fix: Add datasets list access control and fix datasets config display issue (#12533)

Co-authored-by: nite-knite <nkCoding@gmail.com>
2026-05-06 10:28:10 +08:00 · 2025-01-09 17:44:11 +08:00
parent f549d53b68
commit 2e97ba5700
11 changed files with 174 additions and 50 deletions
--- a/web/app/(commonLayout)/datasets/Container.tsx
+++ b/web/app/(commonLayout)/datasets/Container.tsx
@ -4,7 +4,8 @@
 import { useEffect, useMemo, useRef, useState } from 'react'
 import { useRouter } from 'next/navigation'
 import { useTranslation } from 'react-i18next'
-import { useDebounceFn } from 'ahooks'
+import { useBoolean, useDebounceFn } from 'ahooks'
+import { useQuery } from '@tanstack/react-query'

 // Components
 import ExternalAPIPanel from '../../components/datasets/external-api/external-api-panel'
@ -16,7 +17,9 @@ import TabSliderNew from '@/app/components/base/tab-slider-new'
 import TagManagementModal from '@/app/components/base/tag-management'
 import TagFilter from '@/app/components/base/tag-management/filter'
 import Button from '@/app/components/base/button'
+import Input from '@/app/components/base/input'
 import { ApiConnectionMod } from '@/app/components/base/icons/src/vender/solid/development'
+import CheckboxWithLabel from '@/app/components/datasets/create/website/base/checkbox-with-label'

 // Services
 import { fetchDatasetApiBaseUrl } from '@/service/datasets'
@ -26,16 +29,14 @@ import { useTabSearchParams } from '@/hooks/use-tab-searchparams'
 import { useStore as useTagStore } from '@/app/components/base/tag-management/store'
 import { useAppContext } from '@/context/app-context'
 import { useExternalApiPanel } from '@/context/external-api-panel-context'
-// eslint-disable-next-line import/order
-import { useQuery } from '@tanstack/react-query'
-import Input from '@/app/components/base/input'

 const Container = () => {
  const { t } = useTranslation()
  const router = useRouter()
-  const { currentWorkspace } = useAppContext()
+  const { currentWorkspace, isCurrentWorkspaceOwner } = useAppContext()
  const showTagManagementModal = useTagStore(s => s.showTagManagementModal)
  const { showExternalApiPanel, setShowExternalApiPanel } = useExternalApiPanel()
+  const [includeAll, { toggle: toggleIncludeAll }] = useBoolean(false)

  const options = useMemo(() => {
    return [
@ -90,6 +91,14 @@ const Container = () => {
        />
        {activeTab === 'dataset' && (
          <div className='flex items-center justify-center gap-2'>
+            {isCurrentWorkspaceOwner && <CheckboxWithLabel
+              isChecked={includeAll}
+              onChange={toggleIncludeAll}
+              label={t('dataset.allKnowledge')}
+              labelClassName='system-md-regular text-text-secondary'
+              className='mr-2'
+              tooltip={t('dataset.allKnowledgeDescription') as string}
+            />}
            <TagFilter type='knowledge' value={tagFilterValue} onChange={handleTagsChange} />
            <Input
              showLeftIcon
@ -113,7 +122,7 @@ const Container = () => {
      </div>
      {activeTab === 'dataset' && (
        <>
-          <Datasets containerRef={containerRef} tags={tagIDs} keywords={searchKeywords} />
+          <Datasets containerRef={containerRef} tags={tagIDs} keywords={searchKeywords} includeAll={includeAll} />
          <DatasetFooter />
          {showTagManagementModal && (
            <TagManagementModal type='knowledge' show={showTagManagementModal} />
--- a/web/app/(commonLayout)/datasets/Datasets.tsx
+++ b/web/app/(commonLayout)/datasets/Datasets.tsx
@ -6,7 +6,7 @@ import { debounce } from 'lodash-es'
 import { useTranslation } from 'react-i18next'
 import NewDatasetCard from './NewDatasetCard'
 import DatasetCard from './DatasetCard'
-import type { DataSetListResponse } from '@/models/datasets'
+import type { DataSetListResponse, FetchDatasetsParams } from '@/models/datasets'
 import { fetchDatasets } from '@/service/datasets'
 import { useAppContext } from '@/context/app-context'

