fix(event_source): fix regression when working with zero numbers in DynamoDBStreamEvent (#4932)

Fix regression when working with ZERO

Author: leandrodamascena

aws_lambda_powertools/utilities/data_classes/dynamo_db_stream_event.py

@@ -73,7 +73,12 @@ def _deserialize_bool(self, value: bool) -> bool:
         return value
 
     def _deserialize_n(self, value: str) -> Decimal:
+        # value is None or "."? It's zero
+        # then return early
         value = value.lstrip("0")
+        if not value or value == ".":
+            return DYNAMODB_CONTEXT.create_decimal(0)
+
         if len(value) > 38:
             # See: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.NamingRulesDataTypes.html#HowItWorks.DataTypes.Number
             # Calculate the number of trailing zeros after the 38th character

tests/unit/data_classes/required_dependencies/test_dynamo_db_stream_event.py

@@ -75,6 +75,22 @@ def test_dynamodb_stream_record_deserialization_large_int_without_trailing_zeros
     }
 
 
+def test_dynamodb_stream_record_deserialization_zero_value():
+
+    data = {
+        "Keys": {"key1": {"attr1": "value1"}},
+        "NewImage": {
+            "Name": {"S": "Joe"},
+            "Age": {"N": "0"},
+        },
+    }
+    record = StreamRecord(data)
+    assert record.new_image == {
+        "Name": "Joe",
+        "Age": DECIMAL_CONTEXT.create_decimal("0"),
+    }
+
+
 def test_dynamodb_stream_record_deserialization():
     byte_list = [s.encode("utf-8") for s in ["item1", "item2"]]
     decimal_context = Context(