Remove unused code and fix performance issue (#11284)

### What problem does this PR solve?

1. remove redundant code
2. fix miner performance issue

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring

Signed-off-by: Jin Hai <haijin.chn@gmail.com>

Dockerfile

@@ -51,7 +51,9 @@ RUN --mount=type=cache,id=ragflow_apt,target=/var/cache/apt,sharing=locked \
     apt install -y libpython3-dev libgtk-4-1 libnss3 xdg-utils libgbm-dev && \
     apt install -y libjemalloc-dev && \
     apt install -y python3-pip pipx nginx unzip curl wget git vim less && \
-    apt install -y ghostscript
+    apt install -y ghostscript && \
+    apt install -y pandoc && \
+    apt install -y texlive
 
 RUN if [ "$NEED_MIRROR" == "1" ]; then \
         pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple && \

agent/canvas.py

@@ -298,8 +298,6 @@ class Canvas(Graph):
                     for kk, vv in kwargs["webhook_payload"].items():
                         self.components[k]["obj"].set_output(kk, vv)
 
-            self.components[k]["obj"].reset(True)
-
         for k in kwargs.keys():
             if k in ["query", "user_id", "files"] and kwargs[k]:
                 if k == "files":
@@ -408,6 +406,10 @@ class Canvas(Graph):
                     else:
                         yield decorate("message", {"content": cpn_obj.output("content")})
                         cite = re.search(r"\[ID:[ 0-9]+\]",  cpn_obj.output("content"))
+
+                    if isinstance(cpn_obj.output("attachment"), tuple):
+                        yield decorate("message", {"attachment": cpn_obj.output("attachment")})
+                        
                     yield decorate("message_end", {"reference": self.get_reference() if cite else None})
 
                     while partials:

agent/component/base.py

@@ -463,12 +463,15 @@ class ComponentBase(ABC):
         return self._param.outputs.get("_ERROR", {}).get("value")
 
     def reset(self, only_output=False):
-        for k in self._param.outputs.keys():
-            self._param.outputs[k]["value"] = None
+        outputs: dict = self._param.outputs # for better performance
+        for k in outputs.keys():
+            outputs[k]["value"] = None
         if only_output:
             return
-        for k in self._param.inputs.keys():
-            self._param.inputs[k]["value"] = None
+
+        inputs: dict = self._param.inputs # for better performance
+        for k in inputs.keys():
+            inputs[k]["value"] = None
         self._param.debug_inputs = {}
 
     def get_input(self, key: str=None) -> Union[Any, dict[str, Any]]:

agent/component/list_operations.py

@@ -121,10 +121,26 @@ class ListOperations(ComponentBase,ABC):
             return False
 
     def _sort(self):
-        if self._param.sort_method == "asc":
-            self._set_outputs(sorted(self.inputs))
-        elif self._param.sort_method == "desc":
-            self._set_outputs(sorted(self.inputs, reverse=True))
+        items = self.inputs or []
+        method = getattr(self._param, "sort_method", "asc") or "asc"
+        reverse = method == "desc"
+
+        if not items:
+            self._set_outputs([])
+            return
+
+        first = items[0]
+
+        if isinstance(first, dict):
+            outputs = sorted(
+                items,
+                key=lambda x: self._hashable(x),
+                reverse=reverse,
+            )
+        else:
+            outputs = sorted(items, reverse=reverse)
+
+        self._set_outputs(outputs)
 
     def _drop_duplicates(self):
         seen = set()
@@ -145,5 +161,6 @@ class ListOperations(ComponentBase,ABC):
         if isinstance(x, set):
             return tuple(sorted(self._hashable(v) for v in x))
         return x
+    
     def thoughts(self) -> str:
         return "ListOperation in progress"

agent/component/message.py

@@ -17,6 +17,9 @@ import json
 import os
 import random
 import re
+import pypandoc
+import logging
+import tempfile
 from functools import partial
 from typing import Any
 
