feat(file): Add file ancestor directory lookup feature by go (#14037)

### What problem does this PR solve? feat(file): Add file ancestor directory lookup feature by go ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-07-01 08:15:44 +08:00 · 2026-04-14 15:22:03 +08:00
parent 6aec8058bb
commit 1031aebc8f
7 changed files with 224 additions and 133 deletions
--- a/internal/handler/file.go
+++ b/internal/handler/file.go
@@ -219,6 +219,41 @@ func (h *FileHandler) GetAllParentFolders(c *gin.Context) {
 	})
 }

+// GetFileAncestors gets all ancestor folders of a file (matches Python /files/<file_id>/ancestors)
+// @Summary Get File Ancestors
+// @Description Get all ancestor folders in path from file to root
+// @Tags file
+// @Accept json
+// @Produce json
+// @Param id path string true "file ID"
+// @Success 200 {object} map[string]interface{}
+// @Router /api/v1/files/{id}/ancestors [get]
+func (h *FileHandler) GetFileAncestors(c *gin.Context) {
+	_, errorCode, errorMessage := GetUser(c)
+	if errorCode != common.CodeSuccess {
+		jsonError(c, errorCode, errorMessage)
+		return
+	}
+
+	fileID := c.Param("id")
+	if fileID == "" {
+		jsonError(c, common.CodeBadRequest, "file id is required")
+		return
+	}
+
+	parentFolders, err := h.fileService.GetAllParentFolders(fileID)
+	if err != nil {
+		jsonError(c, common.CodeServerError, err.Error())
+		return
+	}
+
+	c.JSON(http.StatusOK, gin.H{
+		"code":    common.CodeSuccess,
+		"data":    gin.H{"parent_folders": parentFolders},
+		"message": common.CodeSuccess.Message(),
+	})
+}
+
 type CreateFolderRequest struct {
 	Name     string `json:"name" binding:"required"`
 	ParentID string `json:"parent_id"`
@@ -385,6 +420,7 @@ type MoveFileRequest struct {
 //   - dest_file_id only: move files to a new folder (names unchanged)
 //   - new_name only: rename a single file in place (no storage operation)
 //   - both: move and rename simultaneously
+//
 // @Tags file
 // @Accept json
 // @Produce json
--- a/internal/router/router.go
+++ b/internal/router/router.go
@@ -196,6 +196,7 @@ func (r *Router) Setup(engine *gin.Engine) {
 				file.GET("", r.fileHandler.ListFiles)
 				file.DELETE("", r.fileHandler.DeleteFiles)
 				file.POST("/move", r.fileHandler.MoveFiles)
+				file.GET("/:id/ancestors", r.fileHandler.GetFileAncestors)
 				file.GET("/:id", r.fileHandler.Download)
 			}

--- a/web/src/locales/en.ts
+++ b/web/src/locales/en.ts
@@ -1507,6 +1507,19 @@ Example: Virtual Hosted Style`,
        author: 'Author',
        sectionTitle: 'Section title',
      },
+      includeHeadingContent: 'Include heading content',
+      includeHeadingContentTip:
+        'When enabled, content directly under a heading is kept as its own chunk. Child chunks keep only the heading path.',
+      hierarchyTip: `Build a heading tree and produce self-contained chunks, each carrying its full ancestor heading path (e.g. Part 1 › Chapter 3 › Section 2 + body text).<br>
+Best for: Documents with independent, structurally significant sections — such as legal statutes, regulations, contracts, and technical specifications — where each chunk must be identifiable by its structural position even without surrounding context.`,
+      groupTip: `Split the document flat at a chosen heading level and automatically merge adjacent small sections to preserve content continuity. No parent-heading path is injected.<br>
+Best for: Documents with flowing, contextually connected content — such as books, manuals, reports, and articles — where adjacent paragraphs should stay together to maintain narrative coherence.`,
+      enableMultiColumn: 'Remove original table of contents',
+      enableMultiColumnTip:
+        'Detect and parse multi-column page layouts to preserve the correct reading order. Turn this on for PDFs or documents with two-column or newspaper-style layouts.',
+      removeToc: 'Remove original table of contents',
+      removeTocTip:
+        'Remove the table of contents included in the original PDF, so it is not parsed as regular content or chunked for retrieval.',
      autoPlay: 'Auto play audio',
      downloadFileTypeTip: 'The file type to download',
      downloadFileType: 'Download file type',
@@ -2244,7 +2257,7 @@ This process aggregates variables from multiple branches into a single variable
      tokenChunkerDescription:
        'Split text into chunks by token length with optional delimiters and overlap.',
      titleChunkerDescription:
-        'Split documents into sections by title hierarchy with regex rules for finer control.',
+        'Split documents into sections by title hierarchy. Define heading levels with regex rules, then choose Hierarchy or Group mode to control how chunks are structured.',
      titleChunker: 'Title Chunker',
      extractor: 'Transformer',
      extractorDescription:
@@ -2267,6 +2280,7 @@ This process aggregates variables from multiple branches into a single variable
      },
      fields: 'Field',
      addParser: 'Add Parser',
+      group: 'Group',
      hierarchy: 'Hierarchy',
      regularExpressions: 'Regular Expressions',
      overlappedPercent: 'Overlapped percent (%)',
--- a/web/src/locales/zh.ts
+++ b/web/src/locales/zh.ts
@@ -1253,6 +1253,19 @@ General：实体和关系提取提示来自 GitHub - microsoft/graphrag：基于
        author: '作者',
        sectionTitle: '章节标题',
      },
+      includeHeadingContent: '包含标题内容',
+      includeHeadingContentTip:
+        '启用后，标题下的直接内容将作为一个独立的块保留。子块仅保留标题路径。',
+      hierarchyTip: `构建标题树并生成独立的块，每个块携带其完整的祖先标题路径（例如 第1部分 › 第3章 › 第2节 + 正文）。<br>
+适用场景：具有独立的、结构性重要章节的文档——如法律条款、法规、合同和技术规范——其中每个块即使没有上下文也能通过其结构位置来识别。`,
+      groupTip: `在选定的标题级别将文档扁平分割，并自动合并相邻的小节以保持内容连续性。不注入父标题路径。<br>
+适用场景：具有流动性的、内容相关联的文档——如书籍、手册、报告和文章——其中相邻段落应保持在一起以维持叙述连贯性。`,
+      enableMultiColumn: '启用多栏',
+      enableMultiColumnTip:
+        '检测并解析多栏页面布局以保持正确的阅读顺序。对于具有双栏或报纸式布局的PDF或文档，请开启此功能。',
+      removeToc: '移除原始目录',
+      removeTocTip:
+        '移除原始PDF中包含的目录，这样它就不会被解析为常规内容或作为检索块。',
      autoPlay: '自动播放',
      downloadFileTypeTip: '文件下载的类型',
      downloadFileType: '文件类型',
@@ -1943,7 +1956,7 @@ General：实体和关系提取提示来自 GitHub - microsoft/graphrag：基于
      tokenChunkerDescription:
        '根据分词器长度将文本拆分成块，并带有可选的分隔符和重叠。',
      titleChunkerDescription:
-        '使用正则表达式规则按标题层次结构将文档拆分成多个部分，以实现更精细的控制。',
+        '按标题层级拆分文档。通过正则表达式定义各级标题，再选择层级或分组模式控制切片方式。',
      titleChunker: '按标题分块',
      extractor: '提取器',
      extractorDescription:
@@ -1966,6 +1979,7 @@ General：实体和关系提取提示来自 GitHub - microsoft/graphrag：基于
      },
      fields: '字段',
      addParser: '增加解析器',
+      group: '聚合',
      hierarchy: '层次结构',
      regularExpressions: '正则表达式',
      overlappedPercent: '重叠百分比（%）',
--- a/web/src/pages/agent/form/parser-form/common-form-fields.tsx
+++ b/web/src/pages/agent/form/parser-form/common-form-fields.tsx
@@ -93,9 +93,10 @@ export function TwoColumnCheckFormField({ prefix }: CommonProps) {
  return (
    <RAGFlowFormItem
      name={buildFieldNameWithPrefix(`enable_multi_column`, prefix)}
-      label={t('flow.enableMultiColumn', 'Enable multi column')}
+      label={t('flow.enableMultiColumn')}
      horizontal={true}
-      labelClassName="w-[200px]"
+      labelClassName="w-full"
+      tooltip={t('flow.enableMultiColumnTip')}
    >
      {(field) => (
        <Checkbox
@@ -114,9 +115,10 @@ export function RmdirFormField({ prefix }: CommonProps) {
  return (
    <RAGFlowFormItem
      name={buildFieldNameWithPrefix(`remove_toc`, prefix)}
-      label={t('flow.remove_toc', 'Remove TOC')}
+      label={t('flow.removeToc')}
      horizontal={true}
-      labelClassName="w-[200px]"
+      tooltip={t('flow.removeTocTip')}
+      labelClassName="w-full"
    >
      {(field) => (
        <Checkbox
--- a/web/src/pages/agent/form/parser-form/index.tsx
+++ b/web/src/pages/agent/form/parser-form/index.tsx
@@ -5,14 +5,13 @@ import {
 import { RAGFlowFormItem } from '@/components/ragflow-form';
 import { BlockButton, Button } from '@/components/ui/button';
 import { Form } from '@/components/ui/form';
-import { MultiSelect } from '@/components/ui/multi-select';
 import { Separator } from '@/components/ui/separator';
 import { cn } from '@/lib/utils';
 import { buildOptions } from '@/utils/form';
 import { zodResolver } from '@hookform/resolvers/zod';
 import { useHover } from 'ahooks';
 import { Trash2 } from 'lucide-react';
-import { memo, useCallback, useEffect, useMemo, useRef } from 'react';
+import { memo, useCallback, useMemo, useRef } from 'react';
 import {
  useFieldArray,
  UseFieldArrayRemove,
@@ -25,8 +24,6 @@ import {
  FileType,
  InitialOutputFormatMap,
  initialParserValues,
-  MAIN_CONTENT_PREPROCESS_VALUE,
-  PreprocessValue,
 } from '../../constant/pipeline';
 import { useFormValues } from '../../hooks/use-form-values';
 import { useWatchFormChange } from '../../hooks/use-watch-form-change';
@@ -49,82 +46,82 @@ import { WordFormFields } from './word-form-fields';

 const outputList = buildOutputList(initialParserValues.outputs);

-type PreprocessOptionConfig = {
-  value: PreprocessValue;
-  required?: boolean;
-};
+// type PreprocessOptionConfig = {
+//   value: PreprocessValue;
+//   required?: boolean;
+// };

-const DefaultPreprocessOptionConfigs: PreprocessOptionConfig[] = [
-  { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-];
+// const DefaultPreprocessOptionConfigs: PreprocessOptionConfig[] = [
+//   { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+// ];

-const PreprocessOptionConfigsMap: Partial<
-  Record<FileType, PreprocessOptionConfig[]>
-> = {
-  [FileType.PDF]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-    { value: PreprocessValue.abstract },
-    { value: PreprocessValue.author },
-    { value: PreprocessValue.section_title },
-  ],
-  [FileType.PowerPoint]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-  ],
-  [FileType.Spreadsheet]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-  ],
-  [FileType.TextMarkdown]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-    { value: PreprocessValue.section_title },
-  ],
-  [FileType.Code]: [{ value: MAIN_CONTENT_PREPROCESS_VALUE, required: true }],
-  [FileType.Html]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-    { value: PreprocessValue.section_title },
-  ],
-  [FileType.Doc]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-    { value: PreprocessValue.section_title },
-  ],
-  [FileType.Docx]: [
-    { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
-    { value: PreprocessValue.section_title },
-  ],
-};
+// const PreprocessOptionConfigsMap: Partial<
+//   Record<FileType, PreprocessOptionConfig[]>
+// > = {
+//   [FileType.PDF]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//     { value: PreprocessValue.abstract },
+//     { value: PreprocessValue.author },
+//     { value: PreprocessValue.section_title },
+//   ],
+//   [FileType.PowerPoint]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//   ],
+//   [FileType.Spreadsheet]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//   ],
+//   [FileType.TextMarkdown]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//     { value: PreprocessValue.section_title },
+//   ],
+//   [FileType.Code]: [{ value: MAIN_CONTENT_PREPROCESS_VALUE, required: true }],
+//   [FileType.Html]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//     { value: PreprocessValue.section_title },
+//   ],
+//   [FileType.Doc]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//     { value: PreprocessValue.section_title },
+//   ],
+//   [FileType.Docx]: [
+//     { value: MAIN_CONTENT_PREPROCESS_VALUE, required: true },
+//     { value: PreprocessValue.section_title },
+//   ],
+// };

-function getPreprocessOptionConfigs(fileType?: FileType) {
-  if (!fileType) {
-    return DefaultPreprocessOptionConfigs;
-  }
+// function getPreprocessOptionConfigs(fileType?: FileType) {
+//   if (!fileType) {
+//     return DefaultPreprocessOptionConfigs;
+//   }

-  return PreprocessOptionConfigsMap[fileType] ?? DefaultPreprocessOptionConfigs;
-}
+//   return PreprocessOptionConfigsMap[fileType] ?? DefaultPreprocessOptionConfigs;
+// }

-function normalizePreprocessValuesByFileType(
-  fileType: FileType | undefined,
-  values: string[] | undefined,
-) {
-  const optionConfigs = getPreprocessOptionConfigs(fileType);
-  const allowedValueSet = new Set(optionConfigs.map((x) => x.value));
-  const requiredValues = optionConfigs
-    .filter((x) => x.required)
-    .map((x) => x.value);
-  const normalizedOptionalValues = (Array.isArray(values) ? values : []).filter(
-    (value) => allowedValueSet.has(value as PreprocessValue),
-  ) as PreprocessValue[];
+// function normalizePreprocessValuesByFileType(
+//   fileType: FileType | undefined,
+//   values: string[] | undefined,
+// ) {
+//   const optionConfigs = getPreprocessOptionConfigs(fileType);
+//   const allowedValueSet = new Set(optionConfigs.map((x) => x.value));
+//   const requiredValues = optionConfigs
+//     .filter((x) => x.required)
+//     .map((x) => x.value);
+//   const normalizedOptionalValues = (Array.isArray(values) ? values : []).filter(
+//     (value) => allowedValueSet.has(value as PreprocessValue),
+//   ) as PreprocessValue[];

-  return Array.from(
-    new Set<PreprocessValue>([...requiredValues, ...normalizedOptionalValues]),
-  );
-}
+//   return Array.from(
+//     new Set<PreprocessValue>([...requiredValues, ...normalizedOptionalValues]),
+//   );
+// }

-function isSameStringArray(a: string[] | undefined, b: string[]) {
-  if (!a || a.length !== b.length) {
-    return false;
-  }
+// function isSameStringArray(a: string[] | undefined, b: string[]) {
+//   if (!a || a.length !== b.length) {
+//     return false;
+//   }

-  return a.every((item, idx) => item === b[idx]);
-}
+//   return a.every((item, idx) => item === b[idx]);
+// }

 const FileFormatWidgetMap = {
  [FileType.PDF]: PdfFormFields,
@@ -151,7 +148,7 @@ export const FormSchema = z.object({
  setups: z.array(
    z.object({
      fileFormat: z.string().nullish(),
-      preprocess: z.array(z.string()).optional(),
+      // preprocess: z.array(z.string()).optional(),
      output_format: z.string().optional(),
      parse_method: z.string().optional(),
      lang: z.string().optional(),
@@ -212,56 +209,56 @@ function ParserItem({
    [form, index],
  );

-  const handlePreprocessChange = useCallback(
-    (value: PreprocessValue[]) => {
-      form.setValue(`setups.${index}.preprocess`, value, {
-        shouldDirty: true,
-        shouldValidate: true,
-        shouldTouch: true,
-      });
-    },
-    [form, index],
-  );
+  // const handlePreprocessChange = useCallback(
+  //   (value: PreprocessValue[]) => {
+  //     form.setValue(`setups.${index}.preprocess`, value, {
+  //       shouldDirty: true,
+  //       shouldValidate: true,
+  //       shouldTouch: true,
+  //     });
+  //   },
+  //   [form, index],
+  // );

-  const preprocessOptions = useMemo(() => {
-    const optionConfigs = getPreprocessOptionConfigs(fileFormat as FileType);
+  // const preprocessOptions = useMemo(() => {
+  //   const optionConfigs = getPreprocessOptionConfigs(fileFormat as FileType);

-    return optionConfigs.map((optionConfig) => {
-      const labelMap: Record<string, string> = {
-        [MAIN_CONTENT_PREPROCESS_VALUE]: t('flow.preprocess.mainContent'),
-        [PreprocessValue.section_title]: t('flow.preprocess.sectionTitle'),
-        [PreprocessValue.abstract]: t('flow.preprocess.abstract'),
-        [PreprocessValue.author]: t('flow.preprocess.author'),
-      };
+  //   return optionConfigs.map((optionConfig) => {
+  //     const labelMap: Record<string, string> = {
+  //       [MAIN_CONTENT_PREPROCESS_VALUE]: t('flow.preprocess.mainContent'),
+  //       [PreprocessValue.section_title]: t('flow.preprocess.sectionTitle'),
+  //       [PreprocessValue.abstract]: t('flow.preprocess.abstract'),
+  //       [PreprocessValue.author]: t('flow.preprocess.author'),
+  //     };

-      const label = labelMap[optionConfig.value] || optionConfig.value;
+  //     const label = labelMap[optionConfig.value] || optionConfig.value;

-      return {
-        value: optionConfig.value,
-        disabled: optionConfig.required,
-        label: label,
-      };
-    });
-  }, [fileFormat, t]);
+  //     return {
+  //       value: optionConfig.value,
+  //       disabled: optionConfig.required,
+  //       label: label,
+  //     };
+  //   });
+  // }, [fileFormat, t]);

-  useEffect(() => {
-    const currentPreprocessValues = form.getValues(
-      `setups.${index}.preprocess`,
-    ) as string[] | undefined;
-    const normalizedPreprocessValues = normalizePreprocessValuesByFileType(
-      fileFormat as FileType,
-      currentPreprocessValues,
-    );
+  // useEffect(() => {
+  //   const currentPreprocessValues = form.getValues(
+  //     `setups.${index}.preprocess`,
+  //   ) as string[] | undefined;
+  //   const normalizedPreprocessValues = normalizePreprocessValuesByFileType(
+  //     fileFormat as FileType,
+  //     currentPreprocessValues,
+  //   );

-    if (
-      !isSameStringArray(currentPreprocessValues, normalizedPreprocessValues)
-    ) {
-      form.setValue(`setups.${index}.preprocess`, normalizedPreprocessValues, {
-        shouldDirty: false,
-        shouldValidate: true,
-      });
-    }
-  }, [fileFormat, form, index]);
+  //   if (
+  //     !isSameStringArray(currentPreprocessValues, normalizedPreprocessValues)
+  //   ) {
+  //     form.setValue(`setups.${index}.preprocess`, normalizedPreprocessValues, {
+  //       shouldDirty: false,
+  //       shouldValidate: true,
+  //     });
+  //   }
+  // }, [fileFormat, form, index]);

  return (
    <section
@@ -301,7 +298,7 @@ function ParserItem({
          fileType={fileFormat as FileType}
        />
      </div>
-      <RAGFlowFormItem
+      {/* <RAGFlowFormItem
        name={buildFieldNameWithPrefix(`preprocess`, prefix)}
        label={t('flow.preprocess.preprocess')}
      >
@@ -320,7 +317,7 @@ function ParserItem({
            options={preprocessOptions}
          ></MultiSelect>
        )}
-      </RAGFlowFormItem>
+      </RAGFlowFormItem> */}
      {index < fieldLength - 1 && <Separator />}
    </section>
  );
@@ -351,10 +348,10 @@ const ParserForm = ({ node }: INextOperatorForm) => {
      parse_method: '',
      lang: '',
      fields: [],
-      llm_id: '',
+      vlm: { llm_id: '' },
      table_result_type: '',
      markdown_image_response_type: '',
-      preprocess: [],
+      // preprocess: [],
    });
  }, [append]);

--- a/web/src/pages/agent/form/title-chunker-form/index.tsx
+++ b/web/src/pages/agent/form/title-chunker-form/index.tsx
@@ -8,7 +8,7 @@ import { Form } from '@/components/ui/form';
 import { Input } from '@/components/ui/input';
 import { zodResolver } from '@hookform/resolvers/zod';
 import { Trash2 } from 'lucide-react';
-import { memo, useEffect, useRef } from 'react';
+import { memo, useEffect, useRef, useState } from 'react';
 import { useFieldArray, useForm, useFormContext } from 'react-hook-form';
 import { useTranslation } from 'react-i18next';
 import { z } from 'zod';
@@ -197,6 +197,7 @@ const TitleChunkerForm = ({ node }: INextOperatorForm) => {
  });
  const isInitialized = useRef(false);
  const initialMode = useRef<string | undefined>(undefined);
+  const [showAllTip, setShowAllTip] = useState(true);

  const method = form.watch('method');
  const name = 'rules';
@@ -210,6 +211,7 @@ const TitleChunkerForm = ({ node }: INextOperatorForm) => {
    }

    if (method !== initialMode.current) {
+      setShowAllTip(true);
      const currentMode = initialMode.current;
      const hierarchyValue = form.getValues('hierarchy');
      const rulesValue = form.getValues('rules');
@@ -290,6 +292,34 @@ const TitleChunkerForm = ({ node }: INextOperatorForm) => {
            ],
          }}
        />
+        {/* <div className={cn("text-xs text-text-secondary w-full border p-1", showAllTip ? "block" : "")}>
+          {method === 'hierarchy' && t('flow.hierarchyTip')}
+          {method === 'group' && t('flow.groupTip')}
+        </div> */}
+        <div
+          className={`text-xs text-text-secondary w-full border rounded-sm p-2 cursor-pointer ${showAllTip ? 'block' : 'truncate'}`}
+          onClick={() => setShowAllTip(!showAllTip)}
+        >
+          <div className="flex flex-col justify-start items-center">
+            <span
+              className="flex self-start"
+              dangerouslySetInnerHTML={{
+                __html:
+                  method === 'hierarchy'
+                    ? t('flow.hierarchyTip')
+                    : method === 'group'
+                      ? t('flow.groupTip')
+                      : '',
+              }}
+            >
+              {/* {method === 'hierarchy' && t('flow.hierarchyTip')}
+              {method === 'group' && t('flow.groupTip')} */}
+            </span>
+            {/* <span className="flex ml-2 text-xs self-center">
+              {showAllTip ? '▲' : ''}
+            </span> */}
+          </div>
+        </div>
        <RAGFlowFormItem name={'hierarchy'} label={''}>
          <SelectWithSearch options={hierarchyOptions}></SelectWithSearch>
        </RAGFlowFormItem>
@@ -297,12 +327,9 @@ const TitleChunkerForm = ({ node }: INextOperatorForm) => {
          <RAGFlowFormItem
            name="include_heading_content"
            label={t('flow.includeHeadingContent', 'Include heading content')}
-            tooltip={t(
-              'flow.includeHeadingContentTip',
-              'When enabled, content directly under a heading is kept as its own chunk. Child chunks keep only the heading path.',
-            )}
+            tooltip={t('flow.includeHeadingContentTip')}
            horizontal={true}
-            labelClassName="w-[200px]"
+            labelClassName="w-full"
          >
            {(field) => (
              <Checkbox