Merge branch 'develop' into dependabot/pip/develop/cdklabs-generative-ai-cdk-constructs-0.1.320

Author: leandrodamascena

aws_lambda_powertools/metrics/metrics.py

@@ -174,7 +174,6 @@ def log_metrics(
         )
 
     def set_default_dimensions(self, **dimensions) -> None:
-        self.provider.set_default_dimensions(**dimensions)
         """Persist dimensions across Lambda invocations
 
         Parameters
@@ -195,9 +194,7 @@ def set_default_dimensions(self, **dimensions) -> None:
             def lambda_handler():
                 return True
         """
-        for name, value in dimensions.items():
-            self.add_dimension(name, value)
-
+        self.provider.set_default_dimensions(**dimensions)
         self.default_dimensions.update(**dimensions)
 
     def clear_default_dimensions(self) -> None:

aws_lambda_powertools/metrics/provider/cloudwatch_emf/cloudwatch.py

@@ -87,7 +87,7 @@ def __init__(
     ):
         self.metric_set = metric_set if metric_set is not None else {}
         self.dimension_set = dimension_set if dimension_set is not None else {}
-        self.default_dimensions = default_dimensions or {}
+        self.default_dimensions = default_dimensions if default_dimensions is not None else {}
         self.namespace = resolve_env_var_choice(choice=namespace, env=os.getenv(constants.METRICS_NAMESPACE_ENV))
         self.service = resolve_env_var_choice(choice=service, env=os.getenv(constants.SERVICE_NAME_ENV))
         self.function_name = function_name
@@ -317,7 +317,7 @@ def add_dimension(self, name: str, value: str) -> None:
             )
             return
 
-        if name in self.dimension_set or name in self.default_dimensions:
+        if name in self.dimension_set and self.dimension_set[name] != value:
             warnings.warn(
                 f"Dimension '{name}' has already been added. The previous value will be overwritten.",
                 category=PowertoolsUserWarning,

tests/functional/metrics/required_dependencies/test_metrics_cloudwatch_emf.py

@@ -1133,6 +1133,105 @@ def test_clear_default_dimensions(namespace):
     assert not my_metrics.default_dimensions
 
 
+def test_set_default_dimensions_no_warning_on_first_call(namespace):
+    # GIVEN a Metrics instance with no dimensions set
+    my_metrics = Metrics(namespace=namespace)
+
+    # WHEN we persist default dimensions for the first time
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter("default")
+        my_metrics.set_default_dimensions(environment="test", log_group="/lambda/test")
+
+    # THEN no overwrite warning should be emitted
+    assert not [warning for warning in w if "has already been added" in str(warning.message)]
+
+
+def test_set_default_dimensions_no_warning_when_unchanged(namespace):
+    # GIVEN a Metrics instance with default dimensions persisted
+    my_metrics = Metrics(namespace=namespace)
+    my_metrics.set_default_dimensions(environment="test", log_group="/lambda/test")
+
+    # WHEN we persist the same default dimensions again e.g., on a warm invocation
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter("default")
+        my_metrics.set_default_dimensions(environment="test", log_group="/lambda/test")
+
+    # THEN no overwrite warning should be emitted
+    assert not [warning for warning in w if "has already been added" in str(warning.message)]
+
+
+def test_set_default_dimensions_warns_when_value_changes(namespace):
+    # GIVEN a Metrics instance with a default dimension persisted
+    my_metrics = Metrics(namespace=namespace)
+    my_metrics.set_default_dimensions(environment="test")
+
+    # WHEN we persist the same default dimension with a different value
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter("default")
+        my_metrics.set_default_dimensions(environment="prod")
+
+    # THEN a single overwrite warning should be emitted
+    assert len([warning for warning in w if "has already been added" in str(warning.message)]) == 1
+
+
+def test_log_metrics_with_default_dimensions_no_warning_across_invocations(namespace, metric, capsys):
+    # GIVEN a Metrics instance with default dimensions persisted
+    my_metrics = Metrics(namespace=namespace)
+    my_metrics.set_default_dimensions(environment="test", log_group="/lambda/test")
+
+    @my_metrics.log_metrics
+    def lambda_handler(evt, ctx):
+        my_metrics.add_metric(**metric)
+
+    # WHEN metrics are flushed across multiple invocations
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter("default")
+        lambda_handler({}, {})
+        lambda_handler({}, {})
+
+    # THEN no overwrite warning should be emitted
+    assert not [warning for warning in w if "has already been added" in str(warning.message)]
+
+
+def test_provider_keeps_provided_default_dimensions_dict(namespace):
+    # GIVEN a provider constructed with an empty default dimensions dict e.g., the one Metrics shares
+    shared_default_dimensions: dict = {}
+    my_provider = AmazonCloudWatchEMFProvider(namespace=namespace, default_dimensions=shared_default_dimensions)
+
+    # WHEN default dimensions are set through the provider
+    my_provider.set_default_dimensions(environment="test")
+
+    # THEN the provided dict remains in use and receives the update
+    assert my_provider.default_dimensions is shared_default_dimensions
+    assert shared_default_dimensions == {"environment": "test"}
+
+
+def test_metrics_shares_default_dimensions_with_provider(namespace):
+    # GIVEN a Metrics instance with the default provider
+    my_metrics = Metrics(namespace=namespace)
+
+    # WHEN default dimensions are set
+    my_metrics.set_default_dimensions(environment="test")
+
+    # THEN Metrics and the provider hold the same dict, both with the update
+    assert my_metrics.default_dimensions is my_metrics.provider.default_dimensions
+    assert my_metrics.default_dimensions == {"environment": "test"}
+
+
+def test_add_dimension_no_warning_when_value_unchanged(namespace):
+    # GIVEN a Metrics instance with a dimension added
+    my_metrics = Metrics(namespace=namespace)
+    my_metrics.add_dimension("environment", "test")
+
+    # WHEN the same dimension is added again with the same value
+    with warnings.catch_warnings(record=True) as w:
+        warnings.simplefilter("default")
+        my_metrics.add_dimension("environment", "test")
+
+    # THEN no overwrite warning should be emitted
+    assert not [warning for warning in w if "has already been added" in str(warning.message)]
+
+
 def test_add_dimensions_with_empty_value(namespace, capsys, metric):
     # GIVEN Metrics is initialized
     my_metrics = Metrics(namespace=namespace)