protect yourself against attacks

rustic folioBOT
#
SkytAsul
[PaperMC/Paper] Issue opened: #12086 ItemMeta content incorrectly deserialized if inside shulker box

Expected behavior

If I serialize a shulker box containing an item with complex ItemMeta (such as CustomModelData) in 1.21.1 and deserialize it in 1.21.4, I expect the ItemMeta to be unchanged (or rather, correctly upgraded if the data format has changed).

For instance, if this item is present in a shulker box in 1.21.1:

  ==: org.bukkit.inventory.ItemStack
  v: 3955
  type: PAPER
  meta: 
    ==: ItemMeta
    meta-type: UNSPECIFIC
    display-name: '{"text":"","extra":[{"text":"Custom Paper","italic":false}]}'
    custom-model-data: 42

I expect the shulker box in 1.21.4 to contain:

item: 
  ==: org.bukkit.inventory.ItemStack
  v: 4189
  type: PAPER
  meta: 
    ==: ItemMeta
    meta-type: UNSPECIFIC
    display-name: '{"text":"","extra":[{"text":"Custom Paper","italic":false}]}'
    custom-model-data: 
      ==: CustomModelData
      floats: 
      - 42.0
      flags: []
      strings: []
      colors: []

Observed/Actual behavior

CustomModelData i...

#
electronicboy
[PaperMC/Paper] New comment on issue: #12086 ItemMeta content incorrectly deserialized if inside shulker box

ItemMeta is pretty much incapable of being run through DFU. This has been a long-term issue, and while random hack fixes have been employed over the years, this is generally just not something we feel is worth trying to chase. We will, at some point, serialise ItemStacks directly using SNBT so that we can run them through DFU/DataConverter. In the meantime, I'd suggest storing item stacks using the API to serialise them as bytes if you care about longevity, you can base64 them to store them as strings if needed

rustic folioBOT
rustic folioBOT
rustic folioBOT
#
kompot7
[PaperMC/Paper] Issue opened: #12089 creative mode

Expected behavior

so that the items from the creative are visually removed, and not so that you have to re-enter every time after cleaning.

Observed/Actual behavior

Hello, I found such a bug that when I go into the creative and clear the inventory through the cross, it is not cleared visually, all things remain there, only after logging into the server the inventory becomes empty. There is no such thing in survival mode if you write through /clear, and then in the creative through / clear, the same thing does not visually disappear.

Steps/models to reproduce

items in creative mode are not visually cleaned even when using a cross or /clear, only after deleting and logging back to the server the inventory is cleared.
if you write a lot of times in the creative / clear, then it says that there are no items in the inventory.

Plugin and Datapack List

missing

Paper version

Paper version 1.21.4-138-main@5395ae3

Other

No response