feat: Extend YAML serialization to allow Python tuples (#7853)

haystack/marshal/yaml.py

@@ -7,11 +7,21 @@
 import yaml
 
 
+# Custom YAML safe loader that supports loading Python tuples
+class YamlLoader(yaml.SafeLoader):  # pylint: disable=too-many-ancestors
+    def construct_python_tuple(self, node: yaml.SequenceNode):
+        """Construct a Python tuple from the sequence."""
+        return tuple(self.construct_sequence(node))
+
+
+YamlLoader.add_constructor("tag:yaml.org,2002:python/tuple", YamlLoader.construct_python_tuple)
+
+
 class YamlMarshaller:
     def marshal(self, dict_: Dict[str, Any]) -> str:
         """Return a YAML representation of the given dictionary."""
         return yaml.dump(dict_)
 
     def unmarshal(self, data_: Union[str, bytes, bytearray]) -> Dict[str, Any]:
         """Return a dictionary from the given YAML data."""
-        return yaml.safe_load(data_)
+        return yaml.load(data_, Loader=YamlLoader)

releasenotes/notes/serialization-tuple-support-ffe176417e7099f5.yaml

@@ -0,0 +1,4 @@
+---
+enhancements:
+  - |
+    Pipeline serialization to YAML now supports tuples as field values.

test/marshal/__init__.py

@@ -0,0 +1,3 @@
+# SPDX-FileCopyrightText: 2022-present deepset GmbH <[email protected]>
+#
+# SPDX-License-Identifier: Apache-2.0

test/marshal/test_yaml.py

@@ -0,0 +1,42 @@
+# SPDX-FileCopyrightText: 2022-present deepset GmbH <[email protected]>
+#
+# SPDX-License-Identifier: Apache-2.0
+import pytest
+
+from haystack.marshal.yaml import YamlMarshaller
+
+
+@pytest.fixture
+def yaml_data():
+    return {"key": "value", 1: 0.221, "list": [1, 2, 3], "tuple": (1, None, True), "dict": {"set": {False}}}
+
+
+@pytest.fixture
+def serialized_yaml_str():
+    return """key: value
+1: 0.221
+list:
+- 1
+- 2
+- 3
+tuple: !!python/tuple
+- 1
+- null
+- true
+dict:
+  set: !!set
+    false: null
+"""
+
+
+def test_yaml_marshal(yaml_data, serialized_yaml_str):
+    marshaller = YamlMarshaller()
+    marshalled = marshaller.marshal(yaml_data)
+    assert isinstance(marshalled, str)
+    assert marshalled.strip().replace("\n", "") == serialized_yaml_str.strip().replace("\n", "")
+
+
+def test_yaml_unmarshal(yaml_data, serialized_yaml_str):
+    marshaller = YamlMarshaller()
+    unmarshalled = marshaller.unmarshal(serialized_yaml_str)
+    assert unmarshalled == yaml_data