Fix nested partial to use attr_name instead of data_key for prefix (#2903)

* Fix nested partial to use attr_name instead of data_key

When building dot-delimited sub_partial for nested schemas, the prefix
was computed from field_name (which is data_key when set) instead of
attr_name. Since users specify partial fields using attribute names like
partial=("nested_field.child",), the prefix needs to match attr_name.

When data_key differs from attr_name, the prefix mismatch caused the
nested partial list to always be empty, so the nested schema never
received the partial option and required fields raised unexpected
validation errors.

* Add test for nested partial with data_key using attr_name prefix

* Update authors and changelog

---------

Co-authored-by: Jared Deckard <jared@shademaps.com>
Co-authored-by: Steven Loria <git@stevenloria.com>

Author: 3 people

AUTHORS.rst

@@ -186,3 +186,4 @@ Contributors (chronological)
 - Emmanuel Ferdman  `@emmanuel-ferdman  <https://github.com/emmanuel-ferdman>`_
 - Fridayworks `@worksbyfriday  <https://github.com/worksbyfriday>`_
 - `@rstar327 <https://github.com/rstar327>`_
+- Kadir Can Ozden  `@bysiber  <https://github.com/bysiber>`_

CHANGELOG.rst

@@ -12,6 +12,8 @@ Bug fixes:
   Thanks :user:`nosnickid` for the report and :user:`worksbyfriday` for the PR.
 - Fix `marshmallow.validate.OneOf` emitting extra pairs when labels outnumber choices (:issue:`2869`).
   Thanks: user:`T90REAL` for the report and :user:`rstar327` for the PR.
+- Fix behavior when passing a dot-delited attribute name to ``partial`` for a key with ``data_key`` set (:pr:`2903`).
+  Thanks :user:`bysiber` for the PR.
 
 
 4.2.2 (2026-02-04)

src/marshmallow/schema.py

@@ -654,7 +654,7 @@ def _deserialize(
                 d_kwargs = {}
                 # Allow partial loading of nested schemas.
                 if partial_is_collection:
-                    prefix = field_name + "."
+                    prefix = attr_name + "."
                     len_prefix = len(prefix)
                     sub_partial = [
                         f[len_prefix:] for f in partial if f.startswith(prefix)

tests/test_deserialization.py

@@ -2307,6 +2307,26 @@ class SchemaB(Schema):
         with pytest.raises(ValidationError):
             SchemaB().load({"z": {"x": 0}})
 
+    def test_nested_partial_tuple_with_data_key(self):
+        class AddressSchema(Schema):
+            zip_code = fields.String(required=True)
+            city = fields.String(required=True)
+
+        class PersonSchema(Schema):
+            address = fields.Nested(
+                AddressSchema, data_key="homeAddress", required=True
+            )
+
+        data = {"homeAddress": {"city": "Springfield"}}
+        # partial uses attr_name ("address.zip_code"), not data_key
+        result = PersonSchema().load(data, partial=("address.zip_code",))
+        assert result["address"]["city"] == "Springfield"
+        assert "zip_code" not in result["address"]
+        # Without partial, it should fail on the missing required field
+        with pytest.raises(ValidationError) as excinfo:
+            PersonSchema().load(data)
+        assert "zip_code" in excinfo.value.messages["homeAddress"]
+
 
 @pytest.mark.parametrize("FieldClass", ALL_FIELDS)
 def test_required_field_failure(FieldClass):