def arrayToDict(arr):
#debug_id = str(uuid.uuid4())
#print >> sys.stderr,"[START:%s] arrayToDict:%s" %(debug_id, (arr))
rs = {}
if arr == None:
print "ERROR: arr is NULL in arrayToDict"
return rs
try:
for obj in arr:
if isinstance(obj, basestring):
if not "basestring_" in rs:
rs['basestring_'] = [obj]
else:
rs['basestring_'].append(obj)
elif isinstance(obj, list):
#print >> sys.stderr, "DEBUG1 [%s] list %s" %(debug_id, str(obj))
subarr = extractArrayValues(obj)
subrs = arrayToDict(subarr)
mergeTwoArrayDict(rs,subrs)
#print >> sys.stderr, "DEBUG2 [%s] list %s" %(debug_id, str(obj))
elif isinstance(obj, dict):
#print >> sys.stderr, "DEBUG1 [%s] dict %s" %(debug_id, str(obj))
for k in obj:
if not k in rs:
rs[k] = [obj[k]]
else:
rs[k].append(obj[k])
#print >> sys.stderr, "DEBUG2 [%s] dict %s" %(debug_id, str(obj))
except Exception as e:
displayErrorMsg('arrayToDict',str(e)+' '+str(arr))
def extractObjectValues(data):
if data == None:
return {}
rs = {} #key : [val1, val2]
for k in data:
if not k in rs:
rs[k] = []
val = data[k]
if isinstance(val, dict):
sub_rs = extractObjectValues(val)
mergeTwoArrayDict(rs, sub_rs)
elif isinstance(val, list) or isinstance(val, tuple):
arr = extractArrayValues(val)
for item in arr:
if isinstance(item, dict):
mergeTwoArrayDict(rs, item)
elif isinstance(item, basestring) or isinstance(item, int) or isinstance(item, float):
rs[k].append(item)
else:
displayErrorMsg('extractObjectValues',\
"shouldn't have other types"+str(type(item)))
elif isinstance(val, basestring) or isinstance(val, int) or isinstance(val, float):
rs[k].append(val)
else:
displayErrorMsg('extractObjectValues',\
"unknown type:"+str(type(val)))
continue
invalid_keys = [k for k in rs if len(rs[k]) == 0]
for k in invalid_keys:
del rs[k]
return rs
def extractObjectValues(data):
if data == None:
return {}
rs = {} #key : [val1, val2]
for k in data:
if not k in rs:
rs[k] = []
val = data[k]
if isinstance(val, dict):
sub_rs = extractObjectValues(val)
mergeTwoArrayDict(rs, sub_rs)
elif isinstance(val, list) or isinstance(val, tuple):
arr = extractArrayValues(val)
for item in arr:
if isinstance(item, dict):
mergeTwoArrayDict(rs, item)
elif isinstance(item, basestring) or isinstance(
item, int) or isinstance(item, float):
rs[k].append(item)
else:
displayErrorMsg('extractObjectValues',\
"shouldn't have other types"+str(type(item)))
elif isinstance(val, basestring) or isinstance(val, int) or isinstance(
val, float):
rs[k].append(val)
else:
displayErrorMsg('extractObjectValues',\
"unknown type:"+str(type(val)))
continue
invalid_keys = [k for k in rs if len(rs[k]) == 0]
for k in invalid_keys:
del rs[k]
return rs
def arrayToDict(arr):
#debug_id = str(uuid.uuid4())
#print >> sys.stderr,"[START:%s] arrayToDict:%s" %(debug_id, (arr))
rs = {}
if arr == None:
print "ERROR: arr is NULL in arrayToDict"
return rs
try:
for obj in arr:
if isinstance(obj, basestring):
if not "basestring_" in rs:
rs['basestring_'] = [obj]
else:
rs['basestring_'].append(obj)
elif isinstance(obj, list):
#print >> sys.stderr, "DEBUG1 [%s] list %s" %(debug_id, str(obj))
subarr = extractArrayValues(obj)
subrs = arrayToDict(subarr)
mergeTwoArrayDict(rs, subrs)
#print >> sys.stderr, "DEBUG2 [%s] list %s" %(debug_id, str(obj))
elif isinstance(obj, dict):
#print >> sys.stderr, "DEBUG1 [%s] dict %s" %(debug_id, str(obj))
for k in obj:
if not k in rs:
rs[k] = [obj[k]]
else:
rs[k].append(obj[k])
#print >> sys.stderr, "DEBUG2 [%s] dict %s" %(debug_id, str(obj))
except Exception as e:
displayErrorMsg('arrayToDict', str(e) + ' ' + str(arr))
def match(self, target_tree):
if not isinstance(target_tree, TemplateTree):
displayErrorMsg('TemplateTree.match,'\
"matching tree, target tree should be TemplateTree" )
return False
if self.key != target_tree.key:
return False
