- Notifications
You must be signed in to change notification settings - Fork 28.8k
/
Copy pathnotification_service_doc_tests.py
385 lines (305 loc) · 13.6 KB
/
notification_service_doc_tests.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
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
# Copyright 2022 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file 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.
importjson
importos
importre
importtime
fromtypingimportDict, List
fromget_ci_error_statisticsimportget_jobs
fromslack_sdkimportWebClient
client=WebClient(token=os.environ["CI_SLACK_BOT_TOKEN"])
defhandle_test_results(test_results):
expressions=test_results.split(" ")
failed=0
success=0
# When the output is short enough, the output is surrounded by = signs: "== OUTPUT =="
# When it is too long, those signs are not present.
time_spent=expressions[-2] if"="inexpressions[-1] elseexpressions[-1]
fori, expressioninenumerate(expressions):
if"failed"inexpression:
failed+=int(expressions[i-1])
if"passed"inexpression:
success+=int(expressions[i-1])
returnfailed, success, time_spent
defextract_first_line_failure(failures_short_lines):
failures= {}
file=None
in_error=False
forlineinfailures_short_lines.split("\n"):
ifre.search(r"_ \[doctest\]", line):
in_error=True
file=line.split(" ")[2]
elifin_errorandnotline.split(" ")[0].isdigit():
failures[file] =line
in_error=False
returnfailures
classMessage:
def__init__(self, title: str, doc_test_results: Dict):
self.title=title
self.n_success=sum(job_result["n_success"] forjob_resultindoc_test_results.values())
self.n_failures=sum(job_result["n_failures"] forjob_resultindoc_test_results.values())
self.n_tests=self.n_success+self.n_failures
# Failures and success of the modeling tests
self.doc_test_results=doc_test_results
@property
deftime(self) ->str:
all_results= [*self.doc_test_results.values()]
time_spent= [r["time_spent"].split(", ")[0] forrinall_resultsiflen(r["time_spent"])]
total_secs=0
fortimeintime_spent:
time_parts=time.split(":")
# Time can be formatted as xx:xx:xx, as .xx, or as x.xx if the time spent was less than a minute.
iflen(time_parts) ==1:
time_parts= [0, 0, time_parts[0]]
hours, minutes, seconds=int(time_parts[0]), int(time_parts[1]), float(time_parts[2])
total_secs+=hours*3600+minutes*60+seconds
hours, minutes, seconds=total_secs//3600, (total_secs%3600) //60, total_secs%60
returnf"{int(hours)}h{int(minutes)}m{int(seconds)}s"
@property
defheader(self) ->Dict:
return {"type": "header", "text": {"type": "plain_text", "text": self.title}}
@property
defno_failures(self) ->Dict:
return {
"type": "section",
"text": {
"type": "plain_text",
"text": f"🌞 There were no failures: all {self.n_tests} tests passed. The suite ran in {self.time}.",
"emoji": True,
},
"accessory": {
"type": "button",
"text": {"type": "plain_text", "text": "Check Action results", "emoji": True},
"url": f"https://github.com/huggingface/transformers/actions/runs/{os.environ['GITHUB_RUN_ID']}",
},
}
@property
deffailures(self) ->Dict:
return {
"type": "section",
"text": {
"type": "plain_text",
"text": (
f"There were {self.n_failures} failures, out of {self.n_tests} tests.\nThe suite ran in"
f" {self.time}."
),
"emoji": True,
},
"accessory": {
"type": "button",
"text": {"type": "plain_text", "text": "Check Action results", "emoji": True},
"url": f"https://github.com/huggingface/transformers/actions/runs/{os.environ['GITHUB_RUN_ID']}",
},
}
@property
defcategory_failures(self) ->List[Dict]:
failure_blocks= []
MAX_ERROR_TEXT=3000-len("The following examples had failures:\n\n\n\n") -len("[Truncated]\n")
line_length=40
category_failures= {k: v["failed"] fork, vindoc_test_results.items() ifisinstance(v, dict)}
defsingle_category_failures(category, failures):
text=""
iflen(failures) ==0:
return""
text+=f"*{category} failures*:".ljust(line_length//2).rjust(line_length//2) +"\n"
foridx, failureinenumerate(failures):
new_text=text+f"`{failure}`\n"
iflen(new_text) >MAX_ERROR_TEXT:
text=text+"[Truncated]\n"
break
text=new_text
returntext
forcategory, failuresincategory_failures.items():
report=single_category_failures(category, failures)
iflen(report) ==0:
continue
block= {
"type": "section",
"text": {
"type": "mrkdwn",
"text": f"The following examples had failures:\n\n\n{report}\n",
},
}
failure_blocks.append(block)
returnfailure_blocks
@property
defpayload(self) ->str:
blocks= [self.header]
ifself.n_failures>0:
blocks.append(self.failures)
ifself.n_failures>0:
blocks.extend(self.category_failures)
ifself.n_failures==0:
blocks.append(self.no_failures)
returnjson.dumps(blocks)
@staticmethod
deferror_out():
payload= [
{
"type": "section",
"text": {
"type": "plain_text",
"text": "There was an issue running the tests.",
},
"accessory": {
"type": "button",
"text": {"type": "plain_text", "text": "Check Action results", "emoji": True},
"url": f"https://github.com/huggingface/transformers/actions/runs/{os.environ['GITHUB_RUN_ID']}",
},
}
]
print("Sending the following payload")
print(json.dumps({"blocks": json.loads(payload)}))
client.chat_postMessage(
channel=SLACK_REPORT_CHANNEL_ID,
text="There was an issue running the tests.",
blocks=payload,
)
defpost(self):
print("Sending the following payload")
print(json.dumps({"blocks": json.loads(self.payload)}))
text=f"{self.n_failures} failures out of {self.n_tests} tests,"ifself.n_failureselse"All tests passed."
self.thread_ts=client.chat_postMessage(
channel=SLACK_REPORT_CHANNEL_ID,
blocks=self.payload,
text=text,
)
defget_reply_blocks(self, job_name, job_link, failures, text):
# `text` must be less than 3001 characters in Slack SDK
# keep some room for adding "[Truncated]" when necessary
MAX_ERROR_TEXT=3000-len("[Truncated]")
failure_text=""
forkey, valueinfailures.items():
new_text=failure_text+f"*{key}*\n_{value}_\n\n"
iflen(new_text) >MAX_ERROR_TEXT:
# `failure_text` here has length <= 3000
failure_text=failure_text+"[Truncated]"
break
# `failure_text` here has length <= MAX_ERROR_TEXT
failure_text=new_text
title=job_name
content= {"type": "section", "text": {"type": "mrkdwn", "text": text}}
ifjob_linkisnotNone:
content["accessory"] = {
"type": "button",
"text": {"type": "plain_text", "text": "GitHub Action job", "emoji": True},
"url": job_link,
}
return [
{"type": "header", "text": {"type": "plain_text", "text": title, "emoji": True}},
content,
{"type": "section", "text": {"type": "mrkdwn", "text": failure_text}},
]
defpost_reply(self):
ifself.thread_tsisNone:
raiseValueError("Can only post reply if a post has been made.")
sorted_dict=sorted(self.doc_test_results.items(), key=lambdat: t[0])
forjob_name, job_resultinsorted_dict:
iflen(job_result["failures"]) >0:
text=f"*Num failures* :{len(job_result['failed'])}\n"
failures=job_result["failures"]
blocks=self.get_reply_blocks(job_name, job_result["job_link"], failures, text=text)
print("Sending the following reply")
print(json.dumps({"blocks": blocks}))
client.chat_postMessage(
channel=SLACK_REPORT_CHANNEL_ID,
text=f"Results for {job_name}",
blocks=blocks,
thread_ts=self.thread_ts["ts"],
)
time.sleep(1)
defretrieve_artifact(name: str):
_artifact= {}
ifos.path.exists(name):
files=os.listdir(name)
forfileinfiles:
try:
withopen(os.path.join(name, file), encoding="utf-8") asf:
_artifact[file.split(".")[0]] =f.read()
exceptUnicodeDecodeErrorase:
raiseValueError(f"Could not open {os.path.join(name, file)}.") frome
return_artifact
defretrieve_available_artifacts():
classArtifact:
def__init__(self, name: str):
self.name=name
self.paths= []
def__str__(self):
returnself.name
defadd_path(self, path: str):
self.paths.append({"name": self.name, "path": path})
_available_artifacts: Dict[str, Artifact] = {}
directories=filter(os.path.isdir, os.listdir())
fordirectoryindirectories:
artifact_name=directory
ifartifact_namenotin_available_artifacts:
_available_artifacts[artifact_name] =Artifact(artifact_name)
_available_artifacts[artifact_name].add_path(directory)
return_available_artifacts
if__name__=="__main__":
SLACK_REPORT_CHANNEL_ID=os.environ["SLACK_REPORT_CHANNEL"]
github_actions_jobs=get_jobs(
workflow_run_id=os.environ["GITHUB_RUN_ID"], token=os.environ["ACCESS_REPO_INFO_TOKEN"]
)
artifact_name_to_job_map= {}
forjobingithub_actions_jobs:
forstepinjob["steps"]:
ifstep["name"].startswith("Test suite reports artifacts: "):
artifact_name=step["name"][len("Test suite reports artifacts: ") :]
artifact_name_to_job_map[artifact_name] =job
break
available_artifacts=retrieve_available_artifacts()
doc_test_results= {}
# `artifact_key` is the artifact path
forartifact_key, artifact_objinavailable_artifacts.items():
artifact_path=artifact_obj.paths[0]
ifnotartifact_path["path"].startswith("doc_tests_gpu_test_reports_"):
continue
# change "_" back to "/" (to show the job name as path)
job_name=artifact_path["path"].replace("doc_tests_gpu_test_reports_", "").replace("_", "/")
# This dict (for each job) will contain all the information relative to each doc test job, in particular:
# - failed: list of failed tests
# - failures: dict in the format 'test': 'error_message'
job_result= {}
doc_test_results[job_name] =job_result
job=artifact_name_to_job_map[artifact_path["path"]]
job_result["job_link"] =job["html_url"]
job_result["category"] ="Python Examples"ifjob_name.startswith("src/") else"MD Examples"
artifact=retrieve_artifact(artifact_path["path"])
if"stats"inartifact:
failed, success, time_spent=handle_test_results(artifact["stats"])
job_result["n_failures"] =failed
job_result["n_success"] =success
job_result["time_spent"] =time_spent[1:-1] +", "
job_result["failed"] = []
job_result["failures"] = {}
all_failures=extract_first_line_failure(artifact["failures_short"])
forlineinartifact["summary_short"].split("\n"):
ifre.search("FAILED", line):
line=line.replace("FAILED ", "")
line=line.split()[0].replace("\n", "")
if"::"inline:
file_path, test=line.split("::")
else:
file_path, test=line, line
job_result["failed"].append(test)
failure=all_failures[test] iftestinall_failureselse"N/A"
job_result["failures"][test] =failure
# Save and to be uploaded as artifact
os.makedirs("doc_test_results", exist_ok=True)
withopen("doc_test_results/doc_test_results.json", "w", encoding="UTF-8") asfp:
json.dump(doc_test_results, fp, ensure_ascii=False, indent=4)
message=Message("🤗 Results of the doc tests.", doc_test_results)
message.post()
message.post_reply()