Fix logging

Author: bpblanken

v03_pipeline/lib/logger.py

@@ -0,0 +1,35 @@
+import logging.config
+
+LOG_CONFIG = {
+    'version': 1,
+    'disable_existing_loggers': True,
+    'propagate': True,
+    'formatters': {
+        'default': {
+            'format': '%(asctime)s - %(module)s - %(levelname)s - %(message)s',
+            'datefmt': '%Y-%m-%d %H:%M:%S',
+        },
+    },
+    'handlers': {
+        'default': {
+            'formatter': 'default',
+            'class': 'logging.StreamHandler',
+        },
+    },
+    'loggers': {
+        '': {
+            'level': 'INFO',
+            'handlers': ['default'],
+            'propagate': True,
+        },
+    },
+}
+
+_CONFIGURED = False
+
+def get_logger(name: str):
+    global _CONFIGURED # noqa: PLW0603
+    if not _CONFIGURED:
+        logging.config.dictConfig(LOG_CONFIG)
+        _CONFIGURED = True
+    return logging.getLogger(name)

v03_pipeline/lib/reference_data/compare_globals.py

@@ -3,6 +3,7 @@
 
 import hail as hl
 
+from v03_pipeline.lib.logger import get_logger
 from v03_pipeline.lib.model import (
     DatasetType,
     ReferenceDatasetCollection,
@@ -16,7 +17,7 @@
     parse_dataset_version,
 )
 
-logger = logging.getLogger(__name__)
+logger = get_logger(__name__)
 
 
 @dataclasses.dataclass

v03_pipeline/lib/tasks/reference_data/updated_reference_dataset_collection.py

@@ -3,6 +3,7 @@
 import hail as hl
 import luigi
 
+from v03_pipeline.lib.logger import get_logger
 from v03_pipeline.lib.model import ReferenceDatasetCollection
 from v03_pipeline.lib.paths import valid_reference_dataset_collection_path
 from v03_pipeline.lib.reference_data.compare_globals import (
@@ -15,7 +16,7 @@
 from v03_pipeline.lib.tasks.base.base_update_task import BaseUpdateTask
 from v03_pipeline.lib.tasks.files import GCSorLocalTarget
 
-logger = logging.getLogger(__name__)
+logger = get_logger(__name__)
 
 
 class UpdatedReferenceDatasetCollectionTask(BaseUpdateTask):