Fix/strip whitespace

openml/_api_calls.py

@@ -296,7 +296,7 @@ def __parse_server_exception(
     if response.status_code == 414:
         raise OpenMLServerError("URI too long! ({})".format(url))
     try:
-        server_exception = xmltodict.parse(response.text)
+        server_exception = xmltodict.parse(response.text, strip_whitespace=False)
     except xml.parsers.expat.ExpatError:
         raise
     except Exception:

openml/base.py

@@ -130,7 +130,7 @@ def publish(self) -> "OpenMLBase":
         response_text = openml._api_calls._perform_api_call(
             call, "post", file_elements=file_elements
         )
-        xml_response = xmltodict.parse(response_text)
+        xml_response = xmltodict.parse(response_text, strip_whitespace=False)
 
         self._parse_publish_response(xml_response)
         return self

openml/datasets/dataset.py

@@ -934,7 +934,9 @@ def _read_features(features_file: str) -> Dict[int, OpenMLDataFeature]:
         with open(features_file, encoding="utf8") as fh:
             features_xml_string = fh.read()
         xml_dict = xmltodict.parse(
-            features_xml_string, force_list=("oml:feature", "oml:nominal_value")
+            features_xml_string,
+            strip_whitespace=False,
+            force_list=("oml:feature", "oml:nominal_value"),
         )
         features_xml = xml_dict["oml:data_features"]
 
@@ -970,7 +972,9 @@ def _read_qualities(qualities_file: str) -> Dict[str, float]:
     except:  # noqa E722
         with open(qualities_file, encoding="utf8") as fh:
             qualities_xml = fh.read()
-        xml_as_dict = xmltodict.parse(qualities_xml, force_list=("oml:quality",))
+        xml_as_dict = xmltodict.parse(
+            qualities_xml, strip_whitespace=False, force_list=("oml:quality",)
+        )
         qualities = xml_as_dict["oml:data_qualities"]["oml:quality"]
         qualities = _check_qualities(qualities)
         with open(qualities_pickle_file, "wb") as fh_binary:

openml/datasets/functions.py

@@ -52,7 +52,7 @@ def list_qualities() -> List[str]:
     """
     api_call = "data/qualities/list"
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    qualities = xmltodict.parse(xml_string, force_list=("oml:quality"))
+    qualities = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:quality"))
     # Minimalistic check if the XML is useful
     if "oml:data_qualities_list" not in qualities:
         raise ValueError("Error in return XML, does not contain " '"oml:data_qualities_list"')
@@ -181,7 +181,7 @@ def _list_datasets(data_id: Optional[List] = None, output_format="dict", **kwarg
 def __list_datasets(api_call, output_format="dict"):
 
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    datasets_dict = xmltodict.parse(xml_string, force_list=("oml:dataset",))
+    datasets_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:dataset",))
 
     # Minimalistic check if the XML is useful
     assert type(datasets_dict["oml:data"]["oml:dataset"]) == list, type(datasets_dict["oml:data"])
@@ -724,7 +724,7 @@ def status_update(data_id, status):
         raise ValueError("Illegal status value. " "Legal values: %s" % legal_status)
     data = {"data_id": data_id, "status": status}
     result_xml = openml._api_calls._perform_api_call("data/status/update", "post", data=data)
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     server_data_id = result["oml:data_status_update"]["oml:id"]
     server_status = result["oml:data_status_update"]["oml:status"]
     if status != server_status or int(data_id) != int(server_data_id):
@@ -832,7 +832,7 @@ def edit_dataset(
     result_xml = openml._api_calls._perform_api_call(
         "data/edit", "post", data=form_data, file_elements=file_elements
     )
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     data_id = result["oml:data_edit"]["oml:id"]
     return int(data_id)
 
@@ -871,7 +871,7 @@ def fork_dataset(data_id: int) -> int:
     # compose data fork parameters
     form_data = {"data_id": data_id}
     result_xml = openml._api_calls._perform_api_call("data/fork", "post", data=form_data)
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     data_id = result["oml:data_fork"]["oml:id"]
     return int(data_id)
 
@@ -891,7 +891,7 @@ def _topic_add_dataset(data_id: int, topic: str):
         raise TypeError("`data_id` must be of type `int`, not {}.".format(type(data_id)))
     form_data = {"data_id": data_id, "topic": topic}
     result_xml = openml._api_calls._perform_api_call("data/topicadd", "post", data=form_data)
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     data_id = result["oml:data_topic"]["oml:id"]
     return int(data_id)
 
