Update error message when kedro-datasets is not installed or `DataS…

…et` spelling is used (#3952)

* Update error message when kedro-datasets is not installed

Signed-off-by: Ankita Katiyar <[email protected]>

* Release notes, update error hint

Signed-off-by: Ankita Katiyar <[email protected]>

* Skip test for python 3.8

Signed-off-by: Ankita Katiyar <[email protected]>

* Ignore test coverage

Signed-off-by: Ankita Katiyar <[email protected]>

---------

Signed-off-by: Ankita Katiyar <[email protected]>
Signed-off-by: Ankita Katiyar <[email protected]>

RELEASE.md

@@ -4,6 +4,7 @@
 
 ## Bug fixes and other changes
 * Updated error message for invalid catalog entries.
+* Updated error message for catalog entries when the dataset class is not found with hints on how to resolve the issue.
 * Fixed a bug in the `DataCatalog` `shallow_copy()` method to ensure it returns the type of the used catalog and doesn't cast it to `DataCatalog`.
 
 ## Breaking changes to the API

kedro/io/core.py

@@ -155,7 +155,7 @@ def from_config(
         except Exception as exc:
             raise DatasetError(
                 f"An exception occurred when parsing config "
-                f"for dataset '{name}':\n{str(exc)}."
+                f"for dataset '{name}':\n{str(exc)}"
             ) from exc
 
         try:
@@ -406,7 +406,23 @@ def parse_dataset_definition(
                 class_obj = tmp
                 break
         else:
-            raise DatasetError(f"Class '{dataset_type}' not found, is this a typo?")
+            hint = ""
+            if "DataSet" in dataset_type:
+                hint = (  # pragma: no cover # To remove when we drop support for python 3.8
+                    "Hint: If you are trying to use a dataset from `kedro-datasets`>=2.0.0, "
+                    "make sure that the dataset name uses the `Dataset` spelling instead of `DataSet`."
+                )
+            else:
+                hint = (
+                    "Hint: If you are trying to use a dataset from `kedro-datasets`, "
+                    "make sure that the package is installed in your current environment. "
+                    "You can do so by running `pip install kedro-datasets` or "
+                    "`pip install kedro-datasets[<dataset-group>]` to install `kedro-datasets` along with "
+                    "related dependencies for the specific dataset group."
+                )
+            raise DatasetError(
+                f"Class '{dataset_type}' not found, is this a typo?" f"\n{hint}"
+            )
 
     if not class_obj:
         class_obj = dataset_type

tests/io/test_data_catalog.py

@@ -1,5 +1,6 @@
 import logging
 import re
+import sys
 from copy import deepcopy
 from datetime import datetime, timezone
 from pathlib import Path
@@ -503,7 +504,24 @@ def test_config_missing_class(self, sane_config):
 
         pattern = (
             "An exception occurred when parsing config for dataset 'boats':\n"
-            "Class 'kedro.io.CSVDatasetInvalid' not found"
+            "Class 'kedro.io.CSVDatasetInvalid' not found, is this a typo?"
+        )
+        with pytest.raises(DatasetError, match=re.escape(pattern)):
+            DataCatalog.from_config(**sane_config)
+
+    @pytest.mark.skipif(
+        sys.version_info < (3, 9),
+        reason="for python 3.8 kedro-datasets version 1.8 is used which has the old spelling",
+    )
+    def test_config_incorrect_spelling(self, sane_config):
+        """Check hint if the type uses the old DataSet spelling"""
+        sane_config["catalog"]["boats"]["type"] = "pandas.CSVDataSet"
+
+        pattern = (
+            "An exception occurred when parsing config for dataset 'boats':\n"
+            "Class 'pandas.CSVDataSet' not found, is this a typo?"
+            "\nHint: If you are trying to use a dataset from `kedro-datasets`>=2.0.0,"
+            " make sure that the dataset name uses the `Dataset` spelling instead of `DataSet`."
         )
         with pytest.raises(DatasetError, match=re.escape(pattern)):
             DataCatalog.from_config(**sane_config)