feat: [AXM-200] Implement user's enrolments status API

[KyryloKireiev]

Description

UserEnrollmentsStatus - implement user's enrolments status API.
UserEnrollmentsStatus - add test case for the testing API.

YouTrack

https://youtrack.raccoongang.com/issue/AXM-200/Active-inactive-status-for-User-enrollments

[monteri]

LGTM, check comments, please

[monteri]

[info] I think it is enough to have before border value, after border value and border value itself. Maybe we need to add 30 also. It is called Boundary Testing principle

[monteri]

But no action needed here

[KyryloKireiev]

I agree, to have before border value, after border value it would be enough.

However, it is possible that we would get a problem if we used the 30 day limit and ran the tests at approximately 00:00. So I removed this value. But this can also be easily handled using a decorator @freeze_time('2018-01-01').

What do you think? Do I need to refactor these tests? It's pretty simple

[monteri]

no, it is alright

[monteri]

Suggested change

	def get(self, request, *args, **kwargs) -> Response:
	"""
	Gets user's enrollments status.
	"""
	active_status_date = datetime.datetime.now(pytz.UTC) - datetime.timedelta(days=30)
	user_enrollments = CourseEnrollment.objects.filter(
	user__username=kwargs.get('username'),
	is_active=True,
	)
	mobile_available = [
	enrollment for enrollment in user_enrollments
	if is_mobile_available_for_user(self.request.user, enrollment.course_overview)
	]
	user_completions_last_month = BlockCompletion.objects.filter(
	user__username=kwargs.get('username'),
	created__gte=active_status_date
	)
	course_ids_where_user_has_completions = [
	str(completion.block_key.course_key) for completion in user_completions_last_month
	]
	enrollments_status = []
	for user_enrollment in mobile_available:
	course_id = str(user_enrollment.course_overview.id)
	enrollments_status.append(
	{
	'course_id': course_id,
	'course_name': user_enrollment.course_overview.display_name,
	'is_active': bool(
	course_id in course_ids_where_user_has_completions
	or user_enrollment.created > active_status_date
	)
	}
	)
	return Response(enrollments_status)
	from django.db.models import Q
	@mobile_view(is_user=True)
	class UserEnrollmentsStatus(views.APIView):
	# ... docstring with more context ...
	def get(self, request, *args, **kwargs) -> Response:
	"""Gets user's enrollments status."""
	enrollments = self._get_enrollments_with_progress(kwargs.get('username'))
	enrollments_status = self._build_enrollments_status(enrollments)
	return Response(enrollments_status)
	def _get_enrollments_with_progress(self, username):
	active_status_date = datetime.datetime.now(pytz.UTC) - datetime.timedelta(days=30)
	return CourseEnrollment.objects.filter(
	user__username=username,
	is_active=True,
	blockcompletion__created__gte=active_status_date
	).select_related('course_overview')
	def _build_enrollments_status(self, enrollments):
	enrollments_status = []
	for enrollment in enrollments:
	if is_mobile_available_for_user(self.request.user, enrollment.course_overview):
	# ... (rest of the logic) ...
	return enrollments_status

This is suggestion from Gemini:

• I agree with them on the point to divide this big function into smaller functions.
• In terms of Q and combining queries I think it requires check if this will work as expected. So it is up to you if you think this is worth to change. It is more like FYI how it can be optimized.

[KyryloKireiev]

Yes, I think it's a good idea. I'm going to divide this big function into smaller functions