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
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
|
# -*- coding: utf8 -*-
# ============LICENSE_START=======================================================
# org.onap.vvp/validation-scripts
# ===================================================================
# Copyright © 2018 AT&T Intellectual Property. All rights reserved.
# ===================================================================
#
# Unless otherwise specified, all software contained herein is licensed
# under the Apache License, Version 2.0 (the "License");
# you may not use this software except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
#
#
# Unless otherwise specified, all documentation contained herein is licensed
# under the Creative Commons License, Attribution 4.0 Intl. (the "License");
# you may not use this documentation except in compliance with the License.
# You may obtain a copy of the License at
#
# https://creativecommons.org/licenses/by/4.0/
#
# Unless required by applicable law or agreed to in writing, documentation
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# ============LICENSE_END============================================
import collections
import csv
import datetime
import hashlib
import io
import json
import os
import sys
import time
import requests
import traceback
import docutils.core
import jinja2
import pytest
from more_itertools import partition
from six import string_types
import xlsxwriter
__path__ = [os.path.dirname(os.path.abspath(__file__))]
resolution_steps_file = "resolution_steps.json"
heat_requirements_file = "heat_requirements.json"
report_columns = [
("Input File", "file"),
("Test", "test_file"),
("Requirements", "req_description"),
("Resolution Steps", "resolution_steps"),
("Error Message", "message"),
("Raw Test Output", "raw_output"),
]
report = collections.OrderedDict(report_columns)
COLLECTION_FAILURES = []
COLLECTION_FAILURE_WARNING = """WARNING: The following unexpected errors occurred
while preparing to validate the the input files. Some validations may not have been
executed. Please refer these issue to the VNF Validation Tool team.
"""
def extract_error_msg(rep):
try:
msg = str(rep.longrepr.reprcrash)
except AttributeError:
msg = str(rep)
if "AssertionError:" in msg:
return msg.split("AssertionError:")[1]
else:
return msg
@pytest.hookimpl(tryfirst=True, hookwrapper=True)
def pytest_runtest_makereport(item, call):
outcome = yield
rep = outcome.get_result()
output_dir = "{}/../output".format(__path__[0])
if rep.outcome == "failed":
if not os.path.exists(output_dir):
os.mkdir(output_dir)
if os.path.exists("{}/failures".format(output_dir)):
with open("{}/failures".format(output_dir), "r") as o:
jdata = json.loads(o.read())
else:
jdata = {}
if hasattr(item.function, "requirement_ids"):
requirement_ids = item.function.requirement_ids
else:
requirement_ids = ""
if "environment_pair" in item.fixturenames:
resolved_pair = "{} environment pair".format(
item.funcargs["environment_pair"]["name"]
)
elif "heat_volume_pair" in item.fixturenames:
resolved_pair = "{} volume pair".format(
item.funcargs["heat_volume_pair"]["name"]
)
elif "heat_templates" in item.fixturenames:
resolved_pair = item.funcargs["heat_templates"]
elif "yaml_files" in item.fixturenames:
resolved_pair = item.funcargs["yaml_files"]
else:
resolved_pair = rep.nodeid.split("[")[1][:-1]
markers = set(m.name for m in item.iter_markers())
base_test = "base" in markers
msg = extract_error_msg(rep)
if base_test:
msg = "!!Base Test Failure!! Halting test suite execution...\n{}".format(
msg
)
jdata[len(jdata)] = {
"file": resolved_pair,
"vnfrqts": requirement_ids,
"test": item.function.__name__,
"test_file": item.function.__module__.split(".")[-1],
"raw_output": str(rep.longrepr),
"message": msg,
}
with open("{}/failures".format(output_dir), "w") as f:
json.dump(jdata, f, indent=4)
if not item.config.option.continue_on_failure and base_test:
pytest.exit(
"{}\n{}\n{}".format(msg, resolved_pair, item.function.__name__)
)
def make_timestamp():
timezone = time.tzname[time.localtime().tm_isdst]
return "{} {}".format(str(datetime.datetime.now()), timezone)
def pytest_sessionfinish(session, exitstatus):
if not session.config.option.template_dir:
return
template_path = os.path.abspath(session.config.option.template_dir[0])
profile_name = session.config.option.validation_profile_name
generate_report(
"{}/../output".format(__path__[0]),
template_path,
profile_name,
session.config.option.report_format,
)
def pytest_collection_modifyitems(session, config, items):
allowed_marks = ["xfail", "base"]
profile = config.option.validation_profile
for item in items:
markers = set(m.name for m in item.iter_markers())
if not profile and markers and set(markers).isdisjoint(allowed_marks):
item.add_marker(
pytest.mark.skip(
reason="No validation profile selected. Skipping tests with marks."
)
)
if (
profile
and markers
and profile not in markers
and set(markers).isdisjoint(allowed_marks)
):
item.add_marker(
pytest.mark.skip(reason="Doesn't match selection validation profile")
)
items.sort(
key=lambda item: 0 if "base" in set(m.name for m in item.iter_markers()) else 1
)
def make_href(path):
paths = [path] if isinstance(path, string_types) else path
links = []
for p in paths:
abs_path = os.path.abspath(p)
name = abs_path if os.path.isdir(abs_path) else os.path.split(abs_path)[1]
links.append(
"<a href='file://{abs_path}' target='_blank'>{name}</a>".format(
abs_path=abs_path, name=name
)
)
return "<br/>".join(links)
def generate_report(outpath, template_path, profile_name, output_format):
failures = "{}/failures".format(outpath)
faildata = None
rdata = None
hdata = None
if os.path.exists(failures):
with open(failures, "r") as f:
faildata = json.loads(f.read())
else:
faildata = {}
resolution_steps = "{}/../{}".format(__path__[0], resolution_steps_file)
if os.path.exists(resolution_steps):
with open(resolution_steps, "r") as f:
rdata = json.loads(f.read())
heat_requirements = "{}/../{}".format(__path__[0], heat_requirements_file)
if os.path.exists(heat_requirements):
with open(heat_requirements, "r") as f:
hdata = json.loads(f.read())
# point requirements at the most recent version
current_version = hdata["current_version"]
hdata = hdata["versions"][current_version]["needs"]
# mapping requirement IDs from failures to requirement descriptions
for k, v in faildata.items():
req_text = ""
if v["vnfrqts"] != "":
for req in v["vnfrqts"]:
if req in hdata:
req_text += "\n\n{}: \n{}".format(req, hdata[req]["description"])
faildata[k]["req_description"] = req_text
# mapping resolution steps to module and test name
for k, v in faildata.items():
# resolution_step = ""
faildata[k]["resolution_steps"] = ""
for rs in rdata:
if v["test_file"] == rs["module"] and v["test"] == rs["function"]:
faildata[k]["resolution_steps"] = "\n{}: \n{}".format(
rs["header"], rs["resolution_steps"]
)
output_format = output_format.lower().strip() if output_format else "html"
if output_format == "html":
generate_html_report(outpath, profile_name, template_path, faildata)
elif output_format == "excel":
generate_excel_report(outpath, profile_name, template_path, faildata)
elif output_format == "csv":
generate_csv_report(outpath, profile_name, template_path, faildata)
else:
raise ValueError("Unsupported output format: " + output_format)
def generate_csv_report(output_dir, profile_name, template_path, faildata):
rows = [["Validation Failures"]]
headers = [
("Profile Selected:", profile_name),
("Report Generated At:", make_timestamp()),
("Directory Validated:", template_path),
("Checksum:", hash_directory(template_path)),
("Total Errors:", len(faildata) + len(COLLECTION_FAILURES)),
]
rows.append([])
for header in headers:
rows.append(header)
rows.append([])
if COLLECTION_FAILURES:
rows.append([COLLECTION_FAILURE_WARNING])
rows.append(["Validation File", "Test", "Fixtures", "Error"])
for failure in COLLECTION_FAILURES:
rows.append(
[
failure["module"],
failure["test"],
";".join(failure["fixtures"]),
failure["error"],
]
)
rows.append([])
# table header
rows.append([col for col, _ in report_columns])
# table content
for data in faildata.values():
rows.append(
[
data.get("file", ""),
data.get("test_file", ""),
data.get("req_description", ""),
data.get("resolution_steps", ""),
data.get("message", ""),
data.get("raw_output", ""),
]
)
output_path = os.path.join(output_dir, "report.csv")
with open(output_path, "w", newline="") as f:
writer = csv.writer(f)
for row in rows:
writer.writerow(row)
def generate_excel_report(output_dir, profile_name, template_path, faildata):
output_path = os.path.join(output_dir, "report.xlsx")
workbook = xlsxwriter.Workbook(output_path)
bold = workbook.add_format({"bold": True})
code = workbook.add_format(({"font_name": "Courier", "text_wrap": True}))
normal = workbook.add_format({"text_wrap": True})
heading = workbook.add_format({"bold": True, "font_size": 18})
worksheet = workbook.add_worksheet("failures")
worksheet.write(0, 0, "Validation Failures", heading)
headers = [
("Profile Selected:", profile_name),
("Report Generated At:", make_timestamp()),
("Directory Validated:", template_path),
("Checksum:", hash_directory(template_path)),
("Total Errors:", len(faildata) + len(COLLECTION_FAILURES)),
]
for row, (header, value) in enumerate(headers, start=2):
worksheet.write(row, 0, header, bold)
worksheet.write(row, 1, value)
worksheet.set_column(0, len(headers) - 1, 40)
worksheet.set_column(len(headers), len(headers), 80)
if COLLECTION_FAILURES:
collection_failures_start = 2 + len(headers) + 2
worksheet.write(collection_failures_start, 0, COLLECTION_FAILURE_WARNING, bold)
collection_failure_headers = ["Validation File", "Test", "Fixtures", "Error"]
for col_num, col_name in enumerate(collection_failure_headers):
worksheet.write(collection_failures_start + 1, col_num, col_name, bold)
for row, data in enumerate(COLLECTION_FAILURES, collection_failures_start + 2):
worksheet.write(row, 0, data["module"])
worksheet.write(row, 1, data["test"])
worksheet.write(row, 2, ",".join(data["fixtures"]))
worksheet.write(row, 3, data["error"], code)
# table header
start_error_table_row = 2 + len(headers) + len(COLLECTION_FAILURES) + 4
worksheet.write(start_error_table_row, 0, "Validation Failures", bold)
for col_num, (col_name, _) in enumerate(report_columns):
worksheet.write(start_error_table_row + 1, col_num, col_name, bold)
# table content
for row, data in enumerate(faildata.values(), start=start_error_table_row + 2):
for col, key in enumerate(report.values()):
if key == "file":
paths = (
[data[key]] if isinstance(data[key], string_types) else data[key]
)
contents = "\n".join(paths)
worksheet.write(row, col, contents, normal)
elif key == "raw_output":
worksheet.write_string(row, col, data[key], code)
else:
worksheet.write(row, col, data[key], normal)
workbook.close()
def generate_html_report(outpath, profile_name, template_path, faildata):
failures = []
for data in faildata.values():
failures.append(
{
"file_links": make_href(data["file"]),
"test_id": data["test_file"],
"error_message": data["message"],
"raw_output": data["raw_output"],
"requirements": docutils.core.publish_parts(
writer_name="html", source=data["req_description"]
)["body"],
"resolution_steps": data["resolution_steps"],
}
)
pkg_dir = os.path.split(__file__)[0]
j2_template_path = os.path.join(pkg_dir, "report.html.jinja2")
with open(j2_template_path, "r") as f:
report_template = jinja2.Template(f.read())
contents = report_template.render(
num_failures=len(failures) + len(COLLECTION_FAILURES),
profile_name=profile_name,
template_dir=make_href(template_path),
checksum=hash_directory(template_path),
timestamp=make_timestamp(),
failures=failures,
collection_failures=COLLECTION_FAILURES,
)
with open(os.path.join(outpath, "report.html"), "w") as f:
f.write(contents)
def pytest_addoption(parser):
"""
Add needed CLI arguments
"""
parser.addoption(
"--template-directory",
dest="template_dir",
action="append",
help="Directory which holds the templates for validation",
)
parser.addoption(
"--self-test",
dest="self_test",
action="store_true",
help="Test the unit tests against their fixtured data",
)
parser.addoption(
"--validation-profile",
dest="validation_profile",
action="store",
help="Runs all unmarked tests plus test with a matching marker",
)
parser.addoption(
"--validation-profile-name",
dest="validation_profile_name",
action="store",
help="Friendly name of the validation profile used in reports",
)
parser.addoption(
"--report-format",
dest="report_format",
action="store",
help="Format of output report (html, csv, excel)",
)
parser.addoption(
"--continue-on-failure",
dest="continue_on_failure",
action="store_true",
help="Continue validation even when structural errors exist in input files",
)
def pytest_configure(config):
"""
Ensure that we are receive either `--self-test` or
`--template-dir=<directory` as CLI arguments
"""
if config.getoption("template_dir") and config.getoption("self_test"):
raise Exception('"--template-dir", and "--self-test"' " are mutually exclusive")
if not (
config.getoption("template_dir")
or config.getoption("self_test")
or config.getoption("help")
):
raise Exception('One of "--template-dir" or' ' "--self-test" must be specified')
def pytest_generate_tests(metafunc):
"""
If a unit test requires an argument named 'filename'
we generate a test for the filenames selected. Either
the files contained in `template_dir` or if `template_dir`
is not specified on the CLI, the fixtures associated with this
test name.
"""
# noinspection PyBroadException
try:
if "filename" in metafunc.fixturenames:
from .parametrizers import parametrize_filename
parametrize_filename(metafunc)
if "filenames" in metafunc.fixturenames:
from .parametrizers import parametrize_filenames
parametrize_filenames(metafunc)
if "template_dir" in metafunc.fixturenames:
from .parametrizers import parametrize_template_dir
parametrize_template_dir(metafunc)
if "environment_pair" in metafunc.fixturenames:
from .parametrizers import parametrize_environment_pair
parametrize_environment_pair(metafunc)
if "heat_volume_pair" in metafunc.fixturenames:
from .parametrizers import parametrize_heat_volume_pair
parametrize_heat_volume_pair(metafunc)
if "yaml_files" in metafunc.fixturenames:
from .parametrizers import parametrize_yaml_files
parametrize_yaml_files(metafunc)
if "env_files" in metafunc.fixturenames:
from .parametrizers import parametrize_environment_files
parametrize_environment_files(metafunc)
if "yaml_file" in metafunc.fixturenames:
from .parametrizers import parametrize_yaml_file
parametrize_yaml_file(metafunc)
if "env_file" in metafunc.fixturenames:
from .parametrizers import parametrize_environment_file
parametrize_environment_file(metafunc)
if "parsed_yaml_file" in metafunc.fixturenames:
from .parametrizers import parametrize_parsed_yaml_file
parametrize_parsed_yaml_file(metafunc)
if "parsed_environment_file" in metafunc.fixturenames:
from .parametrizers import parametrize_parsed_environment_file
parametrize_parsed_environment_file(metafunc)
if "heat_template" in metafunc.fixturenames:
from .parametrizers import parametrize_heat_template
parametrize_heat_template(metafunc)
if "heat_templates" in metafunc.fixturenames:
from .parametrizers import parametrize_heat_templates
parametrize_heat_templates(metafunc)
if "volume_template" in metafunc.fixturenames:
from .parametrizers import parametrize_volume_template
parametrize_volume_template(metafunc)
if "volume_templates" in metafunc.fixturenames:
from .parametrizers import parametrize_volume_templates
parametrize_volume_templates(metafunc)
if "template" in metafunc.fixturenames:
from .parametrizers import parametrize_template
parametrize_template(metafunc)
if "templates" in metafunc.fixturenames:
from .parametrizers import parametrize_templates
parametrize_templates(metafunc)
except Exception as e:
# If an error occurs in the collection phase, then it won't be logged as a
# normal test failure. This means that failures could occur, but not
# be seen on the report resulting in a false positive success message. These
# errors will be stored and reported separately on the report
COLLECTION_FAILURES.append(
{
"module": metafunc.module.__name__,
"test": metafunc.function.__name__,
"fixtures": metafunc.fixturenames,
"error": traceback.format_exc(),
}
)
raise e
def hash_directory(path):
md5 = hashlib.md5()
for dir_path, sub_dirs, filenames in os.walk(path):
for filename in filenames:
file_path = os.path.join(dir_path, filename)
with open(file_path, "rb") as f:
md5.update(f.read())
return md5.hexdigest()
def load_current_requirements():
"""Loads dict of current requirements or empty dict if file doesn't exist"""
url = 'https://onap.readthedocs.io/en/latest/_downloads/needs.json'
r = requests.get(url)
with open('requirements.json', 'wb') as needs:
needs.write(r.content)
path = "heat_requirements.json"
if not os.path.exists(path):
return {}
with io.open(path, encoding="utf8", mode="r") as f:
data = json.load(f)
version = data["current_version"]
return data["versions"][version]["needs"]
def compat_open(path):
"""Invokes open correctly depending on the Python version"""
if sys.version_info.major < 3:
return open(path, "wb")
else:
return open(path, "w", newline="")
def unicode_writerow(writer, row):
if sys.version_info.major < 3:
row = [s.encode("utf8") for s in row]
writer.writerow(row)
def pytest_report_collectionfinish(config, startdir, items):
"""Generates a simple traceability report to output/traceability.csv"""
traceability_path = os.path.join(__path__[0], "../output/traceability.csv")
output_dir = os.path.split(traceability_path)[0]
if not os.path.exists(output_dir):
os.makedirs(output_dir)
requirements = load_current_requirements()
unmapped, mapped = partition(
lambda item: hasattr(item.function, "requirement_ids"), items
)
req_to_test = collections.defaultdict(set)
mapping_errors = set()
for item in mapped:
for req_id in item.function.requirement_ids:
if req_id not in req_to_test:
req_to_test[req_id].add(item)
if req_id not in requirements:
mapping_errors.add(
(req_id, item.function.__module__, item.function.__name__)
)
mapping_error_path = os.path.join(__path__[0], "../output/mapping_errors.csv")
with compat_open(mapping_error_path) as f:
writer = csv.writer(f)
for error in mapping_errors:
unicode_writerow(writer, error)
with compat_open(traceability_path) as f:
out = csv.writer(f)
unicode_writerow(
out,
("Requirement ID", "Requirement", "Section", "Test Module", "Test Name"),
)
for req_id, metadata in requirements.items():
if req_to_test[req_id]:
for item in req_to_test[req_id]:
unicode_writerow(
out,
(
req_id,
metadata["description"],
metadata["section_name"],
item.function.__module__,
item.function.__name__,
),
)
else:
unicode_writerow(
out,
(req_id, metadata["description"], metadata["section_name"], "", ""),
)
# now write out any test methods that weren't mapped to requirements
for item in unmapped:
unicode_writerow(
out, ("", "", "", item.function.__module__, item.function.__name__)
)
|