Add public API endpoint for public collections (#2174)

Fixes #1051 

If org with provided slug doesn't exist or no public collections exist
for that org, return same 404 response with a detail of
"public_profile_not_found" to prevent people from using public endpoint
to determine whether an org exists.

Endpoint is `GET /api/public-collections/<org-slug>` (no auth needed) to
avoid collisions with existing org and collection endpoints.

backend/btrixcloud/colls.py

@@ -30,6 +30,9 @@
     UpdatedResponse,
     SuccessResponse,
     CollectionSearchValuesResponse,
+    OrgPublicCollections,
+    PublicOrgDetails,
+    CollAccessType,
 )
 from .utils import dt_now
 
@@ -395,6 +398,30 @@ async def add_successful_crawl_to_collections(self, crawl_id: str, cid: UUID):
             )
             await self.update_crawl_collections(crawl_id)
 
+    async def get_org_public_collections(self, org_slug: str):
+        """List public collections for org"""
+        try:
+            org = await self.orgs.get_org_by_slug(org_slug)
+        # pylint: disable=broad-exception-caught
+        except Exception:
+            # pylint: disable=raise-missing-from
+            raise HTTPException(status_code=404, detail="public_profile_not_found")
+
+        if not org.enablePublicProfile:
+            raise HTTPException(status_code=404, detail="public_profile_not_found")
+
+        collections, _ = await self.list_collections(
+            org.id, access=CollAccessType.PUBLIC
+        )
+
+        public_org_details = PublicOrgDetails(
+            name=org.name,
+            description=org.publicDescription or "",
+            url=org.publicUrl or "",
+        )
+
+        return OrgPublicCollections(org=public_org_details, collections=collections)
+
 
 # ============================================================================
 # pylint: disable=too-many-locals
@@ -582,4 +609,12 @@ async def download_collection(
     ):
         return await colls.download_collection(coll_id, org)
 
+    @app.get(
+        "/public-collections/{org_slug}",
+        tags=["collections"],
+        response_model=OrgPublicCollections,
+    )
+    async def get_org_public_collections(org_slug: str):
+        return await colls.get_org_public_collections(org_slug)
+
     return colls

backend/btrixcloud/models.py

@@ -1144,6 +1144,24 @@ class RenameOrg(BaseModel):
     slug: Optional[str] = None
 
 
+# ============================================================================
+class PublicOrgDetails(BaseModel):
+    """Model for org details that are available in public profile"""
+
+    name: str
+    description: str = ""
+    url: str = ""
+
+
+# ============================================================================
+class OrgPublicCollections(BaseModel):
+    """Model for listing public collections in org"""
+
+    org: PublicOrgDetails
+
+    collections: List[CollOut] = []
+
+
 # ============================================================================
 class OrgStorageRefs(BaseModel):
     """Input model for setting primary storage + optional replicas"""
@@ -1374,10 +1392,12 @@ class OrgReadOnlyUpdate(BaseModel):
 
 
 # ============================================================================
-class OrgListPublicCollectionsUpdate(BaseModel):
-    """Organization listPublicCollections update"""
+class OrgPublicProfileUpdate(BaseModel):
+    """Organization enablePublicProfile update"""
 
-    listPublicCollections: bool
+    enablePublicProfile: Optional[bool] = None
+    publicDescription: Optional[str] = None
+    publicUrl: Optional[str] = None
 
 
 # ============================================================================
@@ -1448,7 +1468,9 @@ class OrgOut(BaseMongoModel):
     allowedProxies: list[str] = []
     crawlingDefaults: Optional[CrawlConfigDefaults] = None
 
-    listPublicCollections: bool = False
+    enablePublicProfile: bool = False
+    publicDescription: str = ""
+    publicUrl: str = ""
 
 
 # ============================================================================
@@ -1505,7 +1527,9 @@ class Organization(BaseMongoModel):
     allowedProxies: list[str] = []
     crawlingDefaults: Optional[CrawlConfigDefaults] = None
 
-    listPublicCollections: bool = False
+    enablePublicProfile: bool = False
+    publicDescription: Optional[str] = None
+    publicUrl: Optional[str] = None
 
     def is_owner(self, user):
         """Check if user is owner"""

backend/btrixcloud/orgs.py

@@ -78,7 +78,7 @@
     RemovedResponse,
     OrgSlugsResponse,
     OrgImportResponse,
-    OrgListPublicCollectionsUpdate,
+    OrgPublicProfileUpdate,
 )
 from .pagination import DEFAULT_PAGE_SIZE, paginated_format
 from .utils import (
@@ -285,6 +285,14 @@ async def get_org_by_id(self, oid: UUID) -> Organization:
 
         return Organization.from_dict(res)
 