if self.type == 'json' and target_tree.type == 'json':
return True
elif self.type != target_tree.type:
return False
length = len(target_tree.nodes)
for i in range(length):
try:
if self.nodes[i].tag != target_tree.nodes[i].tag:
return False
if self.nodes[i].tag == 'String':
if not self.string_types[str(i)].match(
target_tree.nodes[i].value):
return False
elif self.nodes[i].tag == 'Object':
target_obj = extractObjectValues(
target_tree.nodes[i].value)
for k in target_obj:
if not k in self.object_types[str(i)]:
return False
if isinstance(target_obj[k], list):
for item in target_obj[k]:
if not self.object_types[str(i)][k].match(
item):
return False
else:
if not self.object_types[str(i)][k].match(
target_obj[k]):
return False
elif self.nodes[i].tag == 'Array':
if target_tree.nodes[i].value == None:
continue
target_obj = arrayToDict(target_tree.nodes[i].value)
target_obj = extractObjectValues(target_obj)
for k in target_obj:
if not k in self.array_types[str(i)]:
return False
if isinstance(target_obj[k], list):
for item in target_obj[k]:
if not self.array_types[str(i)][k].match(item):
return False
else:
if not self.array_types[str(i)][k].match(
target_obj[k]):
return False
except Exception as e:
displayErrorMsg('TemplateTree.match', str(e))
return False
return True
def match(self, target_tree):
if not isinstance(target_tree, TemplateTree):
displayErrorMsg('TemplateTree.match,'\
"matching tree, target tree should be TemplateTree" )
return False
if self.key != target_tree.key:
return False
if self.type == 'json' and target_tree.type == 'json':
return True
elif self.type != target_tree.type:
return False
length = len(target_tree.nodes)
for i in range(length):
try:
if self.nodes[i].tag != target_tree.nodes[i].tag:
return False
if self.nodes[i].tag == 'String':
if not self.string_types[str(i)].match(target_tree.nodes[i].value):
return False
elif self.nodes[i].tag == 'Object':
target_obj = extractObjectValues(target_tree.nodes[i].value)
for k in target_obj:
if not k in self.object_types[str(i)]:
return False
if isinstance(target_obj[k], list):
for item in target_obj[k]:
if not self.object_types[str(i)][k].match(item):
return False
else:
if not self.object_types[str(i)][k].match(target_obj[k]):
return False
elif self.nodes[i].tag == 'Array':
if target_tree.nodes[i].value == None:
continue
target_obj = arrayToDict(target_tree.nodes[i].value)
target_obj = extractObjectValues(target_obj)
for k in target_obj:
if not k in self.array_types[str(i)]:
return False
if isinstance(target_obj[k], list):
for item in target_obj[k]:
if not self.array_types[str(i)][k].match(item):
return False
else:
if not self.array_types[str(i)][k].match(target_obj[k]):
return False
except Exception as e:
displayErrorMsg('TemplateTree.match', str(e))
return False
return True
def extractArrayValues(data):
rs = []
for item in data:
if isinstance(item, basestring) or isinstance(item, int) or isinstance(item, float):
rs.append(item)
elif isinstance(item, list) or isinstance(item, tuple):
val = extractArrayValues(item)
rs += val
elif isinstance(item, dict):
rs.append(extractObjectValues(item))
else:
displayErrorMsg('extractArrayValues', "unknown type "+str(type(item)))
return rs
def dumps(self):
try:
obj = {'type': self.type}
if self.type == 'json':
obj['tree'] = self.nodes
else:
obj['tree'] = ','.join([b64encode(x.tag) for x in self.nodes])
obj['string_types_str'] = self.string_types_str
obj['object_types_str'] = self.object_types_str
obj['array_types_str'] = self.array_types_str
return json.dumps(obj)
except Exception as e:
displayErrorMsg("TemplateTree.dumps", str(e))
return None
def dumps(self):
try:
obj = {'type' : self.type}
if self.type == 'json':
obj['tree'] = self.nodes
else:
obj['tree'] = ','.join([b64encode(x.tag) for x in self.nodes])
obj['string_types_str'] = self.string_types_str
obj['object_types_str'] = self.object_types_str
obj['array_types_str'] = self.array_types_str
return json.dumps(obj)
except Exception as e:
displayErrorMsg("TemplateTree.dumps", str(e))
return None
def extractArrayValues(data):
rs = []
for item in data:
if isinstance(item, basestring) or isinstance(item, int) or isinstance(
item, float):
rs.append(item)
elif isinstance(item, list) or isinstance(item, tuple):
val = extractArrayValues(item)
rs += val
elif isinstance(item, dict):
rs.append(extractObjectValues(item))
else:
displayErrorMsg('extractArrayValues',
"unknown type " + str(type(item)))
return rs
def getTreesForDomainFromDB(domain):
tree_strings = fetchTrees(domain)
if tree_strings == None:
return None
tree_dict = {} #{key : TemplateTree}
for key in tree_strings:
tree = TemplateTree(None, None)
try:
tree.loads(tree_strings[key])
tree_dict[key] = tree
except Exception as e:
displayErrorMsg("getTreesForDomainFromDB", str(e))
return tree_dict
def getTreesForDomainFromDB(domain):
tree_strings = fetchTrees(domain)
if tree_strings == None:
return None
tree_dict = {} #{key : TemplateTree}
for key in tree_strings:
tree = TemplateTree(None, None)
try:
tree.loads(tree_strings[key])
tree_dict[key] = tree
except Exception as e:
displayErrorMsg("getTreesForDomainFromDB",str(e))
return tree_dict
def loads(self, obj_str):
try:
obj = json.loads(obj_str)
self.type = obj['type']
if self.type == 'json':
self.nodes = obj['tree']
elif self.type == 'js':
nodes = obj['tree'].split(',')
self.nodes = [ASTOutputNode(b64decode(x)) for x in nodes]
self.calc_key()
self.string_types_str = obj['string_types_str']
for key in self.string_types_str:
node_pattern = NodePattern()
node_pattern.loads(self.string_types_str[key])
self.string_types[key] = node_pattern
#print "successfully loaded %d string patterns" %(len(self.string_types_str))
self.object_types_str = obj['object_types_str']
for index in self.object_types_str:
self.object_types[index] = {}
for key in self.object_types_str[index]:
node_pattern = NodePattern()
node_pattern.loads(self.object_types_str[index][key])
self.object_types[index][key] = node_pattern
#print "successfully loaded %d object patterns" %(len(self.object_types_str))
self.array_types_str = obj['array_types_str']
for index in self.array_types_str:
self.array_types[index] = {}
for key in self.array_types_str[index]:
node_pattern = NodePattern()
node_pattern.loads(self.array_types_str[index][key])
self.array_types[index][key] = node_pattern
#print "successfully loaded %d array patterns" %(len(self.array_types_str))
return True
except Exception as e:
displayErrorMsg("TemplateTree.loads", str(e))
return False
def loads(self, obj_str):
try:
obj = json.loads(obj_str)
self.type = obj['type']
if self.type == 'json':
self.nodes = obj['tree']
elif self.type == 'js':
nodes = obj['tree'].split(',')
self.nodes = [ASTOutputNode(b64decode(x)) for x in nodes]
self.calc_key()
self.string_types_str = obj['string_types_str']
for key in self.string_types_str:
node_pattern = NodePattern()
node_pattern.loads(self.string_types_str[key])
self.string_types[key] = node_pattern
#print "successfully loaded %d string patterns" %(len(self.string_types_str))
self.object_types_str = obj['object_types_str']
for index in self.object_types_str:
self.object_types[index] = {}
for key in self.object_types_str[index]:
node_pattern = NodePattern()
node_pattern.loads(self.object_types_str[index][key])
self.object_types[index][key] = node_pattern
#print "successfully loaded %d object patterns" %(len(self.object_types_str))
self.array_types_str = obj['array_types_str']
for index in self.array_types_str:
self.array_types[index] = {}
for key in self.array_types_str[index]:
node_pattern = NodePattern()
