refactor: Inject isArchivedByLearner into initial API response

kdmccormick · claude · kdmccormick · commit e9b51593ff8b · 2026-05-14T11:36:41.000-04:00
This both:
* demonstrates filter pipeline steps in a practical way which relates to
  the sample archive functionality.
* removes the need to call GET ..../course-archive-status/ in order to
  render the page.

The GET API still exists, as it's called to update the state after a learner
clicks Archive/Unarchive (could probably be factored away, but not important)

This removes the old sample filter pipeline step, which changed the course about
URL into an example.com URL.

Co-Authored-By: Claude Opus 4.7 &lt;noreply@anthropic.com&gt;
diff --git a/backend-plugin-sample/src/openedx_plugin_sample/pipeline.py b/backend-plugin-sample/src/openedx_plugin_sample/pipeline.py
@@ -38,119 +38,53 @@
 """  # pylint: disable=line-too-long
 
 import logging
-import re
 
+import crum
 from openedx_filters.filters import PipelineStep
 
+from .models import CourseArchiveStatus
+
 logger = logging.getLogger(__name__)
 
 
-class ChangeCourseAboutPageUrl(PipelineStep):
+class AddArchiveStatusToLearnerHomeCourseRun(PipelineStep):
     """
-    Filter to customize course about page URLs.
-
-    This filter demonstrates how to intercept and modify course about page URLs,
-    redirecting them to external sites or custom implementations.
-
-    Filter Hook Point:
-    This filter hooks into the course about page URL rendering process.
-    Register it for the filter: org.openedx.learning.course.about.render.started.v1
-
-    Registration Example (in settings/common.py)::
-
-        def plugin_settings(settings):
-            settings.OPEN_EDX_FILTERS_CONFIG = {
-                "org.openedx.learning.course.about.render.started.v1": {
-                    "pipeline": [
-                        "openedx_plugin_sample.pipeline.ChangeCourseAboutPageUrl"
-                    ],
-                    "fail_silently": False,
-                }
-            }
-
-    Filter Documentation:
-    - Available Filters: https://docs.openedx.org/projects/openedx-filters/en/latest/reference/filters.html
-    - PipelineStep: https://docs.openedx.org/projects/openedx-filters/en/latest/reference/filters-tooling.html#openedx_filters.filters.PipelineStep
-
-    Real-World Use Cases:
-    - Redirect to marketing site course pages
-    - Implement custom course discovery interfaces
-    - Add tracking parameters to URLs
-    - Route different course types to different platforms
-    - Implement A/B testing for course pages
+    Customize each courseRun within a Learner Dashboard's /init API response to include the CourseArchiveStatus.
     """  # noqa: E501
 
-    def run_filter(self, url, org, **kwargs):  # pylint: disable=arguments-differ
+    def run_filter(self, serialized_courserun, **kwargs):  # pylint: disable=arguments-differ
         """
-        Modify the course about page URL.
-
-        This method intercepts course about page URL generation and can modify
-        the destination URL based on business logic.
+        Insert `isArchivedByLearner` into one serialized courseRun for the Learner Home /init response.
 
         Args:
-            url (str): The original course about page URL
-            org (str): The organization/institution identifier
-            **kwargs: Additional context data from the platform
+            serialized_courserun (dict): One courseRun from the serializer. Reads
+                `courseId` (a course key string, e.g. "course-v1:edX+DemoX+Demo_Course");
+                all other fields are passed through unchanged.
 
