def seq_process(code_path, preprocess):
nodes = Node_extract(code_path, preprocess)
dd = to_dict(nodes)
num_list = dd[1]
node_list = dd[2]
seq = Sequence_generate(node_list, num_list)
return seq
def find_module(line_number):
if os.path.exists(dir_path + '/jsons/AST.json') and os.path.exists(
dir_path + '/jsons/file_path.json'):
Input_id = line_number
Input_line = int(Input_id[0:8])
Input_file = int(Input_id[8:16])
max = 0
lines = 0
output_list = []
count = 0
with open(dir_path + '/jsons/file_path.json') as f1:
file_path_list = json.load(f1)
f1.close()
if Input_file >= len(file_path_list):
raise SystemExit("The file number is too large")
file_name = file_path_list[Input_file]
node_list = Node_extract(file_name, True)[2]
node_list_new = to_init_dict(node_list, Input_file)[1]
print("###############The Result################")
print(str(file_name) + ":" + str(Input_line))
for i in range(1, len(node_list_new)):
line_range = node_list_new[i]['coord']
if line_range != 'null':
if line_range[0] != 'null' and line_range[1] != 'null':
first_line = line_range[0]
last_line = line_range[1]
line_number1 = int(first_line[0:8])
line_number2 = int(last_line[0:8])
if line_number2 > max:
max = line_number2
if line_number1 != line_number2:
if line_number1 <= Input_line and Input_line <= line_number2:
print("This module's node type is: %s " %
(node_list_new[i]['_nodetype']))
print("This module's coordinate is: %s " %
(node_list_new[i]['coord']))
print("This module's name is: %s " %
(node_list_new[i]['node_name']))
count = count + 1
if lines <= line_number2 - line_number1:
lines = line_number2 - line_number1
output_list = node_list_new[i]['coord']
if count == 0:
print("No module satisfied")
if Input_line > max:
raise SystemExit(
"The line number is too large \n#########################################"
)
print("#########################################")
return output_list
def node_graph(code_path, preprocess, same_list):
"""Show the syntax tree and display the same part of tow trees"""
node_list = Node_extract(code_path, preprocess)
num_list = []
name = re.findall(RE_C, code_path)
for x in range(1, len(node_list)):
num = node_list[x]['_nodetype'].find('-')
num_list.append(num)
cut = re.findall(RE_AZ, node_list[x]['_nodetype'])
if len(cut) > 0:
node_list[x]['_nodetype'] = cut[0]
else:
node_list[x]['_nodetype'] = ''
num_list.insert(0, 0)
if same_list == None:
num_cp = num_list
dot = Digraph(comment=name, format="pdf")
dot.node('0', 'TranslationUnitDecl')
for y in range(1, max(num_cp) + 1, 2):
if y == 1:
for x in range(len(node_list)):
if num_list[x] == y:
dot.node(str(x), node_list[x]['_nodetype'])
dot.edge('0', str(x))
if y > 1:
for x in range(len(num_cp)):
if num_list[x] == y:
dot.node(str(x), node_list[x]['_nodetype'])
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
else:
num_cp = num_list
same_cp = same_list
dot = Digraph(comment=name, format="pdf")
if 0 in same_cp:
dot.node('0', 'TranslationUnitDecl', _attributes={"color": "red"})
else:
dot.node('0', 'TranslationUnitDecl')
for y in range(1, max(num_cp) + 1, 2):
if y == 1:
for x in range(len(node_list)):
if num_list[x] == y:
if x in same_cp:
dot.node(str(x),
node_list[x]['_nodetype'],
_attributes={"color": "red"})
dot.edge('0', str(x))
else:
dot.node(str(x), node_list[x]['_nodetype'])
dot.edge('0', str(x))
if y > 1:
for x in range(len(num_cp)):
if num_list[x] == y:
if x in same_cp:
dot.node(str(x),
node_list[x]['_nodetype'],
_attributes={"color": "red"})
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
else:
dot.node(str(x), node_list[x]['_nodetype'])
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
save_name = './output/' + str(name[0])
dot.view(save_name)
line_number2 = int(last_line[0:8])
