feat(validation): returns output from validate function (#4839)

* Return the output of fastjsonschema.validate()

* Adding documentation

---------

Co-authored-by: Leandro Damascena <[email protected]>
Co-authored-by: Simon Thulbourn <[email protected]>

Author: 3 people

aws_lambda_powertools/utilities/validation/base.py

@@ -14,7 +14,7 @@ def validate_data_against_schema(
     formats: Optional[Dict] = None,
     handlers: Optional[Dict] = None,
     provider_options: Optional[Dict] = None,
-):
+) -> Union[Dict, str]:
     """Validate dict data against given JSON Schema
 
     Parameters
@@ -31,6 +31,12 @@ def validate_data_against_schema(
         Arguments that will be passed directly to the underlying validation call, in this case fastjsonchema.validate.
         For all supported arguments see: https://horejsek.github.io/python-fastjsonschema/#fastjsonschema.validate
 
+    Returns
+    -------
+    Dict
+        The validated event. If the schema specifies a `default` value for fields that are omitted,
+        those default values will be included in the response.
+
     Raises
     ------
     SchemaValidationError
@@ -42,7 +48,13 @@ def validate_data_against_schema(
         formats = formats or {}
         handlers = handlers or {}
         provider_options = provider_options or {}
-        fastjsonschema.validate(definition=schema, data=data, formats=formats, handlers=handlers, **provider_options)
+        return fastjsonschema.validate(
+            definition=schema,
+            data=data,
+            formats=formats,
+            handlers=handlers,
+            **provider_options,
+        )
     except (TypeError, AttributeError, fastjsonschema.JsonSchemaDefinitionException) as e:
         raise InvalidSchemaFormatError(f"Schema received: {schema}, Formats: {formats}. Error: {e}")
     except fastjsonschema.JsonSchemaValueException as e:

aws_lambda_powertools/utilities/validation/validator.py

@@ -172,7 +172,7 @@ def validate(
     provider_options: Optional[Dict] = None,
     envelope: Optional[str] = None,
     jmespath_options: Optional[Dict] = None,
-):
+) -> Any:
     """Standalone function to validate event data using a JSON Schema
 
      Typically used when you need more control over the validation process.
@@ -245,6 +245,12 @@ def handler(event, context):
             validate(event=event, schema=json_schema_dict, envelope="awslogs.powertools_base64_gzip(data) | powertools_json(@).logEvents[*]")
             return event
 
+    Returns
+    -------
+    Dict
+        The validated event. If the schema specifies a `default` value for fields that are omitted,
+        those default values will be included in the response.
+
     Raises
     ------
     SchemaValidationError
@@ -261,7 +267,7 @@ def handler(event, context):
             jmespath_options=jmespath_options,
         )
 
-    validate_data_against_schema(
+    return validate_data_against_schema(
         data=event,
         schema=schema,
         formats=formats,

docs/utilities/validation.md

@@ -67,6 +67,10 @@ It will fail fast with `SchemaValidationError` exception if event or response do
 
 **Validate** standalone function is typically used within the Lambda handler, or any other methods that perform data validation.
 
+???+ info
+    This function returns the validated event as a JSON object. If the schema specifies `default` values for omitted fields,
+    those default values will be included in the response.
+
 You can also gracefully handle schema validation errors by catching `SchemaValidationError` exception.
 
 === "getting_started_validator_standalone_function.py"

tests/functional/validator/_fastjsonschema/test_validator.py

@@ -16,6 +16,12 @@ def test_validate_raw_event(schema, raw_event):
     validate(event=raw_event, schema=schema)
 
 
+def test_validate_raw_event_default(schema_default, raw_event_default):
+    resp = validate(event=raw_event_default, schema=schema_default)
+    assert resp["username"] == "blah blah"
+    assert resp["message"] == "The default message"
+
+
 def test_validate_wrapped_event_raw_envelope(schema, wrapped_event):
     validate(event=wrapped_event, schema=schema, envelope="data.payload")
 

tests/functional/validator/conftest.py

@@ -6,8 +6,8 @@
 @pytest.fixture
 def schema():
     return {
-        "$schema": "http://json-schema.org/draft-07/schema",
-        "$id": "http://example.com/example.json",
+        "$schema": "https://json-schema.org/draft-07/schema",
+        "$id": "https://example.com/example.json",
         "type": "object",
         "title": "Sample schema",
         "description": "The root schema comprises the entire JSON document.",
@@ -30,11 +30,39 @@ def schema():
     }
 
 
+@pytest.fixture
+def schema_default():
+    return {
+        "$schema": "https://json-schema.org/draft-07/schema",
+        "$id": "https://example.com/example.json",
+        "type": "object",
+        "title": "Sample schema",
+        "description": "The root schema comprises the entire JSON document.",
+        "examples": [{"message": "hello world", "username": "lessa"}, {"username": "lessa"}],
+        "required": ["username"],
+        "properties": {
+            "message": {
+                "$id": "#/properties/message",
+                "type": "string",
+                "title": "The message",
+                "examples": ["hello world"],
+                "default": "The default message",
+            },
+            "username": {
+                "$id": "#/properties/username",
+                "type": "string",
+                "title": "The username",
+                "examples": ["lessa"],
+            },
+        },
+    }
+
+
 @pytest.fixture
 def schema_array():
     return {
-        "$schema": "http://json-schema.org/draft-07/schema",
-        "$id": "http://example.com/example.json",
+        "$schema": "https://json-schema.org/draft-07/schema",
+        "$id": "https://example.com/example.json",
         "type": "array",
         "title": "Sample schema",
         "description": "Sample JSON Schema for dummy data in an array",
@@ -71,8 +99,8 @@ def schema_array():
 @pytest.fixture
 def schema_response():
     return {
-        "$schema": "http://json-schema.org/draft-07/schema",
-        "$id": "http://example.com/example.json",
+        "$schema": "https://json-schema.org/draft-07/schema",
+        "$id": "https://example.com/example.json",
         "type": "object",
         "title": "Sample outgoing schema",
         "description": "The root schema comprises the entire JSON document.",
@@ -89,7 +117,7 @@ def schema_response():
 def schema_refs():
     return {
         "ParentSchema": {
-            "$schema": "http://json-schema.org/draft-07/schema",
+            "$schema": "https://json-schema.org/draft-07/schema",
             "$id": "testschema://ParentSchema",
             "type": "object",
             "title": "Sample schema",
@@ -104,7 +132,7 @@ def schema_refs():
             },
         },
         "ChildSchema": {
-            "$schema": "http://json-schema.org/draft-07/schema",
+            "$schema": "https://json-schema.org/draft-07/schema",
             "$id": "testschema://ChildSchema",
             "type": "object",
             "title": "Sample schema",
@@ -137,6 +165,11 @@ def raw_event():
     return {"message": "hello hello", "username": "blah blah"}
 
 
+@pytest.fixture
+def raw_event_default():
+    return {"username": "blah blah"}
+
+
 @pytest.fixture
 def wrapped_event():
     return {"data": {"payload": {"message": "hello hello", "username": "blah blah"}}}
@@ -407,7 +440,7 @@ def cloudwatch_logs_event():
 @pytest.fixture
 def cloudwatch_logs_schema():
     return {
-        "$schema": "http://json-schema.org/draft-07/schema",
+        "$schema": "https://json-schema.org/draft-07/schema",
         "$id": "http://example.com/example.json",
         "type": "array",
         "title": "Sample schema",
@@ -622,7 +655,7 @@ def eventbridge_schema_registry_cloudtrail_v2_s3():
 @pytest.fixture
 def schema_datetime_format():
     return {
-        "$schema": "http://json-schema.org/draft-07/schema",
+        "$schema": "https://json-schema.org/draft-07/schema",
         "$id": "http://example.com/example.json",
         "type": "object",
         "title": "Sample schema with string date-time format",