def save_item_correlation(self, subtype, obj_id, item_id, item_date):
self.update_correlation_daterange(subtype, obj_id, item_date)
# global set
r_serv_metadata.sadd(
'set_{}_{}:{}'.format(self.correlation_name, subtype, obj_id),
item_id)
# daily
r_serv_metadata.hincrby(
'{}:{}:{}'.format(self.correlation_name, subtype, item_date),
obj_id, 1)
# all type
r_serv_metadata.zincrby(
'{}_all:{}'.format(self.correlation_name, subtype), obj_id, 1)
## object_metadata
# item
r_serv_metadata.sadd(
'item_{}_{}:{}'.format(self.correlation_name, subtype, item_id),
obj_id)
# domain
if item_basic.is_crawled(item_id):
domain = item_basic.get_item_domain(item_id)
self.save_domain_correlation(domain, subtype, obj_id)
def add_obj_tag(object_type, object_id, tag, obj_date=None):
if object_type == "item": # # TODO: # FIXME: # REVIEW: rename me !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
if obj_date is None:
raise ValueError("obj_date is None")
# add tag
r_serv_metadata.sadd('tag:{}'.format(object_id), tag)
r_serv_tags.sadd('{}:{}'.format(tag, obj_date), object_id)
# add domain tag
if item_basic.is_crawled(
object_id
) and tag != 'infoleak:submission="crawler"' and tag != 'infoleak:submission="manual"':
domain = item_basic.get_item_domain(object_id)
add_tag("domain", tag, domain)
else:
r_serv_metadata.sadd('tag:{}'.format(object_id), tag)
r_serv_tags.sadd('{}:{}'.format(object_type, tag), object_id)
def get_item_domain(item_id):
return item_basic.get_item_domain(item_id)