Implementation:Infiniflow Ragflow NextSearchHooks

Knowledge Sources	Infiniflow_Ragflow
Domains	Frontend, Search
Last Updated	2026-02-12 06:00 GMT

Overview

Concrete core search hooks with SSE streaming and state management provided by the RAGFlow frontend.

Description

The next-search/hooks module exports hooks that power the search page: `useGetSharedSearchParams` (extracts shared search URL params), `useSearchFetchMindMap` / `useShowMindMapDrawer` (fetches and controls mind map visualization), `useTestChunkRetrieval` / `useTestChunkAllRetrieval` (chunk retrieval testing with pagination), `useTestRetrieval` (combines retrieval with document filtering), `useFetchRelatedQuestions` (fetches related question suggestions), `useSendQuestion` (orchestrates SSE-based AI answer streaming, chunk retrieval, and related questions in parallel), `useSearching` (top-level composition hook that wires everything together including mind map, pagination, and document preview drawer), `useCheckSettings` (validates search app configuration), and `usePendingMindMap` (loading progress timer).

Usage

Import `useSearching` as the primary hook in the search page component. Import individual hooks when building custom search UI compositions or testing utilities.

Code Reference

Source Location

• Repository: Infiniflow_Ragflow
• File: web/src/pages/next-search/hooks.ts
• Lines: 1-604

Signature

export interface ISearchingProps {
  searchText?: string;
  data: ISearchAppDetailProps;
  setIsSearching?: Dispatch<SetStateAction<boolean>>;
  setSearchText?: Dispatch<SetStateAction<string>>;
}

export type ISearchReturnProps = ReturnType<typeof useSearching>;

export const useGetSharedSearchParams = () => { from; sharedId; locale; tenantId; data; visibleAvatar };
export const useSearchFetchMindMap = () => { data; loading; fetchMindMap };
export const useShowMindMapDrawer = (kbIds: string[], question: string, searchId?: string) => { ... };
export const useTestChunkRetrieval = (tenantId?: string) => { data; loading; testChunk };
export const useTestChunkAllRetrieval = (tenantId?: string) => { data; loading; testChunkAll };
export const useTestRetrieval = (kbIds: string[], searchStr: string, sendingLoading: boolean) => { ... };
export const useFetchRelatedQuestions = (tenantId?: string, searchId?: string) => { data; loading; fetchRelatedQuestions };
export const useSendQuestion = (kbIds: string[], tenantId?: string, searchId?: string, related_search?: boolean) => { ... };
export const useSearching = (props: ISearchingProps) => ISearchReturnProps;
export const useCheckSettings = (data: ISearchAppDetailProps) => { openSetting: boolean };
export const usePendingMindMap = () => number;

Import

import { useSearching, useCheckSettings, ISearchReturnProps } from '@/pages/next-search/hooks';

I/O Contract

Inputs

Name	Type	Required	Description
searchText	string	No	Initial search text to trigger on mount
data	ISearchAppDetailProps	Yes	Search application configuration including kb_ids, summary, related_search settings
setIsSearching	Dispatch<SetStateAction<boolean>>	No	Callback to toggle search state in parent
setSearchText	Dispatch<SetStateAction<string>>	No	Callback to clear search text in parent after consumption

Outputs

Name	Type	Description
answer	IAnswer	Current AI-generated answer from SSE stream
sendingLoading	boolean	Whether the AI answer is still being streamed
chunks	IChunk[]	Retrieved document chunks matching the query
total	number	Total number of matching chunks
relatedQuestions	string[]	Suggested related questions (up to 5)
searchStr	string	Current search input value
handleSearch	(value: string) => void	Triggers search with given query text
pagination	object	Current pagination state
onChange	(page: number, pageSize: number) => void	Pagination change handler

Usage Examples

import { useSearching } from '@/pages/next-search/hooks';

const searchProps = useSearching({
  searchText: initialQuery,
  data: searchAppDetail,
  setIsSearching,
});

return <SearchingView {...searchProps} searchData={searchAppDetail} />;

Related Pages

• Environment:Infiniflow_Ragflow_Frontend_Node_Environment