def handle(self, *args, **options):
bf = BloomFilter(n=options.get("n") * 2, p=options.get("p"))
with open(options.get("input"), "r") as inp:
for i in trange(options.get("n")):
line = inp.readline()
if not line:
break
bf.add(line.strip().upper().encode("ascii"))
filename = settings.BASE_PASSWORD_STRENGTH_BLOOM_FILE
tmpfile = f"{filename}.tmp"
logger.info(f"Writing new password bloom filter: {tmpfile}")
with open(tmpfile, "wb") as outp:
bf.write(outp)
logger.info(f"Replacing old password bloom filter: {filename}")
os.rename(tmpfile, filename)
# insert domain into bloomfilter
if not domain.lower().encode("utf8") in bloom:
bloom.add(domain.lower().encode("utf8"))
# insert into family/domain lookup table
dga_lookup_dict[domain] = family
# test with first DGA domain/family pair that should be present in the bloomfilter and the dga_dict
if not (first_entry and test_domain and test_family):
logging.error("Unknown error while creating bloomfilter and DGA dict")
sys.exit(-1)
if test_domain.lower().encode("utf8") in bloom and dga_lookup_dict.get(
test_domain.lower(), "") == test_family:
logging.info("%s (%s)", test_domain, test_family)
logging.info("Bloomfilter and DGA dict successfully created")
else:
logging.error("Unknown error while creating bloomfilter and DGA dict")
sys.exit(-1)
# 5. write bloomfilter and dga dict to a file so that the sandbox can use it in the python signature
logging.info("Write bloomfilter and dga dict to files")
bloom_path = os.path.join(CUCKOO_ROOT, "data", "dga.bloom")
with open(bloom_path, "wb") as f:
bloom.write(f)
lookup_path = os.path.join(CUCKOO_ROOT, "data", "dga_lookup_dict.json.gz")
with gzip.GzipFile(lookup_path, "w") as fout:
fout.write(json.dumps(dga_lookup_dict).encode("utf8"))
logging.info("Successfully generated bloomfilter and dga dict file")