Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

Fix: restrict xxx_ids to limit 100 (#18074)

Wang Qibd876c796ec9
diff --git a/api/apps/restful_apis/agent_api.py b/api/apps/restful_apis/agent_api.py
index 3b93217a6e..703961d101 100644
--- a/api/apps/restful_apis/agent_api.py
+++ b/api/apps/restful_apis/agent_api.py
@@ -64,7 +64,7 @@ from api.utils.api_utils import (
server_error_response,
validate_request,
)
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from common import settings
from common.ssrf_guard import assert_host_is_safe
from common.constants import RetCode
@@ -698,6 +698,10 @@ def list_agents(tenant_id):
canvas_type = request.args.get("canvas_type")
owner_ids = [item for item in request.args.get("owner_ids", "").strip().split(",") if item]
tags = [item for item in request.args.get("tags", "").strip().split(",") if item]
+ try:
+ validate_rest_api_ids(owner_ids, "owner_ids")
+ except ValueError as e:
+ return get_result(code=RetCode.ARGUMENT_ERROR, message=str(e))
page_number = validate_rest_api_page(request.args.get("page", DEFAULT_PAGE))
items_per_page = validate_rest_api_page_size(request.args.get("page_size", DEFAULT_PAGE_SIZE))
diff --git a/api/apps/restful_apis/chat_api.py b/api/apps/restful_apis/chat_api.py
index 140750ba0d..1d9658595a 100644
--- a/api/apps/restful_apis/chat_api.py
+++ b/api/apps/restful_apis/chat_api.py
@@ -51,7 +51,7 @@ from api.utils.api_utils import (
server_error_response,
validate_request,
)
-from api.utils.pagination_utils import validate_rest_api_page, validate_rest_api_page_size, DEFAULT_PAGE, DEFAULT_PAGE_SIZE
+from api.utils.pagination_utils import validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size, DEFAULT_PAGE, DEFAULT_PAGE_SIZE
from common.constants import LLMType, RetCode, StatusEnum
from common import settings
from common.misc_utils import get_uuid, thread_pool_exec
@@ -510,6 +510,11 @@ async def list_chats():
if orderby not in ("create_time", "update_time", "name"):
return get_json_result(code=RetCode.ARGUMENT_ERROR, message=f"invalid orderby field: {orderby}")
+ try:
+ validate_rest_api_ids(owner_ids, "owner_ids")
+ except ValueError as ex:
+ return get_json_result(code=RetCode.ARGUMENT_ERROR, message=str(ex))
+
try:
# Invalid or negative pagination values fall back to defaults
# instead of leaking internal conversion/SQL errors.
diff --git a/api/apps/restful_apis/chunk_api.py b/api/apps/restful_apis/chunk_api.py
index 2d254b8d96..8e72874825 100644
--- a/api/apps/restful_apis/chunk_api.py
+++ b/api/apps/restful_apis/chunk_api.py
@@ -49,7 +49,7 @@ from api.utils.api_utils import (
get_result,
server_error_response,
)
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from api.utils.image_utils import store_chunk_image
from api.utils.reference_metadata_utils import (
enrich_chunks_with_document_metadata,
@@ -478,6 +478,10 @@ async def list_chunks(tenant_id, dataset_id, document_id):
size = validate_rest_api_page_size(req.get("page_size", DEFAULT_PAGE_SIZE))
question = req.get("keywords", "")
chunk_ids = _get_query_id_list(req, "chunk_ids")
+ try:
+ validate_rest_api_ids(chunk_ids, "chunk_ids")
+ except ValueError as e:
+ return get_result(code=RetCode.ARGUMENT_ERROR, message=str(e))
query = {
"doc_ids": [document_id],
"page": page,
diff --git a/api/apps/restful_apis/dataset_api.py b/api/apps/restful_apis/dataset_api.py
index 54e68a3b35..ccd79fd095 100644
--- a/api/apps/restful_apis/dataset_api.py
+++ b/api/apps/restful_apis/dataset_api.py
@@ -20,7 +20,7 @@ from quart import request, make_response
from common.constants import RetCode
from api.apps import login_required, current_user
from api.utils.api_utils import get_error_argument_result, get_error_data_result, get_json_result, get_result, add_tenant_id_to_kwargs
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from api.utils.validation_utils import (
CreateDatasetReq,
DeleteDatasetReq,
@@ -44,6 +44,7 @@ def aggregate_tags(tenant_id):
return get_error_data_result(message="Lack of dataset_ids in query parameters")
try:
+ validate_rest_api_ids(dataset_ids, "dataset_ids")
success, result = dataset_api_service.aggregate_tags(dataset_ids, tenant_id)
if success:
return get_result(data=result)
@@ -66,6 +67,7 @@ def get_flattened_metadata(tenant_id):
return get_error_data_result(message="Lack of dataset_ids in query parameters")
try:
+ validate_rest_api_ids(dataset_ids, "dataset_ids")
success, result = dataset_api_service.get_flattened_metadata(dataset_ids, tenant_id)
if success:
return get_result(data=result)
@@ -602,7 +604,7 @@ async def has_any_wiki(tenant_id, dataset_id):
async def list_wiki_pages(tenant_id, dataset_id):
"""List artifact pages for the dataset Artifact tab.
- GET /api/v1/datasets/<dataset_id>/artifacts?page=1&page_size=200&page_type=entity&topic=topic&keywords=query
+ GET /api/v1/datasets/<dataset_id>/artifacts?page=1&page_size=100&page_type=entity&topic=topic&keywords=query
Success: {"code": 0, "data": {"total": int, "items": [{slug, title, page_type}]}}
"""
try:
@@ -638,7 +640,7 @@ async def list_wiki_pages(tenant_id, dataset_id):
async def list_wiki_topics(tenant_id, dataset_id):
"""List wiki topics for the dataset Artifact tab.
- GET /api/v1/datasets/<dataset_id>/artifacts/topics?page=1&page_size=200&keywords=query
+ GET /api/v1/datasets/<dataset_id>/artifacts/topics?page=1&page_size=100&keywords=query
Success: {"code": 0, "data": {"total": int, "items": [{topic, title, slug}]}}
"""
try:
diff --git a/api/apps/restful_apis/document_api.py b/api/apps/restful_apis/document_api.py
index 3e92e4ecd8..9d25875238 100644
--- a/api/apps/restful_apis/document_api.py
+++ b/api/apps/restful_apis/document_api.py
@@ -61,7 +61,7 @@ from api.utils.api_utils import (
get_error_argument_result,
check_duplicate_ids,
)
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from api.utils.validation_utils import (
UpdateDocumentReq,
format_validation_error_message,
@@ -352,6 +352,11 @@ async def metadata_summary(dataset_id, tenant_id):
# Get doc_ids from query parameters (comma-separated string)
doc_ids_param = request.args.get("doc_ids", "")
doc_ids = doc_ids_param.split(",") if doc_ids_param else None
+ try:
+ validate_rest_api_ids(doc_ids, "doc_ids")
+ except ValueError as e:
+ return get_error_argument_result(str(e))
+
try:
summary = DocMetadataService.get_metadata_summary(dataset_id, doc_ids)
return get_result(data={"summary": summary})
@@ -930,6 +935,10 @@ def _get_docs_with_request(req, dataset_id: str):
return RetCode.DATA_ERROR, f"you don't own the document {doc_name}", [], 0
doc_ids = q.getlist("ids")
+ try:
+ validate_rest_api_ids(doc_ids)
+ except ValueError as e:
+ return RetCode.ARGUMENT_ERROR, str(e), [], 0
if doc_id and len(doc_ids) > 0:
return RetCode.DATA_ERROR, f"Should not provide both 'id':{doc_id} and 'ids'{doc_ids}"
if len(doc_ids) > 0:
@@ -1318,8 +1327,12 @@ def list_thumbnails():
return get_json_result(data=False, message='Lack of "Document ID"', code=RetCode.ARGUMENT_ERROR)
try:
- docs = DocumentService.get_thumbnails(doc_ids)
+ validate_rest_api_ids(doc_ids, "doc_ids")
+ except ValueError as e:
+ return get_error_argument_result(str(e))
+ try:
+ docs = DocumentService.get_thumbnails(doc_ids)
for doc_item in docs:
if doc_item["thumbnail"] and not doc_item["thumbnail"].startswith(IMG_BASE64_PREFIX):
doc_item["thumbnail"] = f"/api/v1/documents/images/{doc_item['kb_id']}-{doc_item['thumbnail']}"
diff --git a/api/apps/restful_apis/mcp_api.py b/api/apps/restful_apis/mcp_api.py
index bb71ed21d9..f2988b0865 100644
--- a/api/apps/restful_apis/mcp_api.py
+++ b/api/apps/restful_apis/mcp_api.py
@@ -21,7 +21,7 @@ from api.db.db_models import MCPServer
from api.db.services.mcp_server_service import MCPServerService
from api.db.services.user_service import TenantService
from api.utils.api_utils import get_data_error_result, get_json_result, get_mcp_tools, get_request_json, server_error_response, validate_request
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from api.utils.web_utils import get_float, safe_json_parse
from common.constants import VALID_MCP_SERVER_TYPES
from common.mcp_tool_call_conn import MCPToolCallSession, close_multiple_mcp_toolcall_sessions
@@ -80,6 +80,13 @@ async def list_mcp() -> Response:
desc = True
mcp_ids = _get_mcp_ids_from_args()
+ try:
+ validate_rest_api_ids(mcp_ids, "mcp_ids")
+ except ValueError as e:
+ from api.utils.api_utils import get_error_argument_result
+
+ return get_error_argument_result(str(e))
+
try:
servers = MCPServerService.get_servers(current_user.id, mcp_ids, 0, 0, orderby, desc, keywords) or []
total = len(servers)
diff --git a/api/apps/restful_apis/memory_api.py b/api/apps/restful_apis/memory_api.py
index 5958924882..90d18c0c6f 100644
--- a/api/apps/restful_apis/memory_api.py
+++ b/api/apps/restful_apis/memory_api.py
@@ -24,7 +24,7 @@ from api.apps import login_required, current_user
from api.utils.api_utils import validate_request, get_request_json, get_error_argument_result, get_json_result
from api.apps.services import memory_api_service
from api.db.joint_services.tenant_model_service import ensure_tenant_model_ids_for_params
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
@manager.route("/memories", methods=["POST"]) # noqa: F821
@@ -143,6 +143,13 @@ async def list_memory():
keywords = request.args.get("keywords")
page = validate_rest_api_page(request.args.get("page", DEFAULT_PAGE))
page_size = validate_rest_api_page_size(request.args.get("page_size", DEFAULT_PAGE_SIZE))
+ try:
+ for field_name in ("owner_ids", "ids"):
+ values = [item.strip() for item in str(filter_params.get(field_name) or "").split(",") if item.strip()]
+ validate_rest_api_ids(values, field_name)
+ except ValueError as e:
+ return get_error_argument_result(str(e))
+
try:
res = await memory_api_service.list_memory(filter_params, keywords, page, page_size)
return get_json_result(message=True, data=res)
@@ -176,6 +183,11 @@ async def get_memory_messages(memory_id):
keywords = keywords.strip()
page = validate_rest_api_page(args.get("page", 1))
page_size = validate_rest_api_page_size(args.get("page_size", 50))
+ try:
+ validate_rest_api_ids(agent_ids, "agent_id")
+ except ValueError as e:
+ return get_error_argument_result(str(e))
+
try:
res = await memory_api_service.get_memory_messages(memory_id, agent_ids, keywords, page, page_size)
return get_json_result(message=True, data=res)
@@ -267,6 +279,7 @@ async def search_message():
