Extracts files inside BFZ archives (Cryo BigFile)

bigfile_extract.py

# wodim, 04/01/2016 -- public domain

import mimetypes
import os
import sys
import zlib

import magic

WORD_SIZE = 4

if len(sys.argv) < 2 or len(sys.argv) > 3:
    print("Usage: %s <raw file> [output folder]" % (sys.argv[0]))
    sys.exit(1)

bigfile = open(sys.argv[1], "rb")

if len(sys.argv) == 3:
    extracted_dir = sys.argv[2]
    if not os.path.isdir(extracted_dir):
        os.mkdir(extracted_dir)
else:
    extracted_dir = None

entries_count = int.from_bytes(bigfile.read(WORD_SIZE), byteorder="little")

for i in range(entries_count):
    dir_offset = bigfile.tell()
    entry = bigfile.read(WORD_SIZE * 5)

    file_name = "file%04x" % (i,)
    offset = int.from_bytes(entry[0:WORD_SIZE], byteorder="little")
    size = int.from_bytes(entry[WORD_SIZE:WORD_SIZE * 2], byteorder="little")
    real_size = int.from_bytes(entry[WORD_SIZE * 2:WORD_SIZE * 3], byteorder="little")
    checksum = int.from_bytes(entry[WORD_SIZE * 3:WORD_SIZE * 4], byteorder="little")
    # last field is just a word of padding

    print("File:\t\t%s" % (file_name,))
    print("Offset:\t\t%04x" % (offset,))
    print("Dir offset:\t%04x" % (dir_offset,))
    print("Size:\t\t%04x" % (size,))
    print("Real size:\t%04x" % (real_size,))
    print("Checksum:\t%04x" % (checksum,))
    print("==================================")

    if extracted_dir: # i.e. if the user wants to extract the file
        file_path = os.path.join(extracted_dir, "%s" % (file_name,))

        # remove existing file
        if os.path.isfile(file_path):
            os.remove(file_path)

        bigfile.seek(offset, 0)
        contents = bigfile.read(size)

        if size != real_size:
            file_path = file_path + "_"
            contents = zlib.decompress(contents)
        
        mimetype = magic.from_buffer(contents, mime=True).decode("utf-8")
        extension = mimetypes.guess_extension(mimetype)
        if extension:
            file_path = file_path + extension

        with open(file_path, "wb") as output_file:
            output_file.write(contents)

    bigfile.seek(dir_offset + WORD_SIZE * 5)

bigfile.close()