Solution which considers private datasets (#439)

* Basic solution for private datasets

* Faulty and not necessary implementation of finally

* Fixed assertRaises call in test_get_data

* Updated get_dataset

* Fixed typo

Author: ArlindKadra

openml/datasets/functions.py

@@ -4,15 +4,16 @@
 import os
 import re
 import shutil
+import six
 
 from oslo_concurrency import lockutils
 import xmltodict
 
 import openml.utils
 import openml._api_calls
 from .dataset import OpenMLDataset
-from ..exceptions import OpenMLCacheException, OpenMLServerNoResult, \
-    OpenMLHashException
+from ..exceptions import OpenMLCacheException, OpenMLServerException, \
+    OpenMLHashException, PrivateDatasetError
 from .. import config
 from .._api_calls import _read_url
 
@@ -315,13 +316,21 @@ def get_dataset(dataset_id):
         did_cache_dir = _create_dataset_cache_directory(dataset_id)
 
         try:
+            remove_dataset_cache = True
             description = _get_dataset_description(did_cache_dir, dataset_id)
             arff_file = _get_dataset_arff(did_cache_dir, description)
             features = _get_dataset_features(did_cache_dir, dataset_id)
             qualities = _get_dataset_qualities(did_cache_dir, dataset_id)
-        except Exception as e:
-            _remove_dataset_cache_dir(did_cache_dir)
-            raise e
+            remove_dataset_cache = False
+        except OpenMLServerException as e:
+            # if there was an exception, check if the user had access to the dataset
+            if e.code == 112:
+                six.raise_from(PrivateDatasetError(e.message), None)
+            else:
+                raise e
+        finally:
+            if remove_dataset_cache:
+                _remove_dataset_cache_dir(did_cache_dir)
 
         dataset = _create_dataset_from_description(
             description, features, qualities, arff_file
@@ -357,9 +366,8 @@ def _get_dataset_description(did_cache_dir, dataset_id):
 
     try:
         return _get_cached_dataset_description(dataset_id)
-    except (OpenMLCacheException):
+    except OpenMLCacheException:
         dataset_xml = openml._api_calls._perform_api_call("data/%d" % dataset_id)
-
         with io.open(description_file, "w", encoding='utf8') as fh:
             fh.write(dataset_xml)
 

openml/exceptions.py

@@ -43,4 +43,10 @@ def __init__(self, message):
 
 class OpenMLHashException(PyOpenMLError):
     """Locally computed hash is different than hash announced by the server."""
-    pass
+    pass
+
+
+class PrivateDatasetError(PyOpenMLError):
+    "Exception thrown when the user has no rights to access the dataset"
+    def __init__(self, message):
+        super(PrivateDatasetError, self).__init__(message)

openml/tasks/functions.py

@@ -7,7 +7,7 @@
 from oslo_concurrency import lockutils
 import xmltodict
 
-from ..exceptions import OpenMLCacheException, OpenMLServerNoResult
+from ..exceptions import OpenMLCacheException
 from ..datasets import get_dataset
 from .task import OpenMLTask, _create_task_cache_dir
 from .. import config

tests/test_datasets/test_dataset_functions.py

@@ -17,7 +17,8 @@
 
 import openml
 from openml import OpenMLDataset
-from openml.exceptions import OpenMLCacheException, PyOpenMLError, OpenMLHashException
+from openml.exceptions import OpenMLCacheException, PyOpenMLError, \
+    OpenMLHashException, PrivateDatasetError
 from openml.testing import TestBase
 from openml.utils import _tag_entity
 
@@ -231,6 +232,11 @@ def test_get_dataset(self):
         self.assertGreater(len(dataset.features), 1)
         self.assertGreater(len(dataset.qualities), 4)
 
+        # Issue324 Properly handle private datasets when trying to access them
+        openml.config.server = self.production_server
+        self.assertRaises(PrivateDatasetError, openml.datasets.get_dataset, 45)
+
+
     def test_get_dataset_with_string(self):
         dataset = openml.datasets.get_dataset(101)
         self.assertRaises(PyOpenMLError, dataset._get_arff, 'arff')