group results by service using itertools

allows us to nicely reuse the existing format_statistics function

app/dao/services_dao.py

@@ -187,6 +187,7 @@ def dao_fetch_todays_stats_for_all_services():
     return db.session.query(
         Notification.notification_type,
         Notification.status,
+        Notification.service_id,
         func.count(Notification.id).label('count')
     ).select_from(
         Service
@@ -199,4 +200,6 @@ def dao_fetch_todays_stats_for_all_services():
         Notification.notification_type,
         Notification.status,
         Notification.service_id
+    ).order_by(
+        Notification.service_id
     )

app/service/rest.py

@@ -1,3 +1,5 @@
+import itertools
+
 from flask import (
     jsonify,
     request,
@@ -21,7 +23,8 @@ from app.dao.services_dao import (
     dao_remove_user_from_service,
     dao_fetch_stats_for_service,
     dao_fetch_todays_stats_for_service,
-    dao_fetch_weekly_historical_stats_for_service
+    dao_fetch_weekly_historical_stats_for_service,
+    dao_fetch_todays_stats_for_all_services
 )
 from app.dao import notifications_dao
 from app.dao.provider_statistics_dao import get_fragment_count
@@ -51,9 +54,9 @@ def get_services():
     user_id = request.args.get('user_id', None)
     if user_id:
         services = dao_fetch_all_services_by_user(user_id)
+    elif request.args.get('detailed') == 'True':
+        return get_detailed_services()
     else:
-        if request.args.get('detailed') == 'True':
-            return get_detailed_services()
         services = dao_fetch_all_services()
     data = service_schema.dump(services, many=True).data
     return jsonify(data=data)
@@ -251,12 +254,10 @@ def get_detailed_service(service_id, today_only=False):
 
 def get_detailed_services():
     services = {service.id: service for service in dao_fetch_all_services()}
-    stats = dao_fetch_todays_stats_for_all_services(service_id)
+    stats = dao_fetch_todays_stats_for_all_services()
 
-    for row in stats:
-        services[row.service_id].statistics
-        # todo: how do we separate rows of statistics by service?
-        service.statistics = statistics.format_statistics(stats)
+    for service_id, rows in itertools.groupby(stats, lambda x: x.service_id):
+        services[service_id].statistics = statistics.format_statistics(rows)
 
-    data = detailed_service_schema.dump(service).data
+    data = detailed_service_schema.dump(service, many=True).data
     return jsonify(data=data)

tests/app/notifications/test_rest.py

@@ -207,7 +207,7 @@ def test_normal_api_key_returns_notifications_created_from_jobs_and_from_api(
 
         notifications = json.loads(response.get_data(as_text=True))['notifications']
         assert len(notifications) == 2
-        assert set(x['id'] for x in notifications) == set([str(sample_notification.id), str(api_notification.id)])
+        assert set(x['id'] for x in notifications) == {str(sample_notification.id), str(api_notification.id)}
 
 
 @pytest.mark.parametrize('key_type', [KEY_TYPE_NORMAL, KEY_TYPE_TEAM, KEY_TYPE_TEST])