-
Notifications
You must be signed in to change notification settings - Fork 0
/
summarise_results.py
217 lines (192 loc) · 8.06 KB
/
summarise_results.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
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
import argparse
import logging
import os
import re
class Result:
def __init__(self, filepath, seed):
self.filepath = filepath
self.seed = seed
self.failures = []
self.successes = []
self.num_tests_run = 0
self.length = ""
def set_num_tests_run(self, num):
self.num_tests_run = num
def set_length(self, length):
self.length = length
def add_failure(self, test_name, p_value):
self.failures.append((test_name, p_value))
def add_success(self, test_name, p_value):
self.successes.append((test_name, p_value))
def parse_testu01_report(seed, filepath):
result = Result(filepath, seed)
with open(filepath, "r") as fp:
logging.debug(filepath)
contents = fp.read()
if "All tests were passed" in contents:
return result
m = re.search(
r"Summary results.*\n(.*\n)+\s*-+\s*\n((.*\n)+)\s*-+\s*\n",
contents,
re.MULTILINE,
)
if not m:
raise Exception("No results summary")
for failure in m.group(2).split("\n"):
if failure.strip():
tokens = failure.split(" ")
result.add_failure(tokens[1].strip(), tokens[2].strip())
return result
def parse_testu01_single_report(seed, filepath):
result = Result(filepath, seed)
if not filepath:
print(f"Warning: seed {seed} output file does not exist")
return None
with open(filepath, "r") as fp:
logging.debug(filepath)
contents = fp.read()
m = re.search(r"p-value of test\s+:\s+(\S+)", contents)
if not m:
print(f"Warning: no p-value in {filepath} for seed {seed}")
return None
test_p_value = float(m.group(1))
if test_p_value > 0.999 or test_p_value < 0.001:
result.add_failure("-", test_p_value)
else:
result.add_success("-", test_p_value)
return result
def parse_practrand_report(seed, filepath):
# Default reports with only anomolous results listed.
result = Result(filepath, seed)
with open(filepath, "r") as fp:
logging.debug(filepath)
contents = fp.read()
last_result_text = contents.strip().split("\n\n")[-1]
if "no anomalies in" in last_result_text:
return result
m = re.search(
r"Test Name.*\n((.*\n)+)\s+\.\.\.and (\d+) test result\(s\) without anomalies",
last_result_text,
re.MULTILINE,
)
if not m:
raise Exception("No results summary")
result.set_num_tests_run(int(m.group(3)))
for failure in m.group(1).strip().split("\n"):
tokens = failure.strip().split(" ")
result.add_failure(tokens[0], "")
return result
def parse_practrand_report_full(seed, filepath):
# Reports with all tests listed (run with -a)
result = Result(filepath, seed)
with open(filepath, "r") as fp:
logging.debug(filepath)
contents = fp.read()
last_result_text = contents.strip().split("\n\n")[-1]
all_result_lines = last_result_text.split("\n")
# length = all_result_lines[1].split(',')[0]
# result.set_length(length)
result_lines = all_result_lines[3:]
result.set_num_tests_run(len(result_lines))
for line in result_lines:
m = re.search(r"p[\s~]=\s*(\S+)", line)
test_p_value = eval(
m.group(1)
) # Use eval since these numbers can include operators such as '1-7.8e-3'
test_name = line.strip().split(" ")[0]
if test_p_value > 0.999 or test_p_value < 0.001:
result.add_failure(test_name, test_p_value)
return result
def parse_gjrand_report(seed, filepath):
result = Result(filepath, seed)
# if not os.path.exists(filepath):
# return result
with open(filepath, "r") as fp:
logging.debug(filepath)
contents = fp.read()
result_texts = [x.strip() for x in contents.strip().split("============")]
for result_text in result_texts:
if "*****" in result_text or "========" in result_text:
continue
result_text_lines = result_text.split("\n")
test_name = result_text_lines[0]
test_p_value = float(result_text_lines[-1].split()[2])
# print('{} {}'.format(test_name, test_result))
if test_p_value > 0.999 or test_p_value < 0.001:
result.add_failure(test_name, test_p_value)
return result
def get_first_sge_file(directory):
"""
Find the first SGE stdout file.
"""
for filename in os.listdir(directory):
if re.search(r"\.o\d+", filename):
return os.path.join(directory, filename)
return None
def main():
# yapf: disable
parser = argparse.ArgumentParser(description='Summarise test results')
parser.add_argument('suite',
choices=['testu01', 'practrand', 'gjrand', 'matrixrank', 'linearcomp'],
help='The test suite run')
parser.add_argument('directory',
nargs='+',
help='The directory containing the results')
parser.add_argument('--verbose',
action='store_true',
help='display information on each seed')
# yapf: enable
args = parser.parse_args()
logging.basicConfig(level=logging.INFO)
for directory in args.directory:
results = []
if not os.path.exists(directory):
print(f"Directory {directory} does not exist")
continue
for seed_dir in os.listdir(directory):
if not seed_dir.startswith("seed_"):
continue
# if args.verbose:
# print(seed_dir)
if args.suite == "testu01" or args.suite == "practrand":
filepath = get_first_sge_file(os.path.join(directory, seed_dir))
if args.suite == "testu01":
results.append(parse_testu01_report(seed_dir[5:], filepath))
if args.suite == "practrand":
results.append(parse_practrand_report_full(seed_dir[5:], filepath))
if args.suite == "gjrand":
# Read the p?report.txt file.
filepath1 = os.path.join(directory, seed_dir, "report.txt")
filepath2 = os.path.join(directory, seed_dir, "preport.txt")
if os.path.exists(filepath1):
results.append(parse_gjrand_report(seed_dir[5:], filepath1))
if os.path.exists(filepath2):
results.append(parse_gjrand_report(seed_dir[5:], filepath2))
if args.suite == "matrixrank" or args.suite == "linearcomp":
filepath = get_first_sge_file(os.path.join(directory, seed_dir))
result = parse_testu01_single_report(seed_dir[5:], filepath)
if result:
results.append(result)
print(f"==== {directory} ====")
# Details of each seed
if args.verbose:
for result in results:
# failing_tests = ', '.join([x[0] for x in result.failures])
pvals = ", ".join(str(pval) for _, pval in result.successes)
print("{:<34} {}".format(hex(int(result.seed)), pvals))
# print('{:<160} {:<36} {:<4} {:<20} {}'.format(result.filepath,
# hex(int(result.seed)),
# len(result.failures),
# result.length,
# failing_tests)) # yapf: disable
# Totals
total_test_failures = sum([len(x.failures) for x in results])
total_seed_failures = sum([1 if len(x.failures) else 0 for x in results])
print("{:<8} results".format(len(results)))
print("{:<8} total seed failure(s)".format(total_seed_failures))
print("{:<8} total test failure(s)".format(total_test_failures))
if args.suite == "practrand" or args.suite == "gjrand":
total_tests_run = sum([x.num_tests_run for x in results])
print("{:<8} total tests run".format(total_tests_run))
if __name__ == "__main__":
main()