refactor(datasets): deprecate "DataSet" type names (svmlight)

Signed-off-by: Deepyaman Datta <[email protected]>

kedro-datasets/docs/source/kedro_datasets.rst

@@ -86,6 +86,7 @@ kedro_datasets
    kedro_datasets.spark.SparkStreamingDataSet
    kedro_datasets.spark.SparkStreamingDataset
    kedro_datasets.svmlight.SVMLightDataSet
+   kedro_datasets.svmlight.SVMLightDataset
    kedro_datasets.tensorflow.TensorFlowModelDataSet
    kedro_datasets.text.TextDataSet
    kedro_datasets.tracking.JSONDataSet

kedro-datasets/kedro_datasets/svmlight/__init__.py

@@ -1,12 +1,15 @@
-"""``AbstractDataSet`` implementation to load/save data from/to a svmlight/
-libsvm sparse data file."""
+"""``AbstractDataset`` implementation to load/save data from/to a
+svmlight/libsvm sparse data file."""
+from __future__ import annotations
+
 from typing import Any
 
 import lazy_loader as lazy
 
 # https://github.com/pylint-dev/pylint/issues/4300#issuecomment-1043601901
-SVMLightDataSet: Any
+SVMLightDataSet: type[SVMLightDataset]
+SVMLightDataset: Any
 
 __getattr__, __dir__, __all__ = lazy.attach(
-    __name__, submod_attrs={"svmlight_dataset": ["SVMLightDataSet"]}
+    __name__, submod_attrs={"svmlight_dataset": ["SVMLightDataSet", "SVMLightDataset"]}
 )

kedro-datasets/kedro_datasets/svmlight/svmlight_dataset.py

@@ -1,7 +1,8 @@
-"""``SVMLightDataSet`` loads/saves data from/to a svmlight/libsvm file using an
+"""``SVMLightDataset`` loads/saves data from/to a svmlight/libsvm file using an
 underlying filesystem (e.g.: local, S3, GCS). It uses sklearn functions
 ``dump_svmlight_file`` to save and ``load_svmlight_file`` to load a file.
 """
+import warnings
 from copy import deepcopy
 from pathlib import PurePosixPath
 from typing import Any, Dict, Optional, Tuple, Union
@@ -12,8 +13,7 @@
 from scipy.sparse.csr import csr_matrix
 from sklearn.datasets import dump_svmlight_file, load_svmlight_file
 
-from .._io import AbstractVersionedDataset as AbstractVersionedDataSet
-from .._io import DatasetError as DataSetError
+from kedro_datasets._io import AbstractVersionedDataset, DatasetError
 
 # NOTE: kedro.extras.datasets will be removed in Kedro 0.19.0.
 # Any contribution to datasets should be made in kedro-datasets
@@ -25,8 +25,8 @@
 _DO = Tuple[csr_matrix, ndarray]
 
 
-class SVMLightDataSet(AbstractVersionedDataSet[_DI, _DO]):
-    """``SVMLightDataSet`` loads/saves data from/to a svmlight/libsvm file using an
+class SVMLightDataset(AbstractVersionedDataset[_DI, _DO]):
+    """``SVMLightDataset`` loads/saves data from/to a svmlight/libsvm file using an
     underlying filesystem (e.g.: local, S3, GCS). It uses sklearn functions
     ``dump_svmlight_file`` to save and ``load_svmlight_file`` to load a file.
 
@@ -46,15 +46,15 @@ class SVMLightDataSet(AbstractVersionedDataSet[_DI, _DO]):
     .. code-block:: yaml
 
         svm_dataset:
-          type: svmlight.SVMLightDataSet
+          type: svmlight.SVMLightDataset
           filepath: data/01_raw/location.svm
           load_args:
             zero_based: False
           save_args:
             zero_based: False
 
         cars:
-          type: svmlight.SVMLightDataSet
+          type: svmlight.SVMLightDataset
           filepath: gcs://your_bucket/cars.svm
           fs_args:
             project: my-project
@@ -69,15 +69,15 @@ class SVMLightDataSet(AbstractVersionedDataSet[_DI, _DO]):
     advanced_data_catalog_usage.html>`_:
     ::
 
-        >>> from kedro_datasets.svmlight import SVMLightDataSet
+        >>> from kedro_datasets.svmlight import SVMLightDataset
         >>> import numpy as np
         >>>
         >>> # Features and labels.
         >>> data = (np.array([[0, 1], [2, 3.14159]]), np.array([7, 3]))
         >>>
-        >>> data_set = SVMLightDataSet(filepath="test.svm")
-        >>> data_set.save(data)
-        >>> reloaded_features, reloaded_labels = data_set.load()
+        >>> dataset = SVMLightDataset(filepath="test.svm")
+        >>> dataset.save(data)
+        >>> reloaded_features, reloaded_labels = dataset.load()
         >>> assert (data[0] == reloaded_features).all()
         >>> assert (data[1] == reloaded_labels).all()
 
@@ -97,7 +97,7 @@ def __init__(
         fs_args: Dict[str, Any] = None,
         metadata: Dict[str, Any] = None,
     ) -> None:
-        """Creates a new instance of SVMLightDataSet to load/save data from a svmlight/libsvm file.
+        """Creates a new instance of SVMLightDataset to load/save data from a svmlight/libsvm file.
 
         Args:
             filepath: Filepath in POSIX format to a text file prefixed with a protocol like `s3://`.
@@ -177,7 +177,7 @@ def _save(self, data: _DI) -> None:
     def _exists(self) -> bool:
         try:
             load_path = get_filepath_str(self._get_load_path(), self._protocol)
-        except DataSetError:
+        except DatasetError:
             return False
 
         return self._fs.exists(load_path)
@@ -190,3 +190,21 @@ def _invalidate_cache(self) -> None:
         """Invalidate underlying filesystem caches."""
         filepath = get_filepath_str(self._filepath, self._protocol)
         self._fs.invalidate_cache(filepath)
+
+
+_DEPRECATED_CLASSES = {
+    "SVMLightDataSet": SVMLightDataset,
+}
+
+
+def __getattr__(name):
+    if name in _DEPRECATED_CLASSES:
+        alias = _DEPRECATED_CLASSES[name]
+        warnings.warn(
+            f"{repr(name)} has been renamed to {repr(alias.__name__)}, "
+            f"and the alias will be removed in Kedro-Datasets 2.0.0",
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return alias
+    raise AttributeError(f"module {repr(__name__)} has no attribute {repr(name)}")