@@ -24,7 +27,8 @@ from agent.component.base import ComponentBase, ComponentParamBase
 from jinja2 import Template as Jinja2Template
 
 from common.connection_utils import timeout
-
+from common.misc_utils import get_uuid
+from common import settings
 
 class MessageParam(ComponentParamBase):
     """
@@ -34,6 +38,7 @@ class MessageParam(ComponentParamBase):
         super().__init__()
         self.content = []
         self.stream = True
+        self.output_format = None  # default output format
         self.outputs = {
             "content": {
                 "type": "str"
@@ -133,6 +138,7 @@ class Message(ComponentBase):
             yield rand_cnt[s: ]
 
         self.set_output("content", all_content)
+        self._convert_content(all_content)
 
     def _is_jinjia2(self, content:str) -> bool:
         patt = [
@@ -164,6 +170,68 @@ class Message(ComponentBase):
             content = re.sub(n, v, content)
 
         self.set_output("content", content)
+        self._convert_content(content)
 
     def thoughts(self) -> str:
         return ""
+
+    def _convert_content(self, content):
+        doc_id = get_uuid()
+        
+        if self._param.output_format.lower() not in {"markdown", "html", "pdf", "docx"}:
+            self._param.output_format = "markdown"
+
+        try:
+            if self._param.output_format in {"markdown", "html"}:
+                if isinstance(content, str):
+                    converted = pypandoc.convert_text(
+                        content,
+                        to=self._param.output_format,
+                        format="markdown",
+                    )
+                else:
+                    converted = pypandoc.convert_file(
+                        content,
+                        to=self._param.output_format,
+                        format="markdown",
+                    )
+
+                binary_content = converted.encode("utf-8")
+
+            else:  # pdf, docx
+                with tempfile.NamedTemporaryFile(suffix=f".{self._param.output_format}", delete=False) as tmp:
+                    tmp_name = tmp.name
+
+                try:
+                    if isinstance(content, str):
+                        pypandoc.convert_text(
+                            content,
+                            to=self._param.output_format,
+                            format="markdown",
+                            outputfile=tmp_name,
+                        )
+                    else:
+                        pypandoc.convert_file(
+                            content,
+                            to=self._param.output_format,
+                            format="markdown",
+                            outputfile=tmp_name,
+                        )
+
+                    with open(tmp_name, "rb") as f:
+                        binary_content = f.read()
+
+                finally:
+                    if os.path.exists(tmp_name):
+                        os.remove(tmp_name)
+
+            settings.STORAGE_IMPL.put(self._canvas._tenant_id, doc_id, binary_content)
+            self.set_output("attachment", {
+                "doc_id":doc_id, 
+                "format":self._param.output_format, 
+                "file_name":f"{doc_id[:8]}.{self._param.output_format}"})
+
+            logging.info(f"Converted content uploaded as {doc_id} (format={self._param.output_format})")
+
+        except Exception as e:
+            logging.error(f"Error converting content to {self._param.output_format}: {e}")

api/apps/document_app.py

@@ -508,6 +508,7 @@ def get(doc_id):
         ext = ext.group(1) if ext else None
         if ext:
             if doc.type == FileType.VISUAL.value:
+
                 content_type = CONTENT_TYPE_MAP.get(ext, f"image/{ext}")
             else:
                 content_type = CONTENT_TYPE_MAP.get(ext, f"application/{ext}")
@@ -517,6 +518,22 @@ def get(doc_id):
         return server_error_response(e)
 
 
+@manager.route("/download/<attachment_id>", methods=["GET"])  # noqa: F821
+@login_required
+def download_attachment(attachment_id):
+    try:
+        ext = request.args.get("ext", "markdown")
+        data = settings.STORAGE_IMPL.get(current_user.id, attachment_id)
+        # data = settings.STORAGE_IMPL.get("eb500d50bb0411f0907561d2782adda5", attachment_id)
+        response = flask.make_response(data)
+        response.headers.set("Content-Type", CONTENT_TYPE_MAP.get(ext, f"application/{ext}"))
+
+        return response
+
+    except Exception as e:
+        return server_error_response(e)
+
+
 @manager.route("/change_parser", methods=["POST"])  # noqa: F821
 @login_required
 @validate_request("doc_id")

docs/guides/accessing_admin_ui.md

@@ -12,6 +12,10 @@ The RAGFlow Admin UI is a web-based interface that provides comprehensive system
 
 To access the RAGFlow admin UI, append `/admin` to the web UI's address, e.g. `http://[RAGFLOW_WEB_UI_ADDR]/admin`, replace `[RAGFLOW_WEB_UI_ADDR]` with real RAGFlow web UI address.
 
