Update before release (#854)

### What problem does this PR solve?

Update version information before release 0.6.0.

### Type of change

- [x] Documentation Update

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

.gitignore

@@ -29,3 +29,4 @@ Cargo.lock
 docker/ragflow-logs/
 /flask_session
 /logs
+rag/res/deepdoc

Dockerfile

@@ -1,4 +1,4 @@
-FROM swr.cn-north-4.myhuaweicloud.com/infiniflow/ragflow-base:v1.0
+FROM infiniflow/ragflow-base:v2.0
 USER  root
 
 WORKDIR /ragflow
@@ -15,6 +15,7 @@ ENV PYTHONPATH=/ragflow/
 ENV HF_ENDPOINT=https://hf-mirror.com
 
 ADD docker/entrypoint.sh ./entrypoint.sh
+ADD docker/.env ./
 RUN chmod +x ./entrypoint.sh
 
 ENTRYPOINT ["./entrypoint.sh"]

Dockerfile.cuda

@@ -1,4 +1,4 @@
-FROM swr.cn-north-4.myhuaweicloud.com/infiniflow/ragflow-base:v1.0
+FROM FROM infiniflow/ragflow-base:v2.0
 USER  root
 
 WORKDIR /ragflow

README.md

@@ -17,8 +17,8 @@
     <a href="https://demo.ragflow.io" target="_blank">
         <img alt="Static Badge" src="https://img.shields.io/badge/Online-Demo-4e6b99"></a>
     <a href="https://hub.docker.com/r/infiniflow/ragflow" target="_blank">
-        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.5.0-brightgreen"
-            alt="docker pull infiniflow/ragflow:v0.5.0"></a>
+        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.6.0-brightgreen"
+            alt="docker pull infiniflow/ragflow:v0.6.0"></a>
       <a href="https://github.com/infiniflow/ragflow/blob/main/LICENSE">
     <img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=1570EF" alt="license">
   </a>
@@ -26,7 +26,20 @@
 
 ## 💡 What is RAGFlow?
 
-[RAGFlow](https://demo.ragflow.io) is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. It offers a streamlined RAG workflow for businesses of any scale, combining LLM (Large Language Models) to provide truthful question-answering capabilities, backed by well-founded citations from various complex formatted data.
+[RAGFlow](https://ragflow.io/) is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. It offers a streamlined RAG workflow for businesses of any scale, combining LLM (Large Language Models) to provide truthful question-answering capabilities, backed by well-founded citations from various complex formatted data.
+
+## 📌 Latest Updates
+
+- 2024-05-21 Supports streaming output and text chunk retrieval API. 
+- 2024-05-15 Integrates OpenAI GPT-4o.
+- 2024-05-08 Integrates LLM DeepSeek-V2.
+- 2024-04-26 Adds file management.
+- 2024-04-19 Supports conversation API ([detail](./docs/conversation_api.md)).
+- 2024-04-16 Integrates an embedding model 'bce-embedding-base_v1' from [BCEmbedding](https://github.com/netease-youdao/BCEmbedding), and [FastEmbed](https://github.com/qdrant/fastembed), which is designed specifically for light and speedy embedding.
+- 2024-04-11 Supports [Xinference](./docs/xinference.md) for local LLM deployment.
+- 2024-04-10 Adds a new layout recognition model for analyzing legal documents.
+- 2024-04-08 Supports [Ollama](./docs/ollama.md) for local LLM deployment.
+- 2024-04-07 Supports Chinese UI.
 
 ## 🌟 Key Features
 
@@ -56,17 +69,6 @@
 - Multiple recall paired with fused re-ranking.
 - Intuitive APIs for seamless integration with business.
 