@ -15,13 +15,15 @@ const getKey = (
  previousPageData: DataSetListResponse,
  tags: string[],
  keyword: string,
+  includeAll: boolean,
 ) => {
  if (!pageIndex || previousPageData.has_more) {
-    const params: any = {
+    const params: FetchDatasetsParams = {
      url: 'datasets',
      params: {
        page: pageIndex + 1,
        limit: 30,
+        include_all: includeAll,
      },
    }
    if (tags.length)
@ -37,16 +39,18 @@ type Props = {
  containerRef: React.RefObject<HTMLDivElement>
  tags: string[]
  keywords: string
+  includeAll: boolean
 }

 const Datasets = ({
  containerRef,
  tags,
  keywords,
+  includeAll,
 }: Props) => {
  const { isCurrentWorkspaceEditor } = useAppContext()
  const { data, isLoading, setSize, mutate } = useSWRInfinite(
-    (pageIndex: number, previousPageData: DataSetListResponse) => getKey(pageIndex, previousPageData, tags, keywords),
+    (pageIndex: number, previousPageData: DataSetListResponse) => getKey(pageIndex, previousPageData, tags, keywords, includeAll),
    fetchDatasets,
    { revalidateFirstPage: false, revalidateAll: true },
  )
--- a/web/app/(commonLayout)/datasets/template/template.en.mdx
+++ b/web/app/(commonLayout)/datasets/template/template.en.mdx
@ -1,5 +1,5 @@
 import { CodeGroup } from '@/app/components/develop/code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '@/app/components/develop/md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty, PropertyInstruction, Paragraph } from '@/app/components/develop/md.tsx'

 # Knowledge API

@ -80,6 +80,27 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
              - <code>max_tokens</code> The maximum length (tokens) must be validated to be shorter than the length of the parent chunk
              - <code>chunk_overlap</code> Define the overlap between adjacent chunks (optional)
      </Property>
+      <PropertyInstruction>When no parameters are set for the knowledge base, the first upload requires the following parameters to be provided; if not provided, the default parameters will be used.</PropertyInstruction>
+      <Property name='retrieval_model' type='object' key='retrieval_model'>
+        Retrieval model
+          - <code>search_method</code> (string) Search method
+            - <code>hybrid_search</code> Hybrid search
+            - <code>semantic_search</code> Semantic search
+            - <code>full_text_search</code> Full-text search
+          - <code>reranking_enable</code> (bool) Whether to enable reranking
+          - <code>reranking_mode</code> (object) Rerank model configuration
+            - <code>reranking_provider_name</code> (string) Rerank model provider
+            - <code>reranking_model_name</code> (string) Rerank model name
+          - <code>top_k</code> (int) Number of results to return
+          - <code>score_threshold_enabled</code> (bool) Whether to enable score threshold
+          - <code>score_threshold</code> (float) Score threshold
+      </Property>
+      <Property name='embedding_model' type='string' key='embedding_model'>
+        Embedding model name
+      </Property>
+      <Property name='embedding_model_provider' type='string' key='embedding_model_provider'>
+        Embedding model provider
+      </Property>
    </Properties>
  </Col>
  <Col sticky>
@ -197,6 +218,27 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
      <Property name='file' type='multipart/form-data' key='file'>
        Files that need to be uploaded.
      </Property>
+      <PropertyInstruction>When no parameters are set for the knowledge base, the first upload requires the following parameters to be provided; if not provided, the default parameters will be used.</PropertyInstruction>
+      <Property name='retrieval_model' type='object' key='retrieval_model'>
+        Retrieval model
+          - <code>search_method</code> (string) Search method
+            - <code>hybrid_search</code> Hybrid search
+            - <code>semantic_search</code> Semantic search
+            - <code>full_text_search</code> Full-text search
+          - <code>reranking_enable</code> (bool) Whether to enable reranking
+          - <code>reranking_mode</code> (object) Rerank model configuration
+            - <code>reranking_provider_name</code> (string) Rerank model provider
+            - <code>reranking_model_name</code> (string) Rerank model name
+          - <code>top_k</code> (int) Number of results to return
+          - <code>score_threshold_enabled</code> (bool) Whether to enable score threshold
+          - <code>score_threshold</code> (float) Score threshold
+      </Property>
+      <Property name='embedding_model' type='string' key='embedding_model'>
+        Embedding model name
+      </Property>
+      <Property name='embedding_model_provider' type='string' key='embedding_model_provider'>
+        Embedding model provider
+      </Property>
    </Properties>
  </Col>
  <Col sticky>
@ -1188,10 +1230,10 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
        - <code>reranking_mode</code> (object) Rerank model configuration, required if reranking is enabled
            - <code>reranking_provider_name</code> (string) Rerank model provider
            - <code>reranking_model_name</code> (string) Rerank model name
-        - <code>weights</code> (double) Semantic search weight setting in hybrid search mode
+        - <code>weights</code> (float) Semantic search weight setting in hybrid search mode
        - <code>top_k</code> (integer) Number of results to return (optional)
        - <code>score_threshold_enabled</code> (bool) Whether to enable score threshold
-        - <code>score_threshold</code> (double) Score threshold
+        - <code>score_threshold</code> (float) Score threshold
      </Property>
      <Property name='external_retrieval_model' type='object' key='external_retrieval_model'>
          Unused field
--- a/web/app/(commonLayout)/datasets/template/template.zh.mdx
+++ b/web/app/(commonLayout)/datasets/template/template.zh.mdx
@ -1,5 +1,5 @@
 import { CodeGroup } from '@/app/components/develop/code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '@/app/components/develop/md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty, PropertyInstruction, Paragraph } from '@/app/components/develop/md.tsx'

