Files
edx-platform/lms/djangoapps/course_api/views.py
Calen Pennington 6c596f7c49 Allow courses api to return data incrementally
Prior to this commit, the course api (/api/courses/v1/courses/)
performed all the work necessary to return all courses available
to the user, and then only actually returned on page's worth of those
courses.

With this change, the api now does the work incrementally, computing
only the data needed to fetch the courses up to and including the page
being returned. This still increases approximately linearly as
the page number accessed being increases, but should be more cache-friendly.
One side effect of this is that the max_page reported by pagination
will be an overestimate (it will include pages that are removed due
to a users access restrictions).

This change also changes the sort-order of courses being returned by the
course_api. By sorting by course-id, rather than course-number, we
can sort in the database, rather than in Python, and defer loading data
from the end of the list until it is requested.

REVMI-90
2019-02-06 10:11:15 -05:00

277 lines
10 KiB
Python

"""
Course API Views
"""
import search
from django.conf import settings
from django.core.exceptions import ValidationError
from rest_framework.generics import ListAPIView, RetrieveAPIView
from rest_framework.throttling import UserRateThrottle
from edx_rest_framework_extensions.paginators import NamespacedPageNumberPagination
from openedx.core.lib.api.view_utils import DeveloperErrorViewMixin, view_auth_classes
from openedx.core.lib.api.view_utils import LazySequence
from . import USE_RATE_LIMIT_2_FOR_COURSE_LIST_API, USE_RATE_LIMIT_10_FOR_COURSE_LIST_API
from .api import course_detail, list_courses
from .forms import CourseDetailGetForm, CourseListGetForm
from .serializers import CourseDetailSerializer, CourseSerializer
@view_auth_classes(is_authenticated=False)
class CourseDetailView(DeveloperErrorViewMixin, RetrieveAPIView):
"""
**Use Cases**
Request details for a course
**Example Requests**
GET /api/courses/v1/courses/{course_key}/
**Response Values**
Body consists of the following fields:
* effort: A textual description of the weekly hours of effort expected
in the course.
* end: Date the course ends, in ISO 8601 notation
* enrollment_end: Date enrollment ends, in ISO 8601 notation
* enrollment_start: Date enrollment begins, in ISO 8601 notation
* id: A unique identifier of the course; a serialized representation
of the opaque key identifying the course.
* media: An object that contains named media items. Included here:
* course_image: An image to show for the course. Represented
as an object with the following fields:
* uri: The location of the image
* name: Name of the course
* number: Catalog number of the course
* org: Name of the organization that owns the course
* overview: A possibly verbose HTML textual description of the course.
Note: this field is only included in the Course Detail view, not
the Course List view.
* short_description: A textual description of the course
* start: Date the course begins, in ISO 8601 notation
* start_display: Readably formatted start of the course
* start_type: Hint describing how `start_display` is set. One of:
* `"string"`: manually set by the course author
* `"timestamp"`: generated from the `start` timestamp
* `"empty"`: no start date is specified
* pacing: Course pacing. Possible values: instructor, self
Deprecated fields:
* blocks_url: Used to fetch the course blocks
* course_id: Course key (use 'id' instead)
**Parameters:**
username (optional):
The username of the specified user for whom the course data
is being accessed. The username is not only required if the API is
requested by an Anonymous user.
**Returns**
* 200 on success with above fields.
* 400 if an invalid parameter was sent or the username was not provided
for an authenticated request.
* 403 if a user who does not have permission to masquerade as
another user specifies a username other than their own.
* 404 if the course is not available or cannot be seen.
Example response:
{
"blocks_url": "/api/courses/v1/blocks/?course_id=edX%2Fexample%2F2012_Fall",
"media": {
"course_image": {
"uri": "/c4x/edX/example/asset/just_a_test.jpg",
"name": "Course Image"
}
},
"description": "An example course.",
"end": "2015-09-19T18:00:00Z",
"enrollment_end": "2015-07-15T00:00:00Z",
"enrollment_start": "2015-06-15T00:00:00Z",
"course_id": "edX/example/2012_Fall",
"name": "Example Course",
"number": "example",
"org": "edX",
"overview: "<p>A verbose description of the course.</p>"
"start": "2015-07-17T12:00:00Z",
"start_display": "July 17, 2015",
"start_type": "timestamp",
"pacing": "instructor"
}
"""
serializer_class = CourseDetailSerializer