-## 📌 Latest Features
-
-- 2024-05-08 Integrates LLM DeepSeek.
-- 2024-04-26 Adds file management.
-- 2024-04-19 Supports conversation API ([detail](./docs/conversation_api.md)).
-- 2024-04-16 Integrates an embedding model 'bce-embedding-base_v1' from [BCEmbedding](https://github.com/netease-youdao/BCEmbedding), and [FastEmbed](https://github.com/qdrant/fastembed), which is designed specifically for light and speedy embedding.
-- 2024-04-11 Supports [Xinference](./docs/xinference.md) for local LLM deployment.
-- 2024-04-10 Adds a new layout recognition model for analyzing Laws documentation.
-- 2024-04-08 Supports [Ollama](./docs/ollama.md) for local LLM deployment.
-- 2024-04-07 Supports Chinese UI.
-
 ## 🔎 System Architecture
 
 <div align="center" style="margin-top:20px;margin-bottom:20px;">
@@ -114,12 +116,14 @@
 
 3. Build the pre-built Docker images and start up the server:
 
+   > Running the following commands automatically downloads the *dev* version RAGFlow Docker image. To download and run a specified Docker version, update `RAGFLOW_VERSION` in **docker/.env** to the intended version, for example `RAGFLOW_VERSION=v0.6.0`, before running the following commands.
+
    ```bash
    $ cd ragflow/docker
    $ chmod +x ./entrypoint.sh
    $ docker compose up -d
    ```
-   > Please note that running the above commands will automatically download the development version docker image of RAGFlow. If you want to download and run a specific version of docker image, please find the RAGFLOW_VERSION variable in the docker/.env file, change it to the corresponding version, for example, RAGFLOW_VERSION=v0.5.0, and run the above commands.
+   
 
    > The core image is about 9 GB in size and may take a while to load.
 
@@ -247,8 +251,32 @@ $ chmod +x ./entrypoint.sh
 $ bash ./entrypoint.sh
 ```
 
+7. Start the WebUI service
+```bash
+$ cd web
+$ npm install --registry=https://registry.npmmirror.com --force
+$ vim .umirc.ts
+# Modify proxy.target to 127.0.0.1:9380
+$ npm run dev 
+```
+
+8. Deploy the WebUI service
+```bash
+$ cd web
+$ npm install --registry=https://registry.npmmirror.com --force
+$ umi build
+$ mkdir -p /ragflow/web
+$ cp -r dist /ragflow/web
+$ apt install nginx -y
+$ cp ../docker/nginx/proxy.conf /etc/nginx
+$ cp ../docker/nginx/nginx.conf /etc/nginx
+$ cp ../docker/nginx/ragflow.conf /etc/nginx/conf.d
+$ systemctl start nginx
+```
+
 ## 📚 Documentation
 
+- [Quickstart](./docs/quickstart.md)
 - [FAQ](./docs/faq.md)
 
 ## 📜 Roadmap

README_ja.md

@@ -17,8 +17,8 @@
     <a href="https://demo.ragflow.io" target="_blank">
         <img alt="Static Badge" src="https://img.shields.io/badge/Online-Demo-4e6b99"></a>
     <a href="https://hub.docker.com/r/infiniflow/ragflow" target="_blank">
-        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.5.0-brightgreen"
-            alt="docker pull infiniflow/ragflow:v0.5.0"></a>
+        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.6.0-brightgreen"
+            alt="docker pull infiniflow/ragflow:v0.6.0"></a>
       <a href="https://github.com/infiniflow/ragflow/blob/main/LICENSE">
     <img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=1570EF" alt="license">
   </a>
@@ -26,7 +26,22 @@
 
 ## 💡 RAGFlow とは？
 
-[RAGFlow](https://demo.ragflow.io) は、深い文書理解に基づいたオープンソースの RAG (Retrieval-Augmented Generation) エンジンである。LLM（大規模言語モデル）を組み合わせることで、様々な複雑なフォーマットのデータから根拠のある引用に裏打ちされた、信頼できる質問応答機能を実現し、あらゆる規模のビジネスに適した RAG ワークフローを提供します。
+[RAGFlow](https://ragflow.io/) は、深い文書理解に基づいたオープンソースの RAG (Retrieval-Augmented Generation) エンジンである。LLM（大規模言語モデル）を組み合わせることで、様々な複雑なフォーマットのデータから根拠のある引用に裏打ちされた、信頼できる質問応答機能を実現し、あらゆる規模のビジネスに適した RAG ワークフローを提供します。
+
+## 📌 最新情報
+
+- 2024-05-21 ストリーミング出力とテキストチャンク取得APIをサポート。
+- 2024-05-15 OpenAI GPT-4oを統合しました。
+- 2024-05-08 LLM DeepSeek-V2を統合しました。
+- 2024-04-26 「ファイル管理」機能を追加しました。
+- 2024-04-19 会話 API をサポートします ([詳細](./docs/conversation_api.md))。
+- 2024-04-16 [BCEmbedding](https://github.com/netease-youdao/BCEmbedding) から埋め込みモデル「bce-embedding-base_v1」を追加します。
+- 2024-04-16 [FastEmbed](https://github.com/qdrant/fastembed) は、軽量かつ高速な埋め込み用に設計されています。
+- 2024-04-11 ローカル LLM デプロイメント用に [Xinference](./docs/xinference.md) をサポートします。
+- 2024-04-10 メソッド「Laws」に新しいレイアウト認識モデルを追加します。
+- 2024-04-08 [Ollama](./docs/ollama.md) を使用した大規模モデルのローカライズされたデプロイメントをサポートします。
+- 2024-04-07 中国語インターフェースをサポートします。
+
 
