feat: container support for tags [FC-0083] (#36484)

* feat: container tag support

* fix: fixes from review

* docs: fix typo

Author: rpenido

openedx/core/djangoapps/content/search/api.py

@@ -17,6 +17,7 @@
 from meilisearch import Client as MeilisearchClient
 from meilisearch.errors import MeilisearchApiError, MeilisearchError
 from meilisearch.models.task import TaskInfo
+from opaque_keys import OpaqueKey
 from opaque_keys.edx.keys import UsageKey
 from opaque_keys.edx.locator import LibraryContainerLocator, LibraryLocatorV2, LibraryCollectionLocator
 from openedx_learning.api import authoring as authoring_api
@@ -42,7 +43,7 @@
     searchable_doc_for_collection,
     searchable_doc_for_container,
     searchable_doc_for_library_block,
-    searchable_doc_for_usage_key,
+    searchable_doc_for_key,
     searchable_doc_collections,
     searchable_doc_tags,
     searchable_doc_tags_for_collection,
@@ -486,8 +487,7 @@ def index_container_batch(batch, num_done, library_key) -> int:
                         container,
                     )
                     doc = searchable_doc_for_container(container_key)
-                    # TODO: when we add container tags
-                    # doc.update(searchable_doc_tags_for_container(container_key))
+                    doc.update(searchable_doc_tags(container_key))
                     docs.append(doc)
                 except Exception as err:  # pylint: disable=broad-except
                     status_cb(f"Error indexing container {container.key}: {err}")
@@ -511,7 +511,7 @@ def index_container_batch(batch, num_done, library_key) -> int:
             collections = authoring_api.get_collections(library.learning_package_id, enabled=True)
             num_collections = collections.count()
             num_collections_done = 0
-            status_cb(f"{num_collections_done + 1}/{num_collections}. Now indexing collections in library {lib_key}")
+            status_cb(f"{num_collections_done}/{num_collections}. Now indexing collections in library {lib_key}")
             paginator = Paginator(collections, 100)
             for p in paginator.page_range:
                 num_collections_done = index_collection_batch(
@@ -620,7 +620,7 @@ def delete_index_doc(usage_key: UsageKey) -> None:
     Args:
         usage_key (UsageKey): The usage key of the XBlock to be removed from the index
     """
-    doc = searchable_doc_for_usage_key(usage_key)
+    doc = searchable_doc_for_key(usage_key)
     _delete_index_doc(doc[Fields.id])
 
 
@@ -811,12 +811,12 @@ def upsert_content_library_index_docs(library_key: LibraryLocatorV2) -> None:
     _update_index_docs(docs)
 
 
-def upsert_block_tags_index_docs(usage_key: UsageKey):
+def upsert_content_object_tags_index_doc(key: OpaqueKey):
     """
-    Updates the tags data in documents for the given Course/Library block
+    Updates the tags data in document for the given Course/Library item
     """
-    doc = {Fields.id: meili_id_from_opaque_key(usage_key)}
-    doc.update(searchable_doc_tags(usage_key))
+    doc = {Fields.id: meili_id_from_opaque_key(key)}
+    doc.update(searchable_doc_tags(key))
     _update_index_docs([doc])
 
 

openedx/core/djangoapps/content/search/documents.py

@@ -8,6 +8,7 @@
 
 from django.core.exceptions import ObjectDoesNotExist
 from django.utils.text import slugify
+from opaque_keys import OpaqueKey
 from opaque_keys.edx.keys import LearningContextKey, UsageKey
 from opaque_keys.edx.locator import LibraryContainerLocator, LibraryLocatorV2
 from openedx_learning.api import authoring as authoring_api
@@ -113,7 +114,7 @@ class PublishStatus:
     modified = "modified"
 
 
-def meili_id_from_opaque_key(usage_key: UsageKey) -> str:
+def meili_id_from_opaque_key(key: OpaqueKey) -> str:
     """
     Meilisearch requires each document to have a primary key that's either an
     integer or a string composed of alphanumeric characters (a-z A-Z 0-9),
@@ -124,7 +125,7 @@ def meili_id_from_opaque_key(usage_key: UsageKey) -> str:
     we could use PublishableEntity's primary key / UUID instead.
     """
     # The slugified key _may_ not be unique so we append a hashed string to make it unique:
-    key_str = str(usage_key)
+    key_str = str(key)
     key_bin = key_str.encode()
 
     suffix = blake2b(key_bin, digest_size=4, usedforsecurity=False).hexdigest()
@@ -140,12 +141,12 @@ def _meili_access_id_from_context_key(context_key: LearningContextKey) -> int:
     return access.id
 
