Added fields for learner endpoints.

analytics_data_api/v0/models.py

@@ -3,7 +3,9 @@ from itertools import groupby
 from django.conf import settings
 from django.db import models
 from django.db.models import Sum
-from elasticsearch_dsl import DocType, Q
+# some fields (e.g. Float, Integer) are dynamic and your IDE may highlight them as unavailable
+from elasticsearch_dsl import Date, DocType, Float, Integer, Q, String
+
 
 from analytics_data_api.constants import country, engagement_entity_types, genders, learner
 
@@ -214,6 +216,27 @@ class Video(BaseVideo):
 
 
 class RosterEntry(DocType):
+
+    course_id = String()
+    username = String()
+    name = String()
+    email = String()
+    enrollment_mode = String()
+    cohort = String()
+    segments = String(fields={'raw': String()})
+    problems_attempted = Integer()
+    problems_completed = Integer()
+    problem_attempts_per_completed = Float()
+    # Useful for ordering problem_attempts_per_completed (because results can include null, which is
+    # different from zero).  attempt_ratio_order is equal to the number of problem attempts if
+    # problem_attempts_per_completed is > 1 and set to -problem_attempts if
+    # problem_attempts_per_completed = 1.
+    attempt_ratio_order = Integer()
+    discussions_contributed = Integer()
+    videos_watched = Integer()
+    enrollment_date = Date()
+    last_updated = Date()
+
     # pylint: disable=old-style-class
     class Meta:
         index = settings.ELASTICSEARCH_LEARNERS_INDEX
@@ -230,8 +253,7 @@ class RosterEntry(DocType):
             course_id,
             segments=None,
             ignore_segments=None,
-            # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-            # cohort=None,
+            cohort=None,
             enrollment_mode=None,
             text_search=None,
             order_by='username',
@@ -251,13 +273,14 @@ class RosterEntry(DocType):
                     segment=segment, segments=', '.join(learner.SEGMENTS)
                 ))
         order_by_options = (
-            'username', 'email', 'discussions_contributed', 'problems_attempted', 'problems_completed', 'videos_viewed'
+            'username', 'email', 'discussions_contributed', 'problems_attempted', 'problems_completed',
+            'attempt_ratio_order', 'videos_viewed'
         )
-        sort_order_options = ('asc', 'desc')
         if order_by not in order_by_options:
             raise ValueError("order_by value '{order_by}' must be one of: ({order_by_options})".format(
                 order_by=order_by, order_by_options=', '.join(order_by_options)
             ))
+        sort_order_options = ('asc', 'desc')
         if sort_order not in sort_order_options:
             raise ValueError("sort_order value '{sort_order}' must be one of: ({sort_order_options})".format(
                 sort_order=sort_order, sort_order_options=', '.join(sort_order_options)
@@ -272,9 +295,8 @@ class RosterEntry(DocType):
         elif ignore_segments:
             for segment in ignore_segments:
                 search = search.query(~Q('term', segments=segment))
-        # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-        # if cohort:
-        #     search = search.query('term', cohort=cohort)
+        if cohort:
+            search = search.query('term', cohort=cohort)
         if enrollment_mode:
             search = search.query('term', enrollment_mode=enrollment_mode)
         if text_search:
@@ -309,8 +331,7 @@ class RosterEntry(DocType):
         search.query = Q('bool', must=[Q('term', course_id=course_id)])
         search.aggs.bucket('enrollment_modes', 'terms', field='enrollment_mode')
         search.aggs.bucket('segments', 'terms', field='segments')
-        # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-        # search.aggs.bucket('group_by_cohorts', 'terms', field='cohort')
+        search.aggs.bucket('cohorts', 'terms', field='cohort')
         response = search.execute()
         # Build up the map of aggregation name to count
         aggregations = {

analytics_data_api/v0/serializers.py

@@ -317,7 +317,7 @@ class VideoTimelineSerializer(ModelSerializerWithCreatedField):
         )
 