 ## 🌟 主な特徴
 
@@ -56,18 +71,6 @@
 - 複数の想起と融合された再ランク付け。
 - 直感的な API によってビジネスとの統合がシームレスに。
 
-## 📌 最新の機能
-
-- 2024-05-08 
-- 2024-04-26 「ファイル管理」機能を追加しました。
-- 2024-04-19 会話 API をサポートします ([詳細](./docs/conversation_api.md))。
-- 2024-04-16 [BCEmbedding](https://github.com/netease-youdao/BCEmbedding) から埋め込みモデル「bce-embedding-base_v1」を追加します。
-- 2024-04-16 [FastEmbed](https://github.com/qdrant/fastembed) は、軽量かつ高速な埋め込み用に設計されています。
-- 2024-04-11 ローカル LLM デプロイメント用に [Xinference](./docs/xinference.md) をサポートします。
-- 2024-04-10 メソッド「Laws」に新しいレイアウト認識モデルを追加します。
-- 2024-04-08 [Ollama](./docs/ollama.md) を使用した大規模モデルのローカライズされたデプロイメントをサポートします。
-- 2024-04-07 中国語インターフェースをサポートします。
-
 ## 🔎 システム構成
 
 <div align="center" style="margin-top:20px;margin-bottom:20px;">
@@ -121,7 +124,7 @@
    $ docker compose up -d
    ```
 
-   > 上記のコマンドを実行すると、RAGFlowの開発版dockerイメージが自動的にダウンロードされます。 特定のバージョンのDockerイメージをダウンロードして実行したい場合は、docker/.envファイルのRAGFLOW_VERSION変数を見つけて、対応するバージョンに変更してください。 例えば、RAGFLOW_VERSION=v0.5.0として、上記のコマンドを実行してください。
+   > 上記のコマンドを実行すると、RAGFlowの開発版dockerイメージが自動的にダウンロードされます。 特定のバージョンのDockerイメージをダウンロードして実行したい場合は、docker/.envファイルのRAGFLOW_VERSION変数を見つけて、対応するバージョンに変更してください。 例えば、RAGFLOW_VERSION=v0.6.0として、上記のコマンドを実行してください。
 