+    async def get_org_by_slug(self, slug: str) -> Organization:
+        """Get an org by id"""
+        res = await self.orgs.find_one({"slug": slug})
+        if not res:
+            raise HTTPException(status_code=400, detail="invalid_org_slug")
+
+        return Organization.from_dict(res)
+
     async def get_default_org(self) -> Organization:
         """Get default organization"""
         res = await self.orgs.find_one({"default": True})
@@ -988,13 +996,18 @@ async def update_read_only_on_cancel(
         )
         return res is not None
 
-    async def update_list_public_collections(
-        self, org: Organization, list_public_collections: bool
+    async def update_public_profile(
+        self, org: Organization, update: OrgPublicProfileUpdate
     ):
-        """Update listPublicCollections field on organization"""
+        """Update or enable/disable organization's public profile"""
+        query = update.dict(exclude_unset=True)
+
+        if len(query) == 0:
+            raise HTTPException(status_code=400, detail="no_update_data")
+
         res = await self.orgs.find_one_and_update(
             {"_id": org.id},
-            {"$set": {"listPublicCollections": list_public_collections}},
+            {"$set": query},
         )
         return res is not None
 
@@ -1555,15 +1568,15 @@ async def update_read_only_on_cancel(
         return {"updated": True}
 
     @router.post(
-        "/list-public-collections",
+        "/public-profile",
         tags=["organizations", "collections"],
         response_model=UpdatedResponse,
     )
-    async def update_list_public_collections(
-        update: OrgListPublicCollectionsUpdate,
+    async def update_public_profile(
+        update: OrgPublicProfileUpdate,
         org: Organization = Depends(org_owner_dep),
     ):
-        await ops.update_list_public_collections(org, update.listPublicCollections)
+        await ops.update_public_profile(org, update)
 
         return {"updated": True}
 

backend/test/test_collections.py

@@ -4,7 +4,7 @@
 from zipfile import ZipFile, ZIP_STORED
 from tempfile import TemporaryFile
 
-from .conftest import API_PREFIX
+from .conftest import API_PREFIX, NON_DEFAULT_ORG_NAME, NON_DEFAULT_ORG_SLUG
 from .utils import read_in_chunks
 
 COLLECTION_NAME = "Test collection"
@@ -15,6 +15,7 @@
 
 _coll_id = None
 _second_coll_id = None
+_public_coll_id = None
 upload_id = None
 modified = None
 
@@ -66,6 +67,7 @@ def test_create_public_collection(
     assert data["added"]
     assert data["name"] == PUBLIC_COLLECTION_NAME
 
+    global _public_coll_id
     _public_coll_id = data["id"]
 
     # Verify that it is public
@@ -725,6 +727,123 @@ def test_filter_sort_collections(
     assert r.json()["detail"] == "invalid_sort_direction"
 