@@ -912,7 +912,7 @@ def _topic_delete_dataset(data_id: int, topic: str):
         raise TypeError("`data_id` must be of type `int`, not {}.".format(type(data_id)))
     form_data = {"data_id": data_id, "topic": topic}
     result_xml = openml._api_calls._perform_api_call("data/topicdelete", "post", data=form_data)
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     data_id = result["oml:data_topic"]["oml:id"]
     return int(data_id)
 
@@ -951,7 +951,7 @@ def _get_dataset_description(did_cache_dir, dataset_id):
         with io.open(description_file, "w", encoding="utf8") as fh:
             fh.write(dataset_xml)
 
-    description = xmltodict.parse(dataset_xml)["oml:data_set_description"]
+    description = xmltodict.parse(dataset_xml, strip_whitespace=False)["oml:data_set_description"]
 
     return description
 
@@ -1205,7 +1205,7 @@ def _get_online_dataset_arff(dataset_id):
     # build a dict from the xml.
     # use the url from the dataset description and return the ARFF string
     return openml._api_calls._download_text_file(
-        xmltodict.parse(dataset_xml)["oml:data_set_description"]["oml:url"],
+        xmltodict.parse(dataset_xml, strip_whitespace=False)["oml:data_set_description"]["oml:url"],
     )
 
 
@@ -1225,4 +1225,6 @@ def _get_online_dataset_format(dataset_id):
     """
     dataset_xml = openml._api_calls._perform_api_call("data/%d" % dataset_id, "get")
     # build a dict from the xml and get the format from the dataset description
-    return xmltodict.parse(dataset_xml)["oml:data_set_description"]["oml:format"].lower()
+    return xmltodict.parse(dataset_xml, strip_whitespace=False)["oml:data_set_description"][
+        "oml:format"
+    ].lower()

openml/evaluations/functions.py

@@ -179,7 +179,7 @@ def _list_evaluations(
 def __list_evaluations(api_call, output_format="object"):
     """Helper function to parse API calls which are lists of runs"""
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    evals_dict = xmltodict.parse(xml_string, force_list=("oml:evaluation",))
+    evals_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:evaluation",))
     # Minimalistic check if the XML is useful
     if "oml:evaluations" not in evals_dict:
         raise ValueError(
@@ -265,7 +265,7 @@ def list_evaluation_measures() -> List[str]:
     """
     api_call = "evaluationmeasure/list"
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    qualities = xmltodict.parse(xml_string, force_list=("oml:measures"))
+    qualities = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:measures"))
     # Minimalistic check if the XML is useful
     if "oml:evaluation_measures" not in qualities:
         raise ValueError("Error in return XML, does not contain " '"oml:evaluation_measures"')

openml/flows/flow.py

@@ -397,7 +397,7 @@ def to_filesystem(self, output_directory: str) -> None:
     def from_filesystem(cls, input_directory) -> "OpenMLFlow":
         with open(os.path.join(input_directory, "flow.xml"), "r") as f:
             xml_string = f.read()
-        return OpenMLFlow._from_dict(xmltodict.parse(xml_string))
+        return OpenMLFlow._from_dict(xmltodict.parse(xml_string, strip_whitespace=False))
 
     def _parse_publish_response(self, xml_response: Dict):
         """ Parse the id from the xml_response and assign it to self. """

openml/flows/functions.py

@@ -256,7 +256,7 @@ def flow_exists(name: str, external_version: str) -> Union[int, bool]:
         "flow/exists", "post", data={"name": name, "external_version": external_version},
     )
 
-    result_dict = xmltodict.parse(xml_response)
+    result_dict = xmltodict.parse(xml_response, strip_whitespace=False)
     flow_id = int(result_dict["oml:flow_exists"]["oml:id"])
     if flow_id > 0:
         return flow_id
@@ -329,7 +329,7 @@ def get_flow_id(
 def __list_flows(api_call: str, output_format: str = "dict") -> Union[Dict, pd.DataFrame]:
 
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    flows_dict = xmltodict.parse(xml_string, force_list=("oml:flow",))
+    flows_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:flow",))
 
     # Minimalistic check if the XML is useful
     assert type(flows_dict["oml:flows"]["oml:flow"]) == list, type(flows_dict["oml:flows"])
@@ -538,4 +538,4 @@ def _create_flow_from_xml(flow_xml: str) -> OpenMLFlow:
     OpenMLFlow
     """
 
-    return OpenMLFlow._from_dict(xmltodict.parse(flow_xml))
+    return OpenMLFlow._from_dict(xmltodict.parse(flow_xml, strip_whitespace=False))

openml/runs/functions.py

@@ -768,9 +768,11 @@ def obtain_field(xml_obj, fieldname, from_server, cast=None):
         else:
             raise AttributeError("Run XML does not contain required (server) " "field: ", fieldname)
 
-    run = xmltodict.parse(xml, force_list=["oml:file", "oml:evaluation", "oml:parameter_setting"])[
-        "oml:run"
-    ]
+    run = xmltodict.parse(
+        xml,
+        strip_whitespace=False,
+        force_list=["oml:file", "oml:evaluation", "oml:parameter_setting"],
+    )["oml:run"]
     run_id = obtain_field(run, "oml:run_id", from_server, cast=int)
     uploader = obtain_field(run, "oml:uploader", from_server, cast=int)
     uploader_name = obtain_field(run, "oml:uploader_name", from_server)
@@ -1096,7 +1098,7 @@ def _list_runs(
 def __list_runs(api_call, output_format="dict"):
     """Helper function to parse API calls which are lists of runs"""
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    runs_dict = xmltodict.parse(xml_string, force_list=("oml:run",))
+    runs_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:run",))
     # Minimalistic check if the XML is useful
     if "oml:runs" not in runs_dict:
         raise ValueError('Error in return XML, does not contain "oml:runs": %s' % str(runs_dict))

openml/runs/trace.py

@@ -303,7 +303,9 @@ def trace_from_xml(cls, xml):
             Object containing the run id and a dict containing the trace
             iterations.
         """
-        result_dict = xmltodict.parse(xml, force_list=("oml:trace_iteration",))["oml:trace"]
+        result_dict = xmltodict.parse(
+            xml, strip_whitespace=False, force_list=("oml:trace_iteration",)
+        )["oml:trace"]
 
         run_id = result_dict["oml:run_id"]
         trace = OrderedDict()

openml/setups/functions.py

@@ -50,7 +50,7 @@ def setup_exists(flow) -> int:
     result = openml._api_calls._perform_api_call(
         "/setup/exists/", "post", file_elements=file_elements
     )
-    result_dict = xmltodict.parse(result)
+    result_dict = xmltodict.parse(result, strip_whitespace=False)
     setup_id = int(result_dict["oml:setup_exists"]["oml:id"])
     if setup_id > 0:
         return setup_id
@@ -65,7 +65,7 @@ def _get_cached_setup(setup_id):
     try:
         setup_file = os.path.join(setup_cache_dir, "description.xml")
         with io.open(setup_file, encoding="utf8") as fh:
-            setup_xml = xmltodict.parse(fh.read())
+            setup_xml = xmltodict.parse(fh.read(), strip_whitespace=False)
             setup = _create_setup_from_xml(setup_xml, output_format="object")
         return setup
 
@@ -103,7 +103,7 @@ def get_setup(setup_id):
         with io.open(setup_file, "w", encoding="utf8") as fh:
             fh.write(setup_xml)
 
-    result_dict = xmltodict.parse(setup_xml)
+    result_dict = xmltodict.parse(setup_xml, strip_whitespace=False)
     return _create_setup_from_xml(result_dict, output_format="object")
 
 
@@ -190,7 +190,7 @@ def _list_setups(setup=None, output_format="object", **kwargs):
 def __list_setups(api_call, output_format="object"):
     """Helper function to parse API calls which are lists of setups"""
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    setups_dict = xmltodict.parse(xml_string, force_list=("oml:setup",))
+    setups_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:setup",))
     openml_uri = "http://openml.org/openml"
     # Minimalistic check if the XML is useful
     if "oml:setups" not in setups_dict:

openml/study/functions.py

@@ -76,7 +76,9 @@ def _get_study(id_: Union[int, str], entity_type) -> BaseStudy:
         "oml:run_id",
         "oml:tag",  # legacy.
     )
-    result_dict = xmltodict.parse(xml_string, force_list=force_list_tags)["oml:study"]
+    result_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=force_list_tags)[
+        "oml:study"
+    ]
     study_id = int(result_dict["oml:id"])
     alias = result_dict["oml:alias"] if "oml:alias" in result_dict else None
     main_entity_type = result_dict["oml:main_entity_type"]
@@ -274,7 +276,7 @@ def update_study_status(study_id: int, status: str) -> None:
         raise ValueError("Illegal status value. " "Legal values: %s" % legal_status)
     data = {"study_id": study_id, "status": status}
     result_xml = openml._api_calls._perform_api_call("study/status/update", "post", data=data)
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     server_study_id = result["oml:study_status_update"]["oml:id"]
     server_status = result["oml:study_status_update"]["oml:status"]
     if status != server_status or int(study_id) != int(server_study_id):
@@ -354,7 +356,7 @@ def attach_to_study(study_id: int, run_ids: List[int]) -> int:
     uri = "study/%d/attach" % study_id
     post_variables = {"ids": ",".join(str(x) for x in run_ids)}
     result_xml = openml._api_calls._perform_api_call(uri, "post", post_variables)
-    result = xmltodict.parse(result_xml)["oml:study_attach"]
+    result = xmltodict.parse(result_xml, strip_whitespace=False)["oml:study_attach"]
     return int(result["oml:linked_entities"])
 
