ADD /task/list/tag/{tag} api call

Author: mfeurer

doc/progress.rst

@@ -24,8 +24,8 @@ API call                                        implemented tested properly test
 
 /task/{task}                                    yes         yes
 /task/list                                      yes         yes
-/task/list/type/{id}
-/task/list/tag/{tag}
+/task/list/type/{id}                            yes         yes
+/task/list/tag/{tag}                            yes         yes
 /task {POST}
 /task/tag
 /task/untag

openml/tasks/__init__.py

@@ -1,6 +1,7 @@
 from .task import OpenMLTask
 from .split import OpenMLSplit
-from .task_functions import get_task, list_tasks, list_tasks_by_type
+from .task_functions import get_task, list_tasks, list_tasks_by_type, \
+    list_tasks_by_tag
 
 __all__ = ['OpenMLTask', 'get_task', 'list_tasks', 'list_tasks_by_type',
-           'OpenMLSplit']
+           'list_tasks_by_tag', 'OpenMLSplit']

openml/tasks/task_functions.py

@@ -106,6 +106,24 @@ def list_tasks_by_type(task_type_id):
     return _list_tasks("task/list/type/%d" % task_type_id)
 
 
+def list_tasks_by_tag(tag):
+    """Return all tasks having the given tag
+
+    Parameters
+    ----------
+    tag : str
+
+    Returns
+    -------
+    list
+        A list of all tasks having a give tag. Every task is represented by
+        a dictionary containing the following information: task id,
+        dataset id, task_type and status. If qualities are calculated for
+        the associated dataset, some of these are also returned.
+    """
+    return _list_tasks("task/list/tag/%s" % tag)
+
+
 def list_tasks():
     """Return a list of all tasks which are on OpenML.
 
@@ -124,9 +142,18 @@ def _list_tasks(api_call):
     return_code, xml_string = _perform_api_call(api_call)
     tasks_dict = xmltodict.parse(xml_string)
     # Minimalistic check if the XML is useful
-    assert tasks_dict['oml:tasks']['@xmlns:oml'] == \
-        'http://openml.org/openml'
-    assert type(tasks_dict['oml:tasks']['oml:task']) == list
+    if 'oml:tasks' not in tasks_dict:
+        raise ValueError('Error in return XML, does not contain "oml:runs": %s'
+                         % str(tasks_dict))
+    elif '@xmlns:oml' not in tasks_dict['oml:tasks']:
+        raise ValueError('Error in return XML, does not contain '
+                         '"oml:runs"/@xmlns:oml: %s'
+                         % str(tasks_dict))
+    elif tasks_dict['oml:tasks']['@xmlns:oml'] != 'http://openml.org/openml':
+        raise ValueError('Error in return XML, value of  '
+                         '"oml:runs"/@xmlns:oml is not '
+                         '"http://openml.org/openml": %s'
+                         % str(tasks_dict))
 
     tasks = []
     procs = get_estimation_procedure_list()

tests/examples/test_OpenMLDemo.py

@@ -45,7 +45,7 @@ def test_notebook(self):
                 msg = 'Error executing the notebook "%s". ' % notebook_filename
                 msg += 'See notebook "%s" for the traceback.\n\n' % notebook_filename_out
                 msg += e.traceback
-                self.fail(msg)
+                 self.fail(msg)
             finally:
                 with open(notebook_filename_out, mode='wt') as f:
                     nbformat.write(nb, f)

tests/test_task.py

@@ -29,6 +29,12 @@ def test_list_tasks_by_type(self):
         for task in tasks:
             self._check_task(task)
 
+    def test_list_tasks_by_tag(self):
+        tasks = openml.tasks.list_tasks_by_tag('basic')
+        self.assertGreaterEqual(len(tasks), 57)
+        for task in tasks:
+            self._check_task(task)
+
     def test_get_task(self):
         task = openml.tasks.get_task(1)
         self.assertTrue(os.path.exists(