node_pattern.loads(self.array_types_str[index][key])
self.array_types[index][key] = node_pattern
#print "successfully loaded %d array patterns" %(len(self.array_types_str))
return True
except Exception as e:
displayErrorMsg("TemplateTree.loads", str(e))
return False
def calc_key(self):
if self.nodes != None:
m = hashlib.md5()
if self.type == "json":
keys = sorted(self.nodes.keys())
for k in keys:
m.update(k)
elif self.type == "js":
for node in self.nodes:
m.update(node.tag)
else:
debug_msg = "TemplateTree nodes format error: %s %d %d" \
%(self.nodes[0].__class__, id(type(self.nodes[0])), id(ASTOutputNode))
displayErrorMsg('TemplateTree.calc_key', debug_msg)
self.key = None
key = m.hexdigest()
else:
displayErrorMsg('TemplateTree.calc_key', "Nodes are None")
self.key = None
self.key = key
def calc_key(self):
if self.nodes != None:
m = hashlib.md5()
if self.type == "json":
keys = sorted(self.nodes.keys())
for k in keys:
m.update(k)
elif self.type == "js":
for node in self.nodes:
m.update(node.tag)
else:
debug_msg = "TemplateTree nodes format error: %s %d %d" \
%(self.nodes[0].__class__, id(type(self.nodes[0])), id(ASTOutputNode))
displayErrorMsg('TemplateTree.calc_key', debug_msg)
self.key = None
key = m.hexdigest()
else:
displayErrorMsg('TemplateTree.calc_key', "Nodes are None")
self.key = None
self.key = key
def generateTemplateBasedOnURLsFromFile(path, dst_path):
f = open(path)
scriptdict = {}
total_script_blocks = 0
total_uniq_script_blocks = 0
debug_dict = {}
static_scripts = 0
dynamic_scripts = 0
for line in f:
url = line.strip()
print "process url " + url
hosts, inlines = fetchScripts(url)
if inlines == None or len(inlines) == 0:
print "no inlines for " + url
continue
for inline in inlines:
is_json = False
rs, sc = analyzeJSCodesFinerBlock(inline)
if rs == None:
rs = analyzeJSON(inline)
is_json = True
if rs == None:
continue
if is_json:
tree = TemplateTree(rs, None)
if not tree.key in scriptdict:
scriptdict[tree.key] = [(inline, url, tree, -1)]
debug_dict[tree.key] = [inline]
else:
debug_dict[tree.key].append(inline)
contents = [x[0] for x in scriptdict[key]]
if not inline in contents:
scriptdict[tree.key].append((inline, url, tree, -1))
total_uniq_script_blocks += 1
total_script_blocks += 1
else:
for index in range(len(rs)):
total_script_blocks += 1
seq = rs[index]
tree = TemplateTree(seq, None)
key = tree.key
if not key in scriptdict:
debug_dict[key] = [sc[index]]
scriptdict[key] = [(sc[index], url, tree, index)]
print " add key %s" % key
else:
contents = [x[0] for x in scriptdict[key]]
debug_dict[key].append(sc[index])
if not sc[index] in contents:
scriptdict[key].append(
(sc[index], url, tree, index))
print " item %s has %d unique scripts" % (
key, len(scriptdict[key]))
total_uniq_script_blocks += 1
fw = open(os.path.join(dst_path, 'debug'), 'w')
for k in debug_dict:
vals = debug_dict[k]
fw.write("%d %s \n" % (len(vals), k))
fw.write(" --EXAMPLE-- %s\n" % vals[0])
fw.close()
#start to analyze trees
#scriptdict[tree_key] = [(script, url, tree, index)]
trees = []
insufficient_urls = {}
keys = sorted(scriptdict.keys(), key=lambda k: len(scriptdict[k]))
for key in keys:
is_static = True
name = "%d_%s" % (len(scriptdict[key]), key)
fw = open(os.path.join(dst_path, name), 'w')
for item in scriptdict[key]:
fw.write(item[1] + "||" + str(item[3]) + " " + str(item[0]) +
"\n")
#make sure all template trees with the same key are the same
script_list = scriptdict[key]
length_list = sorted([len(item[2].nodes) for item in script_list])
seq_length = 0
if length_list[0] != length_list[-1]:
fw.write("[ALERT] seq length is not consistent")
fw.close()
continue
else:
seq_length = length_list[0]
#only handle JavaScript for now
tree = script_list[0][2]
if tree.type == "json":
print "the inline is json!"