 
-def searchable_doc_for_usage_key(usage_key: UsageKey) -> dict:
+def searchable_doc_for_key(key: OpaqueKey) -> dict:
     """
-    Generates a base document identified by its usage key.
+    Generates a base document identified by its opaque key.
     """
     return {
-        Fields.id: meili_id_from_opaque_key(usage_key),
+        Fields.id: meili_id_from_opaque_key(key),
     }
 
 
@@ -244,7 +245,7 @@ class implementation returns only:
     return block_data
 
 
-def _tags_for_content_object(object_id: UsageKey | LearningContextKey) -> dict:
+def _tags_for_content_object(object_id: OpaqueKey) -> dict:
     """
     Given an XBlock, course, library, etc., get the tag data for its index doc.
 
@@ -406,7 +407,7 @@ def searchable_doc_for_library_block(xblock_metadata: lib_api.LibraryXBlockMetad
         block_published = None
         publish_status = PublishStatus.never
 
-    doc = searchable_doc_for_usage_key(xblock_metadata.usage_key)
+    doc = searchable_doc_for_key(xblock_metadata.usage_key)
     doc.update({
         Fields.type: DocType.library_block,
         Fields.breadcrumbs: [],
@@ -427,13 +428,13 @@ def searchable_doc_for_library_block(xblock_metadata: lib_api.LibraryXBlockMetad
     return doc
 
 
-def searchable_doc_tags(usage_key: UsageKey) -> dict:
+def searchable_doc_tags(key: OpaqueKey) -> dict:
     """
     Generate a dictionary document suitable for ingestion into a search engine
     like Meilisearch or Elasticsearch, with the tags data for the given content object.
     """
-    doc = searchable_doc_for_usage_key(usage_key)
-    doc.update(_tags_for_content_object(usage_key))
+    doc = searchable_doc_for_key(key)
+    doc.update(_tags_for_content_object(key))
 
     return doc
 
@@ -443,7 +444,7 @@ def searchable_doc_collections(usage_key: UsageKey) -> dict:
     Generate a dictionary document suitable for ingestion into a search engine
     like Meilisearch or Elasticsearch, with the collections data for the given content object.
     """
-    doc = searchable_doc_for_usage_key(usage_key)
+    doc = searchable_doc_for_key(usage_key)
     doc.update(_collections_for_content_object(usage_key))
 
     return doc
@@ -461,7 +462,7 @@ def searchable_doc_tags_for_collection(
         library_key,
         collection_key,
     )
-    doc = searchable_doc_for_usage_key(collection_usage_key)
+    doc = searchable_doc_for_key(collection_usage_key)
     doc.update(_tags_for_content_object(collection_usage_key))
 
     return doc
@@ -473,7 +474,7 @@ def searchable_doc_for_course_block(block) -> dict:
     like Meilisearch or Elasticsearch, so that the given course block can be
     found using faceted search.
     """
-    doc = searchable_doc_for_usage_key(block.usage_key)
+    doc = searchable_doc_for_key(block.usage_key)
     doc.update({
         Fields.type: DocType.course_block,
     })
@@ -503,7 +504,7 @@ def searchable_doc_for_collection(
         collection_key,
     )
 
-    doc = searchable_doc_for_usage_key(collection_usage_key)
+    doc = searchable_doc_for_key(collection_usage_key)
 
     try:
         collection = collection or lib_api.get_library_collection_from_usage_key(collection_usage_key)

openedx/core/djangoapps/content/search/handlers.py

@@ -8,7 +8,7 @@
 from django.dispatch import receiver
 from opaque_keys import InvalidKeyError
 from opaque_keys.edx.keys import UsageKey
-from opaque_keys.edx.locator import LibraryCollectionLocator
+from opaque_keys.edx.locator import LibraryCollectionLocator, LibraryContainerLocator
 from openedx_events.content_authoring.data import (
     ContentLibraryData,
     ContentObjectChangedData,
@@ -41,7 +41,7 @@
 from .api import (
     only_if_meilisearch_enabled,
     upsert_block_collections_index_docs,
-    upsert_block_tags_index_docs,
+    upsert_content_object_tags_index_doc,
     upsert_collection_tags_index_docs,
 )
 from .tasks import (
@@ -211,23 +211,28 @@ def content_object_associations_changed_handler(**kwargs) -> None:
         return
 
     try:
-        # Check if valid if course or library block
+        # Check if valid course or library block
         usage_key = UsageKey.from_string(str(content_object.object_id))
     except InvalidKeyError:
         try:
-            # Check if valid if library collection
+            # Check if valid library collection
             usage_key = LibraryCollectionLocator.from_string(str(content_object.object_id))
         except InvalidKeyError:
-            log.error("Received invalid content object id")
-            return
+            try:
+                # Check if valid library container
+                usage_key = LibraryContainerLocator.from_string(str(content_object.object_id))
+            except InvalidKeyError:
+                # Invalid content object id
+                log.error("Received invalid content object id")
+                return
 