 
@@ -397,7 +399,7 @@ def detach_from_study(study_id: int, run_ids: List[int]) -> int:
     uri = "study/%d/detach" % study_id
     post_variables = {"ids": ",".join(str(x) for x in run_ids)}
     result_xml = openml._api_calls._perform_api_call(uri, "post", post_variables)
-    result = xmltodict.parse(result_xml)["oml:study_detach"]
+    result = xmltodict.parse(result_xml, strip_whitespace=False)["oml:study_detach"]
     return int(result["oml:linked_entities"])
 
 
@@ -567,7 +569,7 @@ def _list_studies(output_format="dict", **kwargs) -> Union[Dict, pd.DataFrame]:
 
 def __list_studies(api_call, output_format="object") -> Union[Dict, pd.DataFrame]:
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    study_dict = xmltodict.parse(xml_string, force_list=("oml:study",))
+    study_dict = xmltodict.parse(xml_string, strip_whitespace=False, force_list=("oml:study",))
 
     # Minimalistic check if the XML is useful
     assert type(study_dict["oml:study_list"]["oml:study"]) == list, type(

openml/tasks/functions.py

@@ -87,7 +87,7 @@ def _get_estimation_procedure_list():
     url_suffix = "estimationprocedure/list"
     xml_string = openml._api_calls._perform_api_call(url_suffix, "get")
 
-    procs_dict = xmltodict.parse(xml_string)
+    procs_dict = xmltodict.parse(xml_string, strip_whitespace=False)
     # Minimalistic check if the XML is useful
     if "oml:estimationprocedures" not in procs_dict:
         raise ValueError("Error in return XML, does not contain tag " "oml:estimationprocedures.")
@@ -232,7 +232,9 @@ def _list_tasks(task_type=None, output_format="dict", **kwargs):
 
 def __list_tasks(api_call, output_format="dict"):
     xml_string = openml._api_calls._perform_api_call(api_call, "get")
-    tasks_dict = xmltodict.parse(xml_string, force_list=("oml:task", "oml:input"))
+    tasks_dict = xmltodict.parse(
+        xml_string, strip_whitespace=False, force_list=("oml:task", "oml:input")
+    )
     # Minimalistic check if the XML is useful
     if "oml:tasks" not in tasks_dict:
         raise ValueError('Error in return XML, does not contain "oml:runs": %s' % str(tasks_dict))
@@ -405,7 +407,7 @@ def _create_task_from_xml(xml):
     -------
     OpenMLTask
     """
-    dic = xmltodict.parse(xml)["oml:task"]
+    dic = xmltodict.parse(xml, strip_whitespace=False)["oml:task"]
     estimation_parameters = dict()
     inputs = dict()
     # Due to the unordered structure we obtain, we first have to extract

openml/utils.py

@@ -131,7 +131,7 @@ def _tag_entity(entity_type, entity_id, tag, untag=False):
     post_variables = {"%s_id" % entity_type: entity_id, "tag": tag}
     result_xml = openml._api_calls._perform_api_call(uri, "post", post_variables)
 
-    result = xmltodict.parse(result_xml, force_list={"oml:tag"})[main_tag]
+    result = xmltodict.parse(result_xml, strip_whitespace=False, force_list={"oml:tag"})[main_tag]
 
     if "oml:tag" in result:
         return result["oml:tag"]
@@ -173,7 +173,7 @@ def _delete_entity(entity_type, entity_id):
 
     url_suffix = "%s/%d" % (entity_type, entity_id)
     result_xml = openml._api_calls._perform_api_call(url_suffix, "delete")
-    result = xmltodict.parse(result_xml)
+    result = xmltodict.parse(result_xml, strip_whitespace=False)
     if "oml:%s_delete" % entity_type in result:
         return True
     else: