Minecraft Analysis for 1.18 Experimental Snapshots

Project Explanation:

TODO

Project Layout Description:

• chunk_data/: Contains simulated data from the current region files.
• mining_data/: Contains simulated data for each of the [current techniques] for each of the region files.
• regions/: Contains fully generated region files from Minecraft 1.18 Experimental Snapshot 4. These files were generated using the world-pregen datapack. Files are from seed: 3686983410240111049.
• results/: Contains csv files that have been processed and are ready for display.
• graphical_results/: Contains graphs from different simulations.
• src/: Contains the rust source code that runs and manages the simulations.
  • bin/: Contains files related to the executable of the project.
    • mc_analysis.rs: Contains the code to launch the simulations.
    • ui.rs: Contains the code to draw a ui for selecting the simulation type and parameters along with a ui for monitoring simulations.
  • lib.rs: Contains a small amount of code relating to updating simulation status.
  • mining.rs: Contains functions that make up the steps to simulate mining such as mining a 2x1 or 1x1 hole in a horizontal direction.
  • simulations.rs: Contains functions that handle the larger task of simulating a mining technique. These functions go through the process of using techniques to get blocks, trimming possible duplicate blocks (this shouldn't happen?), categorizing blocks, and recording data.
  • techniques.rs: Contains the various techniques that are simulated.
• static/: Contains the files for the website that hosts the analysis.
• index.html:
• results.py:
• ValidBlocks.txt:

Libraries used in the creation of this project:

• anvil-parser: This is the original Python library that was used to get block data from Minecraft worlds.
• anvil-parser (fork): This is my fork of anvil-parser that has changes to allow access to the new block height ranges.
• depreciated_minecraft_analysis: This was the original code for this project before performance concerns and scope creep became relevant.
• hematite-nbt: The Rust library to read the nbt file structure that Minecraft region files use.
• mvp_anvil: A custom Rust library that provides support for getting regions, chunks, and blocks out of the Blobs that hematite-nbt returns for region files. The code is a port of relevant sections of anvil-parser.
• csv: A Rust library for interacting with .csv files.
• threadpool: A library that provides a Struct to manage and launch multithreaded workloads on a fixed number of worker threads.
• tui: A terminal ui library, the crossterm library is being used as a backend for better support on more OS's.

How to install and run:

Using a prebuilt executable:

1. Navigate to the actions tab of this repo.
2. Click the latest workflow run that has a passing build (green checkmark).
3. Under 'Artifacts' download mcAnalysis for Linux or mcAnalysis.exe for Windows. I do not currently have access to a Mac computer to test the build process with.
4. Place the file in a directory and launch it. On Linux you may need to give the file execution permissions with 'chmod +x mcAnalysis'.
5. The ui should launch in a terminal and some directories should be created.
6. Follow the instructions in the ui and place region files in the regions/ directory.
7. Press enter to continue and select whatever simulation you would like to run.

Building yourself:

git clone https://github.com/nuhtan/minecraft_analysis.git

cd minecraft_analysis

cargo build --release

• The executable will be located in 'target/release/'.
• You can now run './target/release/mc_analysis' or /target/release/mc_analysis.exe'.
• Continue from step 5 of the prebuilt executable instructions.

Techniques that are currently included:

• Chunk: Rather than a technique this is a full gathering of target blocks from each y layer of each chunk of each region file.
• Branch:
• Branch w/ Poke Holes:

What still needs to be done:

• Documentation for rust files and results.py.
• Get Nick's R files and include them.
• results.py should be broken into functions for easier reading and reusability.
• Results should also generate stacked bar graphs to compare the different mining strategies directly.
• Create an explanation and comparison for results.
• There should be comparisons within the mining techniques to see what parameters lead to the most efficient mining strategy.
• Test cases and benchmarks.
• Internal documentation for more complex sections of the code.
• UI needs to handle the alternative simulations, chunk and parameter testing.
• Github pages to host conclusions.
• Upload this and mvp_anvil to crates.io.
• Exit during simulation.
• Less destructive exits, files should be saved after each y level not only once the entire range has been simulated.
• Predict finish time for simulations?
• Change UI color choices.
• Adding more monitoring data to running simulations.