def add(self, d, dx, name, sname, regexp_pattern, regexp_exclude_pattern):
for _class in d.get_classes():
if regexp_pattern != None:
if re.match(regexp_pattern, _class.get_name()) == None:
continue
if regexp_exclude_pattern != None:
if re.match(regexp_exclude_pattern, _class.get_name()) != None:
continue
print "\t", _class.get_name()
for method in _class.get_methods():
code = method.get_code()
if code == None:
continue
if method.get_length() < 50 or method.get_name(
) == "<clinit>" or method.get_name() == "<init>":
continue
buff_list = dx.get_method_signature(
method, predef_sign=DEFAULT_SIGNATURE).get_list()
if len(set(buff_list)) == 1:
continue
for e in buff_list:
self.db.add_element(name, sname, _class.get_name(),
method.get_length(), long(simhash(e)))
def percentages_to_graph(self):
info = {"info": [], "nodes": [], "links": []}
N = {}
L = {}
for _class in self.vm.get_classes():
elems_hash = set()
# print _class.get_name()
for method in _class.get_methods():
code = method.get_code()
if code == None:
continue
buff_list = self.vmx.get_method_signature(
method, predef_sign=DEFAULT_SIGNATURE).get_list()
for i in buff_list:
elem_hash = long(simhash(i))
elems_hash.add(elem_hash)
ret = self.db.elems_are_presents(elems_hash)
sort_ret = eval_res_per_class(ret)
if sort_ret != {}:
if _class.get_name() not in N:
info["nodes"].append({
"name":
_class.get_name().split("/")[-1],
"group":
0
})
N[_class.get_name()] = len(N)
for j in sort_ret:
if j not in N:
N[j] = len(N)
info["nodes"].append({"name": j, "group": 1})
key = _class.get_name() + j
if key not in L:
L[key] = {
"source": N[_class.get_name()],
"target": N[j],
"value": 0
}
info["links"].append(L[key])
for k in sort_ret[j]:
if sort_ret[j][k] > L[key]["value"]:
L[key]["value"] = sort_ret[j][k]
return info
def percentages_code(self, exclude_list):
libs = re.compile('|'.join("(" + i + ")" for i in exclude_list))
classes_size = 0
classes_db_size = 0
classes_edb_size = 0
classes_udb_size = 0
for _class in self.vm.get_classes():
class_size = 0
elems_hash = set()
for method in _class.get_methods():
code = method.get_code()
if code == None:
continue
buff_list = self.vmx.get_method_signature(
method, predef_sign=DEFAULT_SIGNATURE).get_list()
for i in buff_list:
elem_hash = long(simhash(i))
elems_hash.add(elem_hash)
class_size += method.get_length()
classes_size += class_size
if class_size == 0:
continue
ret = self.db.elems_are_presents(elems_hash)
sort_ret = eval_res_per_class(ret)
if sort_ret == {}:
if libs.search(_class.get_name()) != None:
classes_edb_size += class_size
else:
classes_udb_size += class_size
else:
classes_db_size += class_size
return (classes_db_size / float(classes_size)) * 100, (
classes_edb_size / float(classes_size)) * 100, (
classes_udb_size / float(classes_size)) * 100
def percentages_to_graph(self) :
info = { "info" : [], "nodes" : [], "links" : []}
N = {}
L = {}
for _class in self.vm.get_classes() :
elems_hash = set()
# print _class.get_name()
for method in _class.get_methods() :
code = method.get_code()
if code == None :
continue
buff_list = self.vmx.get_method_signature( method, predef_sign = DEFAULT_SIGNATURE ).get_list()
for i in buff_list :
elem_hash = long(simhash( i ))
elems_hash.add( elem_hash )
ret = self.db.elems_are_presents( elems_hash )
sort_ret = eval_res_per_class( ret )
if sort_ret != {} :
if _class.get_name() not in N :
info["nodes"].append( { "name" : _class.get_name().split("/")[-1], "group" : 0 } )
N[_class.get_name()] = len(N)
for j in sort_ret :
if j not in N :
N[j] = len(N)