     # This event's changes may contain both "tags" and "collections", but this will happen rarely, if ever.
     # So we allow a potential double "upsert" here.
     if not content_object.changes or "tags" in content_object.changes:
         if isinstance(usage_key, LibraryCollectionLocator):
             upsert_collection_tags_index_docs(usage_key)
         else:
-            upsert_block_tags_index_docs(usage_key)
+            upsert_content_object_tags_index_doc(usage_key)
     if not content_object.changes or "collections" in content_object.changes:
         upsert_block_collections_index_docs(usage_key)
 

openedx/core/djangoapps/content/search/tests/test_api.py

@@ -234,7 +234,6 @@ def setUp(self):
             "modified": created_date.timestamp(),
             "access_id": lib_access.id,
             "breadcrumbs": [{"display_name": "Library"}],
-            # "tags" should be here but we haven't implemented them yet
             # "published" is not set since we haven't published it yet
         }
 
@@ -262,6 +261,7 @@ def test_reindex_meilisearch(self, mock_meilisearch):
         doc_collection = copy.deepcopy(self.collection_dict)
         doc_collection["tags"] = {}
         doc_unit = copy.deepcopy(self.unit_dict)
+        doc_unit["tags"] = {}
 
         api.rebuild_index()
         assert mock_meilisearch.return_value.index.return_value.add_documents.call_count == 4
@@ -292,6 +292,7 @@ def test_reindex_meilisearch_incremental(self, mock_meilisearch):
         doc_collection = copy.deepcopy(self.collection_dict)
         doc_collection["tags"] = {}
         doc_unit = copy.deepcopy(self.unit_dict)
+        doc_unit["tags"] = {}
 
         api.rebuild_index(incremental=True)
         assert mock_meilisearch.return_value.index.return_value.add_documents.call_count == 4
@@ -472,8 +473,7 @@ def test_index_xblock_tags(self, mock_meilisearch):
         """
         Test indexing an XBlock with tags.
         """
-
-        # Tag XBlock (these internally call `upsert_block_tags_index_docs`)
+        # Tag XBlock (these internally call `upsert_content_object_tags_index_doc`)
         tagging_api.tag_object(str(self.sequential.usage_key), self.taxonomyA, ["one", "two"])
         tagging_api.tag_object(str(self.sequential.usage_key), self.taxonomyB, ["three", "four"])
 
@@ -866,3 +866,43 @@ def test_delete_collection(self, mock_meilisearch):
         mock_meilisearch.return_value.index.return_value.update_documents.assert_called_once_with([
             doc_problem_without_collection,
         ])
+
+    @override_settings(MEILISEARCH_ENABLED=True)
+    def test_index_library_container_metadata(self, mock_meilisearch):
+        """
+        Test indexing a Library Container.
+        """
+        api.upsert_library_container_index_doc(self.unit.container_key)
+
+        mock_meilisearch.return_value.index.return_value.update_documents.assert_called_once_with([self.unit_dict])
+
+    @override_settings(MEILISEARCH_ENABLED=True)
+    def test_index_tags_in_containers(self, mock_meilisearch):
+        # Tag collection
+        tagging_api.tag_object(self.unit_key, self.taxonomyA, ["one", "two"])
+        tagging_api.tag_object(self.unit_key, self.taxonomyB, ["three", "four"])
+
+        # Build expected docs with tags at each stage
+        doc_unit_with_tags1 = {
+            "id": "lctorg1libunitunit-1-e4527f7c",
+            "tags": {
+                'taxonomy': ['A'],
+                'level0': ['A > one', 'A > two']
+            }
+        }
+        doc_unit_with_tags2 = {
+            "id": "lctorg1libunitunit-1-e4527f7c",
+            "tags": {
+                'taxonomy': ['A', 'B'],
+                'level0': ['A > one', 'A > two', 'B > four', 'B > three']
+            }
+        }
+
+        assert mock_meilisearch.return_value.index.return_value.update_documents.call_count == 2
+        mock_meilisearch.return_value.index.return_value.update_documents.assert_has_calls(
+            [
+                call([doc_unit_with_tags1]),
+                call([doc_unit_with_tags2]),
+            ],
+            any_order=True,
+        )