Fix: Added table of contents extraction functionality and optimized form item layout #9869 (#10492)

### What problem does this PR solve? Fix: Added table of contents extraction functionality and optimized form item layout #9869 - Added `EnableTocToggle` component to toggle table of contents extraction on and off - Added multiple parser configuration components (such as naive, book, laws, etc.), displaying different parser components based on built-in slicing methods ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-08 20:42:30 +08:00 · 2025-10-11 18:45:55 +08:00
parent 6a0f448419
commit c21cea2038
35 changed files with 694 additions and 173 deletions
--- a/web/src/components/chunk-method-dialog/index.tsx
+++ b/web/src/components/chunk-method-dialog/index.tsx
@ -20,6 +20,7 @@ import { IParserConfig } from '@/interfaces/database/document';
 import { IChangeParserConfigRequestBody } from '@/interfaces/request/document';
 import {
  ChunkMethodItem,
+  EnableTocToggle,
  ParseTypeItem,
 } from '@/pages/dataset/dataset-setting/configuration/common-item';
 import { zodResolver } from '@hookform/resolvers/zod';
@ -113,6 +114,7 @@ export function ChunkMethodDialog({
        auto_keywords: z.coerce.number().optional(),
        auto_questions: z.coerce.number().optional(),
        html4excel: z.boolean().optional(),
+        toc_extraction: z.boolean().optional(),
        // raptor: z
        //   .object({
        //     use_raptor: z.boolean().optional(),
@ -247,7 +249,7 @@ export function ChunkMethodDialog({
  }, [parseType, form]);
  return (
    <Dialog open onOpenChange={hideModal}>
-      <DialogContent className="max-w-[50vw]">
+      <DialogContent className="max-w-[50vw] text-text-primary">
        <DialogHeader>
          <DialogTitle>{t('knowledgeDetails.chunkMethod')}</DialogTitle>
        </DialogHeader>
@ -338,6 +340,7 @@ export function ChunkMethodDialog({
                  show={showAutoKeywords(selectedTag) || showExcelToHtml}
                  className="space-y-3"
                >
+                  <EnableTocToggle />
                  {showAutoKeywords(selectedTag) && (
                    <>
                      <AutoKeywordsFormField></AutoKeywordsFormField>
--- a/web/src/components/chunk-method-dialog/use-default-parser-values.ts
+++ b/web/src/components/chunk-method-dialog/use-default-parser-values.ts
@ -15,6 +15,7 @@ export function useDefaultParserValues() {
      auto_keywords: 0,
      auto_questions: 0,
      html4excel: false,
+      toc_extraction: false,
      // raptor: {
      //   use_raptor: false,
      //   prompt: t('knowledgeConfiguration.promptText'),
--- a/web/src/components/data-pipeline-select/index.tsx
+++ b/web/src/components/data-pipeline-select/index.tsx
@ -1,5 +1,7 @@
 import { AgentCategory } from '@/constants/agent';
+import { FormLayout } from '@/constants/form';
 import { useTranslate } from '@/hooks/common-hooks';
+import { useNavigatePage } from '@/hooks/logic-hooks/navigate-hooks';
 import { useFetchAgentList } from '@/hooks/use-agent-request';
 import { buildSelectOptions } from '@/utils/component-util';
 import { ArrowUpRight } from 'lucide-react';
@ -21,18 +23,27 @@ export interface IDataPipelineSelectNode {
 }

 interface IProps {
-  toDataPipeline?: () => void;
+  showToDataPipeline?: boolean;
  formFieldName: string;
  isMult?: boolean;
  setDataList?: (data: IDataPipelineSelectNode[]) => void;
+  layout?: FormLayout;
 }

 export function DataFlowSelect(props: IProps) {
-  const { toDataPipeline, formFieldName, isMult = false, setDataList } = props;
+  const {
+    showToDataPipeline,
+    formFieldName,
+    isMult = false,
+    setDataList,
+    layout = FormLayout.Vertical,
+  } = props;
+
  const { t } = useTranslate('knowledgeConfiguration');
  const form = useFormContext();
+  const { navigateToAgents } = useNavigatePage();
  const toDataPipLine = () => {
-    toDataPipeline?.();
+    navigateToAgents();
  };
  const { data: dataPipelineOptions } = useFetchAgentList({
    canvas_category: AgentCategory.DataflowCanvas,
@ -69,15 +80,16 @@ export function DataFlowSelect(props: IProps) {
      name={formFieldName}
      render={({ field }) => (
        <FormItem className=" items-center space-y-0 ">
+          {layout === FormLayout.Vertical && (
            <div className="flex flex-col gap-1">
              <div className="flex gap-2 justify-between ">
                <FormLabel
-                tooltip={t('dataFlowTip')}
+                  // tooltip={t('dataFlowTip')}
                  className="text-sm text-text-primary whitespace-wrap "
                >
-                {t('dataPipeline')}
+                  {t('manualSetup')}
                </FormLabel>
-              {toDataPipeline && (
+                {showToDataPipeline && (
                  <div
                    className="text-sm flex text-text-primary cursor-pointer"
                    onClick={toDataPipLine}
@ -110,6 +122,50 @@ export function DataFlowSelect(props: IProps) {
                </FormControl>
              </div>
            </div>
+          )}
+          {layout === FormLayout.Horizontal && (
+            <div className="flex gap-1 items-center">
+              <div className="flex gap-2 justify-between w-1/4">
+                <FormLabel
+                  // tooltip={t('dataFlowTip')}
+                  className="text-sm text-text-secondary whitespace-wrap "
+                >
+                  {t('manualSetup')}
+                </FormLabel>
+              </div>
+
+              <div className="text-muted-foreground w-3/4 flex flex-col items-end">
+                {showToDataPipeline && (
+                  <div
+                    className="text-sm flex text-text-primary cursor-pointer"
+                    onClick={toDataPipLine}
+                  >
+                    {t('buildItFromScratch')}
+                    <ArrowUpRight size={14} />
+                  </div>
+                )}
+                <FormControl>
+                  <>
+                    {!isMult && (
+                      <SelectWithSearch
+                        {...field}
+                        placeholder={t('dataFlowPlaceholder')}
+                        options={options}
+                      />
+                    )}
+                    {isMult && (
+                      <MultiSelect
+                        {...field}
+                        onValueChange={field.onChange}
+                        placeholder={t('dataFlowPlaceholder')}
+                        options={options}
+                      />
+                    )}
+                  </>
+                </FormControl>
+              </div>
+            </div>
+          )}
          <div className="flex pt-1">
            <FormMessage />
          </div>
--- a/web/src/components/delimiter-form-field.tsx
+++ b/web/src/components/delimiter-form-field.tsx
@ -61,7 +61,7 @@ export function DelimiterFormField() {
              <FormLabel
                required
                tooltip={t('knowledgeDetails.delimiterTip')}
-                className="text-sm text-muted-foreground whitespace-break-spaces w-1/4"
+                className="text-sm text-text-secondary whitespace-break-spaces w-1/4"
              >
                {t('knowledgeDetails.delimiter')}
              </FormLabel>
--- a/web/src/components/excel-to-html-form-field.tsx
+++ b/web/src/components/excel-to-html-form-field.tsx
@ -28,7 +28,7 @@ export function ExcelToHtmlFormField() {
            <div className="flex items-center gap-1">
              <FormLabel
                tooltip={t('html4excelTip')}
-                className="text-sm text-muted-foreground whitespace-break-spaces w-1/4"
+                className="text-sm text-text-secondary whitespace-break-spaces w-1/4"
              >
                {t('html4excel')}
              </FormLabel>
--- a/web/src/components/layout-recognize-form-field.tsx
+++ b/web/src/components/layout-recognize-form-field.tsx
@ -79,7 +79,7 @@ export function LayoutRecognizeFormField({
            >
              <FormLabel
                tooltip={t('layoutRecognizeTip')}
-                className={cn('text-sm text-muted-foreground whitespace-wrap', {
+                className={cn('text-sm text-text-secondary whitespace-wrap', {
                  ['w-1/4']: horizontal,
                })}
              >
--- a/web/src/components/max-token-number-from-field.tsx
+++ b/web/src/components/max-token-number-from-field.tsx
@ -17,7 +17,7 @@ export function MaxTokenNumberFormField({ max = 2048, initialValue }: IProps) {
      tooltip={t('chunkTokenNumberTip')}
      max={max}
      defaultValue={initialValue ?? 0}
-      layout={FormLayout.Vertical}
+      layout={FormLayout.Horizontal}
    ></SliderInputFormField>
  );
 }
--- a/web/src/components/slider-input-form-field.tsx
+++ b/web/src/components/slider-input-form-field.tsx
@ -36,7 +36,7 @@ export function SliderInputFormField({
  tooltip,
  defaultValue,
  className,
-  layout = FormLayout.Vertical,
+  layout = FormLayout.Horizontal,
 }: SliderInputFormFieldProps) {
  const form = useFormContext();

--- a/web/src/components/ui/radio.tsx
+++ b/web/src/components/ui/radio.tsx
@ -1,5 +1,4 @@
 import { cn } from '@/lib/utils';
-import { Radio as LucideRadio } from 'lucide-react';
 import React, { useContext, useState } from 'react';

 const RadioGroupContext = React.createContext<{
@ -57,7 +56,7 @@ function Radio({ value, checked, disabled, onChange, children }: RadioProps) {
        onClick={handleClick}
      >
        {isChecked && (
-          <LucideRadio className="h-3 w-3 fill-primary text-primary" />
+          <div className="h-3 w-3 fill-primary text-primary bg-text-primary rounded-full" />
        )}
      </span>
      {children && <span className="text-foreground">{children}</span>}
--- a/web/src/hooks/logic-hooks/navigate-hooks.ts
+++ b/web/src/hooks/logic-hooks/navigate-hooks.ts
@ -19,7 +19,8 @@ export const useNavigatePage = () => {

  const navigateToDataset = useCallback(
    (id: string) => () => {
-      navigate(`${Routes.DatasetBase}${Routes.DataSetOverview}/${id}`);
+      // navigate(`${Routes.DatasetBase}${Routes.DataSetOverview}/${id}`);
+      navigate(`${Routes.Dataset}/${id}`);
    },
    [navigate],
  );
--- a/web/src/locales/en.ts
+++ b/web/src/locales/en.ts
@ -126,8 +126,8 @@ export default {
      startDate: 'Start Date',
      source: 'Source',
      fileName: 'File Name',
-      datasetLogs: 'Dataset Logs',
-      fileLogs: 'File Logs',
+      datasetLogs: 'Dataset',
+      fileLogs: 'File',
      overview: 'Overview',
      success: 'Success',
      failed: 'Failed',
@ -270,6 +270,9 @@ export default {
      reRankModelWaring: 'Re-rank model is very time consuming.',
    },
    knowledgeConfiguration: {
+      tocExtraction: 'toc toggle',
+      tocExtractionTip:
+        " For existing chunks, generate a hierarchical table of contents (one directory per file). During queries, when Directory Enhancement is activated, the system will use a large model to determine which directory items are relevant to the user's question, thereby identifying the relevant chunks.",
      deleteGenerateModalContent: `
        <p>Deleting the generated <strong class='text-text-primary'>{{type}}</strong>  results 
        will remove all derived entities and relationships from this dataset. 
--- a/web/src/locales/zh.ts
+++ b/web/src/locales/zh.ts
@ -114,8 +114,8 @@ export default {
      startDate: '开始时间',
      source: '来源',
      fileName: '文件名',
-      datasetLogs: '数据集日志',
-      fileLogs: '文件日志',
+      datasetLogs: '数据集',
+      fileLogs: '文件',
      overview: '概览',
      success: '成功',
      failed: '失败',
@ -255,6 +255,9 @@ export default {
      theDocumentBeingParsedCannotBeDeleted: '正在解析的文档不能被删除',
    },
    knowledgeConfiguration: {
+      tocExtraction: '目录提取',
+      tocExtractionTip:
+        '对于已有的chunk生成层级结构的目录信息（每个文件一个目录）。在查询时，激活`目录增强`后，系统会用大模型去判断用户问题和哪些目录项相关，从而找到相关的chunk。',
      deleteGenerateModalContent: `
        <p>删除生成的 <strong class='text-text-primary'>{{type}}</strong> 结果
          将从此数据集中移除所有派生实体和关系。
--- a/web/src/pages/dataset/dataset-setting/chunk-method-form.tsx
+++ b/web/src/pages/dataset/dataset-setting/chunk-method-form.tsx
@ -0,0 +1,64 @@
+import { useFormContext, useWatch } from 'react-hook-form';
+
+import { DocumentParserType } from '@/constants/knowledge';
+import { useMemo } from 'react';
+import { AudioConfiguration } from './configuration/audio';
+import { BookConfiguration } from './configuration/book';
+import { EmailConfiguration } from './configuration/email';
+import { KnowledgeGraphConfiguration } from './configuration/knowledge-graph';
+import { LawsConfiguration } from './configuration/laws';
+import { ManualConfiguration } from './configuration/manual';
+import { NaiveConfiguration } from './configuration/naive';
+import { OneConfiguration } from './configuration/one';
+import { PaperConfiguration } from './configuration/paper';
+import { PictureConfiguration } from './configuration/picture';
+import { PresentationConfiguration } from './configuration/presentation';
+import { QAConfiguration } from './configuration/qa';
+import { ResumeConfiguration } from './configuration/resume';
+import { TableConfiguration } from './configuration/table';
+import { TagConfiguration } from './configuration/tag';
+
+const ConfigurationComponentMap = {
+  [DocumentParserType.Naive]: NaiveConfiguration,
+  [DocumentParserType.Qa]: QAConfiguration,
+  [DocumentParserType.Resume]: ResumeConfiguration,
+  [DocumentParserType.Manual]: ManualConfiguration,
+  [DocumentParserType.Table]: TableConfiguration,
+  [DocumentParserType.Paper]: PaperConfiguration,
+  [DocumentParserType.Book]: BookConfiguration,
+  [DocumentParserType.Laws]: LawsConfiguration,
+  [DocumentParserType.Presentation]: PresentationConfiguration,
+  [DocumentParserType.Picture]: PictureConfiguration,
+  [DocumentParserType.One]: OneConfiguration,
+  [DocumentParserType.Audio]: AudioConfiguration,
+  [DocumentParserType.Email]: EmailConfiguration,
+  [DocumentParserType.Tag]: TagConfiguration,
+  [DocumentParserType.KnowledgeGraph]: KnowledgeGraphConfiguration,
+};
+
+function EmptyComponent() {
+  return <div></div>;
+}
+
+export function ChunkMethodForm() {
+  const form = useFormContext();
+
+  const finalParserId: DocumentParserType = useWatch({
+    control: form.control,
+    name: 'parser_id',
+  });
+
+  const ConfigurationComponent = useMemo(() => {
+    return finalParserId
+      ? ConfigurationComponentMap[finalParserId]
+      : EmptyComponent;
+  }, [finalParserId]);
+
+  return (
+    <section className="h-full flex flex-col">
+      <div className="overflow-auto flex-1 min-h-0">
+        <ConfigurationComponent></ConfigurationComponent>
+      </div>
+    </section>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/components/tag-item.tsx
+++ b/web/src/pages/dataset/dataset-setting/components/tag-item.tsx
@ -8,8 +8,9 @@ import {
  FormMessage,
 } from '@/components/ui/form';
 import { MultiSelect } from '@/components/ui/multi-select';
+import { FormLayout } from '@/constants/form';
 import { useFetchKnowledgeList } from '@/hooks/knowledge-hooks';
-import { Flex, Form, InputNumber, Select, Slider, Space } from 'antd';
+import { Form, Select, Space } from 'antd';
 import DOMPurify from 'dompurify';
 import { useFormContext, useWatch } from 'react-hook-form';
 import { useTranslation } from 'react-i18next';
@ -44,7 +45,7 @@ export const TagSetItem = () => {
        <FormItem className=" items-center space-y-0 ">
          <div className="flex items-center">
            <FormLabel
-              className="text-sm text-muted-foreground whitespace-nowrap w-1/4"
+              className="text-sm text-text-secondary whitespace-nowrap w-1/4"
              tooltip={
                <div
                  dangerouslySetInnerHTML={{
@ -116,27 +117,9 @@ export const TopNTagsItem = () => {
      max={10}
      min={1}
      defaultValue={3}
+      layout={FormLayout.Horizontal}
    ></SliderInputFormField>
  );
-
-  return (
-    <Form.Item label={t('knowledgeConfiguration.topnTags')}>
-      <Flex gap={20} align="center">
-        <Flex flex={1}>
-          <Form.Item
-            name={['parser_config', 'topn_tags']}
-            noStyle
-            initialValue={3}
-          >
-            <Slider max={10} min={1} style={{ width: '100%' }} />
-          </Form.Item>
-        </Flex>
-        <Form.Item name={['parser_config', 'topn_tags']} noStyle>
-          <InputNumber max={10} min={1} />
-        </Form.Item>
-      </Flex>
-    </Form.Item>
-  );
 };

 export function TagItems() {
--- a/web/src/pages/dataset/dataset-setting/configuration/audio.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/audio.tsx
@ -0,0 +1,20 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+import { TagItems } from '../components/tag-item';
+
+export function AudioConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </>
+
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/book.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/book.tsx
@ -0,0 +1,28 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+
+export function BookConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </ConfigurationFormContainer>
+      <ConfigurationFormContainer>
+        <TagItems></TagItems>
+      </ConfigurationFormContainer>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/common-item.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/common-item.tsx
@ -42,7 +42,7 @@ export function ChunkMethodItem(props: IProps) {
                'w-1/4 whitespace-pre-wrap': line === 1,
              })}
            >
-              {t('dataPipeline')}
+              {t('builtIn')}
            </FormLabel>
            <div className={line === 1 ? 'w-3/4 ' : 'w-full'}>
              <FormControl>
@ -115,7 +115,7 @@ export function EmbeddingModelItem({ line = 1, isEdit = true }: IProps) {
  );
 }

-export function ParseTypeItem() {
+export function ParseTypeItem({ line = 2 }: { line?: number }) {
  const { t } = useTranslate('knowledgeConfiguration');
  const form = useFormContext();

@ -125,17 +125,26 @@ export function ParseTypeItem() {
      name={'parseType'}
      render={({ field }) => (
        <FormItem className=" items-center space-y-0 ">
-          <div className="">
+          <div
+            className={cn('flex', {
+              ' items-center': line === 1,
+              'flex-col gap-1': line === 2,
+            })}
+          >
            <FormLabel
-              tooltip={t('parseTypeTip')}
-              className="text-sm  whitespace-wrap "
+              // tooltip={t('parseTypeTip')}
+              className={cn('text-sm  whitespace-wrap ', {
+                'w-1/4': line === 1,
+              })}
            >
              {t('parseType')}
            </FormLabel>
-            <div className="text-muted-foreground">
+            <div
+              className={cn('text-muted-foreground', { 'w-3/4': line === 1 })}
+            >
              <FormControl>
                <Radio.Group {...field}>
-                  <div className="w-3/4 flex gap-2 justify-between text-muted-foreground">
+                  <div className="w-1/2 flex gap-2 justify-between text-muted-foreground">
                    <Radio value={1}>{t('builtIn')}</Radio>
                    <Radio value={2}>{t('manualSetup')}</Radio>
                  </div>
@ -144,7 +153,7 @@ export function ParseTypeItem() {
            </div>
          </div>
          <div className="flex pt-1">
-            <div className="w-1/4"></div>
+            <div className={line === 1 ? 'w-1/4' : ''}></div>
            <FormMessage />
          </div>
        </FormItem>
@ -188,3 +197,39 @@ export function EnableAutoGenerateItem() {
    />
  );
 }
+
+export function EnableTocToggle() {
+  const { t } = useTranslate('knowledgeConfiguration');
+  const form = useFormContext();
+
+  return (
+    <FormField
+      control={form.control}
+      name={'parser_config.toc_extraction'}
+      render={({ field }) => (
+        <FormItem className=" items-center space-y-0 ">
+          <div className="flex items-center">
+            <FormLabel
+              tooltip={t('tocExtractionTip')}
+              className="text-sm  whitespace-wrap w-1/4"
+            >
+              {t('tocExtraction')}
+            </FormLabel>
+            <div className="text-muted-foreground w-3/4">
+              <FormControl>
+                <Switch
+                  checked={field.value}
+                  onCheckedChange={field.onChange}
+                />
+              </FormControl>
+            </div>
+          </div>
+          <div className="flex pt-1">
+            <div className="w-1/4"></div>
+            <FormMessage />
+          </div>
+        </FormItem>
+      )}
+    />
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/email.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/email.tsx
@ -0,0 +1,18 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { TagItems } from '../components/tag-item';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function EmailConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </>
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/knowledge-graph.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/knowledge-graph.tsx
@ -0,0 +1,15 @@
+import { DelimiterFormField } from '@/components/delimiter-form-field';
+import { EntityTypesFormField } from '@/components/entity-types-form-field';
+import { MaxTokenNumberFormField } from '@/components/max-token-number-from-field';
+
+export function KnowledgeGraphConfiguration() {
+  return (
+    <>
+      <>
+        <EntityTypesFormField></EntityTypesFormField>
+        <MaxTokenNumberFormField max={8192 * 2}></MaxTokenNumberFormField>
+        <DelimiterFormField></DelimiterFormField>
+      </>
+    </>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/laws.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/laws.tsx
@ -0,0 +1,29 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+
+export function LawsConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <TagItems></TagItems>
+      </ConfigurationFormContainer>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/manual.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/manual.tsx
@ -0,0 +1,27 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+
+export function ManualConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </ConfigurationFormContainer>
+
+      <TagItems></TagItems>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/naive.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/naive.tsx
@ -0,0 +1,33 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { DelimiterFormField } from '@/components/delimiter-form-field';
+import { ExcelToHtmlFormField } from '@/components/excel-to-html-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { MaxTokenNumberFormField } from '@/components/max-token-number-from-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+import { EnableTocToggle } from './common-item';
+
+export function NaiveConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+        <MaxTokenNumberFormField initialValue={512}></MaxTokenNumberFormField>
+        <DelimiterFormField></DelimiterFormField>
+        <EnableTocToggle />
+      </ConfigurationFormContainer>
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+        <ExcelToHtmlFormField></ExcelToHtmlFormField>
+        <TagItems></TagItems>
+      </ConfigurationFormContainer>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/one.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/one.tsx
@ -0,0 +1,21 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function OneConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      <>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </>
+
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/paper.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/paper.tsx
@ -0,0 +1,28 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+
+export function PaperConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </ConfigurationFormContainer>
+      <ConfigurationFormContainer>
+        <TagItems></TagItems>
+      </ConfigurationFormContainer>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/picture.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/picture.tsx
@ -0,0 +1,18 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { TagItems } from '../components/tag-item';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function PictureConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </>
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/presentation.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/presentation.tsx
@ -0,0 +1,29 @@
+import {
+  AutoKeywordsFormField,
+  AutoQuestionsFormField,
+} from '@/components/auto-keywords-form-field';
+import { LayoutRecognizeFormField } from '@/components/layout-recognize-form-field';
+import { TagItems } from '../components/tag-item';
+import {
+  ConfigurationFormContainer,
+  MainContainer,
+} from '../configuration-form-container';
+
+export function PresentationConfiguration() {
+  return (
+    <MainContainer>
+      <ConfigurationFormContainer>
+        <LayoutRecognizeFormField></LayoutRecognizeFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <AutoKeywordsFormField></AutoKeywordsFormField>
+        <AutoQuestionsFormField></AutoQuestionsFormField>
+      </ConfigurationFormContainer>
+
+      <ConfigurationFormContainer>
+        <TagItems></TagItems>
+      </ConfigurationFormContainer>
+    </MainContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/qa.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/qa.tsx
@ -0,0 +1,10 @@
+import { TagItems } from '../components/tag-item';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function QAConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/resume.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/resume.tsx
@ -0,0 +1,10 @@
+import { TagItems } from '../components/tag-item';
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function ResumeConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      <TagItems></TagItems>
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/table.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/table.tsx
@ -0,0 +1,12 @@
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function TableConfiguration() {
+  return (
+    <ConfigurationFormContainer>
+      {/* <ChunkMethodItem></ChunkMethodItem>
+      <EmbeddingModelItem></EmbeddingModelItem>
+
+      <PageRankFormField></PageRankFormField> */}
+    </ConfigurationFormContainer>
+  );
+}
--- a/web/src/pages/dataset/dataset-setting/configuration/tag.tsx
+++ b/web/src/pages/dataset/dataset-setting/configuration/tag.tsx
@ -0,0 +1,5 @@
+import { ConfigurationFormContainer } from '../configuration-form-container';
+
+export function TagConfiguration() {
+  return <ConfigurationFormContainer></ConfigurationFormContainer>;
+}
--- a/web/src/pages/dataset/dataset-setting/form-schema.ts
+++ b/web/src/pages/dataset/dataset-setting/form-schema.ts
@ -1,6 +1,9 @@
+import { t } from 'i18next';
 import { z } from 'zod';

-export const formSchema = z.object({
+export const formSchema = z
+  .object({
+    parseType: z.number(),
    name: z.string().min(1, {
      message: 'Username must be at least 2 characters.',
    }),
@ -25,6 +28,7 @@ export const formSchema = z.object({
        html4excel: z.boolean(),
        tag_kb_ids: z.array(z.string()).nullish(),
        topn_tags: z.number().optional(),
+        toc_extraction: z.boolean().optional(),
        raptor: z
          .object({
            use_raptor: z.boolean().optional(),
@ -73,7 +77,16 @@ export const formSchema = z.object({
      .optional(),
    pagerank: z.number(),
    // icon: z.array(z.instanceof(File)),
-});
+  })
+  .superRefine((data, ctx) => {
+    if (data.parseType === 2 && !data.pipeline_id) {
+      ctx.addIssue({
+        path: ['pipeline_id'],
+        message: t('common.pleaseSelect'),
+        code: 'custom',
+      });
+    }
+  });

 export const pipelineFormSchema = z.object({
  pipeline_id: z.string().optional(),
--- a/web/src/pages/dataset/dataset-setting/index.tsx
+++ b/web/src/pages/dataset/dataset-setting/index.tsx
@ -1,14 +1,18 @@
-import { IDataPipelineSelectNode } from '@/components/data-pipeline-select';
+import {
+  DataFlowSelect,
+  IDataPipelineSelectNode,
+} from '@/components/data-pipeline-select';
 import GraphRagItems from '@/components/parse-configuration/graph-rag-form-fields';
 import RaptorFormFields from '@/components/parse-configuration/raptor-form-fields';
 import { Button } from '@/components/ui/button';
 import Divider from '@/components/ui/divider';
 import { Form } from '@/components/ui/form';
+import { FormLayout } from '@/constants/form';
 import { DocumentParserType } from '@/constants/knowledge';
 import { PermissionRole } from '@/constants/permission';
 import { zodResolver } from '@hookform/resolvers/zod';
 import { useEffect, useState } from 'react';
-import { useForm } from 'react-hook-form';
+import { useForm, useWatch } from 'react-hook-form';
 import { useTranslation } from 'react-i18next';
 import { z } from 'zod';
 import { TopTitle } from '../dataset-title';
@ -16,10 +20,10 @@ import {
  GenerateType,
  IGenerateLogButtonProps,
 } from '../dataset/generate-button/generate';
-import LinkDataPipeline, {
-  IDataPipelineNodeProps,
-} from './components/link-data-pipeline';
+import { ChunkMethodForm } from './chunk-method-form';
+import { IDataPipelineNodeProps } from './components/link-data-pipeline';
 import { MainContainer } from './configuration-form-container';
+import { ChunkMethodItem, ParseTypeItem } from './configuration/common-item';
 import { formSchema } from './form-schema';
 import { GeneralForm } from './general-form';
 import { useFetchKnowledgeConfigurationOnMount } from './hooks';
@ -44,6 +48,7 @@ const enum MethodValue {

 export default function DatasetSettings() {
  const { t } = useTranslation();
+
  const form = useForm<z.infer<typeof formSchema>>({
    resolver: zodResolver(formSchema),
    defaultValues: {
@ -58,6 +63,7 @@ export default function DatasetSettings() {
        auto_questions: 0,
        html4excel: false,
        topn_tags: 3,
+        toc_extraction: false,
        raptor: {
          use_raptor: true,
          max_token: 256,
@ -73,17 +79,17 @@ export default function DatasetSettings() {
        },
      },
      pipeline_id: '',
+      parseType: 1,
      pagerank: 0,
    },
  });
-
  const knowledgeDetails = useFetchKnowledgeConfigurationOnMount(form);
-
  const [pipelineData, setPipelineData] = useState<IDataPipelineNodeProps>();
  const [graphRagGenerateData, setGraphRagGenerateData] =
    useState<IGenerateLogButtonProps>();
  const [raptorGenerateData, setRaptorGenerateData] =
    useState<IGenerateLogButtonProps>();
+
  useEffect(() => {
    console.log('🚀 ~ DatasetSettings ~ knowledgeDetails:', knowledgeDetails);
    if (knowledgeDetails) {
@ -102,8 +108,10 @@ export default function DatasetSettings() {
        finish_at: knowledgeDetails.raptor_task_finish_at,
