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
|
#!/usr/bin/env python3
import os
import argparse
from collections import defaultdict
import matplotlib.pyplot as plt
import numpy as np
from matplotlib import colors
from matplotlib.ticker import PercentFormatter
def main():
# Parse args
script_dir = os.path.dirname(os.path.realpath(__file__))
default_input_path = os.path.join(script_dir, 'duration300.txt')
parser = argparse.ArgumentParser()
parser.add_argument('--input-file', default=default_input_path)
parser.add_argument('--output-dir', default='duration_statistics')
args = parser.parse_args()
os.makedirs(args.output_dir, exist_ok=True)
duration_stat = defaultdict(
lambda: {
'start': None, # Timestamp when start storing
'stored': None, # Timestamp when Onion address is responsed
'address': None, # Onion address
'retrieve': list(), # List of retrieving attemps in (timestamp, success)
'first_success_retrieve': None, # Timestamp when first retrieving occurs
'retrieve_stat': None, # List of (timestamp, success_count, fail_count, success_rate, success_rate_after_first_success_retrieve)
'live_time': 0 # Time difference b/w last and first success retrieving
}
)
# Load input file
with open(args.input_file) as file_input:
for line in file_input:
values = line[:-1].split('\t')
ts, bid, action = values[:3]
ts = float(ts)
if action == 'start':
assert duration_stat[bid]['start'] is None
duration_stat[bid]['start'] = ts
elif action == 'stored':
assert duration_stat[bid]['start'] is not None
assert duration_stat[bid]['stored'] is None
assert duration_stat[bid]['address'] is None
duration_stat[bid]['stored'] = ts
duration_stat[bid]['address'] = values[3]
elif action == 'retrieved':
assert duration_stat[bid]['start'] is not None
assert duration_stat[bid]['stored'] is not None
assert duration_stat[bid]['address'] is not None
duration_stat[bid]['retrieve'].append((ts, True))
elif action == 'failed to retrieve':
assert duration_stat[bid]['start'] is not None
assert duration_stat[bid]['stored'] is not None
assert duration_stat[bid]['address'] is not None
duration_stat[bid]['retrieve'].append((ts, False))
# Compute success rates
drop_count = 0
for bid, stat in duration_stat.items():
if stat['address'] is None:
drop_count += 1
durations = list(
(ts - stat['start'], success)
for ts, success in stat['retrieve']
)
success_count = 0
fail_count = 0
fail_count_after_first = 0
success_rates = list()
first_success_dur = None
last_success_dur = None
for dur, success in durations:
if success:
success_count += 1
last_success_dur = dur
if first_success_dur is None:
first_success_dur = dur
else:
fail_count += 1
if first_success_dur is not None:
fail_count_after_first += 1
success_rate = success_count / (success_count + fail_count)
if success_count + fail_count_after_first > 0:
success_rate_after_first = success_count / (success_count + fail_count_after_first)
else:
success_rate_after_first = None
success_rates.append((success_count, fail_count, success_rate, success_rate_after_first))
if last_success_dur is not None:
assert first_success_dur is not None
live_time = last_success_dur - first_success_dur
else:
live_time = None
retrieve_stat = list(
(ts, succ_count, fail_count, rate, rate_after_first)
for (ts, _success), (succ_count, fail_count, rate, rate_after_first) in zip(durations, success_rates)
)
stat['first_success_retrieve'] = first_success_dur
stat['retrieve_stat'] = retrieve_stat
stat['live_time'] = live_time
drop_rate = drop_count / len(duration_stat)
# Live time diagrams
n_bins = 20
live_time_stat = list(
stat['live_time'] / 60
for bid, stat in duration_stat.items()
if stat['live_time'] is not None
)
fig, axs = plt.subplots(1, 1, sharey=True, tight_layout=True)
axs.hist(live_time_stat, bins=n_bins)
plt.savefig('wtf.png')
print(drop_rate)
if __name__ == '__main__':
main()
|
