vberlier / nbtlib

A python library to read and edit nbt data.
MIT License
121 stars 16 forks source link

Add nbtlib.contrib package? #60

Open vberlier opened 4 years ago

vberlier commented 4 years ago

From https://github.com/vberlier/nbtlib/issues/1#issuecomment-568782953.

It could be interesting to provide the necessary abstractions to deal with common nbt use-cases in a contrib package.

The package could include:

max-ishere commented 3 years ago

Here are 2 functions that allow you to read a region file header and then get the raw data from them

import zlib, gzip

def GetAnvilChunkData(region, x: int, z: int) -> dict:
    """Returns info stored in anvil header about a chunk at chunk coordinates
    Arguments:
    region -- a file object created with open('/path/')
    x, z -- chunk position (regardless if its region-local or global)
    """
    region.seek(4 * ((x % 32) + (z % 32) * 32), 0)

    offset = int.from_bytes(region.read(3), byteorder ='big')
    sectors = int.from_bytes(region.read(1), byteorder ='big')

    region.seek(4 * ((x % 32) + (z % 32) * 32) + 0x1000, 0)

    timestamp = int.from_bytes(region.read(4), byteorder ='big')

    return dict({'offset':    offset,
                'sectors':   sectors,
                 'timestamp': timestamp})

def GetChunkNbtData(region, x: int, z: int):
    """Finds chunk data sectors and returns uncompressed buffer with NBT data"""
    anvil_data = GetAnvilChunkData(region, x, z)

    if anvil_data['offset'] < 2:
        return bytes(0)

    region.seek(anvil_data['offset'] * 0x1000)

    length = int.from_bytes(region.read(4), byteorder ='big')
    compression_type = int.from_bytes(region.read(1), byteorder ='big')

    if compression_type == 3:
        return region.read(length)

    if compression_type == 2:
        return zlib.decompress(region.read(length))

    if compression_type == 1:
        return gzip.decompress(region.read(length))

Once you have those you can use code below to read raw nbt data and make a tree.

#!/bin/python
import nbtlib, io
from mvc.mca import GetChunkNbtData

def main():
    file = open('data/region.mca', 'rb')
    nbt_file = nbtlib.File.from_fileobj(io.BytesIO(GetChunkNbtData(file, 3, 3)))
    print(nbt_file.root.keys())

main()

Seems to work fine with my test data from hermitcraft.com S7:

$ python main.py
dict_keys(['Level', 'DataVersion'])

If this gets added then pls credit me, not that I am willing to do that rn myself... sorry have some stuff i want to finish