fw.write("[TODO]: the inline is json. This is next step\n")
fw.close()
trees.append(tree)
continue
#process String/Object/Array nodes
#script_list: [(script, url, tree, index)]
fw.write("start analyzeing values\n")
script_length = len(script_list)
for i in range(seq_length):
node = script_list[0][2].nodes[i]
try:
if node.tag == "String":
vals = [item[2].nodes[i].value for item in script_list]
encoded_val = [b64encode(x) for x in vals]
#item = 'string%d: %s' %(i, ','.join(encoded_val))
#fw.write(item+"\n")
tree.strings[i] = vals
node_pattern = generateNodePattern(vals)
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
tree.string_types_str[str(i)] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [
item[1] for item in script_list
]
# testing
#node_pattern = NodePattern()
#r = node_pattern.loads(tree.string_types_str[i])
#if r == False:
# print "node_pattern failed to load: "+tree.string_types_str[i]
#else:
# print "successfully loaded tree: "+tree.string_types_str[i]
print "STRING%d: [TYPE:%s] [VALUE:%s]" \
%(i, tree.string_types_str[str(i)],','.join(encoded_val))
if node.tag == "Object":
#debug = "tag:%s val:%s" \
# %(script_list[0][2].nodes[i].tag,str(script_list[0][2].nodes[i].value))
#print "DEBUG: %s" %debug
rs = analyzeObjectResultHelper(script_list, i)
rs = extractObjectValues(rs)
type_dict = {}
for k in rs:
encoded_val = [b64encode(x) for x in rs[k]]
node_pattern = generateNodePattern(rs[k])
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
type_dict[k] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [
item[1] for item in script_list
]
#testing
#node_pattern = NodePattern()
#r = node_pattern.loads(type_dict[k])
#if r == False:
# print "node_pattern failed to load: "+type_dict[k]
#else:
# print "successfully loaded tree: "+type_dict[k]
print "OBJECT%d: [TYPE:%s] [KEY:%s][VALUE:%s]" \
%(i, type_dict[k], k, ','.join(encoded_val))
tree.objects[i] = rs
tree.object_types_str[str(i)] = type_dict
if node.tag == "Array":
rs = analyzeArrayResultHelper(script_list, i)
rs = extractObjectValues(rs)
type_dict = {}
for k in rs:
encoded_val = [b64encode(x) for x in rs[k]]
#fw.write("array%d: %s:%s\n" % (i, k, ','.join(encoded_val)) )
node_pattern = generateNodePattern(rs[k])
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
type_dict[k] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [
item[1] for item in script_list
]
#testing
#node_pattern = NodePattern()
#r = node_pattern.loads(type_dict[k])
#if r == False:
# print "node_pattern failed to load: "+type_dict[k]
#else:
# print "successfully loaded tree: "+type_dict[k]
print "ARRAY%d: [TYPE:%s] [KEY:%s][VALUE:%s]" \
%(i, type_dict[k], k, ','.join(encoded_val))
tree.arrays[i] = rs
tree.array_types_str[str(i)] = type_dict
except Exception as e:
displayErrorMsg("fetchAndProcessScriptsOfURLsFromFile",\
"excpetion in analyzing node %d %s " %(i, str(e)))
if is_static:
static_scripts += script_length
print "Done writing %d items for file %s " % (len(
scriptdict[key]), name)
trees.append(tree)
fw.close()
#store trees
trees = sorted(trees, key=lambda x: x.get_length())
fw = open(os.path.join(dst_path, "trees"), 'w')
