SIENTIAPDE-1316

Update .gitignore and refactor metrics.py, activities.py, and gates.py for improved clarity and consistency. Added coverage.xml and cache directories to .gitignore. Standardized string formatting and parameter handling in metrics and activities classes, enhancing code readability. Removed the deprecated faker.py file and adjusted related tests accordingly.
This commit is contained in:
vitor-aignosi
2025-10-16 13:31:12 -03:00
parent 8bdbf049b8
commit 97eb5bc904
27 changed files with 1101 additions and 1336 deletions

View File

@@ -1,14 +1,16 @@
from temporalio import workflow
with workflow.unsafe.imports_passed_through():
from scouter.activities.activities import Activities
from typing import Any
from datetime import timedelta
from sientia_do.temporal.policies import retry_policy
from typing import Any
from sientia_do.temporal.constants import DATETIME_FORMAT_WITH_TZ
from sientia_do.temporal.policies import retry_policy
from scouter.activities.activities import Activities
@workflow.defn(name="core_scouter")
@workflow.defn(name='core_scouter')
class CoreScouter:
"""
Core data processing workflow that handles data quality, aggregation, and export.
@@ -68,21 +70,17 @@ class CoreScouter:
**metadata,
'filters': input_data['filters'],
'data': input_data['data'],
'model_tags': input_data['model_tags']
'model_tags': input_data['model_tags'],
},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)
grouped_data = await workflow.execute_local_activity_method(
Activities.aggregate_data,
{
**metadata,
'data': filtered_data,
'model_tags': input_data['model_tags']
},
{**metadata, 'data': filtered_data, 'model_tags': input_data['model_tags']},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)
held_data = await workflow.execute_local_activity_method(
@@ -94,10 +92,10 @@ class CoreScouter:
'data': grouped_data,
'model_id': input_data['model_id'],
'model_tags': input_data['model_tags'],
'retention_time': input_data['retention_time']
'retention_time': input_data['retention_time'],
},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)
if held_data == {}:
@@ -110,13 +108,10 @@ class CoreScouter:
'schema': input_data['schema'],
'table_name': input_data['table_name'],
'data': held_data,
'timestamp_conversion': {
'column': 'timestamp',
'format': DATETIME_FORMAT_WITH_TZ
}
'timestamp_conversion': {'column': 'timestamp', 'format': DATETIME_FORMAT_WITH_TZ},
},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)
await workflow.execute_activity_method(
@@ -125,7 +120,7 @@ class CoreScouter:
**metadata,
},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)
if input_data.get('debug_data_package', False):
@@ -139,5 +134,5 @@ class CoreScouter:
'schedule_name': input_data['schedule_name'],
},
retry_policy=retry_policy,
start_to_close_timeout=timedelta(seconds=60)
start_to_close_timeout=timedelta(seconds=60),
)