 
-class LearnerSerializer(serializers.Serializer):
+class LearnerSerializer(serializers.Serializer, DefaultIfNoneMixin):
     username = serializers.CharField()
     enrollment_mode = serializers.CharField()
     name = serializers.CharField()
@@ -325,11 +325,9 @@ class LearnerSerializer(serializers.Serializer):
     email = serializers.CharField()
     segments = serializers.Field(source='segments')
     engagements = serializers.SerializerMethodField('get_engagements')
-
-    # TODO: add these back in when the index returns them
-    # enrollment_date = serializers.DateField(format=settings.DATE_FORMAT, allow_empty=True)
-    # last_updated = serializers.DateField(format=settings.DATE_FORMAT)
-    # cohort = serializers.CharField(allow_none=True)
+    enrollment_date = serializers.DateField(format=settings.DATE_FORMAT)
+    last_updated = serializers.DateField(format=settings.DATE_FORMAT)
+    cohort = serializers.CharField()
 
     def get_account_url(self, obj):
         if settings.LMS_USER_ACCOUNT_BASE_URL:
@@ -342,10 +340,17 @@ class LearnerSerializer(serializers.Serializer):
         Add the engagement totals.
         """
         engagements = {}
-        for entity_type in engagement_entity_types.AGGREGATE_TYPES:
-            for event in engagement_events.EVENTS[entity_type]:
-                metric = '{0}_{1}'.format(entity_type, event)
-                engagements[metric] = getattr(obj, metric, 0)
+
+        # fill in these fields will 0 if values not returned/found
+        default_if_none_fields = ['discussions_contributed', 'problems_attempted',
+                                  'problems_completed', 'videos_viewed']
+        for field in default_if_none_fields:
+            engagements[field] = self.default_if_none(getattr(obj, field, None), 0)
+
+        # preserve null values for problem attempts per completed
+        no_default_field = 'problem_attempts_per_completed'
+        engagements[no_default_field] = getattr(obj, no_default_field, None)
+
         return engagements
 
 
@@ -424,8 +429,7 @@ class EnagementRangeMetricSerializer(serializers.Serializer):
 class CourseLearnerMetadataSerializer(serializers.Serializer):
     enrollment_modes = serializers.SerializerMethodField('get_enrollment_modes')
     segments = serializers.SerializerMethodField('get_segments')
-    # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-    # cohorts = serializers.SerializerMethodField('get_cohorts')
+    cohorts = serializers.SerializerMethodField('get_cohorts')
     engagement_ranges = serializers.SerializerMethodField('get_engagement_ranges')
 
     def get_enrollment_modes(self, obj):
@@ -434,9 +438,8 @@ class CourseLearnerMetadataSerializer(serializers.Serializer):
     def get_segments(self, obj):
         return obj['es_data']['segments']
 
-    # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-    # def get_cohorts(self, obj):
-    #     return obj['es_data']['cohorts']
+    def get_cohorts(self, obj):
+        return obj['es_data']['cohorts']
 
     def get_engagement_ranges(self, obj):
         query_set = obj['engagement_ranges']

analytics_data_api/v0/views/learners.py

@@ -174,14 +174,21 @@ class LearnerListView(CourseViewMixin, generics.ListAPIView):
         """
         self._validate_query_params()
         query_params = self.request.QUERY_PARAMS
+
+        # Ordering by problem_attempts_per_completed can be ambiguous because
+        # it's a ratio values could be infinite (e.g. divide by zero) if no problems
+        # were completed.  Instead, sorting by attempt_ratio_order will produce
+        # a sensible ordering
+        order_by = query_params.get('order_by')
+        order_by = 'attempt_ratio_order' if order_by == 'problem_attempts_per_completed' else order_by
+
         params = {
             'segments': split_query_argument(query_params.get('segments')),
             'ignore_segments': split_query_argument(query_params.get('ignore_segments')),
-            # TODO: enable during https://openedx.atlassian.net/browse/AN-6319
-            # 'cohort': query_params.get('cohort'),
+            'cohort': query_params.get('cohort'),
             'enrollment_mode': query_params.get('enrollment_mode'),
             'text_search': query_params.get('text_search'),
-            'order_by': query_params.get('order_by'),
+            'order_by': order_by,
             'sort_order': query_params.get('sort_order')
         }
         # Remove None values from `params` so that we don't overwrite default
@@ -211,6 +218,7 @@ class EngagementTimelineView(CourseViewMixin, generics.ListAPIView):
                 * problems_completed: Unique number of problems completed.
                 * discussions_contributed: Number of discussions participated in (e.g. forum posts)
                 * videos_viewed: Number of videos watched.
+                * problem_attempts_per_completed: TBD
 
     **Parameters**