        task_id: knowledgeDetails.raptor_task_id,
      } as IGenerateLogButtonProps);
+      form.setValue('parseType', knowledgeDetails.pipeline_id ? 2 : 1);
+      form.setValue('pipeline_id', knowledgeDetails.pipeline_id || '');
    }
-  }, [knowledgeDetails]);
+  }, [knowledgeDetails, form]);

  async function onSubmit(data: z.infer<typeof formSchema>) {
    try {
@ -137,6 +145,22 @@ export default function DatasetSettings() {
      } as IGenerateLogButtonProps);
    }
  };
+
+  const parseType = useWatch({
+    control: form.control,
+    name: 'parseType',
+    defaultValue: knowledgeDetails.pipeline_id ? 2 : 1,
+  });
+  const selectedTag = useWatch({
+    name: 'parser_id',
+    control: form.control,
+  });
+  useEffect(() => {
+    if (parseType === 1) {
+      form.setValue('pipeline_id', '');
+    }
+    console.log('parseType', parseType);
+  }, [parseType, form]);
  return (
    <section className="p-5 h-full flex flex-col">
      <TopTitle
@ -167,10 +191,30 @@ export default function DatasetSettings() {
                  onDelete={() => handleDeletePipelineTask(GenerateType.Raptor)}
                ></RaptorFormFields>
                <Divider />
-                <LinkDataPipeline
+                <ParseTypeItem line={1} />
+                {parseType === 1 && (
+                  <ChunkMethodItem line={1}></ChunkMethodItem>
+                )}
+                {parseType === 2 && (
+                  <DataFlowSelect
+                    isMult={false}
+                    showToDataPipeline={true}
+                    formFieldName="pipeline_id"
+                    layout={FormLayout.Horizontal}
+                  />
+                )}
+
+                <Divider />
+                {parseType === 1 && (
+                  <ChunkMethodForm
+                    selectedTag={selectedTag as DocumentParserType}
+                  />
+                )}
+
+                {/* <LinkDataPipeline
                  data={pipelineData}
                  handleLinkOrEditSubmit={handleLinkOrEditSubmit}
-                />
+                /> */}
              </MainContainer>
            </div>
            <div className="text-right items-center flex justify-end gap-3 w-[768px]">
--- a/web/src/pages/dataset/dataset-setting/saving-button.tsx
+++ b/web/src/pages/dataset/dataset-setting/saving-button.tsx
@ -62,6 +62,7 @@ export function SavingButton() {
            if (beValid) {
              form.handleSubmit(async (values) => {
                console.log('saveKnowledgeConfiguration: ', values);
+                delete values['parseType'];
                // delete values['avatar'];
                await saveKnowledgeConfiguration({
                  kb_id,
--- a/web/src/pages/dataset/sidebar/index.tsx
+++ b/web/src/pages/dataset/sidebar/index.tsx
@ -29,11 +29,6 @@ export function SideBar({ refreshCount }: PropType) {

  const items = useMemo(() => {
    const list = [
-      {
-        icon: <DatabaseZap className="size-4" />,
-        label: t(`knowledgeDetails.overview`),
-        key: Routes.DataSetOverview,
-      },
      {
        icon: <FolderOpen className="size-4" />,
        label: t(`knowledgeDetails.subbarFiles`),
@ -44,6 +39,11 @@ export function SideBar({ refreshCount }: PropType) {
        label: t(`knowledgeDetails.testing`),
        key: Routes.DatasetTesting,
      },
+      {
+        icon: <DatabaseZap className="size-4" />,
+        label: t(`knowledgeDetails.overview`),
+        key: Routes.DataSetOverview,
+      },
      {
        icon: <Banknote className="size-4" />,
        label: t(`knowledgeDetails.configuration`),
--- a/web/src/pages/datasets/dataset-creating-dialog.tsx
+++ b/web/src/pages/datasets/dataset-creating-dialog.tsx
@ -16,6 +16,7 @@ import {
  FormMessage,
 } from '@/components/ui/form';
 import { Input } from '@/components/ui/input';
+import { FormLayout } from '@/constants/form';
 import { useNavigatePage } from '@/hooks/logic-hooks/navigate-hooks';
 import { IModalProps } from '@/interfaces/common';
 import { zodResolver } from '@hookform/resolvers/zod';
@ -137,8 +138,9 @@ export function InputForm({ onOk }: IModalProps<any>) {
        {parseType === 2 && (
          <DataFlowSelect
            isMult={false}
-            toDataPipeline={navigateToAgents}
+            showToDataPipeline={true}
            formFieldName="pipeline_id"
+            layout={FormLayout.Vertical}
          />
        )}
      </form>