style: make all mutable function arguments "None" and init in function body

Author: jfyuen

dataikuapi/dss/macro.py

@@ -33,7 +33,7 @@ def get_definition(self):
         return self.definition
 
 
-    def run(self, params={}, admin_params={}, wait=True):
+    def run(self, params=None, admin_params=None, wait=True):
         """
         Run the macro from the project
 
@@ -43,6 +43,10 @@ def run(self, params={}, admin_params={}, wait=True):
         :param wait: if True, the call blocks until the run is finished
         :returns: a run identifier to use to abort or retrieve results
         """
+        if params is None:
+            params = {}
+        if admin_params is None:
+            admin_params = {}
         return self.client._perform_json(
             "POST", "/projects/%s/runnables/%s" % (self.project_key, self.runnable_type), 
             params={'wait':wait}, body={'params':params, 'adminParams':admin_params})['runId']

dataikuapi/dss/ml.py

@@ -1255,7 +1255,7 @@ def train(self, session_name=None, session_description=None):
         self.wait_train_complete()
         return self.get_trained_models_ids(session_id = train_ret["sessionId"])
 
-    def ensemble(self, model_ids=[], method=None):
+    def ensemble(self, model_ids=None, method=None):
         """
         Create an ensemble model of a set of models
         
@@ -1273,6 +1273,8 @@ def ensemble(self, model_ids=[], method=None):
         :return: A model identifier
         :rtype: string
         """
+        if model_ids is None:
+            model_ids = []
         train_ret = self.start_ensembling(model_ids, method)
         self.wait_train_complete()
         return train_ret
@@ -1296,7 +1298,7 @@ def start_train(self, session_name=None, session_description=None):
                 "POST", "/projects/%s/models/lab/%s/%s/train" % (self.project_key, self.analysis_id, self.mltask_id), body=session_info)
 
 
-    def start_ensembling(self, model_ids=[], method=None):
+    def start_ensembling(self, model_ids=None, method=None):
         """
         Creates asynchronously a new ensemble models of a set of models.
 
@@ -1308,6 +1310,8 @@ def start_ensembling(self, model_ids=[], method=None):
         :return: the model identifier of the ensemble
         :rtype: string
         """
+        if model_ids is None:
+            model_ids = []
         ensembling_request = {
                             "method" : method,
                             "modelsIds" : model_ids

dataikuapi/dss/project.py

@@ -49,7 +49,7 @@ def delete(self, drop_data=False):
     # Project export
     ########################################################
 
-    def get_export_stream(self, options = {}):
+    def get_export_stream(self, options=None):
         """
         Return a stream of the exported project
         You need to close the stream after download. Failure to do so will result in the DSSClient becoming unusable.
@@ -70,15 +70,19 @@ def get_export_stream(self, options = {}):
         :returns: a file-like obbject that is a stream of the export archive
         :rtype: file-like
         """
+        if options is None:
+            options = {}
         return self.client._perform_raw(
             "POST", "/projects/%s/export" % self.project_key, body=options).raw
 
-    def export_to_file(self, path, options={}):
+    def export_to_file(self, path, options=None):
         """
         Export the project to a file
         
         :param str path: the path of the file in which the exported project should be saved
         """
+        if options is None:
+            options = {}
         with open(path, 'wb') as f:
             export_stream = self.client._perform_raw(
                 "POST", "/projects/%s/export" % self.project_key, body=options)
@@ -98,7 +102,7 @@ def duplicate(self, target_project_key,
                   export_saved_models=True,
                   export_git_repository=True,
                   export_insights_data=True,
-                  remapping={},
+                  remapping=None,
                   target_project_folder=None):
         """
         Duplicate the project
@@ -116,7 +120,8 @@ def duplicate(self, target_project_key,
         :returns: A dict containing the original and duplicated project's keys
         :rtype: :class:`ProjectDuplicateResult`
         """
-
+        if remapping is None:
+            remapping = {}
         obj = {
             "targetProjectName": target_project_name,
             "targetProjectKey": target_project_key,
@@ -211,7 +216,7 @@ def get_dataset(self, dataset_name):
         return DSSDataset(self.client, self.project_key, dataset_name)
 
     def create_dataset(self, dataset_name, type,
-                params={}, formatType=None, formatParams={}):
+                params=None, formatType=None, formatParams=None):
         """
         Create a new dataset in the project, and return a handle to interact with it.
 
