Need help, LZMA compressed file with unknown compression algorithm

[wq223]

I try to decompress the compressed file of a certain game, LZMA compressed file
Because it has a four-byte LZMA identifier followed by a four-byte decompression size
I skipped these bytes and tried to use python's lzma library to decompress, but it said that the compressed file was not in the correct format
Does it require special treatment? Can you discuss the specific decompression logic? Thank you very much. I attach a compressed file here.
Additional: This game is Ace Racing from NetEase, developed using messiah. The other compression algorithm zzz4 (lz4) has been successfully processed and decompressed in python by referring to other scripts, but lzma keeps giving me the error message that it cannot be decompressed, so I need Understand his compression algorithm and logic

lzma.zip

[ikskoks]

Works fine in quickbms. try this script:

 

idstring "LZMA"
comtype lzma
get UNCOMPRESSED_SIZE long
get ARCHIVE_SIZE asize
xmath COMPRESSED_SIZE "ARCHIVE_SIZE - 8"
SavePos DATA_OFFSET

get NAME basename
get EXTENSION extension
string NAME + ".decompressed." + EXTENSION
clog NAME DATA_OFFSET COMPRESSED_SIZE UNCOMPRESSED_SIZE

So maybe something is wrong with Python's lzma library that you're using.

[wq223]

There are many different variants of lzma, and I considered that I was using the wrong processing method, or incorrect modules, and was unable to decompress them in python, because I didn't know what version he was using, because I referred to the common lzma header, can't seem to find a matching type

[Rabatini]

39 minutes ago, wq223 said:

There are many different variants of lzma, and I considered that I was using the wrong processing method, or incorrect modules, and was unable to decompress them in python, because I didn't know what version he was using, because I referred to the common lzma header, can't seem to find a matching type

It uses standard one.
you can use it with pylzma module.

here, the decompressor and compressor in python.

need install pylzma module and tkinter because of viadagem com janelinha that i did"

 

import os
import struct
import pylzma
import tkinter as tk
from tkinter import filedialog, messagebox

# Function to decompress LZMA
def decompress_lzma(input_file):
    with open(input_file, 'rb') as f:
        # Check if the file has the 'LZMA' magic name
        magic_name = f.read(4)
        if magic_name != b'LZMA':
            raise ValueError("File does not have the 'LZMA' magic name")
        # Get the uncompressed size (UNCOMPRESSED_SIZE)
        uncompressed_size = struct.unpack('<I', f.read(4))[0]
        # Get the total file size (ARCHIVE_SIZE)
        f.seek(0, os.SEEK_END)
        archive_size = f.tell()
        # Calculate the compressed size (COMPRESSED_SIZE)
        compressed_size = archive_size - 8
        # Save the starting position of the compressed data (DATA_OFFSET)
        data_offset = f.seek(8, os.SEEK_SET)
        # Create the output file name
        basename, extension = os.path.splitext(input_file)
        output_file = f"{basename}.decompressed{extension}"
        # Read the compressed data
        f.seek(data_offset)
        compressed_data = f.read(compressed_size)
        # Decompress the data
        decompressed_data = pylzma.decompress(compressed_data, maxlength=uncompressed_size)
        # Write the decompressed data to the output file
        with open(output_file, 'wb') as out_f:
            out_f.write(decompressed_data)
        return output_file

# Function to compress a file using LZMA
def compress_lzma(input_file):
    with open(input_file, 'rb') as f:
        # Read the input file data
        data = f.read()

        # Compress the data using the desired parameters
        compressed_data = pylzma.compress(
            data,
            eos=True,
            dictionary=23,  # 2 MB dictionary size (within the valid range of 0-27)
            fastBytes=32,
            literalContextBits=3, #3 same used in your game 4 you get better compression
            literalPosBits=0,
            algorithm = 2
        )

        # Create the output file name
        basename, extension = os.path.splitext(input_file)
        output_file = f"{basename}.lzma"

        # Write the LZMA header and compressed data
        with open(output_file, 'wb') as out_f:
            # Write the 'LZMA' magic name
            out_f.write(b'LZMA')
            # Write the uncompressed size
            out_f.write(struct.pack('<I', len(data)))
            # Write the compressed data
            out_f.write(compressed_data)

        return output_file
    
# Function to handle file selection and decompression
def decompress_file():
    input_file = filedialog.askopenfilename(filetypes=[("LZMA Files", "*.lzma"), ("All Files", "*.*")])
    if input_file:
        try:
            output_file = decompress_lzma(input_file)
            messagebox.showinfo("Success", f"File decompressed and saved as: {output_file}")
        except Exception as e:
            messagebox.showerror("Error", str(e))

# Function to handle file selection and compression
def compress_file():
    input_file = filedialog.askopenfilename(filetypes=[("All Files", "*.*")])
    if input_file:
        try:
            output_file = compress_lzma(input_file)
            messagebox.showinfo("Success", f"File compressed and saved as: {output_file}")
        except Exception as e:
            messagebox.showerror("Error", str(e))

# Create the main application window
root = tk.Tk()
root.title("LZMA Compressor/Decompressor")

# Set the size of the window
root.geometry("400x100")  # width x height


# Create and place the buttons
compress_button = tk.Button(root, text="Compress File", command=compress_file)
compress_button.pack(pady=10)
decompress_button = tk.Button(root, text="Decompress File", command=decompress_file)
decompress_button.pack(pady=10)

# Start the Tkinter event loop
root.mainloop()

 

lzma.py

[wq223]

Thank you very much for your script, now I finally understand the logic