PandaSpigot

Fork of Paper for 1.8.8 focused on improved performance and stability.

Highlights

• Backported API enhancements from newer versions

  • ServerTickStartEvent & ServerTickEndEvent
  • PlayerChunkLoadEvent & PlayerChunkUnloadEvent
  • PlayerHandshakeEvent
  • EntityMoveEvent

• Greatly improved network performance

  • Updating to Netty 4.1 offers the ability to use newer Java versions with epoll on Linux.
  • Improved flush handling to massively improve entity tracker performance.
  • Support for Unix domain sockets to avoid the overhead of TCP when using a proxy on the same machine.

• More configuration options, such as:

  • Customizable knockback
  • World and player data saving

See a full list of patches here.

Using

You can download the latest pre-built server JAR by clicking the download button below.

For support, please join our Discord.

API

See our API patches here.

Maven

<repositories>
    <repository>
        <id>hpfxd-repo</id>
        <url>https://repo.hpfxd.com/releases/</url>
    </repository>
</repositories>

<dependencies>
    <dependency>
        <groupId>com.hpfxd.pandaspigot</groupId>
        <artifactId>pandaspigot-api</artifactId>
        <version>1.8.8-R0.1-SNAPSHOT</version>
        <scope>provided</scope>
    </dependency>
</dependencies>

Gradle (kts)

repositories {
    mavenCentral()
    maven(url = "https://repo.hpfxd.com/releases/")
}

dependencies {
    compileOnly("com.hpfxd.pandaspigot:pandaspigot-api:1.8.8-R0.1-SNAPSHOT")
}

Building

To compile PandaSpigot, you'll need:

• JDK 17 (required to run the decompiler)
• Git
• Bash

🧩 Although JDK 17 is required for building, the compiled JAR remains fully compatible with Java 8.

Building, patching, and compiling are all done through the main panda script.

PandaSpigot can be built by running ./panda jar, and you will find the final Paperclip jar in paperclip.jar

Panda Bash build script

The panda Bash script is the entry point for setup, patching, and builds. Run it from a Bash-compatible shell on your platform. Run ./panda without arguments to print its command list.

Normal commands:

Command	Description
rb, rebuild	Rebuild patches, can be called from anywhere.
setup	Remap, decompile, and patch Minecraft. Can be run from anywhere.
p, patch	Apply all patches to the project without building it. Can be run from anywhere.
j, jar	Apply all patches and build the project, paperclip.jar will be output. Can be run from anywhere.
c, clean	Removes all generated files, PandaSpigot-API, PandaSpigot-Server, and work.
con, continue	Shortcut command for running git am --continue or git rebase --continue.

Commands that require . ./panda install first:

Command	Description
r, root	Change directory to the root of the project.
a, api	Move to the PandaSpigot-API directory.
s, server	Move to the PandaSpigot-Server directory.
e, edit	Use to edit a specific patch, give it the argument server or api respectively to edit the correct project. Use the argument continue after the changes have been made to finish and rebuild patches. Can be called from anywhere.
install	Add an alias to $RCPATH to allow full functionality of this script.

Install the default panda alias with:

. ./panda install

You can also provide a custom alias name:

. ./panda install example

PandaSpigot-API and PandaSpigot-Server are generated work trees created by the build scripts. The patch creation process follows Paper's contributing guide, which is linked below; ./panda rb is the PandaSpigot command that saves those work tree commits back into patches/api and patches/server.

Patch workflow

PandaSpigot uses a patch-based workflow. Changes are committed once inside the generated work tree, then committed again from the repository root after the patch files are rebuilt.

Use this order:

1. Make your changes in PandaSpigot-API or PandaSpigot-Server.
2. Commit those changes inside that generated work tree.
3. Return to the repository root and run ./panda rb.
4. Review the regenerated files in patches/api or patches/server.
5. Commit the regenerated patch files from the repository root.

Only push commits from the repository root. Do not push or pull from PandaSpigot-API or PandaSpigot-Server; their commits are only used to create patch files. Commit work tree changes before rebuilding or reapplying patches, because those commands can reset the generated work trees.

Contributing

You can mostly follow Paper's contributing guide, just remember:

• Multi-line changes start with // PandaSpigot start and end with // PandaSpigot end
• If the change isn't obvious, add a small explanation like this: // PandaSpigot start - reason
• One-line changes should have // PandaSpigot at the end of the line.
• Follow Java code style (aka. Oracle style), with some exceptions:
  • If you are modifying upstream files, keep your diff size minimal. Going over 80 characters per line is fine to make this happen.
  • When in doubt or the code around your change is in a clearly different style, use the same style as the surrounding code.

When contributing, please think about the side effects of any changes you write. Plugin compatibility is important, and we wish to minimize any breakage.

Please do not open pull requests for features that you cannot justify the existence of, and the added maintenance costs of that come along with them. If you are thinking of adding a feature that may be controversial, please open an issue first!