Decompresses SystemPlayReport .msgpack.lz4 files if needed and converts msgpack to json

decompress_msgpack_lz4_convert_to_json.py

#!/usr/bin/env python3

import struct
import sys
import json
import base64
import os
from pathlib import Path
import msgpack
import lz4.block

if len(sys.argv) != 2:
    sys.exit('Usage: ' + sys.argv[0] + ' [dirname]')

MSGPACK_PATH = Path(sys.argv[1])
if not MSGPACK_PATH.exists() or not MSGPACK_PATH.is_dir():
    sys.exit('Path "' + str(MSGPACK_PATH) + '" does not exist or is not a directory.')

for filepath in MSGPACK_PATH.iterdir():
    if not filepath.is_file():
        continue

    decomp_filepath = None
    decomp = bytes()
    with filepath.open(mode='rb') as f:
        data = f.read()
    file_size = len(data)
    print('File ' + filepath.name + ' size: ', file_size)

    if filepath.suffix == '.lz4':
        i = 0
        while i < file_size:
            chunk_size = struct.unpack('<I', data[i:i+4])[0]
            i += 4
            decomp += lz4.block.decompress(data[i:i+chunk_size], uncompressed_size=0x400000)
            i += chunk_size

        decomp_filepath = filepath.with_suffix('')
        print('Decompressed size: ', len(decomp))
        with decomp_filepath.open(mode='wb') as f:
            f.write(decomp)

    if len(decomp) == 0:
        decomp = data
    if decomp_filepath or filepath.suffix == '.msgpack':
        def ext_hook(_, hook_data):
            return hook_data.hex()

        class Base64Encoder(json.JSONEncoder):
            # pylint: disable=method-hidden
            def default(self, o):
                if isinstance(o, bytes):
                    return base64.b64encode(o).decode()
                return json.JSONEncoder.default(self, o)

        with filepath.with_suffix('').with_suffix('.json').open(mode='w') as f:
            decomp = msgpack.unpackb(decomp, raw=False, ext_hook=ext_hook)
            json.dump(decomp, f, cls=Base64Encoder)
            f.flush()
            print('Final JSON size: ', os.fstat(f.fileno()).st_size)