         Returns:
-            dict: Dictionary with same parameter names as input
-                - url (str): Modified or original URL
-                - org (str): Organization identifier (usually unchanged)
-
-        Raises:
-            FilterException: If processing should be halted
-
-        Filter Requirements:
-        - Must return dictionary with keys matching input parameters
-        - Return None to skip this filter (let other filters run)
-        - Raise FilterException to halt pipeline execution
-        - Handle all input scenarios gracefully
-
-        URL Pattern Matching:
-        This implementation looks for Open edX course keys in the format:
-        course-v1:ORG+COURSE+RUN (e.g., course-v1:edX+DemoX+Demo_Course)
-
-        Documentation:
-        - run_filter method: https://docs.openedx.org/projects/openedx-filters/en/latest/reference/filters-tooling.html#openedx_filters.filters.PipelineStep.run_filter
+            dict: ``{"serialized_courserun": <updated dict>}``. The updated dict has the
+                same keys as the input plus `isArchivedByLearner` (bool) -- True iff a
+                CourseArchiveStatus row exists for the current request user and this
+                courseId with `is_archived=True`; False otherwise (including when no row
+                exists).
+
+        The current user is read from the active request via `crum`, so this filter only
+        runs meaningfully inside a request cycle. Note that `isArchivedByLearner` is
+        distinct from `isArchived`, which the platform sets based on whether the course
+        run itself has ended.
         """  # noqa: E501
-        # Extract course ID using Open edX course key pattern
-        # Course keys follow the format: course-v1:ORG+COURSE+RUN
-        pattern = r'(?P<course_id>course-v1:[^/]+)'
-
-        match = re.search(pattern, url)
-        if match:
-            course_id = match.group('course_id')
-
-            # Example: Redirect to external marketing site
-            new_url = f"https://example.com/new_about_page/{course_id}"
-
-            logger.debug(
-                f"Redirecting course about page for {course_id} from {url} to {new_url}"
-            )
-
-            # Return modified data
-            return {"url": new_url, "org": org}
-
-        # No course ID found - return original data unchanged
-        logger.debug(f"No course ID found in URL {url}, leaving unchanged")
-        return {"url": url, "org": org}
-
-        # Alternative patterns for different business logic:
-
-        # Organization-based routing:
-        # if org == "special_org":
-        #     new_url = f"https://special-site.com/courses/{course_id}"
-        #     return {"url": new_url, "org": org}
-
-        # Course type-based routing:
-        # if "MicroMasters" in course_id:
-        #     new_url = f"https://micromasters.example.com/{course_id}"
-        #     return {"url": new_url, "org": org}
-
-        # A/B testing implementation:
-        # import random
-        # if random.choice([True, False]):
-        #     new_url = f"https://variant-a.example.com/{course_id}"
-        # else:
-        #     new_url = f"https://variant-b.example.com/{course_id}"
-        # return {"url": new_url, "org": org}
+        request = crum.get_current_request()
+        if not (request and request.user):
+            return serialized_courserun
+        try:
+            is_archived_by_learner = CourseArchiveStatus.objects.get(
+                user=request.user, course_id=serialized_courserun["courseId"]
+            ).is_archived
+        except CourseArchiveStatus.DoesNotExist:
+            is_archived_by_learner = False
+        return {
+            "serialized_courserun": {
+                **serialized_courserun,
+                "isArchivedByLearner": is_archived_by_learner,
+            },
+        }
diff --git a/backend-plugin-sample/src/openedx_plugin_sample/settings/common.py b/backend-plugin-sample/src/openedx_plugin_sample/settings/common.py
@@ -11,7 +11,8 @@
 configuration that integrates seamlessly with the platform.
 
 Official Documentation:
-- Plugin Settings: https://docs.openedx.org/projects/edx-django-utils/en/latest/plugins/how_tos/how_to_create_a_plugin_app.html#plugin-settings
+- Plugin Settings:
+  https://docs.openedx.org/projects/edx-django-utils/en/latest/plugins/how_tos/how_to_create_a_plugin_app.html#plugin-settings
 - Django Settings: https://docs.djangoproject.com/en/stable/topics/settings/
 
 Settings Organization:
@@ -96,8 +97,8 @@ def _configure_openedx_filters(settings):
     filters_config = getattr(settings, 'OPEN_EDX_FILTERS_CONFIG', {})
 
     # Filter we want to register
-    filter_name = "org.openedx.learning.course_about.page.url.requested.v1"
-    our_pipeline_step = "openedx_plugin_sample.pipeline.ChangeCourseAboutPageUrl"
+    filter_name = "org.openedx.learning.home.courserun.api.rendered.started.v1"
+    our_pipeline_step = "openedx_plugin_sample.pipeline.AddArchiveStatusToLearnerHomeCourseRun"
 
