Reduce repetitive logs in pipeline_options. Driveby: fix soft-delete warning.

Author: tvalentyn

sdks/python/apache_beam/io/gcp/gcsio.py

@@ -642,7 +642,7 @@ def _updated_to_seconds(updated):
 
   def is_soft_delete_enabled(self, gcs_path):
     try:
-      bucket_name, _ = parse_gcs_path(gcs_path)
+      bucket_name, _ = parse_gcs_path(gcs_path, object_optional=True)
       bucket = self.get_bucket(bucket_name)
       if (bucket.soft_delete_policy is not None and
           bucket.soft_delete_policy.retention_duration_seconds > 0):

sdks/python/apache_beam/options/pipeline_options.py

@@ -37,6 +37,7 @@
 from apache_beam.options.value_provider import StaticValueProvider
 from apache_beam.options.value_provider import ValueProvider
 from apache_beam.transforms.display import HasDisplayData
+from apache_beam.utils import logs
 from apache_beam.utils import proto_utils
 
 __all__ = [
@@ -422,9 +423,10 @@ def get_all_options(
     known_args, unknown_args = parser.parse_known_args(self._flags)
     if retain_unknown_options:
       if unknown_args:
-        _LOGGER.warning(
-            'Unknown pipeline options received: %s. Ignore if flags are '
-            'used for internal purposes.' % (','.join(unknown_args)))
+        if logs.allow_log_once():
+          _LOGGER.warning(
+              'Unknown pipeline options received: %s. Ignore if flags are '
+              'used for internal purposes.' % (','.join(unknown_args)))
 
       seen = set()
 
@@ -466,7 +468,8 @@ def add_new_arg(arg, **kwargs):
       parsed_args, _ = parser.parse_known_args(self._flags)
     else:
       if unknown_args:
-        _LOGGER.warning("Discarding unparseable args: %s", unknown_args)
+        if logs.allow_log_once():
+          _LOGGER.warning("Discarding unparseable args: %s", unknown_args)
       parsed_args = known_args
     result = vars(parsed_args)
 
@@ -1064,19 +1067,21 @@ def _warn_if_soft_delete_policy_enabled(self, arg_name):
       return
 
     gcs_path = getattr(self, arg_name, None)
-    try:
-      from apache_beam.io.gcp import gcsio
-      if gcsio.GcsIO().is_soft_delete_enabled(gcs_path):
+    if logs.allow_log_once(gcs_path):
+      try:
+        from apache_beam.io.gcp import gcsio
+        if gcsio.GcsIO().is_soft_delete_enabled(gcs_path):
+          _LOGGER.warning(
+              "Bucket specified in %s has soft-delete policy enabled."
+              " To avoid being billed for unnecessary storage costs, turn"
+              " off the soft delete feature on buckets that your Dataflow"
+              " jobs use for temporary and staging storage. For more"
+              " information, see"
+              " https://cloud.google.com/storage/docs/use-soft-delete"
+              "#remove-soft-delete-policy." % arg_name)
+      except ImportError:
         _LOGGER.warning(
-            "Bucket specified in %s has soft-delete policy enabled."
-            " To avoid being billed for unnecessary storage costs, turn"
-            " off the soft delete feature on buckets that your Dataflow"
-            " jobs use for temporary and staging storage. For more"
-            " information, see"
-            " https://cloud.google.com/storage/docs/use-soft-delete"
-            "#remove-soft-delete-policy." % arg_name)
-    except ImportError:
-      _LOGGER.warning('Unable to check soft delete policy due to import error.')
+            'Unable to check soft delete policy due to import error.')
 
   # If either temp or staging location has an issue, we use the valid one for
   # both locations. If both are bad we return an error.