+### Default Credentials
+| Username | Password |
+|----------|----------|
+| admin@ragflow.io   | admin |
 
 ## Admin UI Overview
 

pyproject.toml

@@ -145,6 +145,7 @@ dependencies = [
     "markdownify>=1.2.0",
     "captcha>=0.7.1",
     "pip>=25.2",
+    "pypandoc>=1.16",
 ]
 
 [dependency-groups]

uv.lock

@@ -4892,6 +4892,14 @@ wheels = [
     { url = "https://pypi.tuna.tsinghua.edu.cn/packages/80/28/2659c02301b9500751f8d42f9a6632e1508aa5120de5e43042b8b30f8d5d/pyopenssl-25.1.0-py3-none-any.whl", hash = "sha256:2b11f239acc47ac2e5aca04fd7fa829800aeee22a2eb30d744572a157bd8a1ab", size = 56771, upload-time = "2025-05-17T16:28:29.197Z" },
 ]
 
+[[package]]
+name = "pypandoc"
+version = "1.16"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/24/77/af1fc54740a0712988f9518e629d38edc7b8ffccd7549203f19c3d8a2db6/pypandoc-1.16-py3-none-any.whl", hash = "sha256:868f390d48388743e7a5885915cbbaa005dea36a825ecdfd571f8c523416c822", size = 19425, upload-time = "2025-11-08T15:44:38.429Z" },
+]
+
 [[package]]
 name = "pyparsing"
 version = "3.2.3"
@@ -5292,6 +5300,7 @@ dependencies = [
     { name = "pyicu" },
     { name = "pymysql" },
     { name = "pyodbc" },
+    { name = "pypandoc" },
     { name = "pypdf" },
     { name = "pypdf2" },
     { name = "python-calamine" },
@@ -5447,6 +5456,7 @@ requires-dist = [
     { name = "pyicu", specifier = ">=2.15.3,<3.0.0" },
     { name = "pymysql", specifier = ">=1.1.1,<2.0.0" },
     { name = "pyodbc", specifier = ">=5.2.0,<6.0.0" },
+    { name = "pypandoc", specifier = ">=1.16" },
     { name = "pypdf", specifier = "==6.0.0" },
     { name = "pypdf2", specifier = ">=3.0.1,<4.0.0" },
     { name = "python-calamine", specifier = ">=0.4.0" },

web/src/components/next-message-item/index.tsx

@@ -18,8 +18,10 @@ import { cn } from '@/lib/utils';
 import { AgentChatContext } from '@/pages/agent/context';
 import { WorkFlowTimeline } from '@/pages/agent/log-sheet/workflow-timeline';
 import { IMessage } from '@/pages/chat/interface';
+import { downloadFile } from '@/services/file-manager-service';
+import { downloadFileFromBlob } from '@/utils/file-util';
 import { isEmpty } from 'lodash';
-import { Atom, ChevronDown, ChevronUp } from 'lucide-react';
+import { Atom, ChevronDown, ChevronUp, Download } from 'lucide-react';
 import MarkdownContent from '../next-markdown-content';
 import { RAGFlowAvatar } from '../ragflow-avatar';
 import { useTheme } from '../theme-provider';
@@ -245,6 +247,32 @@ function MessageItem({
             {isUser && (
               <UploadedMessageFiles files={item.files}></UploadedMessageFiles>
             )}
+            {isAssistant && item.attachment && item.attachment.doc_id && (
+              <div className="w-full flex items-center justify-end">
+                <Button
+                  variant="link"
+                  className="p-1 m-0 h-auto text-text-sub-title-invert"
+                  onClick={async () => {
+                    if (item.attachment?.doc_id) {
+                      try {
+                        const response = await downloadFile({
+                          docId: item.attachment.doc_id,
+                          ext: item.attachment.format,
+                        });
+                        const blob = new Blob([response.data], {
+                          type: response.data.type,
+                        });
+                        downloadFileFromBlob(blob, item.attachment.file_name);
+                      } catch (error) {
+                        console.error('Download failed:', error);
+                      }
+                    }
+                  }}
+                >
+                  <Download size={16} />
+                </Button>
+              </div>
+            )}
           </section>
         </div>
       </section>

web/src/hooks/use-send-message.ts

@@ -44,9 +44,14 @@ export interface IInputData {
   inputs: Record<string, BeginQuery>;
   tips: string;
 }
-
+export interface IAttachment {
+  doc_id: string;
+  format: string;
+  file_name: string;
+}
 export interface IMessageData {
   content: string;
+  outputs: any;
   start_to_think?: boolean;
   end_to_think?: boolean;
 }

web/src/interfaces/database/chat.ts

@@ -1,4 +1,5 @@
 import { MessageType } from '@/constants/chat';
+import { IAttachment } from '@/hooks/use-send-message';
 
 export interface PromptConfig {
   empty_response: string;
@@ -97,6 +98,7 @@ export interface Message {
   data?: any;
   files?: File[];
   chatBoxId?: string;
+  attachment?: IAttachment;
 }
 
 export interface IReferenceChunk {
@@ -126,6 +128,7 @@ export interface IReferenceObject {
 
 export interface IAnswer {
   answer: string;
+  attachment?: IAttachment;
   reference?: IReference;
   conversationId?: string;
   prompt?: string;

web/src/locales/en.ts

@@ -1009,6 +1009,8 @@ Example: general/v2/`,
       pleaseUploadAtLeastOneFile: 'Please upload at least one file',
     },
     flow: {
+      downloadFileTypeTip: 'The file type to download',
+      downloadFileType: 'Download file type',
       formatTypeError: 'Format or type error',
       variableNameMessage:
         'Variable name can only contain letters and underscores',

web/src/locales/zh.ts

@@ -956,6 +956,8 @@ General：实体和关系提取提示来自 GitHub - microsoft/graphrag：基于
       pleaseUploadAtLeastOneFile: '请上传至少一个文件',
     },
     flow: {
+      downloadFileTypeTip: '文件下载的类型',
+      downloadFileType: '文件类型',
       formatTypeError: '格式或类型错误',
       variableNameMessage: '名称只能包含字母和下划线',
       variableDescription: '变量的描述',

web/src/pages/agent/chat/use-send-agent-message.ts

@@ -5,6 +5,7 @@ import {
   useSelectDerivedMessages,
 } from '@/hooks/logic-hooks';
 import {
+  IAttachment,
   IEventList,
   IInputEvent,
   IMessageEndData,
@@ -75,9 +76,13 @@ export function findMessageFromList(eventList: IEventList) {
     nextContent += '</think>';
   }
 
+  const workflowFinished = eventList.find(
+    (x) => x.event === MessageEventType.WorkflowFinished,
+  ) as IMessageEvent;
   return {
     id: eventList[0]?.message_id,
     content: nextContent,
+    attachment: workflowFinished?.data?.outputs?.attachment || {},
   };
 }
 
@@ -388,12 +393,13 @@ export const useSendAgentMessage = ({
   }, [sendMessageInTaskMode]);
 
   useEffect(() => {
-    const { content, id } = findMessageFromList(answerList);
+    const { content, id, attachment } = findMessageFromList(answerList);
     const inputAnswer = findInputFromList(answerList);
     const answer = content || getLatestError(answerList);
     if (answerList.length > 0) {
       addNewestOneAnswer({
         answer: answer ?? '',
+        attachment: attachment as IAttachment,
         id: id,
         ...inputAnswer,
       });

web/src/pages/agent/constant/index.tsx

@@ -417,6 +417,7 @@ export const initialIterationValues = {
   items_ref: '',
   outputs: {},
 };
+
 export const initialIterationStartValues = {
   outputs: {
     item: {
@@ -845,3 +846,10 @@ export enum JsonSchemaDataType {
   Array = 'array',
   Object = 'object',
 }
+
+export enum ExportFileType {
+  PDF = 'pdf',
+  HTML = 'html',
+  Markdown = 'md',
+  DOCX = 'docx',
+}

web/src/pages/agent/form/message-form/index.tsx

@@ -8,12 +8,14 @@ import {
   FormLabel,
   FormMessage,
 } from '@/components/ui/form';
+import { RAGFlowSelect } from '@/components/ui/select';
 import { zodResolver } from '@hookform/resolvers/zod';
 import { X } from 'lucide-react';
 import { memo } from 'react';
 import { useFieldArray, useForm } from 'react-hook-form';
 import { useTranslation } from 'react-i18next';
 import { z } from 'zod';
+import { ExportFileType } from '../../constant';
 import { INextOperatorForm } from '../../interface';
 import { FormWrapper } from '../components/form-wrapper';
 import { PromptEditor } from '../components/prompt-editor';
@@ -33,10 +35,14 @@ function MessageForm({ node }: INextOperatorForm) {
         }),
       )
       .optional(),
+    output_format: z.string().optional(),
   });
 
   const form = useForm({
-    defaultValues: values,
+    defaultValues: {
+      ...values,
+      output_format: values.output_format,
+    },
     resolver: zodResolver(FormSchema),
   });
 
@@ -50,6 +56,39 @@ function MessageForm({ node }: INextOperatorForm) {
   return (
     <Form {...form}>
       <FormWrapper>
+        <FormContainer>
+          <FormItem>
+            <FormLabel tooltip={t('flow.downloadFileTypeTip')}>
+              {t('flow.downloadFileType')}
+            </FormLabel>
+            <FormField
+              control={form.control}
+              name={`output_format`}
+              render={({ field }) => (
+                <FormItem className="flex-1">
+                  <FormControl>
+                    <RAGFlowSelect
+                      options={Object.keys(ExportFileType).map(
+                        (key: string) => {
+                          return {
+                            value:
+                              ExportFileType[
+                                key as keyof typeof ExportFileType
+                              ],
+                            label: key,
+                          };
+                        },
+                      )}
+                      {...field}
+                      onValueChange={field.onChange}
+                      placeholder={t('flow.messagePlaceholder')}
+                    ></RAGFlowSelect>
+                  </FormControl>
+                </FormItem>
+              )}
+            />
+          </FormItem>
+        </FormContainer>
         <FormContainer>
           <FormItem>
             <FormLabel tooltip={t('flow.msgTip')}>{t('flow.msg')}</FormLabel>

web/src/pages/agent/form/message-form/use-values.ts

@@ -1,7 +1,7 @@
 import { RAGFlowNodeType } from '@/interfaces/database/flow';
 import { isEmpty } from 'lodash';
 import { useMemo } from 'react';
-import { initialMessageValues } from '../../constant';
+import { ExportFileType, initialMessageValues } from '../../constant';
 import { convertToObjectArray } from '../../utils';
 
 export function useValues(node?: RAGFlowNodeType) {
@@ -15,6 +15,7 @@ export function useValues(node?: RAGFlowNodeType) {
     return {
       ...formData,
       content: convertToObjectArray(formData.content),
+      output_format: formData.output_format || ExportFileType.PDF,
     };
   }, [node]);
 

web/src/services/file-manager-service.ts

@@ -13,6 +13,7 @@ const {
   get_document_file,
   getFile,
   moveFile,
+  get_document_file_download,
 } = api;
 
 const methods = {
@@ -65,4 +66,10 @@ const fileManagerService = registerServer<keyof typeof methods>(
   request,
 );
 
+export const downloadFile = (data: { docId: string; ext: string }) => {
+  return request.get(get_document_file_download(data.docId), {
+    params: { ext: data.ext },
+    responseType: 'blob',
+  });
+};
 export default fileManagerService;

web/src/utils/api.ts

@@ -100,6 +100,8 @@ export default {
   document_change_parser: `${api_host}/document/change_parser`,
   document_thumbnails: `${api_host}/document/thumbnails`,
   get_document_file: `${api_host}/document/get`,
+  get_document_file_download: (docId: string) =>
+    `${api_host}/document/download/${docId}`,
   document_upload: `${api_host}/document/upload`,
   web_crawl: `${api_host}/document/web_crawl`,
   document_infos: `${api_host}/document/infos`,