Add included_languages M2M field to contentnodes.

kolibri/core/content/api.py

@@ -258,12 +258,19 @@ def list(self, request, *args, **kwargs):
         return super(RemoteViewSet, self).list(request, *args, **kwargs)
 
 
+class CharInFilter(BaseInFilter, CharFilter):
+    pass
+
+
 class ChannelMetadataFilter(FilterSet):
     available = BooleanFilter(method="filter_available", label="Available")
     contains_exercise = BooleanFilter(
         method="filter_contains_exercise", label="Has exercises"
     )
     contains_quiz = BooleanFilter(method="filter_contains_quiz", label="Has quizzes")
+    languages = CharInFilter(
+        field_name="included_languages", label="Languages", distinct=True
+    )
 
     class Meta:
         model = models.ChannelMetadata
@@ -417,10 +424,6 @@ class UUIDInFilter(BaseInFilter, UUIDFilter):
     pass
 
 
-class CharInFilter(BaseInFilter, CharFilter):
-    pass
-
-
 contentnode_filter_fields = [
     "parent",
     "parent__isnull",
@@ -470,7 +473,7 @@ class ContentNodeFilter(FilterSet):
     learner_needs = CharFilter(method="bitmask_contains_and")
     keywords = CharFilter(method="filter_keywords")
     channels = UUIDInFilter(field_name="channel_id")
-    languages = CharInFilter(field_name="lang_id")
+    languages = CharInFilter(field_name="included_languages")
     categories__isnull = BooleanFilter(field_name="categories", lookup_expr="isnull")
     lft__gt = NumberFilter(field_name="lft", lookup_expr="gt")
     rght__lt = NumberFilter(field_name="rght", lookup_expr="lt")
@@ -671,10 +674,11 @@ def get_queryset(self):
         return models.ContentNode.objects.filter(available=True)
 
     def get_related_data_maps(self, items, queryset):
+        ids = [item["id"] for item in items]
         assessmentmetadata_map = {
             a["contentnode"]: a
             for a in models.AssessmentMetaData.objects.filter(
-                contentnode__in=queryset
+                contentnode__in=ids
             ).values(
                 "assessment_item_ids",
                 "number_of_assessments",
@@ -688,7 +692,7 @@ def get_related_data_maps(self, items, queryset):
         files_map = {}
 
         files = list(
-            models.File.objects.filter(contentnode__in=queryset).values(
+            models.File.objects.filter(contentnode__in=ids).values(
                 "id",
                 "contentnode",
                 "local_file__id",
@@ -723,7 +727,7 @@ def get_related_data_maps(self, items, queryset):
         tags_map = {}
 
         for t in (
-            models.ContentTag.objects.filter(tagged_content__in=queryset)
+            models.ContentTag.objects.filter(tagged_content__in=ids)
             .values(
                 "tag_name",
                 "tagged_content",

kolibri/core/content/contentschema/versions/content_schema_current.py

@@ -2,6 +2,7 @@
 from sqlalchemy import BigInteger
 from sqlalchemy import Boolean
 from sqlalchemy import CHAR
+from sqlalchemy import CheckConstraint
 from sqlalchemy import Column
 from sqlalchemy import Float
 from sqlalchemy import ForeignKey
@@ -45,6 +46,36 @@ class ContentLocalfile(Base):
 class ContentContentnode(Base):
     __tablename__ = "content_contentnode"
     __table_args__ = (
+        CheckConstraint(
+            """
+            "lft" >= 0),
+            "rght" integer unsigned NOT NULL CHECK ("rght" >= 0),
+            "tree_id" integer unsigned NOT NULL CHECK ("tree_id" >= 0),
+            "level" integer unsigned NOT NULL CHECK ("level" >= 0),
+            "lang_id" varchar(14) NULL REFERENCES "content_language" ("id") DEFERRABLE INITIALLY DEFERRED,
+            "license_description" text NULL,
+            "license_name" varchar(50) NULL,
+            "coach_content" bool NOT NULL,
+            "num_coach_contents" integer NULL,
+            "on_device_resources" integer NULL,
+            "options" text NULL,
+            "accessibility_labels" text NULL,
+            "categories" text NULL,
+            "duration" integer unsigned NULL CHECK ("duration" >= 0),
+            "grade_levels" text NULL,
+            "learner_needs" text NULL,
+            "learning_activities" text NULL,
+            "resource_types" text NULL,
+            "accessibility_labels_bitmask_0" bigint NULL,
+            "categories_bitmask_0" bigint NULL,
+            "grade_levels_bitmask_0" bigint NULL,
+            "learner_needs_bitmask_0" bigint NULL,
+            "learning_activities_bitmask_0" bigint NULL,
+            "ancestors" text NULL,
+            "admin_imported" bool NULL,
+            "parent_id" char(32) NULL REFERENCES "content_contentnode" ("id") DEFERRABLE INITIALLY DEFERRED
+            """
+        ),
         Index(
             "content_contentnode_level_channel_id_available_29f0bb18_idx",
             "level",
@@ -69,10 +100,10 @@ class ContentContentnode(Base):
     author = Column(String(200), nullable=False)
     kind = Column(String(200), nullable=False)
     available = Column(Boolean, nullable=False)
-    lft = Column(Integer, nullable=False, index=True)
-    rght = Column(Integer, nullable=False, index=True)
+    lft = Column(Integer, nullable=False)
+    rght = Column(Integer, nullable=False)
     tree_id = Column(Integer, nullable=False, index=True)
-    level = Column(Integer, nullable=False, index=True)
+    level = Column(Integer, nullable=False)
     lang_id = Column(ForeignKey("content_language.id"), index=True)
     license_description = Column(Text)
     license_name = Column(String(50))
@@ -118,6 +149,11 @@ class ContentAssessmentmetadata(Base):
 
 class ContentChannelmetadata(Base):
     __tablename__ = "content_channelmetadata"
+    __table_args__ = (
+        CheckConstraint(
+            '"order" >= 0), "tagline" varchar(150) NULL, "partial" bool NULL, "public" bool NULL'
+        ),
+    )
 
     id = Column(CHAR(32), primary_key=True)
     name = Column(String(200), nullable=False)
@@ -131,9 +167,9 @@ class ContentChannelmetadata(Base):
     published_size = Column(BigInteger)
     total_resource_count = Column(Integer)
     order = Column(Integer)
-    public = Column(Boolean)
     tagline = Column(String(150))
     partial = Column(Boolean)
+    public = Column(Boolean)
 
     root = relationship("ContentContentnode")
 
@@ -167,6 +203,27 @@ class ContentContentnodeHasPrerequisite(Base):
     )
 
 
+class ContentContentnodeIncludedLanguages(Base):
+    __tablename__ = "content_contentnode_included_languages"
+    __table_args__ = (
+        Index(
+            "content_contentnode_included_languages_contentnode_id_language_id_7d14ec8b_uniq",
+            "contentnode_id",
+            "language_id",
+            unique=True,
+        ),
+    )
+
+    id = Column(Integer, primary_key=True)
+    contentnode_id = Column(
+        ForeignKey("content_contentnode.id"), nullable=False, index=True
+    )
+    language_id = Column(ForeignKey("content_language.id"), nullable=False, index=True)
+
+    contentnode = relationship("ContentContentnode")
+    language = relationship("ContentLanguage")
+
+
 class ContentContentnodeRelated(Base):
     __tablename__ = "content_contentnode_related"
     __table_args__ = (

kolibri/core/content/migrations/0039_contentnode_included_languages.py

@@ -0,0 +1,23 @@
+# Generated by Django 3.2.25 on 2024-12-14 18:33
+from django.db import migrations
+from django.db import models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("content", "0038_alter_localfile_extension"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="contentnode",
+            name="included_languages",
+            field=models.ManyToManyField(
+                blank=True,
+                related_name="contentnodes",
+                to="content.Language",
+                verbose_name="languages",
+            ),
+        ),
+    ]

kolibri/core/content/models.py

@@ -214,6 +214,16 @@ class ContentNode(base_models.ContentNode):
     # needs a subsequent Kolibri upgrade step to backfill these values.
     admin_imported = models.BooleanField(null=True)
 
+    # Languages that are in this node and/or any descendant nodes of this node
+    # for non-topic nodes, this is the language of the node itself
+    # for topic nodes, this is the union of all languages of all descendant nodes
+    # and any language set on the topic node itself
+    # We do this to allow filtering of a topic tree by a specific language for
+    # multi-language channels.
+    included_languages = models.ManyToManyField(
+        "Language", related_name="contentnodes", verbose_name="languages", blank=True
+    )
+
     objects = ContentNodeManager()
 
     class Meta: