Refactor: migrate document change status API

api/apps/document_app.py

@@ -201,76 +201,6 @@ def thumbnails():
         return server_error_response(e)
 
 
-@manager.route("/change_status", methods=["POST"])  # noqa: F821
-@login_required
-@validate_request("doc_ids", "status")
-async def change_status():
-    req = await get_request_json()
-    doc_ids = req.get("doc_ids", [])
-    status = str(req.get("status", ""))
-
-    if status not in ["0", "1"]:
-        return get_json_result(data=False, message='"Status" must be either 0 or 1!', code=RetCode.ARGUMENT_ERROR)
-
-    result = {}
-    has_error = False
-    for doc_id in doc_ids:
-        if not DocumentService.accessible(doc_id, current_user.id):
-            result[doc_id] = {"error": "No authorization."}
-            has_error = True
-            continue
-
-        try:
-            e, doc = DocumentService.get_by_id(doc_id)
-            if not e:
-                result[doc_id] = {"error": "No authorization."}
-                has_error = True
-                continue
-            e, kb = KnowledgebaseService.get_by_id(doc.kb_id)
-            if not e:
-                result[doc_id] = {"error": "Can't find this dataset!"}
-                has_error = True
-                continue
-            current_status = str(doc.status)
-            if current_status == status:
-                result[doc_id] = {"status": status}
-                continue
-            if not DocumentService.update_by_id(doc_id, {"status": str(status)}):
-                result[doc_id] = {"error": "Database error (Document update)!"}
-                has_error = True
-                continue
-
-            status_int = int(status)
-            if getattr(doc, "chunk_num", 0) > 0:
-                try:
-                    ok = settings.docStoreConn.update(
-                        {"doc_id": doc_id},
-                        {"available_int": status_int},
-                        search.index_name(kb.tenant_id),
-                        doc.kb_id,
-                    )
-                except Exception as exc:
-                    msg = str(exc)
-                    if "3022" in msg:
-                        result[doc_id] = {"error": "Document store table missing."}
-                    else:
-                        result[doc_id] = {"error": f"Document store update failed: {msg}"}
-                    has_error = True
-                    continue
-                if not ok:
-                    result[doc_id] = {"error": "Database error (docStore update)!"}
-                    has_error = True
-                    continue
-            result[doc_id] = {"status": status}
-        except Exception as e:
-            result[doc_id] = {"error": f"Internal server error: {str(e)}"}
-            has_error = True
-
-    if has_error:
-        return get_json_result(data=result, message="Partial failure", code=RetCode.SERVER_ERROR)
-    return get_json_result(data=result)
-
-
 @manager.route("/run", methods=["POST"])  # noqa: F821
 @login_required
 @validate_request("doc_ids", "run")

api/apps/restful_apis/document_api.py

@@ -1019,3 +1019,122 @@ async def update_metadata(tenant_id, dataset_id):
     target_doc_ids = list(target_doc_ids)
     updated = DocMetadataService.batch_update_metadata(dataset_id, target_doc_ids, updates, deletes)
     return get_result(data={"updated": updated, "matched_docs": len(target_doc_ids)})
+
+
+@manager.route("/datasets/<dataset_id>/documents/batch-update-status", methods=["POST"])  # noqa: F821
+@login_required
+@add_tenant_id_to_kwargs
+async def batch_update_document_status(tenant_id, dataset_id):
+    """
+    Batch update status of documents within a dataset.
+    ---
+    tags:
+      - Documents
+    security:
+      - ApiKeyAuth: []
+    parameters:
+      - in: path
+        name: dataset_id
+        type: string
+        required: true
+        description: ID of the dataset.
+      - in: header
+        name: Authorization
+        type: string
+        required: true
+        description: Bearer token for authentication.
+      - in: body
+        name: body
+        description: Document status update parameters.
+        required: true
+        schema:
+          type: object
+          required:
+            - doc_ids
+            - status
+          properties:
+            doc_ids:
+              type: array
+              items:
+                type: string
+              description: List of document IDs to update.
+            status:
+              type: string
+              enum: ["0", "1"]
+              description: New status (0 = disabled, 1 = enabled).
+    responses:
+      200:
+        description: Document statuses updated successfully.
+    """
+    from common import settings
+    from rag.nlp import search
+
+    req = await get_request_json()
+    doc_ids = req.get("doc_ids", [])
+    status = str(req.get("status", -1))
+
+    if status not in ["0", "1"]:
+        return get_error_argument_result(message=f'"Status" must be either 0 or 1:{status}!')
+
+    # Verify dataset ownership
+    if not KnowledgebaseService.query(id=dataset_id, tenant_id=tenant_id):
+        return get_error_data_result(message="You don't own the dataset.")
+
+    e, kb = KnowledgebaseService.get_by_id(dataset_id)
+    if not e:
+        return get_error_data_result(message="Can't find this dataset!")
+
+    result = {}
+    has_error = False
+    for doc_id in doc_ids:
+        try:
+            e, doc = DocumentService.get_by_id(doc_id)
+            if not e:
+                result[doc_id] = {"error": "Document not found"}
+                has_error = True
+                continue
+
+            if doc.kb_id != dataset_id:
+                logging.warning(f"Document {doc.kb_id} not in dataset {dataset_id}")
+                result[doc_id] = {"error": "Document not found in this dataset."}
+                has_error = True
+                continue
+
+            current_status = str(doc.status)
+            if current_status == status:
+                result[doc_id] = {"status": status}
+                continue
+            if not DocumentService.update_by_id(doc_id, {"status": str(status)}):
+                result[doc_id] = {"error": "Database error (Document update)!"}
+                has_error = True
+                continue
+
+            status_int = int(status)
+            if getattr(doc, "chunk_num", 0) > 0:
+                try:
+                    ok = settings.docStoreConn.update(
+                        {"doc_id": doc_id},
+                        {"available_int": status_int},
+                        search.index_name(kb.tenant_id),
+                        doc.kb_id,
+                    )
+                except Exception as exc:
+                    msg = str(exc)
+                    if "3022" in msg:
+                        result[doc_id] = {"error": "Document store table missing."}
+                    else:
+                        result[doc_id] = {"error": f"Document store update failed: {msg}"}
+                    has_error = True
+                    continue
+                if not ok:
+                    result[doc_id] = {"error": "Database error (docStore update)!"}
+                    has_error = True
+                    continue
+            result[doc_id] = {"status": status}
+        except Exception as e:
+            result[doc_id] = {"error": f"Internal server error: {str(e)}"}
+            has_error = True
+
+    if has_error:
+        return get_json_result(data=result, message="Partial failure", code=RetCode.SERVER_ERROR)
+    return get_json_result(data=result)

test/testcases/test_web_api/test_common.py

@@ -423,8 +423,16 @@ def document_update_metadata_setting(auth, dataset_id, doc_id, payload=None, *,
     return res.json()
 
 
-def document_change_status(auth, payload=None, *, headers=HEADERS, data=None):
-    res = requests.post(url=f"{HOST_ADDRESS}{DOCUMENT_APP_URL}/change_status", headers=headers, auth=auth, json=payload, data=data)
+def document_change_status(auth, dataset_id, payload=None, *, headers=HEADERS, data=None):
+    """
+    Batch update document status within a dataset.
+
+    Args:
+        auth: Authentication credentials
+        dataset_id: ID of the dataset
+        payload: Request body containing doc_ids and status
+    """
+    res = requests.post(url=f"{HOST_ADDRESS}{DATASETS_URL}/{dataset_id}/documents/batch-update-status", headers=headers, auth=auth, json=payload, data=data)
     return res.json()