FIX #152 do not return status code

Author: mfeurer

openml/_api_calls.py

@@ -84,7 +84,7 @@ def _read_url_files(url, data=None, file_dictionary=None, file_elements=None):
     if 'Content-Encoding' not in response.headers or \
             response.headers['Content-Encoding'] != 'gzip':
         warnings.warn('Received uncompressed content from OpenML for %s.' % url)
-    return response.status_code, response.text
+    return response.text
 
 
 def _read_url(url, data=None):
@@ -101,4 +101,4 @@ def _read_url(url, data=None):
     if 'Content-Encoding' not in response.headers or \
             response.headers['Content-Encoding'] != 'gzip':
         warnings.warn('Received uncompressed content from OpenML for %s.' % url)
-    return response.status_code, response.text
+    return response.text

openml/datasets/dataset.py

@@ -318,9 +318,8 @@ def publish(self):
         if self.data_file is not None:
             file_dictionary['dataset'] = self.data_file
 
-        return_code, return_value = _perform_api_call(
-            "/data/", file_dictionary=file_dictionary,
-            file_elements=file_elements)
+        return_value = _perform_api_call("/data/", file_dictionary=file_dictionary,
+                                         file_elements=file_elements)
 
         self.dataset_id = int(xmltodict.parse(return_value)['oml:upload_data_set']['oml:id'])
         return self

openml/datasets/functions.py

@@ -159,7 +159,7 @@ def list_datasets(offset=None, size=None, tag=None):
 
 def _list_datasets(api_call):
     # TODO add proper error handling here!
-    return_code, xml_string = _perform_api_call(api_call)
+    xml_string = _perform_api_call(api_call)
     datasets_dict = xmltodict.parse(xml_string)
 
     # Minimalistic check if the XML is useful
@@ -301,8 +301,7 @@ def _get_dataset_description(did_cache_dir, dataset_id):
     try:
         return _get_cached_dataset_description(dataset_id)
     except (OpenMLCacheException):
-        return_code, dataset_xml = _perform_api_call(
-            "data/%d" % dataset_id)
+        dataset_xml = _perform_api_call("data/%d" % dataset_id)
 
         with io.open(description_file, "w", encoding='utf8') as fh:
             fh.write(dataset_xml)
@@ -347,7 +346,7 @@ def _get_dataset_arff(did_cache_dir, description):
         pass
 
     url = description['oml:url']
-    return_code, arff_string = _read_url(url)
+    arff_string = _read_url(url)
 
     with io.open(output_file_path, "w", encoding='utf8') as fh:
         fh.write(arff_string)
@@ -382,8 +381,7 @@ def _get_dataset_features(did_cache_dir, dataset_id):
         with io.open(features_file, encoding='utf8') as fh:
             features_xml = fh.read()
     except (OSError, IOError):
-        return_code, features_xml = _perform_api_call(
-            "data/features/%d" % dataset_id)
+        features_xml = _perform_api_call("data/features/%d" % dataset_id)
 
         with io.open(features_file, "w", encoding='utf8') as fh:
             fh.write(features_xml)
@@ -417,8 +415,7 @@ def _get_dataset_qualities(did_cache_dir, dataset_id):
         with io.open(qualities_file, encoding='utf8') as fh:
             qualities_xml = fh.read()
     except (OSError, IOError):
-        return_code, qualities_xml = _perform_api_call(
-            "data/qualities/%d" % dataset_id)
+        qualities_xml = _perform_api_call("data/qualities/%d" % dataset_id)
 
         with io.open(qualities_file, "w", encoding='utf8') as fh:
             fh.write(qualities_xml)

openml/flows/flow.py

@@ -335,8 +335,7 @@ def publish(self):
 
         xml_description = self._to_xml()
         file_elements = {'description': xml_description}
-        return_code, return_value = _perform_api_call(
-            "flow/", file_elements=file_elements)
+        return_value = _perform_api_call("flow/", file_elements=file_elements)
         self.flow_id = int(xmltodict.parse(return_value)['oml:upload_flow']['oml:id'])
         return self
 
@@ -351,7 +350,7 @@ def _ensure_flow_exists(self):
         flow_id : int
             Flow id on the server.
         """
-        _, _, flow_id = _check_flow_exists(self.name, self.external_version)
+        _, flow_id = _check_flow_exists(self.name, self.external_version)
         # TODO add numpy and scipy version!
 
         if int(flow_id) == -1:
@@ -385,15 +384,12 @@ def _check_flow_exists(name, version):
     if not (type(version) is str and len(version) > 0):
         raise ValueError('Argument \'version\' should be a non-empty string')
 
