Merge branch 'master' into fix/modelbuilder-accept-eula

Author: nargokul

sagemaker-core/src/sagemaker/core/processing.py

@@ -77,6 +77,8 @@
 from sagemaker.core.workflow.execution_variables import ExecutionVariables
 from sagemaker.core.workflow.functions import Join
 from sagemaker.core.workflow.pipeline_context import runnable_by_pipeline
+from sagemaker.core.telemetry.telemetry_logging import _telemetry_emitter
+from sagemaker.core.telemetry.constants import Feature
 
 from sagemaker.core._studio import _append_project_tags
 from sagemaker.core.config.config_utils import _append_sagemaker_config_tags
@@ -771,6 +773,7 @@ def __init__(
             network_config=network_config,
         )
 
+    @_telemetry_emitter(feature=Feature.PROCESSING, func_name="ScriptProcessor.run")
     @runnable_by_pipeline
     def run(
         self,
@@ -1171,6 +1174,7 @@ def _package_code(
             os.unlink(tmp.name)
             return s3_uri
 
+    @_telemetry_emitter(feature=Feature.PROCESSING, func_name="FrameworkProcessor.run")
     @runnable_by_pipeline
     def run(
         self,

sagemaker-core/src/sagemaker/core/telemetry/constants.py

@@ -29,6 +29,7 @@ class Feature(Enum):
     MODEL_CUSTOMIZATION = 15
     MLOPS = 16
     FEATURE_STORE = 17
+    PROCESSING = 18
 
     def __str__(self):  # pylint: disable=E0307
         """Return the feature name."""

sagemaker-core/src/sagemaker/core/telemetry/telemetry_logging.py

@@ -61,6 +61,7 @@
     str(Feature.MODEL_CUSTOMIZATION): 15,
     str(Feature.MLOPS): 16,
     str(Feature.FEATURE_STORE): 17,
+    str(Feature.PROCESSING): 18,
 }
 
 STATUS_TO_CODE = {

sagemaker-train/src/sagemaker/ai_registry/evaluator.py

@@ -382,7 +382,7 @@ def _create_lambda_function(cls, name: str, source_file: str, role: Optional[str
         # Create Lambda function
         lambda_client = boto3.client("lambda")
         function_name = f"SageMaker-evaluator-{name}-{datetime.now().strftime('%Y%m%d_%H%M%S')}"
-        handler_name = f"{os.path.splitext(os.path.basename(source_file))[0]}.lambda_handler"
+        handler_name = "lambda_function.lambda_handler"
 
         try:
             lambda_response = lambda_client.create_function(

sagemaker-train/tests/integ/ai_registry/conftest.py

@@ -56,6 +56,22 @@ def sample_jsonl_file():
     os.unlink(f.name)
 
 
+@pytest.fixture
+def sample_lambda_py_file():
+    """Create a raw Python Lambda file with a non-default filename to test handler derivation."""
+    code = '''import json
+def lambda_handler(event, context):
+    return {"statusCode": 200, "body": json.dumps({"score": 0.9})}
+'''
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', prefix='my_custom_evaluator_', delete=False) as f:
+        f.write(code)
+        f.flush()
+        os.fsync(f.fileno())
+        fname = f.name
+    yield fname
+    os.unlink(fname)
+
+
 @pytest.fixture
 def sample_lambda_code():
     """Create sample Lambda function code as zip."""

sagemaker-train/tests/integ/ai_registry/test_evaluator.py

@@ -81,6 +81,47 @@ def test_create_reward_function_from_local_code(self, unique_name, sample_lambda
         assert evaluator.method == EvaluatorMethod.BYOC
         assert evaluator.reference is not None
 
+    def test_create_reward_function_from_local_py_file_and_invoke(
+        self, unique_name, sample_lambda_py_file, test_role, cleanup_list
+    ):
+        """End-to-end test: create evaluator from a raw .py file with non-default name and invoke it.
+
+        Regression test for the handler name bug where the Lambda was created with an incorrect
+        handler derived from the source filename instead of 'lambda_function.lambda_handler'.
+        """
+        import json
+        import boto3
+
+        evaluator = Evaluator.create(
+            name=unique_name,
+            type=REWARD_FUNCTION,
+            source=sample_lambda_py_file,
+            role=test_role,
+            wait=True,  # wait for Lambda to be active
+        )
+        cleanup_list.append(evaluator)
+        assert evaluator.method == EvaluatorMethod.BYOC
+        assert evaluator.reference is not None
+
+        # Wait for Lambda to become Active before invoking
+        lambda_client = boto3.client("lambda")
+        waiter = lambda_client.get_waiter("function_active_v2")
+        waiter.wait(FunctionName=evaluator.reference)
+
+        # Invoke the Lambda directly to verify the handler is correct
+        lambda_client = boto3.client("lambda")
+        response = lambda_client.invoke(
+            FunctionName=evaluator.reference,
+            InvocationType="RequestResponse",
+            Payload=json.dumps({"input": "test"}).encode(),
+        )
+        assert response["StatusCode"] == 200
+        assert "FunctionError" not in response, (
+            f"Lambda invocation failed with error: {response.get('FunctionError')}"
+        )
+        result = json.loads(response["Payload"].read())
+        assert result.get("statusCode") == 200
+
     def test_get_evaluator(self, unique_name, sample_prompt_file, cleanup_list):
         """Test retrieving evaluator by name."""
         try:

sagemaker-train/tests/unit/ai_registry/test_evaluator.py

@@ -77,6 +77,9 @@ def test_create_with_byoc(self, mock_air_hub, mock_boto3):
 
         assert evaluator.method == EvaluatorMethod.BYOC
         mock_air_hub.upload_to_s3.assert_called_once()
+        mock_lambda_client.create_function.assert_called_once()
+        call_kwargs = mock_lambda_client.create_function.call_args[1]
+        assert call_kwargs["Handler"] == "lambda_function.lambda_handler"
 
     @patch('sagemaker.ai_registry.evaluator.AIRHub')
     def test_get_all(self, mock_air_hub):