def get_object(self):
"""
Return the requested course object, if the user has appropriate
permissions.
"""
requested_params = self.request.query_params.copy()
requested_params.update({'course_key': self.kwargs['course_key_string']})
form = CourseDetailGetForm(requested_params, initial={'requesting_user': self.request.user})
if not form.is_valid():
raise ValidationError(form.errors)
return course_detail(
self.request,
form.cleaned_data['username'],
form.cleaned_data['course_key'],
)
class CourseListUserThrottle(UserRateThrottle):
"""Limit the number of requests users can make to the course list API."""
# The course list endpoint is likely being inefficient with how it's querying
# various parts of the code and can take courseware down, it needs to be rate
# limited until optimized. LEARNER-5527
THROTTLE_RATES = {
'user': '20/minute',
'staff': '40/minute',
}
def check_for_switches(self):
if USE_RATE_LIMIT_2_FOR_COURSE_LIST_API.is_enabled():
self.THROTTLE_RATES = {
'user': '2/minute',
'staff': '10/minute',
}
elif USE_RATE_LIMIT_10_FOR_COURSE_LIST_API.is_enabled():
self.THROTTLE_RATES = {
'user': '10/minute',
'staff': '20/minute',
}
def allow_request(self, request, view):
self.check_for_switches()
# Use a special scope for staff to allow for a separate throttle rate
user = request.user
if user.is_authenticated and (user.is_staff or user.is_superuser):
self.scope = 'staff'
self.rate = self.get_rate()
self.num_requests, self.duration = self.parse_rate(self.rate)
return super(CourseListUserThrottle, self).allow_request(request, view)
@view_auth_classes(is_authenticated=False)
class CourseListView(DeveloperErrorViewMixin, ListAPIView):
"""
**Use Cases**
Request information on all courses visible to the specified user.
**Example Requests**
GET /api/courses/v1/courses/
**Response Values**
Body comprises a list of objects as returned by `CourseDetailView`.
**Parameters**
search_term (optional):
Search term to filter courses (used by ElasticSearch).
username (optional):
The username of the specified user whose visible courses we
want to see. The username is not required only if the API is
requested by an Anonymous user.
org (optional):
If specified, visible `CourseOverview` objects are filtered
such that only those belonging to the organization with the
provided org code (e.g., "HarvardX") are returned.
Case-insensitive.
mobile (optional):
If specified, only visible `CourseOverview` objects that are
designated as mobile_available are returned.
**Returns**
* 200 on success, with a list of course discovery objects as returned
by `CourseDetailView`.
* 400 if an invalid parameter was sent or the username was not provided
for an authenticated request.
* 403 if a user who does not have permission to masquerade as
another user specifies a username other than their own.
* 404 if the specified user does not exist, or the requesting user does
not have permission to view their courses.
Example response:
[
{
"blocks_url": "/api/courses/v1/blocks/?course_id=edX%2Fexample%2F2012_Fall",
"media": {
"course_image": {
"uri": "/c4x/edX/example/asset/just_a_test.jpg",
"name": "Course Image"
}
},
"description": "An example course.",
"end": "2015-09-19T18:00:00Z",
"enrollment_end": "2015-07-15T00:00:00Z",
"enrollment_start": "2015-06-15T00:00:00Z",
"course_id": "edX/example/2012_Fall",
"name": "Example Course",
"number": "example",
"org": "edX",
"start": "2015-07-17T12:00:00Z",
"start_display": "July 17, 2015",
"start_type": "timestamp"
}
]
"""
pagination_class = NamespacedPageNumberPagination
pagination_class.max_page_size = 100
serializer_class = CourseSerializer
throttle_classes = (CourseListUserThrottle,)
# Return all the results, 10K is the maximum allowed value for ElasticSearch.
# We should use 0 after upgrading to 1.1+:
# - https://github.com/elastic/elasticsearch/commit/8b0a863d427b4ebcbcfb1dcd69c996c52e7ae05e
results_size_infinity = 10000
def get_queryset(self):
"""
Yield courses visible to the user.
"""
form = CourseListGetForm(self.request.query_params, initial={'requesting_user': self.request.user})
if not form.is_valid():
raise ValidationError(form.errors)
db_courses = list_courses(
self.request,
form.cleaned_data['username'],
org=form.cleaned_data['org'],
filter_=form.cleaned_data['filter_'],
)
if not settings.FEATURES['ENABLE_COURSEWARE_SEARCH'] or not form.cleaned_data['search_term']:
return db_courses
search_courses = search.api.course_discovery_search(
form.cleaned_data['search_term'],
size=self.results_size_infinity,
)
search_courses_ids = {course['data']['id'] for course in search_courses['results']}
return LazySequence(
(
course for course in db_courses
if unicode(course.id) in search_courses_ids
),
est_len=len(db_courses)
)