-    return_code, xml_response = _perform_api_call(
-        "flow/exists", data={'name': name, 'external_version': version})
-    # TODO check with latest version of code if this raises an exception
-    if return_code != 200:
-        # fixme raise appropriate error
-        raise ValueError("api call failed: %s" % xml_response)
+    xml_response = _perform_api_call("flow/exists",
+                                     data={'name': name, 'external_version': version})
+
     xml_dict = xmltodict.parse(xml_response)
     flow_id = xml_dict['oml:flow_exists']['oml:id']
-    return return_code, xml_response, flow_id
+    return xml_response, flow_id
 
 
 def _add_if_nonempty(dic, key, value):

openml/flows/functions.py

@@ -18,7 +18,7 @@ def get_flow(flow_id):
     except:
         raise ValueError("Flow ID must be an int, got %s." % str(flow_id))
 
-    return_code, flow_xml = _perform_api_call("flow/%d" % flow_id)
+    flow_xml = _perform_api_call("flow/%d" % flow_id)
 
     flow_dict = xmltodict.parse(flow_xml)
     flow = OpenMLFlow._from_dict(flow_dict)
@@ -71,7 +71,7 @@ def list_flows(offset=None, size=None, tag=None):
 
 def _list_datasets(api_call):
     # TODO add proper error handling here!
-    return_code, xml_string = _perform_api_call(api_call)
+    xml_string = _perform_api_call(api_call)
     flows_dict = xmltodict.parse(xml_string)
 
     # Minimalistic check if the XML is useful

openml/runs/functions.py

@@ -159,7 +159,7 @@ def get_run(run_id):
         return _get_cached_run(run_id)
     except (OpenMLCacheException):
         try:
-            return_code, run_xml = _perform_api_call("run/%d" % run_id)
+            run_xml = _perform_api_call("run/%d" % run_id)
         except (URLError, UnicodeEncodeError) as e:
             # TODO logger.debug
             print(e)
@@ -328,7 +328,7 @@ def list_runs(offset=None, size=None, id=None, task=None,
 def _list_runs(api_call):
     """Helper function to parse API calls which are lists of runs"""
 
-    return_code, xml_string = _perform_api_call(api_call)
+    xml_string = _perform_api_call(api_call)
 
     runs_dict = xmltodict.parse(xml_string)
     # Minimalistic check if the XML is useful

openml/runs/run.py

@@ -146,8 +146,7 @@ def publish(self):
             trace_arff = arff.dumps(self._generate_trace_arff_dict(self.model))
             file_elements['trace'] = ("trace.arff", trace_arff)
 
-        return_code, return_value = _perform_api_call(
-            "/run/", file_elements=file_elements)
+        return_value = _perform_api_call("/run/", file_elements=file_elements)
         run_id = int(xmltodict.parse(return_value)['oml:upload_run']['oml:run_id'])
         self.run_id = run_id
         return self

openml/tasks/functions.py

@@ -58,8 +58,7 @@ def _get_estimation_procedure_list():
         task type id, name, type, repeats, folds, stratified.
     """
 
-    return_code, xml_string = _perform_api_call(
-        "estimationprocedure/list")
+    xml_string = _perform_api_call("estimationprocedure/list")
     procs_dict = xmltodict.parse(xml_string)
     # Minimalistic check if the XML is useful
     if 'oml:estimationprocedures' not in procs_dict:
@@ -128,7 +127,7 @@ def list_tasks(task_type_id=None, offset=None, size=None, tag=None):
 
 
 def _list_tasks(api_call):
-    return_code, xml_string = _perform_api_call(api_call)
+    xml_string = _perform_api_call(api_call)
     with open('/tmp/list_tasks.xml', 'w') as fh:
         fh.write(xml_string)
     tasks_dict = xmltodict.parse(xml_string)
@@ -203,8 +202,7 @@ def get_task(task_id):
     except (OSError, IOError):
 
         try:
-            return_code, task_xml = _perform_api_call(
-                "task/%d" % task_id)
+            task_xml = _perform_api_call("task/%d" % task_id)
         except (URLError, UnicodeEncodeError) as e:
             print(e)
             raise e

openml/tasks/task.py

@@ -71,7 +71,7 @@ def _download_split(self, cache_file):
         except (OSError, IOError):
             split_url = self.estimation_procedure["data_splits_url"]
             try:
-                return_code, split_arff = _read_url(split_url)
+                split_arff = _read_url(split_url)
             except (URLError, UnicodeEncodeError) as e:
                 print(e, split_url)
                 raise e

tests/test_flows/test_flow.py

@@ -105,7 +105,7 @@ def test_from_xml_to_xml(self):
         # Get the raw xml thing
         # TODO maybe get this via get_flow(), which would have to be refactored to allow getting only the xml dictionary
         for flow_id in [1185, 1244, 1196, 1112, ]:
-            flow_xml = _perform_api_call("flow/%d" % flow_id)[1]
+            flow_xml = _perform_api_call("flow/%d" % flow_id)
             flow_dict = xmltodict.parse(flow_xml)
 
             flow = openml.OpenMLFlow._from_dict(flow_dict)