2022-08-19 14:53:31 +00:00
|
|
|
#!/usr/bin/env python3
|
|
|
|
# -*-coding:UTF-8 -*
|
|
|
|
|
|
|
|
import os
|
|
|
|
import sys
|
|
|
|
|
|
|
|
sys.path.append(os.environ['AIL_BIN'])
|
|
|
|
##################################
|
|
|
|
# Import Project packages
|
|
|
|
##################################
|
|
|
|
from lib.ConfigLoader import ConfigLoader
|
|
|
|
|
|
|
|
config_loader = ConfigLoader()
|
|
|
|
r_metadata = config_loader.get_db_conn("Kvrocks_Correlations")
|
|
|
|
config_loader = None
|
|
|
|
|
|
|
|
##################################
|
|
|
|
# CORRELATION MIGRATION
|
|
|
|
##################################
|
|
|
|
#
|
|
|
|
# MIGRATE TO KVROCKS + Rename correlation Keys
|
|
|
|
# => Add support for correlations between subtypes
|
|
|
|
# => Common correlation engine for each objects
|
|
|
|
#
|
|
|
|
# Objects Iterations: -screenshot
|
|
|
|
# -decoded
|
|
|
|
# -subtypes
|
|
|
|
# -domains
|
|
|
|
#
|
|
|
|
# /!\ Handle reinsertion /!\
|
|
|
|
#
|
|
|
|
#
|
|
|
|
# CORRELATION DB ????? => purge if needed
|
|
|
|
#
|
|
|
|
#
|
|
|
|
#
|
|
|
|
#
|
|
|
|
#
|
|
|
|
##################################
|
|
|
|
# CORRELATION MIGRATION
|
|
|
|
##################################
|
|
|
|
|
|
|
|
CORRELATION_TYPES_BY_OBJ = {
|
2022-11-28 14:01:40 +00:00
|
|
|
"cryptocurrency": ["domain", "item"],
|
2022-10-25 14:25:19 +00:00
|
|
|
"cve": ["domain", "item"],
|
|
|
|
"decoded": ["domain", "item"],
|
|
|
|
"domain": ["cve", "cryptocurrency", "decoded", "item", "pgp", "username", "screenshot"],
|
|
|
|
"item": ["cve", "cryptocurrency", "decoded", "domain", "pgp", "username", "screenshot"],
|
2022-11-28 14:01:40 +00:00
|
|
|
"pgp": ["domain", "item"],
|
2022-10-25 14:25:19 +00:00
|
|
|
"username": ["domain", "item"],
|
|
|
|
"screenshot": ["domain", "item"],
|
2022-08-19 14:53:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
def get_obj_correl_types(obj_type):
|
|
|
|
return CORRELATION_TYPES_BY_OBJ.get(obj_type)
|
|
|
|
|
|
|
|
def sanityze_obj_correl_types(obj_type, correl_types):
|
|
|
|
obj_correl_types = get_obj_correl_types(obj_type)
|
|
|
|
if correl_types:
|
|
|
|
correl_types = set(correl_types).intersection(obj_correl_types)
|
|
|
|
if not correl_types:
|
|
|
|
correl_types = obj_correl_types
|
|
|
|
return correl_types
|
|
|
|
|
|
|
|
def get_nb_correlation_by_correl_type(obj_type, subtype, obj_id, correl_type):
|
|
|
|
return r_metadata.scard(f'correlation:obj:{obj_type}:{subtype}:{correl_type}:{obj_id}')
|
|
|
|
|
|
|
|
def get_nb_correlations(obj_type, subtype, obj_id, filter_types=[]):
|
|
|
|
if subtype is None:
|
|
|
|
subtype = ''
|
2022-11-22 09:47:15 +00:00
|
|
|
obj_correlations = {}
|
2022-08-19 14:53:31 +00:00
|
|
|
filter_types = sanityze_obj_correl_types(obj_type, filter_types)
|
|
|
|
for correl_type in filter_types:
|
2022-11-22 09:47:15 +00:00
|
|
|
obj_correlations[correl_type] = get_nb_correlation_by_correl_type(obj_type, subtype, obj_id, correl_type)
|
2022-08-19 14:53:31 +00:00
|
|
|
return obj_correlations
|
|
|
|
|
|
|
|
def get_correlation_by_correl_type(obj_type, subtype, obj_id, correl_type):
|
|
|
|
return r_metadata.smembers(f'correlation:obj:{obj_type}:{subtype}:{correl_type}:{obj_id}')
|
|
|
|
|
|
|
|
def get_correlations(obj_type, subtype, obj_id, filter_types=[]):
|
|
|
|
if subtype is None:
|
|
|
|
subtype = ''
|
|
|
|
obj_correlations = {}
|
|
|
|
filter_types = sanityze_obj_correl_types(obj_type, filter_types)
|
|
|
|
for correl_type in filter_types:
|
|
|
|