@@ -232,6 +237,10 @@ def create_dataset(self, dataset_name, type,
         Returns:
             A :class:`dataikuapi.dss.dataset.DSSDataset` dataset handle
         """
+        if params is None:
+            params = {}
+        if formatParams is None:
+            formatParams = {}
         obj = {
             "name" : dataset_name,
             "projectKey" : self.project_key,
@@ -676,7 +685,7 @@ def get_scenario(self, scenario_id):
         """
         return DSSScenario(self.client, self.project_key, scenario_id)
 
-    def create_scenario(self, scenario_name, type, definition={'params': {}}):
+    def create_scenario(self, scenario_name, type, definition=None):
         """
         Create a new scenario in the project, and return a handle to interact with it
 
@@ -688,6 +697,8 @@ def create_scenario(self, scenario_name, type, definition={'params': {}}):
 
         :returns: a :class:`.scenario.DSSScenario` handle to interact with the newly-created scenario
         """
+        if definition is None:
+            definition = {'params': {}}
         definition['type'] = type
         definition['name'] = scenario_name
         scenario_id = self.client._perform_json("POST", "/projects/%s/scenarios/" % self.project_key,
@@ -790,12 +801,14 @@ def get_tags(self):
         """
         return self.client._perform_json("GET", "/projects/%s/tags" % self.project_key)
 
-    def set_tags(self, tags={}):
+    def set_tags(self, tags=None):
         """
         Set the tags of this project.
         @param obj: must be a modified version of the object returned by list_tags
         """
-        return self.client._perform_empty("PUT", "/projects/%s/tags" % self.project_key, body = tags)
+        if tags is None:
+            tags = {}
+        return self.client._perform_empty("PUT", "/projects/%s/tags" % self.project_key, body=tags)
 
 
     ########################################################

dataikuapi/dss/scenario.py

@@ -20,12 +20,14 @@ def abort(self):
         return self.client._perform_json(
             "POST", "/projects/%s/scenarios/%s/abort" % (self.project_key, self.id))
 
-    def run(self, params={}):
+    def run(self, params=None):
         """
         Requests a run of the scenario, which will start after a few seconds.
 
         :params dict params: additional parameters that will be passed to the scenario through trigger params
         """
+        if params is None:
+            params = {}
         trigger_fire = self.client._perform_json(
             "POST", "/projects/%s/scenarios/%s/run" % (self.project_key, self.id), body=params)
         return DSSTriggerFire(self, trigger_fire)
@@ -47,7 +49,7 @@ def get_trigger_fire(self, trigger_id, trigger_run_id):
             })
         return DSSTriggerFire(self, trigger_fire)
 
-    def run_and_wait(self, params={}, no_fail=False):
+    def run_and_wait(self, params=None, no_fail=False):
         """
         Requests a run of the scenario, which will start after a few seconds. Wait the end of the run to complete.
 
@@ -57,6 +59,8 @@ def run_and_wait(self, params={}, no_fail=False):
         Returns:
             A :class:`dataikuapi.dss.admin.DSSScenarioRun` run handle
         """
+        if params is None:
+            params = {}
         trigger_fire = self.run(params)
         scenario_run = trigger_fire.wait_for_scenario_run(no_fail)
         waiter = DSSScenarioRunWaiter(scenario_run, trigger_fire)

dataikuapi/dssclient.py

@@ -245,7 +245,7 @@ def get_plugin(self, plugin_id):
     # SQL queries
     ########################################################
 
-    def sql_query(self, query, connection=None, database=None, dataset_full_name=None, pre_queries=None, post_queries=None, type='sql', extra_conf={}, script_steps=None, script_input_schema=None, script_output_schema=None, script_report_location=None, read_timestamp_without_timezone_as_string=True, read_date_as_string=False):
+    def sql_query(self, query, connection=None, database=None, dataset_full_name=None, pre_queries=None, post_queries=None, type='sql', extra_conf=None, script_steps=None, script_input_schema=None, script_output_schema=None, script_report_location=None, read_timestamp_without_timezone_as_string=True, read_date_as_string=False):
         """
         Initiate a SQL, Hive or Impala query and get a handle to retrieve the results of the query.
         Internally, the query is run by DSS. The  database to run the query on is specified either by 
@@ -262,6 +262,8 @@ def sql_query(self, query, connection=None, database=None, dataset_full_name=Non
         
         :returns: A :class:`dataikuapi.dss.sqlquery.DSSSQLQuery` query handle
         """
+        if extra_conf is None:
+            extra_conf = {}
         return DSSSQLQuery(self, query, connection, database, dataset_full_name, pre_queries, post_queries, type, extra_conf, script_steps, script_input_schema, script_output_schema, script_report_location, read_timestamp_without_timezone_as_string, read_date_as_string)
 
     ########################################################
@@ -290,7 +292,7 @@ def get_user(self, login):
         """
         return DSSUser(self, login)
 
-    def create_user(self, login, password, display_name='', source_type='LOCAL', groups=[], profile='DATA_SCIENTIST'):
+    def create_user(self, login, password, display_name='', source_type='LOCAL', groups=None, profile='DATA_SCIENTIST'):
         """
         Create a user, and return a handle to interact with it
 
@@ -305,6 +307,8 @@ def create_user(self, login, password, display_name='', source_type='LOCAL', gro
 
         :return: A :class:`dataikuapi.dss.admin.DSSUser` user handle
         """
+        if groups is None:
+            groups = []
         resp = self._perform_text(
                "POST", "/admin/users/", body={
                    "login" : login,
@@ -386,7 +390,7 @@ def get_connection(self, name):
         """
         return DSSConnection(self, name)
 
-    def create_connection(self, name, type, params={}, usable_by='ALL', allowed_groups=[]):
+    def create_connection(self, name, type, params=None, usable_by='ALL', allowed_groups=None):
         """
         Create a connection, and return a handle to interact with it
 
@@ -402,6 +406,10 @@ def create_connection(self, name, type, params={}, usable_by='ALL', allowed_grou
         
         :returns: A :class:`dataikuapi.dss.admin.DSSConnection` connection handle
         """
+        if params is None:
+            params = {}
+        if allowed_groups is None:
+            allowed_groups = []
         resp = self._perform_text(
                "POST", "/admin/connections/", body={
                    "name" : name,
@@ -659,13 +667,15 @@ def get_log(self, name):
         return self._perform_json(
             "GET", "/admin/logs/%s" % name)
 
-    def log_custom_audit(self, custom_type, custom_params={}):
+    def log_custom_audit(self, custom_type, custom_params=None):
         """
         Log a custom entry to the audit trail
         
         :param str custom_type value for customMsgType in audit trail item
         :param dict custom_params value for customMsgParams in audit trail item
         """
+        if custom_params is None:
+            custom_params = {}
         return self._perform_empty("POST",
             "/admin/audit/custom/%s" % custom_type,
             body = custom_params)
@@ -780,10 +790,12 @@ def get_apideployer(self):
     # Data Catalog
     ########################################################
 
-    def catalog_index_connections(self, connection_names=[], all_connections=False, indexing_mode="FULL"):
+    def catalog_index_connections(self, connection_names=None, all_connections=False, indexing_mode="FULL"):
         """
         Triggers an indexing of multiple connections in the data catalog
         """
+        if connection_names is None:
+            connection_names = []
         return self._perform_json("POST", "/catalog/index", body={
             "connectionNames": connection_names,
             "indexAllConnections": all_connections,
@@ -939,7 +951,7 @@ def __init__(self, client, import_id):
         self.client = client
         self.import_id = import_id
 
-    def execute(self, settings = {}):
+    def execute(self, settings=None):
         """
         Executes the import with provided settings.
 
@@ -966,7 +978,7 @@ def execute(self, settings = {}):
         @warning: You must check the 'success' flag
         """
         # Empty JSON dicts can't be parsed properly
-        if settings == {}:
-            settings["_"] = "_"
+        if settings is None:
+            settings = {"_": "_"}
         return self.client._perform_json("POST", "/projects/import/%s/process" % (self.import_id),
             body = settings)