forked from TristanBilot/phishGNN
-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_dataset.py
169 lines (140 loc) · 4.85 KB
/
test_dataset.py
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
import pytest
import os
import sys
from typing import List, Tuple
import pandas as pd
import torch
from torch import Tensor
sys.path.append(os.path.join(os.getcwd(), 'phishGNN'))
from dataset import PhishingDataset
def dataframe_mock(rows: List[Tuple[str, List, str]]):
refs = [[{"url": ref, "nb_edges": 1} for ref in row[1]] for row in rows]
urls = [row[0] for row in rows]
features = [
# 'depth',
'is_phishing',
'redirects',
'is_https',
'is_ip_address',
'is_error_page',
'url_length',
'domain_url_depth',
'domain_url_length',
'has_sub_domain',
'has_at_symbol',
'dashes_count',
'path_starts_with_url',
'is_valid_html',
'anchors_count',
'forms_count',
'javascript_count',
'self_anchors_count',
'has_form_with_url',
'has_iframe',
'use_mouseover',
'is_cert_valid',
'has_dns_record',
'has_whois',
'cert_reliability',
'domain_age',
'domain_end_period',
]
features = {feat: [-1 for _ in range(len(rows))] for feat in features}
features["redirects"] = [row[2] for row in rows]
data = {
'url': urls,
**features,
'refs': refs,
}
df = pd.DataFrame(data=data)
df = df.set_index('url')
df.at[rows[0][0], "is_phishing"] = 1
return df
def test_dataset_easy1():
df = dataframe_mock([
("root", ["a", "b"], 2),
("a", [], 3),
("b", [], 4),
])
idx_should_be = [[0., 0.], [1., 2.]]
x_should_be = [2, 3, 4]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_easy2():
df = dataframe_mock([
("root", ["a", "b"], 2),
("a", ["c", "d"], 3),
("b", [], 4),
("c", [], 5),
("d", ["e"], 6),
("e", [], 7),
])
idx_should_be = [[0., 0., 1., 1., 4.], [1., 2., 3., 4., 5.]]
x_should_be = [2, 3, 4, 5, 6, 7]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_loop1():
df = dataframe_mock([
("root", ["a", "b"], 2),
("a", ["root"], 3),
("b", [], 4),
])
idx_should_be = [[0., 0., 1.], [1., 2., 0.]]
x_should_be = [2, 3, 4]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_loop2():
df = dataframe_mock([
("root", ["root", "b", "root", "root"], 2),
("a", ["root"], 3),
("b", ["root"], 4),
])
idx_should_be = [[0., 0., 0., 0., 1.], [0., 1., 0., 0., 0.]]
x_should_be = [2, 4]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_missing1():
df = dataframe_mock([
("root", ["a"], 2),
])
idx_should_be = [[0.], [1.]]
x_should_be = [2, -1]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_missing2():
df = dataframe_mock([
("root", ["a", "b", "c", "d"], 2),
("a", ["root"], 3),
])
idx_should_be = [[0., 0., 0., 0., 1.], [1., 2., 3., 4., 0.]]
x_should_be = [2, 3, -1, -1, -1]
dataset = PhishingDataset("root")
edge_index, x, edge_attr, y, _ = dataset._build_tensors('root', df)
assert torch.all(torch.eq(edge_index, torch.tensor(idx_should_be)))
assert [int(i[0]) for i in x] == x_should_be
assert y == 1.
def test_dataset_normalize_url():
dataset = PhishingDataset("root")
assert dataset._normalize_url('http://test.com') == "http://www.test.com"
assert dataset._normalize_url('https://test.com') == "https://www.test.com"
assert dataset._normalize_url('https://www.test.com') == "https://www.test.com"
assert dataset._normalize_url('www.test.com') == "http://www.test.com"
assert dataset._normalize_url('test.com') == "http://www.test.com"
assert dataset._normalize_url("http://www.test.com/") == "http://www.test.com"
assert dataset._normalize_url("http://www.test.com") == "http://www.test.com"
test_dataset_easy1()