obj_correlations[correl_type] = get_correlation_by_correl_type(obj_type, subtype, obj_id, correl_type)
|
|
|
|
return obj_correlations
|
|
|
|
|
|
|
|
def exists_obj_correlation(obj_type, subtype, obj_id, obj2_type):
|
|
|
|
if subtype is None:
|
|
|
|
subtype = ''
|
|
|
|
return r_metadata.exists(f'correlation:obj:{obj_type}:{subtype}:{obj2_type}:{obj_id}')
|
|
|
|
|
|
|
|
def is_obj_correlated(obj_type, subtype, obj_id, obj2_type, subtype2, obj2_id):
|
|
|
|
if subtype is None:
|
|
|
|
subtype = ''
|
|
|
|
if subtype2 is None:
|
|
|
|
subtype2 = ''
|
2022-11-28 14:01:40 +00:00
|
|
|
return r_metadata.sismember(f'correlation:obj:{obj_type}:{subtype}:{obj2_type}:{obj_id}', f'{subtype2}:{obj2_id}')
|
2022-08-19 14:53:31 +00:00
|
|
|
|
|
|
|
def add_obj_correlation(obj1_type, subtype1, obj1_id, obj2_type, subtype2, obj2_id):
|
|
|
|
if subtype1 is None:
|
|
|
|
subtype1 = ''
|
|
|
|
if subtype2 is None:
|
|
|
|
subtype2 = ''
|
|
|
|
r_metadata.sadd(f'correlation:obj:{obj1_type}:{subtype1}:{obj2_type}:{obj1_id}', f'{subtype2}:{obj2_id}')
|
|
|
|
r_metadata.sadd(f'correlation:obj:{obj2_type}:{subtype2}:{obj1_type}:{obj2_id}', f'{subtype1}:{obj1_id}')
|
|
|
|
|
|
|
|
|
|
|
|
def delete_obj_correlation(obj1_type, subtype1, obj1_id, obj2_type, subtype2, obj2_id):
|
|
|
|
if subtype1 is None:
|
|
|
|
subtype1 = ''
|
|
|
|
if subtype2 is None:
|
|
|
|
subtype2 = ''
|
2022-11-28 14:01:40 +00:00
|
|
|
r_metadata.srem(f'correlation:obj:{obj1_type}:{subtype1}:{obj2_type}:{obj1_id}', f'{subtype2}:{obj2_id}')
|
|
|
|
r_metadata.srem(f'correlation:obj:{obj2_type}:{subtype2}:{obj1_type}:{obj2_id}', f'{subtype1}:{obj1_id}')
|
2022-08-19 14:53:31 +00:00
|
|
|
|
|
|
|
|
2022-11-28 14:01:40 +00:00
|
|
|
def get_obj_str_id(obj_type, subtype, obj_id):
|
2022-08-19 14:53:31 +00:00
|
|
|
if subtype is None:
|
|
|
|
subtype = ''
|
|
|
|
return f'{obj_type};{subtype};{obj_id}'
|
|
|
|
|
|
|
|
def get_correlations_graph_nodes_links(obj_type, subtype, obj_id, filter_types=[], max_nodes=300, level=1, flask_context=False):
|
|
|
|
links = set()
|
|
|
|
nodes = set()
|
|
|
|
|
|
|
|
obj_str_id = get_obj_str_id(obj_type, subtype, obj_id)
|
|
|
|
|
2022-09-20 14:11:48 +00:00
|
|
|
_get_correlations_graph_node(links, nodes, obj_type, subtype, obj_id, level, max_nodes, filter_types=filter_types, previous_str_obj='')
|
2022-08-19 14:53:31 +00:00
|
|
|
return obj_str_id, nodes, links
|
|
|
|
|
|
|
|
|
|
|
|
def _get_correlations_graph_node(links, nodes, obj_type, subtype, obj_id, level, max_nodes, filter_types=[], previous_str_obj=''):
|
|
|
|
obj_str_id = get_obj_str_id(obj_type, subtype, obj_id)
|
|
|
|
nodes.add(obj_str_id)
|
|
|
|
|
2022-09-20 14:11:48 +00:00
|
|
|
obj_correlations = get_correlations(obj_type, subtype, obj_id, filter_types=filter_types)
|
2022-11-28 14:01:40 +00:00
|
|
|
# print(obj_correlations)
|
2022-08-19 14:53:31 +00:00
|
|
|
for correl_type in obj_correlations:
|
|
|
|
for str_obj in obj_correlations[correl_type]:
|
|
|
|
subtype2, obj2_id = str_obj.split(':', 1)
|
|
|
|
obj2_str_id = get_obj_str_id(correl_type, subtype2, obj2_id)
|
|
|
|
|
|
|
|
if obj2_str_id == previous_str_obj:
|
|
|
|
continue
|
|
|
|
|
|
|
|
if len(nodes) > max_nodes:
|
|
|
|
break
|
|
|
|
nodes.add(obj2_str_id)
|
|
|
|
links.add((obj_str_id, obj2_str_id))
|
|
|
|
|
|
|
|
if level > 0:
|
|
|
|
next_level = level - 1
|
|
|
|
_get_correlations_graph_node(links, nodes, correl_type, subtype2, obj2_id, next_level, max_nodes, filter_types=filter_types, previous_str_obj=obj_str_id)
|