     # Check if this filter already has configuration
     if filter_name in filters_config:
diff --git a/backend-plugin-sample/tests/test_pipeline.py b/backend-plugin-sample/tests/test_pipeline.py
@@ -0,0 +1,101 @@
+#!/usr/bin/env python
+# pylint: disable=redefined-outer-name
+"""
+Tests for the `sample-plugin` Open edX Filters pipeline steps.
+"""
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+from django.contrib.auth import get_user_model
+from opaque_keys.edx.keys import CourseKey
+
+from openedx_plugin_sample.models import CourseArchiveStatus
+from openedx_plugin_sample.pipeline import AddArchiveStatusToLearnerHomeCourseRun
+
+User = get_user_model()
+
+
+@pytest.fixture
+def user():
+    """
+    Create and return a test user.
+    """
+    return User.objects.create_user(
+        username="testuser", email="testuser@example.com", password="password123"
+    )
+
+
+@pytest.fixture
+def course_key():
+    """
+    Create and return a test course key.
+    """
+    return CourseKey.from_string("course-v1:edX+DemoX+Demo_Course")
+
+
+@pytest.fixture
+def serialized_courserun(course_key):
+    """
+    Return a minimal courseRun dict like the learner home /init API would emit.
+    """
+    return {
+        "courseId": str(course_key),
+        "courseNumber": "DemoX",
+    }
+
+
+@pytest.fixture
+def mock_current_request(user):
+    """
+    Patch crum.get_current_request so the filter sees `user` as the requester.
+
+    The filter relies on `crum` to find the current user, which is set by middleware
+    in a real request cycle. In unit tests we stub it directly.
+    """
+    request = MagicMock()
+    request.user = user
+    with patch(
+        "openedx_plugin_sample.pipeline.crum.get_current_request",
+        return_value=request,
+    ):
+        yield request
+
+
+@pytest.mark.django_db
+def test_archived_courserun_gets_is_archived_by_learner_true(
+    user, course_key, serialized_courserun, mock_current_request  # pylint: disable=unused-argument
+):
+    """
+    Test that the filter adds isArchivedByLearner=True when the learner has
+    archived this course.
+    """
+    CourseArchiveStatus.objects.create(
+        course_id=course_key, user=user, is_archived=True
+    )
+
+    result = AddArchiveStatusToLearnerHomeCourseRun(
+        filter_type="org.openedx.learning.home.courserun.api.rendering.started.v1",
+        running_pipeline=[],
+    ).run_filter(serialized_courserun=serialized_courserun)
+
+    assert result["serialized_courserun"]["isArchivedByLearner"] is True
+    # Existing fields on the courseRun are preserved.
+    assert result["serialized_courserun"]["courseId"] == str(course_key)
+    assert result["serialized_courserun"]["courseNumber"] == "DemoX"
+
+
+@pytest.mark.django_db
+def test_courserun_with_no_archive_record_defaults_to_false(
+    serialized_courserun, mock_current_request  # pylint: disable=unused-argument
+):
+    """
+    Test that the filter defaults isArchivedByLearner to False when the learner
+    has no CourseArchiveStatus row for the course.
+    """
+    result = AddArchiveStatusToLearnerHomeCourseRun(
+        filter_type="org.openedx.learning.home.courserun.api.rendering.started.v1",
+        running_pipeline=[],
+    ).run_filter(serialized_courserun=serialized_courserun)
+
+    assert result["serialized_courserun"]["isArchivedByLearner"] is False
diff --git a/frontend-plugin-sample/README.md b/frontend-plugin-sample/README.md
@@ -90,33 +90,35 @@ const courses = courseListData.visibleList;
 
 **Slot Props**: Each slot provides specific data. For CourseListSlot, see the [slot documentation](https://github.com/openedx/frontend-app-learner-dashboard/tree/master/src/plugin-slots/CourseListSlot#plugin-props).
 
-#### 2. Backend API Integration
+#### 2. Backend Data via the Filter Pipeline
 
-```jsx
-useEffect(() => {
-  const fetchArchivedCourses = async () => {
-    const client = getAuthenticatedHttpClient();
-    const lmsBaseUrl = getConfig().LMS_BASE_URL;
-
-    const response = await client.get(
-      `${lmsBaseUrl}/sample-plugin/api/v1/course-archive-status/`,
-      { params: { is_archived: true } }
-    );
-
-    const archivedCourseIds = new Set(
-      response.data.results.map((item) => item.course_id)
-    );
-    setArchivedCourses(archivedCourseIds);
-  };
+Rather than firing an extra GET to `course-archive-status/` on every dashboard
+load, the initial archive state is read directly off the slot props. The backend
+plugin uses an Open edX filter (see [`pipeline.py`](../backend-plugin-sample/src/openedx_plugin_sample/pipeline.py))
+to inject `isArchivedByLearner` into each courseRun in the Learner Home `/init`
+API response, so it arrives alongside the rest of the course data:
 
-  fetchArchivedCourses();
-}, []);
+```jsx
+const [archivedCourses, setArchivedCourses] = useState(() => {
+  const initial = new Set();
+  (courseListData?.visibleList || []).forEach((courseData) => {
+    if (courseData.courseRun?.isArchivedByLearner) {
+      initial.add(courseData.courseRun.courseId);
+    }
+  });
+  return initial;
+});
 ```
 
+**Why this pattern**: One fewer round-trip per dashboard load, and the archive
+state is consistent with the rest of the course data from the same response.
+The REST API is still used for writes (archive/unarchive) — see the toggle
+handler below.
+
 **Key Patterns:**
-- **Authentication**: `getAuthenticatedHttpClient()` handles Open edX auth
+- **Filter-injected data**: Read `courseRun.isArchivedByLearner` straight from slot props
+- **Authentication** (for writes): `getAuthenticatedHttpClient()` handles Open edX auth
 - **Configuration**: `getConfig().LMS_BASE_URL` gets platform URLs
-- **Error Handling**: Try/catch blocks for API failures
 
 #### 3. Open edX UI Components
 
diff --git a/frontend-plugin-sample/src/plugin.jsx b/frontend-plugin-sample/src/plugin.jsx
