How to Scan Files Against a Malware Hash List in Python

import hashlib
from pathlib import Path

# Mock file content (in real usage, read from disk)
MOCK_FILE_CONTENT = b"print('hello world')"

KNOWN_MALWARE_HASHES = {
    "8d969eef6ecad3c29a3a629280e686cf0c3f5d5a86aff3ca12020c923adc6c92",
    "5e884898da28047151d0e56f8dc6292773603d0d6aabbdd62a11ef721d1542d8",
}

def sha256_hex(data: bytes) -> str:
    return hashlib.sha256(data).hexdigest()

def scan_file(file_path: Path, malware_hashes: set[str]) -> str:
    content = file_path.read_bytes()
    file_hash = sha256_hex(content)
    if file_hash in malware_hashes:
        return f"MALWARE DETECTED: {file_path} (hash: {file_hash})"
    return f"CLEAN: {file_path} (hash: {file_hash})"

if __name__ == "__main__":
    # Simulate a file scan
    temp_file = Path("mock_sample.py")
    temp_file.write_bytes(MOCK_FILE_CONTENT)
    print(scan_file(temp_file, KNOWN_MALWARE_HASHES))
    temp_file.unlink()  # cleanup