Skip to content

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

Refactor: add page_size support (#18876)

Wang Qif74e04581ad7
diff --git a/api/apps/restful_apis/bot_api.py b/api/apps/restful_apis/bot_api.py
index e42950f359..2a428f6f1d 100644
--- a/api/apps/restful_apis/bot_api.py
+++ b/api/apps/restful_apis/bot_api.py
@@ -337,7 +337,7 @@ async def ask_about_embedded(tenant_id=None):
async def retrieval_test_embedded(tenant_id=None):
req = await get_request_json()
page = validate_rest_api_page(req.get("page", DEFAULT_PAGE))
- size = validate_rest_api_page_size(req.get("size", DEFAULT_PAGE_SIZE))
+ size = validate_rest_api_page_size(req.get("page_size", req.get("size", DEFAULT_PAGE_SIZE)))
question = req["question"]
kb_ids = req["kb_id"]
if isinstance(kb_ids, str):
diff --git a/api/apps/restful_apis/dataset_api.py b/api/apps/restful_apis/dataset_api.py
index 4f956df457..69e5cc6075 100644
--- a/api/apps/restful_apis/dataset_api.py
+++ b/api/apps/restful_apis/dataset_api.py
@@ -513,7 +513,7 @@ async def search_datasets(tenant_id):
"""Search (retrieval test) across multiple datasets.
POST /api/v1/datasets/search
- JSON body: {"dataset_ids": list[str] (required), "question": str (required), "doc_ids": list[str], "knn_top_k": int (default 1024), "knn_num_candidates": int (default 2048), "page": int, "size": int,
+ JSON body: {"dataset_ids": list[str] (required), "question": str (required), "doc_ids": list[str], "knn_top_k": int (default 1024), "knn_num_candidates": int (default 2048), "page": int, "page_size": int, "size": int (fallback),
"similarity_threshold": float, "vector_similarity_weight": float, "use_kg": bool,
"cross_languages": list[str], "keyword": bool, "meta_data_filter": dict, "include_knowledge_compilation": bool (default true)}
The legacy "top_k" parameter is accepted as an alias for "knn_top_k".
@@ -538,7 +538,7 @@ async def search(tenant_id, dataset_id):
"""Search (retrieval test) within a dataset.
POST /api/v1/datasets/<dataset_id>/search
- JSON body: {"question": str (required), "doc_ids": list[str], "knn_top_k": int (default 1024), "knn_num_candidates": int (default 2048), "page": int, "size": int,
+ JSON body: {"question": str (required), "doc_ids": list[str], "knn_top_k": int (default 1024), "knn_num_candidates": int (default 2048), "page": int, "page_size": int, "size": int (fallback),
"similarity_threshold": float, "vector_similarity_weight": float, "use_kg": bool,
"cross_languages": list[str], "keyword": bool, "meta_data_filter": dict, "include_knowledge_compilation": bool (default true)}
The legacy "top_k" parameter is accepted as an alias for "knn_top_k".
diff --git a/api/apps/services/dataset_api_service.py b/api/apps/services/dataset_api_service.py
index 505d7b677a..a6cd3409fc 100644
--- a/api/apps/services/dataset_api_service.py
+++ b/api/apps/services/dataset_api_service.py
@@ -1054,7 +1054,7 @@ async def search(dataset_id: str, tenant_id: str, req: dict):
)
page = int(req.get("page", 1))
- size = int(req.get("size", 30))
+ size = int(req.get("page_size") or req.get("size", 30))
rerank_candidates_count = int(req.get("rerank_candidates_count", 64))
question = req.get("question", "")
doc_ids = req.get("doc_ids", [])
@@ -1439,7 +1439,7 @@ async def search_datasets(tenant_id: str, req: dict):
kb_ids = req.get("dataset_ids", [])
page = int(req.get("page", 1))
- size = int(req.get("size", 30))
+ size = int(req.get("page_size") or req.get("size", 30))
rerank_candidates_count = int(req.get("rerank_candidates_count", 64))
question = req.get("question", "")
doc_ids = req.get("doc_ids", [])
diff --git a/api/utils/validation_utils.py b/api/utils/validation_utils.py
index d61915266b..7e52e4726f 100644
--- a/api/utils/validation_utils.py
+++ b/api/utils/validation_utils.py
@@ -981,7 +981,8 @@ class SearchDatasetReq(BaseModel):
question: Annotated[str, StringConstraints(strip_whitespace=True, min_length=1), Field(...)]
doc_ids: Annotated[list[str], Field(default=[])]
page: Annotated[int, Field(default=1, ge=1)]
- size: Annotated[int, Field(default=30, ge=1)]
+ page_size: Annotated[int | None, Field(default=None, ge=1, le=100)]
+ size: Annotated[int, Field(default=30, ge=1, le=100)]
rerank_candidates_count: Annotated[int, Field(default=64, ge=1)]
knn_top_k: Annotated[int, Field(default=1024, ge=1, le=2048, validation_alias=AliasChoices("knn_top_k", "top_k"))]
knn_num_candidates: Annotated[int, Field(default=2048, ge=1, le=10000)]
@@ -1012,7 +1013,8 @@ class SearchDatasetsReq(BaseModel):
question: Annotated[str, StringConstraints(strip_whitespace=True, min_length=1), Field(...)]
doc_ids: Annotated[list[str], Field(default=[])]
page: Annotated[int, Field(default=1, ge=1)]
- size: Annotated[int, Field(default=30, ge=1)]
+ page_size: Annotated[int | None, Field(default=None, ge=1, le=100)]
+ size: Annotated[int, Field(default=30, ge=1, le=100)]
rerank_candidates_count: Annotated[int, Field(default=64, ge=1)]
knn_top_k: Annotated[int, Field(default=1024, ge=1, le=2048, validation_alias=AliasChoices("knn_top_k", "top_k"))]
knn_num_candidates: Annotated[int, Field(default=2048, ge=1, le=10000)]