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 | def run_v2_task_non_parallel(
*,
images: list[dict[str, Any]],
zarr_dir: str,
task: TaskV2,
wftask: WorkflowTaskV2,
workflow_dir_local: Path,
workflow_dir_remote: Path,
runner: BaseRunner,
get_runner_config: Callable[
[
WorkflowTaskV2,
Literal["non_parallel", "parallel"],
Optional[Path],
int,
],
Any,
],
dataset_id: int,
history_run_id: int,
task_type: Literal["non_parallel", "converter_non_parallel"],
) -> tuple[dict[int, SubmissionOutcome], int]:
"""
This runs server-side (see `executor` argument)
"""
if task_type not in ["non_parallel", "converter_non_parallel"]:
raise ValueError(
f"Invalid {task_type=} for `run_v2_task_non_parallel`."
)
# Get TaskFiles object
task_files = TaskFiles(
root_dir_local=workflow_dir_local,
root_dir_remote=workflow_dir_remote,
task_order=wftask.order,
task_name=wftask.task.name,
component="",
prefix=SUBMIT_PREFIX,
)
runner_config = get_runner_config(
wftask=wftask,
which_type="non_parallel",
)
function_kwargs = {
"zarr_dir": zarr_dir,
**(wftask.args_non_parallel or {}),
}
if task_type == "non_parallel":
function_kwargs["zarr_urls"] = [img["zarr_url"] for img in images]
# Database History operations
with next(get_sync_db()) as db:
if task_type == "non_parallel":
zarr_urls = function_kwargs["zarr_urls"]
elif task_type == "converter_non_parallel":
zarr_urls = []
history_unit = HistoryUnit(
history_run_id=history_run_id,
status=HistoryUnitStatus.SUBMITTED,
logfile=task_files.log_file_local,
zarr_urls=zarr_urls,
)
db.add(history_unit)
db.commit()
db.refresh(history_unit)
logger.debug(
"[run_v2_task_non_parallel] Created `HistoryUnit` with "
f"{history_run_id=}."
)
history_unit_id = history_unit.id
bulk_upsert_image_cache_fast(
db=db,
list_upsert_objects=[
dict(
workflowtask_id=wftask.id,
dataset_id=dataset_id,
zarr_url=zarr_url,
latest_history_unit_id=history_unit_id,
)
for zarr_url in history_unit.zarr_urls
],
)
result, exception = runner.submit(
functools.partial(
run_single_task,
command=task.command_non_parallel,
workflow_task_order=wftask.order,
workflow_task_id=wftask.task_id,
task_name=wftask.task.name,
),
parameters=function_kwargs,
task_type=task_type,
task_files=task_files,
history_unit_id=history_unit_id,
config=runner_config,
)
positional_index = 0
num_tasks = 1
outcome = {
positional_index: _process_task_output(
result=result,
exception=exception,
)
}
# NOTE: Here we don't have to handle the
# `outcome[0].exception is not None` branch, since for non_parallel
# tasks it was already handled within submit
if outcome[0].invalid_output:
with next(get_sync_db()) as db:
update_status_of_history_unit(
history_unit_id=history_unit_id,
status=HistoryUnitStatus.FAILED,
db_sync=db,
)
return outcome, num_tasks
|