if line_number2 > max:
max = line_number2
if line_number1 != line_number2:
if line_number1 <= Input_line and Input_line <= line_number2:
print("This module's node type is: %s " %
(node_list_new[i]['_nodetype']))
print("This module's coordinate is: %s " %
(node_list_new[i]['coord']))
print("This module's name is: %s " %
(node_list_new[i]['node_name']))
count = count + 1
if lines <= line_number2 - line_number1:
lines = line_number2 - line_number1
output_list = node_list_new[i]['coord']
if count == 0:
print("No module satisfied")
if Input_line > max:
raise SystemExit(
"The line number is too large \n#########################################"
)
print("#########################################")
return output_list
if __name__ == "__main__":
code_path = '/home/jinzhenghui/ovs/vswitchd/bridge.c'
node_list = Node_extract(code_path, True)
node_list_new = to_dict(node_list[0], 0)[0]
print(node_list_new)
code_path_list = []
names = []
json_files_dir = dir_path + '/jsons'
if not os.path.exists(json_files_dir):
os.mkdir(json_files_dir)
# json_file1 is a nested dict
json_file1 = json_files_dir + '/AST.json'
# json file2 is the trace of each node
json_file2 = json_files_dir + '/trace.json'
# json file3 restore the name of each module
json_file3 = json_files_dir + '/Module_names.json'
# json file4 restore the unnested dict
json_file4 = json_files_dir + '/uAST.json'
if '.c' in code_path or '.cpp' in code_path or '.h' in code_path:
code_path_list.append(code_path)
node_list, name_all, node_list1 = Node_extract(
code_path, sel)
node_list.insert(0, code_path)
names.append(name_all)
to_json(node_list, json_file1, json_file2, False)
uAST = to_init_dict(node_list1, 0)[1]
print('The total amount of the nodes is {}'.format(
len(node_list1)))
with open(json_file3, 'w+') as f:
json.dump(names, f, ensure_ascii=False, indent=4)
f.close()
with open(json_file4, 'w+') as f1:
json.dump(uAST, f1, ensure_ascii=False, indent=4)
f1.close()
print("The json file path: " + json_files_dir)
else:
json_list = []
raise SystemExit("Usage: python main.py --view c_file_dir True/False ")
elif args[0] == '--tojson':
if len(args[1:]) == 2:
if args[2] != 'True' and args[2] != 'False':
raise SystemExit("Please use 'True' or 'False' to choose whether you need to remove the headers")
else:
if args[2] == 'True':
sel = True
else:
sel = False
if os.path.exists(args[1]):
code_path = args[1]
code_path_list = []
if '.c' in code_path or '.cpp' in code_path:
code_path_list.append(code_path)
node_list = Node_extract(code_path, sel)
print ('The total amount of the nodes is {}'.format(len(node_list)))
json_files_dir = dir_path + '/jsons'
if not os.path.exists(json_files_dir):
os.mkdir(json_files_dir)
json_file1 = json_files_dir + '/AST.json'
json_file2 = json_files_dir + '/trace.json'
to_json(node_list, json_file1, json_file2, False)
print ("The json file path: "+json_files_dir)
else:
json_list = []
node_len = 0
g = os.walk(code_path)
for path, di, filelist in g:
for filename in filelist:
k = os.path.join(path, filename)
if len(args[1:]) == 1:
if os.path.exists(args[1]):
code_path_a = args[1]
sel = True
seq1 = AST_Compare.seq_process(code_path_a, sel)
node_graph(code_path_a, sel, None)
else:
raise SystemExit("Error: Could not find c/c++ file")
else:
raise SystemExit("Usage: python main.py --view c_file_dir")
elif args[0] == '--tojson':
if len(args[1:]) == 1:
if os.path.exists(args[1]):
code_path_a = args[1]
sel = True
node_list = Node_extract(code_path_a, sel)
json_files_dir = dir_path + '/jsons'
if not os.path.exists(json_files_dir):
os.mkdir(json_files_dir)
json_file = json_files_dir + '/output.json'
to_json(node_list, json_file)
print("The json file path: " + json_files_dir)
else:
raise SystemExit("Error: Could not find c/c++ file")
else:
raise SystemExit("Usage: python main.py --tojson c_file_dir")
else:
raise SystemExit(
"Usage: python main.py --compare c_file_dir1 c_file_dir2 \n python main.py --view c_file_dir \n python main.py --tojson c_file_dir "
)
else:
def node_graph(code_path, preprocess, same_list):
"""Show the syntax tree and display the same part of tow trees"""
if '.c' not in code_path or '.cpp' not in code_path:
code_path_list = []
g = os.walk(code_path)
for path, di, filelist in g:
for filename in filelist:
k = os.path.join(path, filename)
if '.c' in k or '.cpp' in k:
code_path_list.append(k)
dot = Digraph("G", format="pdf")
dot.node('Start', 'Start')
for i in range(len(code_path_list)):
node_list = Node_extract(code_path_list[i], preprocess)
num_list = []
name = re.findall(RE_C, code_path_list[i])
for x in range(1, len(node_list)):
num = node_list[x]['_nodetype'].find('-')
num_list.append(num)
cut = re.findall(RE_AZ, node_list[x]['_nodetype'])
if len(cut) > 0:
node_list[x]['_nodetype'] = cut[0]
else:
node_list[x]['_nodetype'] = ''
num_list.insert(0, 0)
if same_list == None:
num_cp = num_list
sub = Digraph(comment=name)
sub.node(str(i) + 'a0', str(name))
for y in range(1, max(num_cp) + 1, 2):
if y == 1:
for x in range(len(node_list)):
if num_list[x] == y:
sub.node(
str(i) + 'b' + str(x),
node_list[x]['_nodetype'])
sub.edge(str(i) + 'a0', str(i) + 'b' + str(x))
if y > 1:
for x in range(len(num_cp)):
if num_list[x] == y:
sub.node(
str(i) + 'b' + str(x),
node_list[x]['_nodetype'])
for z in range(x, 0, -1):
if num_list[z] == y - 2:
sub.edge(
str(i) + 'b' + str(z),
str(i) + 'b' + str(x))
break
dot.subgraph(sub)
dot.edge('Start', str(i) + 'a0')
save_name = './output/combine'
dot.view(save_name)
if os.path.getsize(save_name + '.pdf') > 200000:
raise SystemExit("Error: The graph is too large for evince")
else:
node_list = Node_extract(code_path, preprocess)
num_list = []
name = re.findall(RE_C, code_path)
for x in range(1, len(node_list)):
num = node_list[x]['_nodetype'].find('-')
num_list.append(num)
cut = re.findall(RE_AZ, node_list[x]['_nodetype'])
if len(cut) > 0:
node_list[x]['_nodetype'] = cut[0]
else:
node_list[x]['_nodetype'] = ''
num_list.insert(0, 0)
if same_list == None:
num_cp = num_list
dot = Digraph(comment=name, format="pdf")
dot.node('0', 'TranslationUnitDecl')
for y in range(1, max(num_cp) + 1, 2):
if y == 1:
for x in range(len(node_list)):
if num_list[x] == y:
dot.node(str(x), node_list[x]['_nodetype'])
dot.edge('0', str(x))
if y > 1:
for x in range(len(num_cp)):
if num_list[x] == y:
dot.node(str(x), node_list[x]['_nodetype'])
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
else:
num_cp = num_list
same_cp = same_list
dot = Digraph(comment=name, format="pdf")
if 0 in same_cp:
dot.node('0',
'TranslationUnitDecl',
_attributes={"color": "red"})
else:
dot.node('0', 'TranslationUnitDecl')
for y in range(1, max(num_cp) + 1, 2):
if y == 1:
for x in range(len(node_list)):
if num_list[x] == y:
if x in same_cp:
dot.node(str(x),
node_list[x]['_nodetype'],
_attributes={"color": "red"})
dot.edge('0', str(x))
else:
dot.node(str(x), node_list[x]['_nodetype'])
dot.edge('0', str(x))
if y > 1:
for x in range(len(num_cp)):
if num_list[x] == y:
if x in same_cp:
dot.node(str(x),
node_list[x]['_nodetype'],
_attributes={"color": "red"})
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
else:
dot.node(str(x), node_list[x]['_nodetype'])
for z in range(x, 0, -1):
if num_list[z] == y - 2:
dot.edge(str(z), str(x))
break
save_name = './output/' + str(name[0])
dot.view(save_name)
if os.path.getsize(save_name + '.pdf') > 200000:
raise SystemExit("Error: The graph is too large for evince")
