-
Notifications
You must be signed in to change notification settings - Fork 5
Expand file tree
/
Copy pathjsonparse.py
More file actions
188 lines (158 loc) · 8.65 KB
/
jsonparse.py
File metadata and controls
188 lines (158 loc) · 8.65 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
import json
import os
from tqdm import tqdm
from pathlib import Path
import torch
def get_adj_node2node(h, edge_index, edge_attr):
indices = edge_index.to('cuda')
values = torch.ones((len(edge_index[0]))).to('cuda')
adjacency = torch.sparse.FloatTensor(indices, values, torch.Size((len(h),len(h)))).to_dense()
node2node_features = torch.zeros(len(h)*len(h),edge_attr.size()[1]).to('cuda')
for i in range(len(edge_index[0])):
node2node_features[len(h)*edge_index[0][i]+edge_index[1][i]] = edge_attr[i]
return adjacency, node2node_features
def saveAllDataToRam(sourceCodePath,jsonVecPath):
ramData = {} #save all data to a dict. i.e. {"jsonVecID1":[lines, features, edge_index, edge_attr], "jsonVecID2":[lines, features, edge_index, edge_attr],...}
faildFileNum = 0
count=0
for root, dirs, files in os.walk(sourceCodePath):
for file in tqdm(files):
try:
sourceCodeFolderID = file.split(".")[0][-1]
CodePath = sourceCodePath + sourceCodeFolderID +'/'+ file
jsonPath = jsonVecPath + file + ".json"
#for codedata
data = json.load(open(jsonPath))
nodes = []
features = []
edgeSrc = []
edgeTag = []
edgesAttr = []
hidden = 16
max_node_token_num = 0
for node in data["jsonNodesVec"]:
#print("len(data[jsonNodesVec][node])",len(data["jsonNodesVec"][node]))
if len(data["jsonNodesVec"][node]) > max_node_token_num:
max_node_token_num = len(data["jsonNodesVec"][node])
for i in range(len(data["jsonNodesVec"])):
nodes.append(i)
node_features = []
for list in data["jsonNodesVec"][str(i)]:
if list != None:
node_features.append(list)
if len(node_features)==0:
#print("node 000000000000000000000000000", jsonPath," ",i)
node_features = [[0 for i in range(hidden)]]
if len(node_features) < max_node_token_num:
for i in range(max_node_token_num-len(node_features)):
node_features.append([0 for i in range(hidden)])
#print("node_features",len(node_features))
features.append(node_features) # multi vecs offen
for edge in data["jsonEdgesVec"]:
#print(len(data["jsonEdgesVec"][edge]))
if data["jsonEdgesVec"][edge][0][0] == 1 and data["jsonEdgesVec"][edge][0][1] == 1 and data["jsonEdgesVec"][edge][0][3] ==1:
edgeSrc.append(int(edge.split("->")[0]))
edgeTag.append(int(edge.split("->")[1]))
edgesAttr.append([0 for i in range(hidden)])
#continue
else:
edgeSrc.append(int(edge.split("->")[0]))
edgeTag.append(int(edge.split("->")[1]))
edgesAttr.append(data["jsonEdgesVec"][edge][0]) # one vec always
for i in range(len(nodes)):
edgeSrc.append(i)
edgeTag.append(i)
#edgesAttr.append([0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536])
edgesAttr.append([0 for i in range(hidden)])
edge_index = [edgeSrc, edgeTag]
features = torch.tensor(features, dtype=torch.float32).to('cuda')
edge_index = torch.tensor(edge_index, dtype=torch.long).to('cuda')
edgesAttr = torch.tensor(edgesAttr, dtype=torch.float32).to('cuda')
adjacency, node2node_features = get_adj_node2node(features, edge_index, edgesAttr)
ramData[CodePath] = [features, edge_index, edgesAttr, adjacency, node2node_features]
count+=1
except:
faildFileNum+=1
print(count, faildFileNum)
print("ramData", len(ramData))
return ramData #i.e. {"jsonVecID1":[lines, features, edge_index, edge_attr], "jsonVecID2":[lines, features, edge_index, edge_attr],...}
def saveTestDataToRam(testList,sourceCodePath,jsonVecPath):
ramData = {} #save all data to a dict. i.e. {"jsonVecID1":[lines, features, edge_index, edge_attr], "jsonVecID2":[lines, features, edge_index, edge_attr],...}
faildFileNum = 0
count=0
for root, dirs, files in os.walk(sourceCodePath):
for file in tqdm(files):
try:
sourceCodeFolderID = file.split(".")[0][-1]
CodePath = sourceCodePath + sourceCodeFolderID +'/'+ file
if CodePath not in testList:
continue
jsonPath = jsonVecPath + file + ".json"
#for codedata
data = json.load(open(jsonPath))
nodes = []
features = []
edgeSrc = []
edgeTag = []
edgesAttr = []
hidden = 16
max_node_token_num = 0
for node in data["jsonNodesVec"]:
#print("len(data[jsonNodesVec][node])",len(data["jsonNodesVec"][node]))
if len(data["jsonNodesVec"][node]) > max_node_token_num:
max_node_token_num = len(data["jsonNodesVec"][node])
for i in range(len(data["jsonNodesVec"])):
nodes.append(i)
node_features = []
for list in data["jsonNodesVec"][str(i)]:
if list != None:
node_features.append(list)
if len(node_features)==0:
#print("node 000000000000000000000000000", jsonPath," ",i)
node_features = [[0 for i in range(hidden)]]
if len(node_features) < max_node_token_num:
for i in range(max_node_token_num-len(node_features)):
node_features.append([0 for i in range(hidden)])
#print("node_features",len(node_features))
features.append(node_features) # multi vecs offen
for edge in data["jsonEdgesVec"]:
#print(len(data["jsonEdgesVec"][edge]))
if data["jsonEdgesVec"][edge][0][0] == 1 and data["jsonEdgesVec"][edge][0][1] == 1 and data["jsonEdgesVec"][edge][0][3] ==1:
edgeSrc.append(int(edge.split("->")[0]))
edgeTag.append(int(edge.split("->")[1]))
edgesAttr.append([0 for i in range(hidden)])
#continue
else:
edgeSrc.append(int(edge.split("->")[0]))
edgeTag.append(int(edge.split("->")[1]))
edgesAttr.append(data["jsonEdgesVec"][edge][0]) # one vec always
for i in range(len(nodes)):
edgeSrc.append(i)
edgeTag.append(i)
#edgesAttr.append([0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536, 0.3536])
edgesAttr.append([0 for i in range(hidden)])
edge_index = [edgeSrc, edgeTag]
features = torch.as_tensor(features, dtype=torch.float32).to('cuda')
edge_index = torch.as_tensor(edge_index, dtype=torch.long).to('cuda')
edgesAttr = torch.as_tensor(edgesAttr, dtype=torch.float32).to('cuda')
adjacency, node2node_features = get_adj_node2node(features, edge_index, edgesAttr)
ramData[CodePath] = [features, edge_index, edgesAttr, adjacency, node2node_features]
count+=1
except:
faildFileNum+=1
print(count, faildFileNum)
return ramData
def getCodePairDataList(ramData, pathlist):
datalist = []
notFindNum = 0
for line in pathlist:
try:
pairinfo = line.split()
codedata1 = ramData[pairinfo[0]]
codedata2 = ramData[pairinfo[1]]
label = int(pairinfo[2])
pairdata = [codedata1[0], codedata2[0], codedata1[1], codedata2[1], codedata1[2], codedata2[2], codedata1[3], codedata2[3], codedata1[4], codedata2[4], label]
datalist.append(pairdata)
except:
notFindNum += 1
return datalist