🐛 Force return on load for old style datasets (#592) (#593)

* 🐛 Force return on load for old style datasets (#592)

* skip test for kedro<0.19.7

* fix test assert in new mlflow version

kedro_mlflow/io/artifacts/mlflow_artifact_dataset.py

@@ -137,7 +137,7 @@ def _load(self) -> Any:  # pragma: no cover
                 if getattr(super().load, "__loadwrapped__", False):  # modern dataset
                     return super().load.__wrapped__(self)
                 else:  # legacy dataset
-                    super()._load()
+                    return super()._load()
 
         # rename the class
         parent_name = dataset_obj.__name__

tests/framework/cli/test_cli_modelify.py

@@ -466,7 +466,7 @@ def test_modelify_with_infer_input_example(
         "artifact_path": "input_example.json",
         "pandas_orient": "split",
         "type": "dataframe",
-        "serving_input_path": "serving_input_payload.json",
+        "serving_input_path": "serving_input_example.json",
     }
 
 

tests/io/artifacts/test_mlflow_artifact_dataset.py

@@ -3,6 +3,7 @@
 import mlflow
 import pandas as pd
 import pytest
+from kedro import __version__ as kedro_version
 from kedro.io import AbstractDataset
 from kedro_datasets.pandas import CSVDataset
 from kedro_datasets.partitions import PartitionedDataset
@@ -11,6 +12,8 @@
 
 from kedro_mlflow.io.artifacts import MlflowArtifactDataset
 
+KEDRO_VERSION = tuple(int(x) for x in kedro_version.split("."))
+
 
 @pytest.fixture
 def df1():
@@ -249,7 +252,7 @@ def test_artifact_dataset_load_with_run_id_and_artifact_path(
 
 
 @pytest.mark.parametrize("artifact_path", [None, "partitioned_data"])
-def test_partitioned_dataset_save_and_reload(
+def test_artifact_dataset_partitioned_dataset_save_and_reload(
     tmp_path, mlflow_client, artifact_path, df1, df2
 ):
     mlflow_dataset = MlflowArtifactDataset(
@@ -292,7 +295,10 @@ def test_partitioned_dataset_save_and_reload(
         pd.testing.assert_frame_equal(df, reloaded_data[k])
 
 
-def test_modern_dataset(tmp_path, mlflow_client, df1):
+@pytest.mark.skipif(
+    KEDRO_VERSION < (0, 19, 7), reason="modern datasets were introduced in kedro 0.19.7"
+)
+def test_artifact_dataset_modern_dataset(tmp_path, mlflow_client, df1):
     class MyOwnDatasetWithoutUnderscoreMethods(AbstractDataset):
         def __init__(self, filepath):
             self._filepath = Path(filepath)
@@ -332,7 +338,7 @@ def _describe(self):
     assert df1.equals(mlflow_dataset.load())
 
 
-def test_legacy_dataset(tmp_path, mlflow_client, df1):
+def test_artifact_dataset_legacy_dataset(tmp_path, mlflow_client, df1):
     class MyOwnDatasetWithUnderscoreMethods(AbstractDataset):
         def __init__(self, filepath):
             self._filepath = Path(filepath)