Add the node's function name to NodeSnapshot (#5663)

* Add the node's function name to NodeSnapshot

Signed-off-by: Jitendra Gundaniya <jitendra_gundaniya@mckinsey.com>

* Make func_name keyword-only in NodeSnapshot and add tests for its behavior

Signed-off-by: Jitendra Gundaniya <jitendra_gundaniya@mckinsey.com>

* Enforce func_name as a keyword-only argument in NodeSnapshot and add a test for TypeError

Signed-off-by: Jitendra Gundaniya <jitendra_gundaniya@mckinsey.com>

---------

Signed-off-by: Jitendra Gundaniya <jitendra_gundaniya@mckinsey.com>
Co-authored-by: Ravi Kumar Pilla <ravi_kumar_pilla@mckinsey.com>

Author: jitu5

RELEASE.md

@@ -1,5 +1,7 @@
 # Upcoming Release
 ## Major features and improvements
+* Added the node function name to project inspection snapshots as `NodeSnapshot.func_name`.
+
 ## Bug fixes and other changes
 ## Documentation changes
 ## Community contributions

docs/extend/serving.md

@@ -81,6 +81,7 @@ curl http://127.0.0.1:8000/snapshot
       "nodes": [
         {
           "name": "split_data_node",
+          "func_name": "split_data",
           "inputs": ["example_iris_data"],
           "outputs": ["X_train", "X_test"],
           "tags": [],

docs/inspect/inspect-project.md

@@ -90,13 +90,14 @@ default_pipeline = next(p for p in snapshot.pipelines if p.name == "__default__"
 
 for node in default_pipeline.nodes:
     print(node.name)
+    print("  func_name:", node.func_name)
     print("  namespace:", node.namespace)
     print("  inputs:   ", node.inputs)
     print("  outputs:  ", node.outputs)
     print("  tags:     ", node.tags)
 ```
 
-Each node is represented as a [`NodeSnapshot`][kedro.inspection.models.NodeSnapshot] with `name`, `namespace`, `inputs`, `outputs`, and `tags` fields.
+Each node is represented as a [`NodeSnapshot`][kedro.inspection.models.NodeSnapshot] with `name`, `func_name`, `namespace`, `inputs`, `outputs`, and `tags` fields.
 
 ## How to inspect catalog datasets
 

kedro/inspection/models.py

@@ -57,13 +57,15 @@ class NodeSnapshot:
 
     Attributes:
         name: Fully-qualified node name (includes namespace prefix if present).
+        func_name: Readable name of the node's underlying function.
         namespace: Node namespace, or ``None`` if the node has no namespace.
         tags: Sorted list of tags assigned to the node.
         inputs: Ordered list of input dataset names.
         outputs: Ordered list of output dataset names.
     """
 
     name: str
+    func_name: str = field(kw_only=True)
     namespace: str | None = None
     tags: list[str] = field(default_factory=list)
     inputs: list[str] = field(default_factory=list)

kedro/inspection/snapshot.py

@@ -79,6 +79,7 @@ def _node_to_snapshot(node: Node) -> NodeSnapshot:
     """
     return NodeSnapshot(
         name=node.name,
+        func_name=node._func_name,  # Matches Node.__str__ and registry describe.
         namespace=node.namespace,
         tags=sorted(node.tags),
         inputs=node.inputs,

tests/inspection/test_helper.py

@@ -12,7 +12,7 @@
 
 
 def _make_node_snapshot(inputs: list[str], outputs: list[str]) -> NodeSnapshot:
-    return NodeSnapshot(name="n", inputs=inputs, outputs=outputs)
+    return NodeSnapshot(name="n", func_name="identity", inputs=inputs, outputs=outputs)
 
 
 def _make_pipeline_snapshot(nodes: list[NodeSnapshot]) -> PipelineSnapshot:

tests/inspection/test_node_pipeline_snapshot.py

@@ -2,6 +2,8 @@
 
 from __future__ import annotations
 
+from functools import partial
+
 import pytest
 
 from kedro.inspection.models import NodeSnapshot, PipelineSnapshot
@@ -42,18 +44,28 @@ def simple_pipeline(simple_node):
 
 class TestNodeSnapshot:
     def test_instantiation_defaults(self):
-        snapshot = NodeSnapshot(name="my_node")
+        snapshot = NodeSnapshot(name="my_node", func_name="my_func")
         assert snapshot.name == "my_node"
+        assert snapshot.func_name == "my_func"
         assert snapshot.namespace is None
         assert snapshot.tags == []
         assert snapshot.inputs == []
         assert snapshot.outputs == []
 
+    def test_func_name_is_keyword_only(self):
+        snapshot = NodeSnapshot("my_node", "my_namespace", func_name="my_func")
+        assert snapshot.func_name == "my_func"
+        assert snapshot.namespace == "my_namespace"
+
+        with pytest.raises(TypeError, match="func_name"):
+            NodeSnapshot("my_node", "my_namespace")
+
 
 class TestNodeToSnapshot:
     def test_populates_all_fields(self, simple_node):
         snapshot = _node_to_snapshot(simple_node)
         assert snapshot.name == simple_node.name
+        assert snapshot.func_name == "_identity"
         assert snapshot.namespace == simple_node.namespace
         assert snapshot.inputs == simple_node.inputs
         assert snapshot.outputs == simple_node.outputs
@@ -69,10 +81,26 @@ def test_namespace_populated(self, namespaced_node):
     def test_returns_node_snapshot_instance(self, simple_node):
         assert isinstance(_node_to_snapshot(simple_node), NodeSnapshot)
 
+    def test_named_partial_func_name_warns(self):
+        partial_node = node(
+            partial(_identity),
+            inputs="raw",
+            outputs="processed",
+            name="partial_node",
+        )
+
+        with pytest.warns(UserWarning, match="made from a 'partial' function"):
+            snapshot = _node_to_snapshot(partial_node)
+
+        assert snapshot.name == "partial_node"
+        assert snapshot.func_name == "<partial>"
+
 
 class TestPipelineSnapshot:
     def test_instantiation(self):
-        node_snap = NodeSnapshot(name="n", inputs=["a"], outputs=["b"])
+        node_snap = NodeSnapshot(
+            name="n", func_name="identity", inputs=["a"], outputs=["b"]
+        )
         snapshot = PipelineSnapshot(name="my_pipe", nodes=[node_snap])
         assert snapshot.name == "my_pipe"
         assert snapshot.nodes == [node_snap]

tests/inspection/test_project_snapshot.py

@@ -43,6 +43,7 @@ def metadata_snapshot():
 def node_snapshot():
     return NodeSnapshot(
         name="node_a",
+        func_name="process_companies",
         inputs=["companies"],
         outputs=["processed_companies"],
     )

tests/server/test_snapshot_endpoint.py

@@ -23,6 +23,7 @@ def _make_snapshot() -> ProjectSnapshot:
                 nodes=[
                     NodeSnapshot(
                         name="my_node",
+                        func_name="process_data",
                         namespace="ns",
                         tags=["tag1"],
                         inputs=["raw_data"],
@@ -76,6 +77,7 @@ def test_snapshot_response_contains_all_fields(self, mocker, make_http_server):
         assert len(pipelines) == 1
         assert pipelines[0]["name"] == "__default__"
         assert pipelines[0]["nodes"][0]["name"] == "my_node"
+        assert pipelines[0]["nodes"][0]["func_name"] == "process_data"
 
         assert "raw_data" in payload["datasets"]
         assert payload["datasets"]["raw_data"]["type"] == "pandas.CSVDataset"