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
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
|
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
import functools
from collections import deque
from .base import BaseFormatter
from .process import strstatus
def output_subtests(func):
@functools.wraps(func)
def inner(self, data):
if self.subtests_count:
return self._format_subtests(data.get("component")) + func(self, data)
else:
return func(self, data)
return inner
class TbplFormatter(BaseFormatter):
"""Formatter that formats logs in the legacy formatting format used by TBPL
This is intended to be used to preserve backward compatibility with existing tools
hand-parsing this format.
"""
def __init__(self, compact=False):
self.suite_start_time = None
self.test_start_times = {}
self.buffer = None
self.compact = compact
self.subtests_count = 0
@property
def compact(self):
return self._compact
@compact.setter
def compact(self, value):
self._compact = value
if value:
self.buffer = deque([], 10)
else:
self.buffer = None
def __call__(self, data):
return getattr(self, data["action"])(data)
def _format_subtests(self, component, subtract_context=False):
count = self.subtests_count
if subtract_context:
count -= len(self.buffer)
self.subtests_count = 0
return self._log({"level": "INFO",
"message": "." * count,
"component": component})
@output_subtests
def log(self, data):
return self._log(data)
def _log(self, data):
if data.get('component'):
message = "%s %s" % (data["component"], data["message"])
else:
message = data["message"]
if "stack" in data:
message += "\n%s" % data["stack"]
return "%s\n" % message
@output_subtests
def process_output(self, data):
return "PROCESS | %(process)s | %(data)s\n" % data
@output_subtests
def process_start(self, data):
msg = "TEST-INFO | started process %s" % data['process']
if 'command' in data:
msg = '%s (%s)' % (msg, data['command'])
return msg + '\n'
@output_subtests
def process_exit(self, data):
return "TEST-INFO | %s: %s\n" % (data['process'],
strstatus(data['exitcode']))
@output_subtests
def crash(self, data):
id = data["test"] if "test" in data else "pid: %s" % data["process"]
signature = data["signature"] if data["signature"] else "unknown top frame"
rv = ["PROCESS-CRASH | %s | application crashed [%s]" % (id, signature)]
if data.get("minidump_path"):
rv.append("Crash dump filename: %s" % data["minidump_path"])
if data.get("stackwalk_stderr"):
rv.append("stderr from minidump_stackwalk:")
rv.append(data["stackwalk_stderr"])
elif data.get("stackwalk_stdout"):
rv.append(data["stackwalk_stdout"])
if data.get("stackwalk_returncode", 0) != 0:
rv.append("minidump_stackwalk exited with return code %d" %
data["stackwalk_returncode"])
if data.get("stackwalk_errors"):
rv.extend(data.get("stackwalk_errors"))
rv = "\n".join(rv)
if not rv[-1] == "\n":
rv += "\n"
return rv
def suite_start(self, data):
self.suite_start_time = data["time"]
return "SUITE-START | Running %i tests\n" % len(data["tests"])
def test_start(self, data):
self.test_start_times[self.test_id(data["test"])] = data["time"]
return "TEST-START | %s\n" % data["test"]
def test_status(self, data):
if self.compact:
if "expected" in data:
rv = []
rv.append(self._format_subtests(data.get("component"), subtract_context=True))
rv.extend(self._format_status(item) for item in self.buffer)
rv.append(self._format_status(data))
self.buffer.clear()
return "".join(rv)
else:
self.subtests_count += 1
self.buffer.append(data)
else:
return self._format_status(data)
def _format_status(self, data):
message = "- " + data["message"] if "message" in data else ""
if "stack" in data:
message += "\n%s" % data["stack"]
if message and message[-1] == "\n":
message = message[:-1]
if "expected" in data:
if not message:
message = "- expected %s" % data["expected"]
failure_line = "TEST-UNEXPECTED-%s | %s | %s %s\n" % (
data["status"], data["test"], data["subtest"],
message)
if data["expected"] != "PASS":
info_line = "TEST-INFO | expected %s\n" % data["expected"]
return failure_line + info_line
return failure_line
return "TEST-%s | %s | %s %s\n" % (
data["status"], data["test"], data["subtest"],
message)
def test_end(self, data):
rv = []
if self.compact and self.subtests_count:
print_context = "expected" in data
rv.append(self._format_subtests(data.get("component"),
subtract_context=print_context))
if print_context:
rv.extend(self._format_status(item) for item in self.buffer)
self.buffer.clear()
test_id = self.test_id(data["test"])
duration_msg = ""
if test_id in self.test_start_times:
start_time = self.test_start_times.pop(test_id)
time = data["time"] - start_time
duration_msg = "took %ims" % time
if "expected" in data:
message = data.get("message", "")
if not message:
message = "expected %s" % data["expected"]
if "stack" in data:
message += "\n%s" % data["stack"]
if message and message[-1] == "\n":
message = message[:-1]
extra = data.get("extra", {})
if "reftest_screenshots" in extra:
screenshots = extra["reftest_screenshots"]
if len(screenshots) == 3:
message += ("\nREFTEST IMAGE 1 (TEST): data:image/png;base64,%s\n"
"REFTEST IMAGE 2 (REFERENCE): data:image/png;base64,%s") % (
screenshots[0]["screenshot"],
screenshots[2]["screenshot"])
elif len(screenshots) == 1:
message += "\nREFTEST IMAGE: data:image/png;base64,%(image1)s" \
% screenshots[0]["screenshot"]
failure_line = "TEST-UNEXPECTED-%s | %s | %s\n" % (
data["status"], test_id, message)
if data["expected"] not in ("PASS", "OK"):
expected_msg = "expected %s | " % data["expected"]
else:
expected_msg = ""
info_line = "TEST-INFO %s%s\n" % (expected_msg, duration_msg)
return failure_line + info_line
sections = ["TEST-%s" % data['status'], test_id]
if duration_msg:
sections.append(duration_msg)
rv.append(' | '.join(sections) + '\n')
return "".join(rv)
def suite_end(self, data):
start_time = self.suite_start_time
time = int((data["time"] - start_time) / 1000)
return "SUITE-END | took %is\n" % time
def test_id(self, test_id):
if isinstance(test_id, (str, unicode)):
return test_id
else:
return tuple(test_id)
@output_subtests
def valgrind_error(self, data):
rv = "TEST-UNEXPECTED-VALGRIND-ERROR | " + data['primary'] + "\n"
for line in data['secondary']:
rv = rv + line + "\n"
return rv
def lint(self, data):
fmt = "TEST-UNEXPECTED-{level} | {path}:{lineno}{column} | {message} ({rule})"
data["column"] = ":%s" % data["column"] if data["column"] else ""
data['rule'] = data['rule'] or data['linter'] or ""
return fmt.append(fmt.format(**data))
|