 
+def test_list_public_collections(
+    crawler_auth_headers,
+    admin_auth_headers,
+    default_org_id,
+    non_default_org_id,
+    crawler_crawl_id,
+    admin_crawl_id,
+):
+    # Create new public collection
+    r = requests.post(
+        f"{API_PREFIX}/orgs/{default_org_id}/collections",
+        headers=crawler_auth_headers,
+        json={
+            "crawlIds": [crawler_crawl_id],
+            "name": "Second public collection",
+            "access": "public",
+        },
+    )
+    assert r.status_code == 200
+    second_public_coll_id = r.json()["id"]
+
+    # Get default org slug
+    r = requests.get(
+        f"{API_PREFIX}/orgs/{default_org_id}",
+        headers=crawler_auth_headers,
+    )
+    assert r.status_code == 200
+    data = r.json()
+    org_slug = data["slug"]
+    org_name = data["name"]
+
+    # Verify that public profile isn't enabled
+    assert data["enablePublicProfile"] is False
+    assert data["publicDescription"] == ""
+    assert data["publicUrl"] == ""
+
+    # Try listing public collections without org public profile enabled
+    r = requests.get(f"{API_PREFIX}/public-collections/{org_slug}")
+    assert r.status_code == 404
+    assert r.json()["detail"] == "public_profile_not_found"
+
+    # Enable public profile on org
+    public_description = "This is a test public org!"
+    public_url = "https://example.com"
+
+    r = requests.post(
+        f"{API_PREFIX}/orgs/{default_org_id}/public-profile",
+        headers=admin_auth_headers,
+        json={
+            "enablePublicProfile": True,
+            "publicDescription": public_description,
+            "publicUrl": public_url,
+        },
+    )
+    assert r.status_code == 200
+    assert r.json()["updated"]
+
+    r = requests.get(
+        f"{API_PREFIX}/orgs/{default_org_id}",
+        headers=admin_auth_headers,
+    )
+    assert r.status_code == 200
+    data = r.json()
+    assert data["enablePublicProfile"]
+    assert data["publicDescription"] == public_description
+    assert data["publicUrl"] == public_url
+
+    # List public collections with no auth (no public profile)
+    r = requests.get(f"{API_PREFIX}/public-collections/{org_slug}")
+    assert r.status_code == 200
+    data = r.json()
+
+    org_data = data["org"]
+    assert org_data["name"] == org_name
+    assert org_data["description"] == public_description
+    assert org_data["url"] == public_url
+
+    collections = data["collections"]
+    assert len(collections) == 2
+    for collection in collections:
+        assert collection["id"] in (_public_coll_id, second_public_coll_id)
+        assert collection["access"] == "public"
+
+    # Test non-existing slug - it should return a 404 but not reveal
+    # whether or not an org exists with that slug
+    r = requests.get(f"{API_PREFIX}/public-collections/nonexistentslug")
+    assert r.status_code == 404
+    assert r.json()["detail"] == "public_profile_not_found"
+
+
+def test_list_public_collections_no_colls(non_default_org_id, admin_auth_headers):
+    # Test existing org that's not public - should return same 404 as
+    # if org doesn't exist
+    r = requests.get(f"{API_PREFIX}/public-collections/{NON_DEFAULT_ORG_SLUG}")
+    assert r.status_code == 404
+    assert r.json()["detail"] == "public_profile_not_found"
+
+    # Enable public profile on org with zero public collections
+    r = requests.post(
+        f"{API_PREFIX}/orgs/{non_default_org_id}/public-profile",
+        headers=admin_auth_headers,
+        json={
+            "enablePublicProfile": True,
+        },
+    )
+    assert r.status_code == 200
+    assert r.json()["updated"]
+
+    # List public collections with no auth - should still get profile even
+    # with no public collections
+    r = requests.get(f"{API_PREFIX}/public-collections/{NON_DEFAULT_ORG_SLUG}")
+    assert r.status_code == 200
+    data = r.json()
+    assert data["org"]["name"] == NON_DEFAULT_ORG_NAME
+    assert data["collections"] == []
+
+
 def test_delete_collection(crawler_auth_headers, default_org_id, crawler_crawl_id):
     # Delete second collection
     r = requests.delete(

backend/test/test_org.py

@@ -17,7 +17,7 @@
 def test_ensure_only_one_default_org(admin_auth_headers):
     r = requests.get(f"{API_PREFIX}/orgs", headers=admin_auth_headers)
     data = r.json()
-    assert data["total"] == 1
+    assert data["total"] == 2
 
     orgs = data["items"]
     default_orgs = [org for org in orgs if org["default"]]
@@ -697,24 +697,6 @@ def test_update_read_only(admin_auth_headers, default_org_id):
     assert data["readOnlyReason"] == ""
 
 
-def test_update_list_public_collections(admin_auth_headers, default_org_id):
-    # Test that default is false
-    r = requests.get(f"{API_PREFIX}/orgs/{default_org_id}", headers=admin_auth_headers)
-    assert r.json()["listPublicCollections"] is False
-
-    # Update
-    r = requests.post(
-        f"{API_PREFIX}/orgs/{default_org_id}/list-public-collections",
-        headers=admin_auth_headers,
-        json={"listPublicCollections": True},
-    )
-    assert r.json()["updated"]
-
-    # Test update is reflected in GET response
-    r = requests.get(f"{API_PREFIX}/orgs/{default_org_id}", headers=admin_auth_headers)
-    assert r.json()["listPublicCollections"]
-
-
 def test_sort_orgs(admin_auth_headers):
     # Create a few new orgs for testing
     r = requests.post(