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 | def collect_local(
*,
task_group_activity_id: int,
task_group_id: int,
) -> None:
"""
Collect a task package.
This function is run as a background task, therefore exceptions must be
handled.
NOTE: by making this function sync, it runs within a thread - due to
starlette/fastapi handling of background tasks (see
https://github.com/encode/starlette/blob/master/starlette/background.py).
Arguments:
task_group_id:
task_group_activity_id:
"""
with TemporaryDirectory() as tmpdir:
log_file_path = get_log_path(Path(tmpdir))
logger = set_logger(
logger_name=LOGGER_NAME,
log_file_path=log_file_path,
)
with next(get_sync_db()) as db:
# Get main objects from db
activity = db.get(TaskGroupActivityV2, task_group_activity_id)
task_group = db.get(TaskGroupV2, task_group_id)
if activity is None or task_group is None:
# Use `logging` directly
logging.error(
"Cannot find database rows with "
f"{task_group_id=} and {task_group_activity_id=}:\n"
f"{task_group=}\n{activity=}. Exit."
)
return
# Log some info
logger.debug("START")
for key, value in task_group.model_dump().items():
logger.debug(f"task_group.{key}: {value}")
# Check that the (local) task_group path does exist
if Path(task_group.path).exists():
error_msg = f"{task_group.path} already exists."
logger.error(error_msg)
fail_and_cleanup(
task_group=task_group,
task_group_activity=activity,
logger_name=LOGGER_NAME,
log_file_path=log_file_path,
exception=FileExistsError(error_msg),
db=db,
)
return
try:
# Create task_group.path folder
Path(task_group.path).mkdir(parents=True)
logger.debug(f"Created {task_group.path}")
# Copy wheel file into task group path
if task_group.wheel_path:
new_wheel_path = _copy_wheel_file_local(
task_group=task_group
)
task_group.wheel_path = new_wheel_path
task_group = add_commit_refresh(obj=task_group, db=db)
# Prepare replacements for templates
replacements = get_collection_replacements(
task_group=task_group,
python_bin=get_python_interpreter_v2(
python_version=task_group.python_version
),
)
# Prepare common arguments for `_customize_and_run_template``
common_args = dict(
replacements=replacements,
script_dir=(
Path(task_group.path) / SCRIPTS_SUBFOLDER
).as_posix(),
prefix=(
f"{int(time.time())}_"
f"{TaskGroupActivityActionV2.COLLECT}_"
),
logger_name=LOGGER_NAME,
)
# Set status to ONGOING and refresh logs
activity.status = TaskGroupActivityStatusV2.ONGOING
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
# Run script 1
stdout = _customize_and_run_template(
template_filename="1_create_venv.sh",
**common_args,
)
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
# Run script 2
stdout = _customize_and_run_template(
template_filename="2_pip_install.sh",
**common_args,
)
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
# Run script 3
pip_freeze_stdout = _customize_and_run_template(
template_filename="3_pip_freeze.sh",
**common_args,
)
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
# Run script 4
stdout = _customize_and_run_template(
template_filename="4_pip_show.sh",
**common_args,
)
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
# Run script 5
venv_info = _customize_and_run_template(
template_filename="5_get_venv_size_and_file_number.sh",
**common_args,
)
venv_size, venv_file_number = venv_info.split()
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
pkg_attrs = parse_script_pip_show_stdout(stdout)
for key, value in pkg_attrs.items():
logger.debug(f"Parsed from pip-show: {key}={value}")
# Check package_name match between pip show and task-group
task_group = db.get(TaskGroupV2, task_group_id)
package_name_pip_show = pkg_attrs.get("package_name")
package_name_task_group = task_group.pkg_name
compare_package_names(
pkg_name_pip_show=package_name_pip_show,
pkg_name_task_group=package_name_task_group,
logger_name=LOGGER_NAME,
)
# Extract/drop parsed attributes
package_name = package_name_task_group
python_bin = pkg_attrs.pop("python_bin")
package_root_parent = pkg_attrs.pop("package_root_parent")
# TODO : Use more robust logic to determine `package_root`.
# Examples: use `importlib.util.find_spec`, or parse the
# output of `pip show --files {package_name}`.
package_name_underscore = package_name.replace("-", "_")
package_root = (
Path(package_root_parent) / package_name_underscore
).as_posix()
# Read and validate manifest file
manifest_path = pkg_attrs.pop("manifest_path")
logger.info(f"now loading {manifest_path=}")
with open(manifest_path) as json_data:
pkg_manifest_dict = json.load(json_data)
logger.info(f"loaded {manifest_path=}")
logger.info("now validating manifest content")
pkg_manifest = ManifestV2(**pkg_manifest_dict)
logger.info("validated manifest content")
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
logger.info("_prepare_tasks_metadata - start")
task_list = _prepare_tasks_metadata(
package_manifest=pkg_manifest,
package_version=task_group.version,
package_root=Path(package_root),
python_bin=Path(python_bin),
)
check_task_files_exist(task_list=task_list)
logger.info("_prepare_tasks_metadata - end")
activity.log = get_current_log(log_file_path)
activity = add_commit_refresh(obj=activity, db=db)
logger.info("create_db_tasks_and_update_task_group - " "start")
create_db_tasks_and_update_task_group(
task_list=task_list,
task_group_id=task_group.id,
db=db,
)
logger.info("create_db_tasks_and_update_task_group - end")
# Update task_group data
logger.info(
"Add pip_freeze, venv_size and venv_file_number "
"to TaskGroupV2 - start"
)
task_group.pip_freeze = pip_freeze_stdout
task_group.venv_size_in_kB = int(venv_size)
task_group.venv_file_number = int(venv_file_number)
task_group = add_commit_refresh(obj=task_group, db=db)
logger.info(
"Add pip_freeze, venv_size and venv_file_number "
"to TaskGroupV2 - end"
)
# Finalize (write metadata to DB)
logger.debug("finalising - START")
activity.status = TaskGroupActivityStatusV2.OK
activity.timestamp_ended = get_timestamp()
activity = add_commit_refresh(obj=activity, db=db)
logger.debug("finalising - END")
logger.debug("END")
except Exception as collection_e:
# Delete corrupted package dir
try:
logger.info(f"Now delete folder {task_group.path}")
shutil.rmtree(task_group.path)
logger.info(f"Deleted folder {task_group.path}")
except Exception as rm_e:
logger.error(
"Removing folder failed.\n"
f"Original error:\n{str(rm_e)}"
)
fail_and_cleanup(
task_group=task_group,
task_group_activity=activity,
logger_name=LOGGER_NAME,
log_file_path=log_file_path,
exception=collection_e,
db=db,
)
return
|