fix: cast dimension value to str to avoid issue where EMF silently fails (#52)

* fix: cast dimension value to str to avoid issue where EMF silently fails to collect the metrics

* fix: correct type hint in fixture

* fix: remove surplus assertions and logic from new test

* chore: fix formatting

Author: cakepietoast

aws_lambda_powertools/metrics/base.py

@@ -203,7 +203,14 @@ def add_dimension(self, name: str, value: str):
             Dimension value
         """
         logger.debug(f"Adding dimension: {name}:{value}")
-        self.dimension_set[name] = value
+
+        # 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
+        if isinstance(value, str):
+            self.dimension_set[name] = value
+        else:
+            self.dimension_set[name] = str(value)
 
     def __extract_metric_unit_value(self, unit: Union[str, MetricUnit]) -> str:
         """Return metric value from metric unit whether that's str or MetricUnit enum

tests/functional/test_metrics.py

@@ -1,5 +1,5 @@
 import json
-from typing import Dict, List
+from typing import Any, Dict, List
 
 import pytest
 
@@ -48,6 +48,14 @@ def dimensions() -> List[Dict[str, str]]:
     ]
 
 
+@pytest.fixture
+def non_str_dimensions() -> List[Dict[str, Any]]:
+    return [
+        {"name": "test_dimension", "value": True},
+        {"name": "test_dimension_2", "value": 3},
+    ]
+
+
 @pytest.fixture
 def namespace() -> Dict[str, str]:
     return {"name": "test_namespace"}
@@ -380,3 +388,27 @@ def lambda_handler(evt, context):
 
     # THEN metric set should be empty after function has been run
     assert my_metrics.metric_set == {}
+
+
+def test_log_metrics_non_string_dimension_values(capsys, metrics, non_str_dimensions, namespace):
+    # GIVEN Metrics is initialized and dimensions with non-string values are added
+    my_metrics = Metrics()
+    my_metrics.add_namespace(**namespace)
+    for metric in metrics:
+        my_metrics.add_metric(**metric)
+    for dimension in non_str_dimensions:
+        my_metrics.add_dimension(**dimension)
+
+    # WHEN we utilize log_metrics to serialize
+    # and flush all metrics at the end of a function execution
+    @my_metrics.log_metrics
+    def lambda_handler(evt, ctx):
+        return True
+
+    lambda_handler({}, {})
+    output = json.loads(capsys.readouterr().out.strip())
+
+    # THEN we should have no exceptions
+    # and dimension values hould be serialized as strings
+    for dimension in non_str_dimensions:
+        assert isinstance(output[dimension["name"]], str)