[FC-005] feat: add necessary models for Openedx survey report

lms/envs/common.py

@@ -3239,6 +3239,9 @@ def _make_locale_paths(settings):  # pylint: disable=missing-function-docstring
     # Agreements
     'openedx.core.djangoapps.agreements',
 
+    # Survey reports
+    'openedx.features.survey_report',
+
     # User and group management via edx-django-utils
     'edx_django_utils.user',
 

lms/envs/production.py

@@ -1080,3 +1080,6 @@ def get_env_setting(setting):
     "SECRET": ENV_TOKENS.get('BIG_BLUE_BUTTON_GLOBAL_SECRET', None),
     "URL": ENV_TOKENS.get('BIG_BLUE_BUTTON_GLOBAL_URL', None),
 }
+
+############## Settings for survey report ##############
+SURVEY_REPORT_EXTRA_DATA = ENV_TOKENS.get('SURVEY_REPORT_EXTRA_DATA', {})

lms/envs/test.py

@@ -671,3 +671,6 @@
         "LOGO_URL": "https://courses.example.com/yourmfe-logo.png",
     },
 }
+
+############## Settings for survey report ##############
+SURVEY_REPORT_EXTRA_DATA = {}

openedx/features/survey_report/README.rst

@@ -0,0 +1,11 @@
+Survey Report
+--------------------
+This Django app was created for the purpose of gathering aggregated, anonymized data
+about Open edX courses at scale, so that we can begin to track the growth
+and trends in Open edX usage over time, namely in the annual Open edX
+Impact Report.
+
+You could find in this directory some methods to manage survey
+reports, one command to generate the report, some queries to get the
+information from database and one method to send the report to openedx
+api.

openedx/features/survey_report/api.py

@@ -0,0 +1,37 @@
+"""
+Contains the logic to manage survey report model.
+"""
+
+from django.conf import settings
+
+from openedx.features.survey_report.models import SurveyReport
+from openedx.features.survey_report.queries import (
+    get_course_enrollments,
+    get_recently_active_users,
+    get_generated_certificates,
+    get_registered_learners,
+    get_unique_courses_offered
+)
+
+MAX_WEEKS_SINCE_LAST_LOGIN: int = 4
+
+
+def generate_report() -> None:
+    """ Generate a report with relevant data."""
+    courses_offered = get_unique_courses_offered()
+    learners = get_recently_active_users(weeks=MAX_WEEKS_SINCE_LAST_LOGIN)
+    registered_learners = get_registered_learners()
+    certificates = get_generated_certificates()
+    enrollments = get_course_enrollments()
+    extra_data = settings.SURVEY_REPORT_EXTRA_DATA
+
+    survey_report = SurveyReport(
+        courses_offered=courses_offered,
+        learners=learners,
+        registered_learners=registered_learners,
+        generated_certificates=certificates,
+        enrollments=enrollments,
+        extra_data=extra_data,
+    )
+
+    survey_report.save()

openedx/features/survey_report/apps.py

@@ -0,0 +1,12 @@
+"""
+Survey Report App Configuration.
+"""
+from django.apps import AppConfig
+
+
+class SurveyReportConfig(AppConfig):
+    """
+    Configuration for the survey report Django app.
+    """
+    default_auto_field = 'django.db.models.BigAutoField'
+    name = 'openedx.features.survey_report'

openedx/features/survey_report/docs/decisions/0001-addition-to-core-repo.rst

@@ -0,0 +1,29 @@
+Addition of the Survey Report App to edx-platform
+=================================================
+
+Status
+------
+Accepted
+
+Context
+-------
+The transition to a more modular architecture for edx-platorm has been
+strengthened by the acceptance of the `No new Django apps ADR`_.
+
+.. _No new Django apps ADR: https://github.com/openedx/edx-platform/tree/master/docs/decisions/0014-no-new-apps.rst
+
+
+Rationale
+---------
+
+This feature was considered for inclusion into the edx-platform code because it
+imports several models from the inner workings of the core functionality in
+order to query them. This goes in accordance with the section further guidance
+of the ADR.
+
+
+Decision
+--------
+
+Locate the Survey Report Application in the edx-platform repository under
+`openedx/features`.

openedx/features/survey_report/management/commands/generate_report.py

@@ -0,0 +1,31 @@
+"""
+CLI command to generate survey report.
+"""
+
+from django.core.management.base import BaseCommand, CommandError
+
+from openedx.features.survey_report.api import generate_report
+
+
+class Command(BaseCommand):
+    """
+    Management command to generate a new survey report with
+    non-sensitive data.
+    """
+
+    help = """
+        This command will create a new survey report using some
+        models to get:
+        - Total number of courses offered
+        - Currently active learners
+        ...
+        learners ever registered, and generated certificates.
+        """
+
+    def handle(self, *_args, **_options):
+        try:
+            generate_report()
+        except Exception as error:
+            raise CommandError(f'An error has occurred while survey report was generating. {error}') from error
+
+        self.stdout.write(self.style.SUCCESS('Survey report has been generated successfully.'))

openedx/features/survey_report/management/commands/tests/test_generate_report.py

@@ -0,0 +1,44 @@
+"""
+Test for generate_report command.
+"""
+
+from io import StringIO
+from unittest import mock
+
+from django.core.management import call_command
+from django.test import TestCase, override_settings
+
+from openedx.features.survey_report.models import SurveyReport
+
+
+class GenerateReportTest(TestCase):
+    """
+    Test for generate_report command.
+    """
+    @override_settings(SURVEY_REPORT_EXTRA_DATA={'extra_data': 'extra_data'})
+    @mock.patch('openedx.features.survey_report.queries.get_course_enrollments')
+    @mock.patch('openedx.features.survey_report.queries.get_generated_certificates')
+    @mock.patch('openedx.features.survey_report.queries.get_registered_learners')
+    @mock.patch('openedx.features.survey_report.queries.get_recently_active_users')
+    @mock.patch('openedx.features.survey_report.queries.get_unique_courses_offered')
+    def test_generate_report(self, mock_get_unique_courses_offered, mock_get_recently_active_users,
+                             mock_get_registered_learners, mock_get_generated_certificates,
+                             mock_get_course_enrollments):
+        """
+        Test that generate_report command creates a survey report.
+        """
+        mock_get_unique_courses_offered.return_value = 1
+        mock_get_recently_active_users.return_value = 2
+        mock_get_registered_learners.return_value = 3
+        mock_get_generated_certificates.return_value = 4
+        mock_get_course_enrollments.return_value = 5
+        out = StringIO()
+        call_command('generate_report', stdout=out)
+
+        survey_report = SurveyReport.objects.last()
+        assert survey_report.courses_offered == 1
+        assert survey_report.learners == 2
+        assert survey_report.registered_learners == 3
+        assert survey_report.generated_certificates == 4
+        assert survey_report.enrollments == 5
+        assert survey_report.extra_data == {'extra_data': 'extra_data'}

openedx/features/survey_report/migrations/0001_initial.py

@@ -0,0 +1,32 @@
+# Generated by Django 3.2.16 on 2022-11-03 20:07
+
+from django.db import migrations, models
+import jsonfield.fields
+
+
+class Migration(migrations.Migration):
+
+    initial = True
+
+    dependencies = [
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='SurveyReport',
+            fields=[
+                ('id', models.BigAutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
+                ('courses_offered', models.BigIntegerField()),
+                ('learners', models.BigIntegerField()),
+                ('registered_learners', models.BigIntegerField()),
+                ('enrollments', models.BigIntegerField()),
+                ('generated_certificates', models.BigIntegerField()),
+                ('extra_data', jsonfield.fields.JSONField(blank=True, default=dict, help_text='Extra information for instance data')),
+                ('created_at', models.DateTimeField(auto_now=True)),
+            ],
+            options={
+                'ordering': ['-created_at'],
+                'get_latest_by': 'created_at',
+            },
+        ),
+    ]

openedx/features/survey_report/models.py

@@ -0,0 +1,37 @@
+"""
+Survey Report models.
+"""
+
+from django.db import models
+from jsonfield import JSONField
+
+
+class SurveyReport(models.Model):
+    """
+    This model stores information to automate the way of gathering impact data from the openedx project.
+
+    .. no_pii:
+
+    fields:
+    - courses_offered: Total number of active unique courses.
+    - learner: Recently active users with login in some weeks.
+    - registered_learners: Total number of users ever registered in the platform.
+    - enrollments: Total number of active enrollments in the platform.
+    - generated_certificates: Total number of generated certificates.
+    - extra_data: Extra information that will be saved in the report, E.g: site_name, openedx-release.
+    """
+    courses_offered = models.BigIntegerField()
+    learners = models.BigIntegerField()
+    registered_learners = models.BigIntegerField()
+    enrollments = models.BigIntegerField()
+    generated_certificates = models.BigIntegerField()
+    extra_data = JSONField(
+        blank=True,
+        default=dict,
+        help_text="Extra information for instance data",
+    )
+    created_at = models.DateTimeField(auto_now=True)
+
+    class Meta:
+        ordering = ["-created_at"]
+        get_latest_by = 'created_at'

openedx/features/survey_report/queries.py

@@ -0,0 +1,92 @@
+"""
+Queries to get data from database.
+"""
+
+import logging
+from datetime import datetime, timedelta
+
+from django.contrib.auth.models import User  # lint-amnesty, pylint: disable=imported-auth-user
+from django.db.models import Count, OuterRef, Q, Subquery
+
+from common.djangoapps.util.query import read_replica_or_default
+from common.djangoapps.student.models import CourseEnrollment
+from lms.djangoapps.grades.models import PersistentCourseGrade
+from openedx.core.djangoapps.content.course_overviews.models import CourseOverview
+
+log = logging.getLogger(__name__)
+
+MIN_ENROLLS_ACTIVE_COURSE: int = 5
+
+
+def get_unique_courses_offered() -> int:
+    """
+    Get total number of unique course that started before today and have an open date,
+    or have not finished yet, whose number of enrollments is greater than MIN_ENROLLS_ACTIVE_COURSE.
+    """
+    log.info("Getting the total number of unique courses offered...")
+    total = CourseOverview.objects.annotate(
+        count=Subquery(
+            CourseEnrollment.objects
+                            .filter(course_id=OuterRef('id'))
+                            .values('course_id')
+                            .annotate(count=Count('course_id'))
+                            .values('count')
+        ))\
+        .filter(count__gt=MIN_ENROLLS_ACTIVE_COURSE)\
+        .filter(start__lt=datetime.now())\
+        .filter(Q(end__isnull=True) | Q(end__gt=datetime.now()))\
+        .using(read_replica_or_default())\
+        .count()
+    log.info("Getting the total number of unique courses offered... DONE")
+    return total
+
+
+def get_recently_active_users(weeks: int) -> int:
+    """
+    Get total number of users with last login in the last weeks.
+
+    Args:
+        weeks (int): number of weeks since the last login to considerate as an active learner.
+    """
+    log.info("Getting the total number of recently active users...")
+    total = User.objects.filter(last_login__gte=datetime.now() - timedelta(weeks=weeks))\
+        .using(read_replica_or_default())\
+        .count()
+    log.info("Getting total number of recently active users... DONE")
+    return total
+
+
+def get_registered_learners() -> int:
+    """
+    Get total number of active learners registered.
+    """
+    log.info("Getting the total number of ever registered learners...")
+    total = User.objects.filter(is_active=True)\
+        .using(read_replica_or_default())\
+        .count()
+    log.info("Getting the total number of ever registered learners... DONE")
+    return total
+
+
+def get_generated_certificates() -> int:
+    """
+    Get total number of generated certificates.
+    """
+    log.info("Getting the total number of generated certificates...")
+    total = PersistentCourseGrade.objects.filter(passed_timestamp__isnull=False)\
+        .using(read_replica_or_default())\
+        .count()
+    log.info("Getting the total number of generated certificates... DONE")
+    return total
+
+
+def get_course_enrollments() -> int:
+    """
+    Get total number of enrollments from users that aren't staff. Course staff members will be included.
+    """
+    log.info("Getting the total number of course enrollments...")
+    total = CourseEnrollment.objects.filter(is_active=True, user__is_superuser=False, user__is_staff=False)\
+        .using(read_replica_or_default())\
+        .count()
+    log.info("Getting the total number of course enrollments... DONE")
+    return total