fw_json = open(os.path.join(dst_path, "jsons"), 'w')
for i in range(len(trees)):
tree_val = trees[i].dumps()
url = scriptdict[trees[i].key][0][1]
storeTree(url, trees[i].key, tree_val)
fw.write("1 %.3d: %s\n" % (i, tree_val))
new_tree = TemplateTree(None, None)
new_tree.loads(tree_val)
if trees[i].type == "js":
fw.write("2 %.3d: %s\n" % (i, getTreeSeq(new_tree.nodes)))
elif trees[i].type == 'json':
fw.write("2 %.3d: %s\n" % (i, json.dumps(new_tree.nodes)))
fw.close()
fw_json.close()
print "generate %d trees for %d scripts uniqe[%d]" \
%(len(trees), total_script_blocks, total_uniq_script_blocks)
print "static_scripts:%d dynamic_scripts:%d" % (static_scripts,
dynamic_scripts)
return insufficient_urls
def generateTemplateBasedOnURLsFromFile(path, dst_path):
f = open(path)
scriptdict = {}
total_script_blocks = 0
total_uniq_script_blocks = 0
debug_dict = {}
static_scripts = 0
dynamic_scripts = 0
for line in f:
url = line.strip()
print "process url "+url
hosts, inlines = fetchScripts(url)
if inlines==None or len(inlines) ==0:
print "no inlines for "+url
continue
for inline in inlines:
is_json = False
rs, sc = analyzeJSCodesFinerBlock(inline)
if rs == None:
rs = analyzeJSON(inline)
is_json = True
if rs == None:
continue
if is_json:
tree = TemplateTree(rs, None)
if not tree.key in scriptdict:
scriptdict[tree.key] = [(inline, url, tree, -1)]
debug_dict[tree.key] = [inline]
else:
debug_dict[tree.key].append(inline)
contents = [x[0] for x in scriptdict[key]]
if not inline in contents:
scriptdict[tree.key].append((inline, url, tree, -1))
total_uniq_script_blocks += 1
total_script_blocks += 1
else:
for index in range(len(rs)):
total_script_blocks += 1
seq = rs[index]
tree = TemplateTree(seq, None)
key = tree.key
if not key in scriptdict:
debug_dict[key] = [sc[index]]
scriptdict[key] = [(sc[index], url, tree, index)]
print " add key %s" %key
else:
contents = [x[0] for x in scriptdict[key]]
debug_dict[key].append(sc[index])
if not sc[index] in contents:
scriptdict[key].append((sc[index],url, tree, index))
print " item %s has %d unique scripts" %(key, len(scriptdict[key]))
total_uniq_script_blocks += 1
fw = open(os.path.join(dst_path,'debug'),'w')
for k in debug_dict:
vals = debug_dict[k]
fw.write("%d %s \n" %(len(vals),k))
fw.write(" --EXAMPLE-- %s\n" %vals[0])
fw.close()
#start to analyze trees
#scriptdict[tree_key] = [(script, url, tree, index)]
trees = []
insufficient_urls = {}
keys = sorted(scriptdict.keys(), key=lambda k:len(scriptdict[k]))
for key in keys:
is_static = True
name = "%d_%s" %(len(scriptdict[key]),key)
fw = open(os.path.join(dst_path,name), 'w')
for item in scriptdict[key]:
fw.write(item[1]+"||"+str(item[3])+" "+str(item[0])+"\n")
#make sure all template trees with the same key are the same
script_list = scriptdict[key]
length_list = sorted([len(item[2].nodes) for item in script_list])
seq_length = 0
if length_list[0] != length_list[-1]:
fw.write("[ALERT] seq length is not consistent")
fw.close()
continue
else:
seq_length = length_list[0]
#only handle JavaScript for now
tree = script_list[0][2]
if tree.type == "json":
print "the inline is json!"
fw.write("[TODO]: the inline is json. This is next step\n")
fw.close()
trees.append(tree)
continue
#process String/Object/Array nodes
#script_list: [(script, url, tree, index)]
fw.write("start analyzeing values\n")
script_length = len(script_list)
for i in range(seq_length):
node = script_list[0][2].nodes[i]
try:
if node.tag == "String":
vals = [item[2].nodes[i].value for item in script_list]
encoded_val = [b64encode(x) for x in vals]
#item = 'string%d: %s' %(i, ','.join(encoded_val))
#fw.write(item+"\n")
tree.strings[i] = vals
node_pattern = generateNodePattern(vals)
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
tree.string_types_str[str(i)] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [item[1] for item in script_list]
# testing
#node_pattern = NodePattern()
#r = node_pattern.loads(tree.string_types_str[i])
#if r == False:
# print "node_pattern failed to load: "+tree.string_types_str[i]
#else:
# print "successfully loaded tree: "+tree.string_types_str[i]
print "STRING%d: [TYPE:%s] [VALUE:%s]" \
%(i, tree.string_types_str[str(i)],','.join(encoded_val))
if node.tag == "Object":
#debug = "tag:%s val:%s" \
# %(script_list[0][2].nodes[i].tag,str(script_list[0][2].nodes[i].value))
#print "DEBUG: %s" %debug
rs = analyzeObjectResultHelper(script_list, i)
rs = extractObjectValues(rs)
type_dict = {}
for k in rs:
encoded_val = [b64encode(x) for x in rs[k]]
node_pattern = generateNodePattern(rs[k])
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
type_dict[k] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [item[1] for item in script_list]
#testing
#node_pattern = NodePattern()
#r = node_pattern.loads(type_dict[k])
#if r == False:
# print "node_pattern failed to load: "+type_dict[k]
#else:
# print "successfully loaded tree: "+type_dict[k]
print "OBJECT%d: [TYPE:%s] [KEY:%s][VALUE:%s]" \
%(i, type_dict[k], k, ','.join(encoded_val))
tree.objects[i] = rs
tree.object_types_str[str(i)] = type_dict
if node.tag == "Array":
rs = analyzeArrayResultHelper(script_list, i)
rs = extractObjectValues(rs)
type_dict = {}
for k in rs:
encoded_val = [b64encode(x) for x in rs[k]]
#fw.write("array%d: %s:%s\n" % (i, k, ','.join(encoded_val)) )
node_pattern = generateNodePattern(rs[k])
if is_static and \
((node_pattern.tp!=StringType.CONST) and (node_pattern.tp!=StringType.INSUFFICIENT)):
is_static = False
dynamic_scripts += script_length
type_dict[k] = node_pattern.dumps()
if node_pattern.is_insufficient():
if not key in insufficient_urls:
insufficient_urls[key] = \
[item[1] for item in script_list]
else:
insufficient_urls[key] += [item[1] for item in script_list]
#testing
#node_pattern = NodePattern()
#r = node_pattern.loads(type_dict[k])
#if r == False:
# print "node_pattern failed to load: "+type_dict[k]
#else:
# print "successfully loaded tree: "+type_dict[k]
print "ARRAY%d: [TYPE:%s] [KEY:%s][VALUE:%s]" \
%(i, type_dict[k], k, ','.join(encoded_val))
tree.arrays[i] = rs
tree.array_types_str[str(i)] = type_dict
except Exception as e:
displayErrorMsg("fetchAndProcessScriptsOfURLsFromFile",\
"excpetion in analyzing node %d %s " %(i, str(e)))
if is_static:
static_scripts += script_length
print "Done writing %d items for file %s " %(len(scriptdict[key]), name)
trees.append(tree)
fw.close()
#store trees
trees = sorted(trees, key=lambda x:x.get_length())
fw = open(os.path.join(dst_path,"trees"), 'w')
fw_json = open(os.path.join(dst_path,"jsons"), 'w')
for i in range(len(trees)):
tree_val = trees[i].dumps()
url = scriptdict[trees[i].key][0][1]
storeTree(url,trees[i].key, tree_val)
fw.write( "1 %.3d: %s\n" %(i, tree_val))
new_tree = TemplateTree(None, None)
new_tree.loads(tree_val)
if trees[i].type == "js":
fw.write( "2 %.3d: %s\n" %(i, getTreeSeq(new_tree.nodes)))
elif trees[i].type == 'json':
fw.write("2 %.3d: %s\n" % (i, json.dumps(new_tree.nodes)))
fw.close()
fw_json.close()
print "generate %d trees for %d scripts uniqe[%d]" \
%(len(trees), total_script_blocks, total_uniq_script_blocks)
print "static_scripts:%d dynamic_scripts:%d" %(static_scripts, dynamic_scripts)
return insufficient_urls