similarity_threshold = float(args.get("similarity_threshold", 0.2))
keywords_similarity_weight = float(args.get("keywords_similarity_weight", 0.7))
try:
+ validate_rest_api_ids(memory_ids, "memory_id")
top_n = validate_rest_api_page_size(int(args.get("top_n", 5)))
except ValueError as exc:
return get_error_argument_result(str(exc))
@@ -290,6 +303,7 @@ async def get_messages():
agent_id = args.get("agent_id", "")
session_id = args.get("session_id", "")
try:
+ validate_rest_api_ids(memory_ids, "memory_id")
limit = validate_rest_api_page_size(int(args.get("limit", 10)))
except ValueError as exc:
return get_error_argument_result(str(exc))
diff --git a/api/apps/restful_apis/search_api.py b/api/apps/restful_apis/search_api.py
index ef85626aab..939a8ca488 100644
--- a/api/apps/restful_apis/search_api.py
+++ b/api/apps/restful_apis/search_api.py
@@ -31,7 +31,7 @@ from api.db.services.user_service import TenantService, UserTenantService
from common.misc_utils import get_uuid
from common.constants import RetCode, StatusEnum
from api.utils.api_utils import get_data_error_result, get_json_result, get_request_json, server_error_response, validate_request
-from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_page, validate_rest_api_page_size
+from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
def _full_text_weight(vector_similarity_weight):
@@ -84,6 +84,11 @@ def list_searches():
desc = request.args.get("desc", "true").lower() != "false"
owner_ids = request.args.getlist("owner_ids")
+ try:
+ validate_rest_api_ids(owner_ids, "owner_ids")
+ except ValueError as e:
+ return get_json_result(code=RetCode.ARGUMENT_ERROR, message=str(e))
+
try:
if not owner_ids:
tenants = []
diff --git a/api/utils/pagination_utils.py b/api/utils/pagination_utils.py
index 2edf5fd074..03ddf2e818 100644
--- a/api/utils/pagination_utils.py
+++ b/api/utils/pagination_utils.py
@@ -17,6 +17,7 @@
DEFAULT_PAGE = 1
DEFAULT_PAGE_SIZE = 30
REST_API_MAX_PAGE_SIZE = 100
+REST_API_MAX_IDS = 100
def validate_rest_api_page(page) -> int:
@@ -41,3 +42,10 @@ def validate_rest_api_page_size(page_size) -> int:
if int_page_size > REST_API_MAX_PAGE_SIZE:
raise ValueError(f"page_size must be less than or equal to {REST_API_MAX_PAGE_SIZE}")
return int_page_size
+
+
+def validate_rest_api_ids(ids: list | None, field_name: str = "ids") -> list | None:
+ """Validate REST API ID lists against the public maximum."""
+ if ids is not None and len(ids) > REST_API_MAX_IDS:
+ raise ValueError(f"{field_name} must contain at most {REST_API_MAX_IDS} IDs")
+ return ids
diff --git a/api/utils/validation_utils.py b/api/utils/validation_utils.py
index 0f7b22f970..ddb455762c 100644
--- a/api/utils/validation_utils.py
+++ b/api/utils/validation_utils.py
@@ -30,7 +30,7 @@ from werkzeug.exceptions import BadRequest, UnsupportedMediaType
from api.constants import DATASET_NAME_LIMIT, FILE_NAME_LEN_LIMIT
from api.db import FileType
-from api.utils.pagination_utils import validate_rest_api_page_size
+from api.utils.pagination_utils import REST_API_MAX_IDS, validate_rest_api_page_size
from common.constants import RetCode
@@ -1043,7 +1043,7 @@ class BaseListReq(BaseModel):
class ListDatasetReq(BaseListReq):
"""Request model for listing datasets."""
- ids: Annotated[list[str] | None, Field(default=None)]
+ ids: Annotated[list[str] | None, Field(default=None, max_length=REST_API_MAX_IDS)]
include_parsing_status: Annotated[bool, Field(default=False)]
ext: Annotated[dict, Field(default={})]