fix: Pipeline TypeError: can only concatenate list (not "NoneType") to list Using Sou (aws#5518)

Author: aviruthen

sagemaker-core/src/sagemaker/core/workflow/utilities.py

@@ -173,6 +173,8 @@ def get_code_hash(step: Entity) -> str:
         source_code = model_trainer.source_code
         if source_code:
             source_dir = source_code.source_dir
+            # requirements may be None when SourceCode.requirements is not set;
+            # get_training_code_hash handles None dependencies gracefully
             requirements = source_code.requirements
             entry_point = source_code.entry_script
             return get_training_code_hash(entry_point, source_dir, requirements)
@@ -197,6 +199,7 @@ def get_processing_dependencies(dependency_args: List[List[str]]) -> List[str]:
 
 
 def get_processing_code_hash(code: str, source_dir: str, dependencies: List[str]) -> str:
+    dependencies = dependencies or []
     """Get the hash of a processing step's code artifact(s).
 
     Args:

sagemaker-core/tests/unit/workflow/test_utilities.py

@@ -214,6 +214,35 @@ def test_get_processing_dependencies_multiple_lists(self):
 
         assert result == ["dep1", "dep2", "dep3", "dep4", "dep5"]
 
+    def test_get_processing_code_hash_with_none_dependencies(self):
+        """Test get_processing_code_hash does not raise TypeError when dependencies is None"""
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".py", delete=False) as f:
+            f.write("print('hello')")
+            temp_file = f.name
+
+        try:
+            # Should not raise TypeError
+            result = get_processing_code_hash(code=temp_file, source_dir=None, dependencies=None)
+
+            assert result is not None
+            assert len(result) == 64
+        finally:
+            os.unlink(temp_file)
+
+    def test_get_processing_code_hash_with_none_dependencies_and_source_dir(self):
+        """Test get_processing_code_hash with source_dir and None dependencies"""
+        with tempfile.TemporaryDirectory() as temp_dir:
+            code_file = Path(temp_dir, "script.py")
+            code_file.write_text("print('hello')")
+
+            # Should not raise TypeError
+            result = get_processing_code_hash(
+                code=str(code_file), source_dir=temp_dir, dependencies=None
+            )
+
+            assert result is not None
+            assert len(result) == 64
+
     def test_get_processing_code_hash_with_source_dir(self):
         """Test get_processing_code_hash with source_dir"""
         with tempfile.TemporaryDirectory() as temp_dir:
@@ -264,6 +293,34 @@ def test_get_processing_code_hash_with_dependencies(self):
 
             assert result is not None
 
+    def test_get_training_code_hash_with_none_dependencies_and_source_dir(self):
+        """Test get_training_code_hash with source_dir and None dependencies does not raise"""
+        with tempfile.TemporaryDirectory() as temp_dir:
+            entry_file = Path(temp_dir, "train.py")
+            entry_file.write_text("print('training')")
+
+            # Should not raise TypeError
+            result = get_training_code_hash(
+                entry_point=str(entry_file), source_dir=temp_dir, dependencies=None
+            )
+
+            assert result is not None
+            assert len(result) == 64
+
+    def test_get_training_code_hash_with_none_dependencies_and_entry_point(self):
+        """Test get_training_code_hash with entry_point only and None dependencies does not raise"""
+        with tempfile.TemporaryDirectory() as temp_dir:
+            entry_file = Path(temp_dir, "train.py")
+            entry_file.write_text("print('training')")
+
+            # Should not raise TypeError
+            result = get_training_code_hash(
+                entry_point=str(entry_file), source_dir=None, dependencies=None
+            )
+
+            assert result is not None
+            assert len(result) == 64
+
     def test_get_training_code_hash_with_source_dir(self):
         """Test get_training_code_hash with source_dir"""
         with tempfile.TemporaryDirectory() as temp_dir: