Release 0.77.11

.github/workflows/ci.yml

@@ -12,7 +12,7 @@ jobs:
       # Label used to access the service container
       db:
         # Docker Hub image
-        image: postgres:12.22
+        image: postgres:15
         # Set health checks to wait until postgres has started
         options: >-
           --health-cmd pg_isready
@@ -31,6 +31,20 @@ jobs:
         ports:
           - 6379:6379
 
+      starrocks:
+        # All-in-one FE+BE image, for CI/local dev only — see
+        # learning_resources.lib.warehouse for what this backs.
+        image: starrocks/allin1-ubuntu:4.1.4
+        options: >-
+          --health-cmd "bash -c 'echo > /dev/tcp/127.0.0.1/9030'"
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 30
+          --health-start-period 30s
+        ports:
+          - 9030:9030
+          - 8030:8030
+
     steps:
       - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v6.0.1
         with:
@@ -78,6 +92,9 @@ jobs:
           MAILGUN_SENDER_DOMAIN: other.fake.site
           OPENSEARCH_INDEX: testindex
           INDEXING_API_USERNAME: mitodl
+          STARROCKS_HOST: localhost
+          STARROCKS_PORT: "9030"
+          STARROCKS_USER: root
           MITOL_COOKIE_DOMAIN: localhost
           MITOL_COOKIE_NAME: cookie_monster
           UNSUBSCRIBE_SECRET_KEY: fake_unsubscribe_secret

RELEASE.rst

@@ -1,6 +1,30 @@
 Release Notes
 =============
 
+Version 0.77.11
+---------------
+
+- fix: bump version for arithmix game (#3832)
+- vector search completeness penalty (#3805)
+- feat(cohort-1): StarRocks warehouse-pull ETL machinery (#3807)
+- Update dependency llama-index-llms-openai to >=0.7.10,<0.8 (#3826)
+- Update dependency posthog to v7 (#2765)
+- Update Django to 5.2 (#3768)
+- fix: bullet lists and mailto links in About section  (#3802)
+- Revert broken Renovate merges (#3812-#3816) (#3822)
+- Update dependency tiktoken to >=0.13,<0.14 (#3816)
+- Update dependency ruff to v0.16.2 (#3815)
+- Update dependency llama-index-llms-openai to >=0.7.10,<0.8 (#3814)
+- Update dependency litellm to v1.95.0 (#3813)
+- Update dependency drf-spectacular to >=0.30,<0.31 (#3812)
+- Update apache/apisix Docker tag to v3.17.0 (#2708)
+- Drive React Query staleTime from the CDN TTL (#3804)
+- get rid of duplicate course checks (#3803)
+- feat: add a Failed filter tab to the B2B contract admin page (#3778)
+- Clear redis cache on a schedule (#3789)
+- adding fix for hiding offeror facets in hybrid search (#3786)
+- chore(deps): mitol-django-observability 2026.3.11 -> 2026.8.19 (#3796)
+
 Version 0.77.10 (Released August 24, 2026)
 ---------------
 

channels/migrations/0017_channelgrouprole_index.py

@@ -0,0 +1,17 @@
+# Generated by Django 5.2.16 on 2026-08-13 14:09
+
+from django.db import migrations
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("channels", "0016_alter_channel_featured_list"),
+    ]
+
+    operations = [
+        migrations.RenameIndex(
+            model_name="channelgrouprole",
+            new_name="channelgrouprole_ch_role_idx",
+            old_fields=("channel", "role"),
+        ),
+    ]

channels/models.py

@@ -279,7 +279,11 @@ class ChannelGroupRole(TimestampedModel):
 
     class Meta:
         unique_together = (("channel", "group", "role"),)
-        index_together = (("channel", "role"),)
+        indexes = [
+            models.Index(
+                fields=["channel", "role"], name="channelgrouprole_ch_role_idx"
+            )
+        ]
 
     def __str__(self):
         return (

docker-compose.services.yml

@@ -39,6 +39,22 @@ services:
       retries: 10
     ports:
       - "6379"
+  starrocks:
+    profiles:
+      - backend
+    # All-in-one FE+BE image, for local dev/test only — production StarRocks
+    # is a separately-operated cluster (see learning_resources.lib.warehouse).
+    image: starrocks/allin1-ubuntu:4.1.4
+    ports:
+      - "9030:9030" # FE MySQL-protocol query port
+      - "8030:8030" # FE HTTP port
+    healthcheck:
+      test: ["CMD-SHELL", "bash -c 'echo > /dev/tcp/127.0.0.1/9030'"]
+      interval: 5s
+      timeout: 5s
+      retries: 60
+      start_period: 30s
+
   qdrant:
     image: qdrant/qdrant:latest
     ports:
@@ -123,7 +139,7 @@ services:
   apigateway:
     profiles:
       - apisix
-    image: apache/apisix:3.13.0-debian # versions above this drop the local port on redirects
+    image: apache/apisix:3.17.0-debian@sha256:0e5377839f4ff5e322a5686ab6ce6797ba768008aca1bfc9b71149c3b326c4df # versions above this drop the local port on redirects
     <<: *default-extra-hosts
     environment:
       - CSRF_COOKIE_DOMAIN=${CSRF_COOKIE_DOMAIN:-.odl.local}

fixtures/common.py

@@ -50,6 +50,14 @@ def warnings_as_errors():
             module=".*(api_jwt|api_jws|rest_framework_jwt|astroid|bs4|celery|factory|botocore|posthog|pydantic).*",
             category=DeprecationWarning,
         )
+        # django-safedelete 1.4.1 (latest) still calls LogEntry.objects.log_action(),
+        # deprecated in Django 5.2. Drop once upstream moves to log_actions().
+        warnings.filterwarnings(
+            "ignore",
+            message=r"LogEntryManager\.log_action\(\) is deprecated.*",
+            module=r"safedelete\.admin",
+            category=DeprecationWarning,
+        )
         yield
     finally:
         warnings.resetwarnings()

frontends/main/Dockerfile.web

@@ -20,7 +20,7 @@
 #     -e NEXT_PUBLIC_MITX_ONLINE_LEGACY_BASE_URL=http://mitxonline.odl.local:8065/ \
 #     -e NEXT_PUBLIC_MITX_ONLINE_BASE_URL=https://api.learn.mit.edu/mitxonline \
 #     -e NEXT_PUBLIC_MITX_ONLINE_CSRF_COOKIE_NAME=csrf_mitxonline \
-#     -e NEXT_CACHE_S_MAXAGE_SECONDS=7200 \
+#     -e NEXT_PUBLIC_CACHE_S_MAXAGE_SECONDS=7200 \
 #     -e NEXT_PUBLIC_MITOL_AXIOS_WITH_CREDENTIALS=true \
 #     mit-learn-nextjs:test
 

frontends/main/package.json

@@ -15,7 +15,7 @@
     "@emotion/cache": "^11.13.1",
     "@emotion/styled": "^11.11.0",
     "@floating-ui/react": "^0.27.16",
-    "@mitodl/arithmix": "^0.2.4",
+    "@mitodl/arithmix": "^0.2.5",
     "@mitodl/course-search-utils": "^3.5.2",
     "@mitodl/hacksnack": "^0.1.2",
     "@mitodl/mitxonline-api-axios": "2026.8.18",

frontends/main/src/app-pages/ChannelPage/ChannelSearch.test.tsx

@@ -7,7 +7,10 @@ import {
   user,
 } from "@/test-utils"
 import { setMockResponse, urls, factories, makeRequest } from "api/test-utils"
-import type { LearningResourcesSearchResponse } from "api"
+import type {
+  LearningResourcesSearchResponse,
+  PaginatedLearningResourceOfferorDetailList,
+} from "api"
 import invariant from "tiny-invariant"
 import type { Channel } from "api/v0"
 import { ChannelTypeEnum } from "api/v0"
@@ -16,9 +19,11 @@ import ChannelPage from "./ChannelPage"
 const setMockApiResponses = ({
   search,
   channelPatch = {},
+  offerors,
 }: {
   search?: Partial<LearningResourcesSearchResponse>
   channelPatch?: Partial<Channel>
+  offerors?: PaginatedLearningResourceOfferorDetailList
 }) => {
   const channel = factories.channels.channel(channelPatch)
   const urlParams = new URLSearchParams(channelPatch?.search_filter)
@@ -58,7 +63,7 @@ const setMockApiResponses = ({
 
   setMockResponse.get(
     urls.offerors.list(),
-    factories.learningResources.offerors({ count: 5 }),
+    offerors ?? factories.learningResources.offerors({ count: 5 }),
   )
 
   setMockResponse.get(expect.stringContaining(urls.search.resources()), {
@@ -201,6 +206,68 @@ describe("ChannelSearch", () => {
     expect(apiSearchParams.get("q")).toBe("python")
     expect(apiSearchParams.get("topic")).toBe("Economics")
   })
+
+  test("Hybrid search 'Offered By' facet only shows facets with 'display_facet' set to true", async () => {
+    const offerors = factories.learningResources.offerors({ count: 3 })
+    offerors.results[0]!.display_facet = true
+    offerors.results[1]!.display_facet = false
+    offerors.results[2]!.display_facet = false
+
+    const resources = factories.learningResources.resources({
+      count: 3,
+    }).results
+    resources.forEach((resource, i) => {
+      resource.professional = true
+      resource.offered_by = {
+        code: offerors.results[i]!.code,
+        name: offerors.results[i]!.name,
+        channel_url: null,
+      }
+    })
+
+    const { channel } = setMockApiResponses({
+      channelPatch: { channel_type: ChannelTypeEnum.Topic },
+      offerors,
+      search: {
+        count: resources.length,
+        results: resources,
+        metadata: {
+          aggregations: {
+            offered_by: offerors.results.map((o, i) => ({
+              key: o.code,
+              doc_count: 10 + i,
+            })),
+          },
+          suggestions: [],
+        },
+      },
+    })
+
+    renderWithProviders(<ChannelPage />, {
+      url: `/c/${channel.channel_type}/${channel.name}?q=python&professional=true`,
+    })
+
+    const showFacetButton = await screen.findByRole("button", {
+      name: /Offered By/i,
+    })
+    await user.click(showFacetButton)
+
+    const offeror0 = await screen.findByRole("checkbox", {
+      name: new RegExp(`^${offerors.results[0]!.name}`),
+    })
+    expect(offeror0).toBeVisible()
+    expect(
+      screen.queryByRole("checkbox", {
+        name: new RegExp(`^${offerors.results[1]!.name}`),
+      }),
+    ).not.toBeInTheDocument()
+    expect(
+      screen.queryByRole("checkbox", {
+        name: new RegExp(`^${offerors.results[2]!.name}`),
+      }),
+    ).not.toBeInTheDocument()
+  })
+
   test.each([
     {
       searchFilter: "offered_by=ocw",