 # 知识库 API

@ -80,6 +80,27 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
              - <code>max_tokens</code> 最大长度 (token) 需要校验小于父级的长度
              - <code>chunk_overlap</code> 分段重叠指的是在对数据进行分段时，段与段之间存在一定的重叠部分（选填）
      </Property>
+      <PropertyInstruction>当知识库未设置任何参数的时候，首次上传需要提供以下参数，未提供则使用默认选项：</PropertyInstruction>
+      <Property name='retrieval_model' type='object' key='retrieval_model'>
+        检索模式
+          - <code>search_method</code> (string) 检索方法
+            - <code>hybrid_search</code> 混合检索
+            - <code>semantic_search</code> 语义检索
+            - <code>full_text_search</code> 全文检索
+          - <code>reranking_enable</code> (bool) 是否开启rerank
+          - <code>reranking_model</code> (object) Rerank 模型配置
+            - <code>reranking_provider_name</code> (string) Rerank 模型的提供商
+            - <code>reranking_model_name</code> (string) Rerank 模型的名称
+          - <code>top_k</code> (int) 召回条数
+          - <code>score_threshold_enabled</code> (bool)是否开启召回分数限制
+          - <code>score_threshold</code> (float) 召回分数限制
+      </Property>
+      <Property name='embedding_model' type='string' key='embedding_model'>
+        Embedding 模型名称
+      </Property>
+      <Property name='embedding_model_provider' type='string' key='embedding_model_provider'>
+        Embedding 模型供应商
+      </Property>
    </Properties>
  </Col>
  <Col sticky>
@ -197,6 +218,27 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
      <Property name='file' type='multipart/form-data' key='file'>
        需要上传的文件。
      </Property>
+      <PropertyInstruction>当知识库未设置任何参数的时候，首次上传需要提供以下参数，未提供则使用默认选项：</PropertyInstruction>
+      <Property name='retrieval_model' type='object' key='retrieval_model'>
+        检索模式
+          - <code>search_method</code> (string) 检索方法
+            - <code>hybrid_search</code> 混合检索
+            - <code>semantic_search</code> 语义检索
+            - <code>full_text_search</code> 全文检索
+          - <code>reranking_enable</code> (bool) 是否开启rerank
+          - <code>reranking_model</code> (object) Rerank 模型配置
+            - <code>reranking_provider_name</code> (string) Rerank 模型的提供商
+            - <code>reranking_model_name</code> (string) Rerank 模型的名称
+          - <code>top_k</code> (int) 召回条数
+          - <code>score_threshold_enabled</code> (bool)是否开启召回分数限制
+          - <code>score_threshold</code> (float) 召回分数限制
+      </Property>
+      <Property name='embedding_model' type='string' key='embedding_model'>
+        Embedding 模型名称
+      </Property>
+      <Property name='embedding_model_provider' type='string' key='embedding_model_provider'>
+        Embedding 模型供应商
+      </Property>
    </Properties>
  </Col>
  <Col sticky>
@ -1186,13 +1228,13 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
          - <code>full_text_search</code> 全文检索
          - <code>hybrid_search</code> 混合检索
        - <code>reranking_enable</code> (bool) 是否启用 Reranking，非必填，如果检索模式为 semantic_search 模式或者 hybrid_search 则传值
-        - <code>reranking_mode</code> (object) Rerank模型配置，非必填，如果启用了 reranking 则传值
+        - <code>reranking_mode</code> (object) Rerank 模型配置，非必填，如果启用了 reranking 则传值
            - <code>reranking_provider_name</code> (string) Rerank 模型提供商
            - <code>reranking_model_name</code> (string) Rerank 模型名称
-        - <code>weights</code> (double) 混合检索模式下语意检索的权重设置
+        - <code>weights</code> (float) 混合检索模式下语意检索的权重设置
        - <code>top_k</code> (integer) 返回结果数量，非必填
        - <code>score_threshold_enabled</code> (bool) 是否开启 score 阈值
-        - <code>score_threshold</code> (double) Score 阈值
+        - <code>score_threshold</code> (float) Score 阈值
      </Property>
      <Property name='external_retrieval_model' type='object' key='external_retrieval_model'>
          未启用字段