fix(metrics): add warning for invalid dimension values; prevent their addition to EMF blobs

aws_lambda_powertools/metrics/provider/cloudwatch_emf/cloudwatch.py

@@ -272,10 +272,18 @@ def add_dimension(self, name: str, value: str) -> None:
             raise SchemaValidationError(
                 f"Maximum number of dimensions exceeded ({MAX_DIMENSIONS}): Unable to add dimension {name}.",
             )
-        # Cast value to str according to EMF spec
-        # Majority of values are expected to be string already, so
-        # checking before casting improves performance in most cases
-        self.dimension_set[name] = value if isinstance(value, str) else str(value)
+
+        if not name or not value:
+            warnings.warn(
+                f"The dimension {name} doesn't meet the requirements and won't be added. "
+                "Ensure the dimension name and value are non empty strings",
+                stacklevel=2,
+            )
+        else:
+            # Cast value to str according to EMF spec
+            # Majority of values are expected to be string already, so
+            # checking before casting improves performance in most cases
+            self.dimension_set[name] = value if isinstance(value, str) else str(value)
 
     def add_metadata(self, key: str, value: Any) -> None:
         """Adds high cardinal metadata for metrics object

tests/functional/metrics/required_dependencies/test_metrics_cloudwatch_emf.py

@@ -1045,6 +1045,25 @@ def test_clear_default_dimensions(namespace):
     assert not my_metrics.default_dimensions
 
 
+def test_add_dimensions_with_empty_value(namespace, capsys, metric):
+    # GIVEN Metrics is initialized
+    my_metrics = Metrics(namespace=namespace)
+
+    my_dimension = "my_empty_dimension"
+
+    # WHEN we try to add a dimension with empty value
+    with pytest.warns(UserWarning, match=f"The dimension {my_dimension} doesn't meet the requirements *"):
+        my_metrics.add_dimension(name="my_empty_dimension", value="")
+
+    my_metrics.add_metric(**metric)
+    my_metrics.flush_metrics()
+
+    output = capture_metrics_output(capsys)
+
+    # THEN the serialized metric should not contain this dimension
+    assert my_dimension not in output
+
+
 def test_get_and_set_namespace_and_service_properties(namespace, service, metrics, capsys):
     # GIVEN Metrics instance is initialized without namespace and service
     my_metrics = Metrics()