info["nodes"].append( { "name" : j, "group" : 1 } )
key = _class.get_name() + j
if key not in L :
L[ key ] = { "source" : N[_class.get_name()], "target" : N[j], "value" : 0 }
info["links"].append( L[ key ] )
for k in sort_ret[j] :
if sort_ret[j][k] > L[ key ]["value"] :
L[ key ]["value"] = sort_ret[j][k]
return info
def percentages_code(self, exclude_list) :
libs = re.compile('|'.join( "(" + i + ")" for i in exclude_list))
classes_size = 0
classes_db_size = 0
classes_edb_size = 0
classes_udb_size = 0
for _class in self.vm.get_classes() :
class_size = 0
elems_hash = set()
for method in _class.get_methods() :
code = method.get_code()
if code == None :
continue
buff_list = self.vmx.get_method_signature( method, predef_sign = DEFAULT_SIGNATURE ).get_list()
for i in buff_list :
elem_hash = long(simhash( i ))
elems_hash.add( elem_hash )
class_size += method.get_length()
classes_size += class_size
if class_size == 0 :
continue
ret = self.db.elems_are_presents( elems_hash )
sort_ret = eval_res_per_class( ret )
if sort_ret == {} :
if libs.search(_class.get_name()) != None :
classes_edb_size += class_size
else :
classes_udb_size += class_size
else :
classes_db_size += class_size
return (classes_db_size/float(classes_size)) * 100, (classes_edb_size/float(classes_size)) * 100, (classes_udb_size/float(classes_size)) * 100
def percentages(self, vm, vmx, threshold=10) :
elems_hash = set()
for _class in vm.get_classes() :
for method in _class.get_methods() :
code = method.get_code()
if code == None :
continue
buff_list = vmx.get_method_signature( method, predef_sign = DEFAULT_SIGNATURE ).get_list()
for i in buff_list :
elem_hash = long(simhash( i ))
elems_hash.add( elem_hash )
ret, info = self.db.elems_are_presents( elems_hash )
sorted_ret = self.eval_res(ret, info, threshold)
info = {}
for i in sorted_ret :
v = sorted(sorted_ret[i], key=lambda x: x[1])
v.reverse()
info[i] = []
for j in v :
info[i].append( [j[0], j[1]] )
info_name = self.db.classes_are_presents( vm.get_classes_names() )
for i in info_name :
if i not in info :
info[i] = None
return info
def percentages(self, vm, vmx, threshold=10):
elems_hash = set()
for _class in vm.get_classes():
for method in _class.get_methods():
code = method.get_code()
if code == None:
continue
buff_list = vmx.get_method_signature(
method, predef_sign=DEFAULT_SIGNATURE).get_list()
for i in buff_list:
elem_hash = long(simhash(i))
elems_hash.add(elem_hash)
ret, info = self.db.elems_are_presents(elems_hash)
sorted_ret = self.eval_res(ret, info, threshold)
info = {}
for i in sorted_ret:
v = sorted(sorted_ret[i], key=lambda x: x[1])
v.reverse()
info[i] = []
for j in v:
info[i].append([j[0], j[1]])
info_name = self.db.classes_are_presents(vm.get_classes_names())
for i in info_name:
if i not in info:
info[i] = None
return info
def add(self, d, dx, name, sname, regexp_pattern, regexp_exclude_pattern):
for _class in d.get_classes():
if regexp_pattern != None:
if re.match(regexp_pattern, _class.get_name()) == None:
continue
if regexp_exclude_pattern != None:
if re.match(regexp_exclude_pattern, _class.get_name()) != None:
continue
print "\t", _class.get_name()
for method in _class.get_methods():
code = method.get_code()
if code == None:
continue
if method.get_length() < 50 or method.get_name() == "<clinit>" or method.get_name() == "<init>":
continue
buff_list = dx.get_method_signature( method, predef_sign = DEFAULT_SIGNATURE ).get_list()
if len(set(buff_list)) == 1:
continue
for e in buff_list:
self.db.add_element( name, sname, _class.get_name(), method.get_length(), long(simhash(e)) )
