feat(datasets): Update __repr__ for PartitionedDataset (#782)

* Implemented __repr__ Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Added dummy object creation needed for class __repr__ Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Updated RELEASE.md Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Implemented test for repr Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Debug output Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Updated filepath Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Fixed lint Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> * Removed redundant function call Signed-off-by: Elena Khaustova <ymax70rus@gmail.com> --------- Signed-off-by: Elena Khaustova <ymax70rus@gmail.com>
kedro-org · Jul 30, 2024 · da66452 · da66452
1 parent 24cd01c
commit da66452
Show file tree

Hide file tree

Showing 3 changed files with 27 additions and 0 deletions.
diff --git a/kedro-datasets/RELEASE.md b/kedro-datasets/RELEASE.md
@@ -1,5 +1,7 @@
 # Upcoming Release
 ## Major features and improvements
+* Improved PartitionedDataset representation when printing.
+
 ## Bug fixes and other changes
 ## Breaking Changes
 ## Community contributions

diff --git a/kedro-datasets/kedro_datasets/partitions/partitioned_dataset.py b/kedro-datasets/kedro_datasets/partitions/partitioned_dataset.py
@@ -327,6 +327,22 @@ def _describe(self) -> dict[str, Any]:
             "dataset_config": clean_dataset_config,
         }
 
+    def __repr__(self) -> str:
+        object_description = self._describe()
+
+        # Dummy object to call _pretty_repr
+        # Only clean_dataset_config parameters are exposed
+        kwargs = deepcopy(self._dataset_config)
+        kwargs[self._filepath_arg] = ""
+        dataset = self._dataset_type(**kwargs)  # type: ignore
+
+        object_description_repr = {
+            "filepath": object_description["path"],
+            "dataset": dataset._pretty_repr(object_description["dataset_config"]),
+        }
+
+        return self._pretty_repr(object_description_repr)
+
     def _invalidate_caches(self) -> None:
         self._partition_cache.clear()
         self._filesystem.invalidate_cache(self._normalized_path)

diff --git a/kedro-datasets/tests/partitions/test_partitioned_dataset.py b/kedro-datasets/tests/partitions/test_partitioned_dataset.py
@@ -57,6 +57,15 @@ class FakeDataset:  # pylint: disable=too-few-public-methods
 
 
 class TestPartitionedDatasetLocal:
+    @pytest.mark.parametrize("dataset", ["pandas.ParquetDataset", ParquetDataset])
+    def test_repr(self, dataset):
+        pds = PartitionedDataset(path="", dataset=dataset)
+        assert (
+            repr(pds)
+            == """kedro_datasets.partitions.partitioned_dataset.PartitionedDataset(filepath='', """
+            """dataset='kedro_datasets.pandas.parquet_dataset.ParquetDataset()')"""
+        )
+
     @pytest.mark.parametrize("dataset", LOCAL_DATASET_DEFINITION)
     @pytest.mark.parametrize(
         "suffix,expected_num_parts", [("", 5), (".csv", 3), ("p4", 1)]