    > コアイメージのサイズは約 9 GB で、ロードに時間がかかる場合があります。
 
@@ -183,7 +186,7 @@
 ```bash
 $ git clone https://github.com/infiniflow/ragflow.git
 $ cd ragflow/
-$ docker build -t infiniflow/ragflow:v0.5.0 .
+$ docker build -t infiniflow/ragflow:v0.6.0 .
 $ cd ragflow/docker
 $ chmod +x ./entrypoint.sh
 $ docker compose up -d
@@ -251,6 +254,7 @@ $ bash ./entrypoint.sh
 
 ## 📚 ドキュメンテーション
 
+- [Quickstart](./docs/quickstart.md)
 - [FAQ](./docs/faq.md)
 
 ## 📜 ロードマップ

README_zh.md

@@ -17,8 +17,8 @@
     <a href="https://demo.ragflow.io" target="_blank">
         <img alt="Static Badge" src="https://img.shields.io/badge/Online-Demo-4e6b99"></a>
     <a href="https://hub.docker.com/r/infiniflow/ragflow" target="_blank">
-        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.5.0-brightgreen"
-            alt="docker pull infiniflow/ragflow:v0.5.0"></a>
+        <img src="https://img.shields.io/badge/docker_pull-ragflow:v0.6.0-brightgreen"
+            alt="docker pull infiniflow/ragflow:v0.6.0"></a>
       <a href="https://github.com/infiniflow/ragflow/blob/main/LICENSE">
     <img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=1570EF" alt="license">
   </a>
@@ -26,7 +26,20 @@
 
 ## 💡 RAGFlow 是什么？
 
-[RAGFlow](https://demo.ragflow.io) 是一款基于深度文档理解构建的开源 RAG（Retrieval-Augmented Generation）引擎。RAGFlow 可以为各种规模的企业及个人提供一套精简的 RAG 工作流程，结合大语言模型（LLM）针对用户各类不同的复杂格式数据提供可靠的问答以及有理有据的引用。
+[RAGFlow](https://ragflow.io/) 是一款基于深度文档理解构建的开源 RAG（Retrieval-Augmented Generation）引擎。RAGFlow 可以为各种规模的企业及个人提供一套精简的 RAG 工作流程，结合大语言模型（LLM）针对用户各类不同的复杂格式数据提供可靠的问答以及有理有据的引用。
+
+## 📌 近期更新
+
+- 2024-05-21 支持流式结果输出和文本块获取API。
+- 2024-05-15 集成大模型 OpenAI GPT-4o。
+- 2024-05-08 集成大模型 DeepSeek。
+- 2024-04-26 增添了'文件管理'功能。
+- 2024-04-19 支持对话 API ([更多](./docs/conversation_api.md))。
+- 2024-04-16 集成嵌入模型 [BCEmbedding](https://github.com/netease-youdao/BCEmbedding) 和 专为轻型和高速嵌入而设计的 [FastEmbed](https://github.com/qdrant/fastembed)。
+- 2024-04-11 支持用 [Xinference](./docs/xinference.md) 本地化部署大模型。
+- 2024-04-10 为‘Laws’版面分析增加了底层模型。
+- 2024-04-08 支持用 [Ollama](./docs/ollama.md) 本地化部署大模型。
+- 2024-04-07 支持中文界面。
 
 ## 🌟 主要功能
 
@@ -56,17 +69,6 @@
 - 基于多路召回、融合重排序。
 - 提供易用的 API，可以轻松集成到各类企业系统。
 
-## 📌 新增功能
-
-- 2024-05-08 集成大模型 DeepSeek
-- 2024-04-26 增添了'文件管理'功能.
-- 2024-04-19 支持对话 API ([更多](./docs/conversation_api.md)).
-- 2024-04-16 集成嵌入模型 [BCEmbedding](https://github.com/netease-youdao/BCEmbedding) 和 专为轻型和高速嵌入而设计的 [FastEmbed](https://github.com/qdrant/fastembed) 。
-- 2024-04-11 支持用 [Xinference](./docs/xinference.md) 本地化部署大模型。
-- 2024-04-10 为‘Laws’版面分析增加了底层模型。
-- 2024-04-08 支持用 [Ollama](./docs/ollama.md) 本地化部署大模型。
-- 2024-04-07 支持中文界面。
-
 ## 🔎 系统架构
 
 <div align="center" style="margin-top:20px;margin-bottom:20px;">
@@ -120,7 +122,7 @@
    $ docker compose -f docker-compose-CN.yml up -d
    ```
 
-   > 请注意，运行上述命令会自动下载 RAGFlow 的开发版本 docker 镜像。如果你想下载并运行特定版本的 docker 镜像，请在 docker/.env 文件中找到 RAGFLOW_VERSION 变量，将其改为对应版本。例如 RAGFLOW_VERSION=v0.5.0，然后运行上述命令。
+   > 请注意，运行上述命令会自动下载 RAGFlow 的开发版本 docker 镜像。如果你想下载并运行特定版本的 docker 镜像，请在 docker/.env 文件中找到 RAGFLOW_VERSION 变量，将其改为对应版本。例如 RAGFLOW_VERSION=v0.6.0，然后运行上述命令。
 
    > 核心镜像文件大约 9 GB，可能需要一定时间拉取。请耐心等待。
 
@@ -182,7 +184,7 @@
 ```bash
 $ git clone https://github.com/infiniflow/ragflow.git
 $ cd ragflow/
-$ docker build -t infiniflow/ragflow:v0.5.0 .
+$ docker build -t infiniflow/ragflow:v0.6.0 .
 $ cd ragflow/docker
 $ chmod +x ./entrypoint.sh
 $ docker compose up -d
@@ -247,9 +249,31 @@ $ docker compose -f docker-compose-base.yml up -d
 $ chmod +x ./entrypoint.sh
 $ bash ./entrypoint.sh
 ```
+7. 启动WebUI服务
+```bash
+$ cd web
+$ npm install --registry=https://registry.npmmirror.com --force
+$ vim .umirc.ts
+# 修改proxy.target为127.0.0.1:9380
+$ npm run dev 
+```
 
+8. 部署WebUI服务
+```bash
+$ cd web
+$ npm install --registry=https://registry.npmmirror.com --force
+$ umi build
+$ mkdir -p /ragflow/web
+$ cp -r dist /ragflow/web
+$ apt install nginx -y
+$ cp ../docker/nginx/proxy.conf /etc/nginx
+$ cp ../docker/nginx/nginx.conf /etc/nginx
+$ cp ../docker/nginx/ragflow.conf /etc/nginx/conf.d
+$ systemctl start nginx
+```
 ## 📚 技术文档
 
+- [Quickstart](./docs/quickstart.md)
 - [FAQ](./docs/faq.md)
 
 ## 📜 路线图

api/apps/api_app.py

@@ -13,19 +13,23 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 #
+import json
 import os
 import re
 from datetime import datetime, timedelta
-from flask import request
+from flask import request, Response
 from flask_login import login_required, current_user
 
 from api.db import FileType, ParserType
-from api.db.db_models import APIToken, API4Conversation
+from api.db.db_models import APIToken, API4Conversation, Task
 from api.db.services import duplicate_name
 from api.db.services.api_service import APITokenService, API4ConversationService
 from api.db.services.dialog_service import DialogService, chat
 from api.db.services.document_service import DocumentService
+from api.db.services.file2document_service import File2DocumentService
+from api.db.services.file_service import FileService
 from api.db.services.knowledgebase_service import KnowledgebaseService
+from api.db.services.task_service import queue_tasks, TaskService
 from api.db.services.user_service import UserTenantService
 from api.settings import RetCode
 from api.utils import get_uuid, current_timestamp, datetime_format
@@ -35,6 +39,9 @@ from itsdangerous import URLSafeTimedSerializer
 from api.utils.file_utils import filename_type, thumbnail
 from rag.utils.minio_conn import MINIO
 
+from rag.utils.es_conn import ELASTICSEARCH
+from rag.nlp import search
+from elasticsearch_dsl import Q
 
 def generate_confirmation_token(tenent_id):
     serializer = URLSafeTimedSerializer(tenent_id)
@@ -164,6 +171,7 @@ def completion():
     e, conv = API4ConversationService.get_by_id(req["conversation_id"])
     if not e:
         return get_data_error_result(retmsg="Conversation not found!")
+    if "quote" not in req: req["quote"] = False
 
     msg = []
     for m in req["messages"]:
@@ -180,13 +188,48 @@ def completion():
             return get_data_error_result(retmsg="Dialog not found!")
         del req["conversation_id"]
         del req["messages"]
-        ans = chat(dia, msg, **req)
+
         if not conv.reference:
             conv.reference = []
-        conv.reference.append(ans["reference"])
-        conv.message.append({"role": "assistant", "content": ans["answer"]})
-        API4ConversationService.append_message(conv.id, conv.to_dict())
-        return get_json_result(data=ans)
+        conv.message.append({"role": "assistant", "content": ""})
+        conv.reference.append({"chunks": [], "doc_aggs": []})
+
+        def fillin_conv(ans):
+            nonlocal conv
+            if not conv.reference:
+                conv.reference.append(ans["reference"])
+            else: conv.reference[-1] = ans["reference"]
+            conv.message[-1] = {"role": "assistant", "content": ans["answer"]}
+
+        def stream():
+            nonlocal dia, msg, req, conv
+            try:
+                for ans in chat(dia, msg, True, **req):
+                    fillin_conv(ans)
+                    yield "data:"+json.dumps({"retcode": 0, "retmsg": "", "data": ans}, ensure_ascii=False) + "\n\n"
+                API4ConversationService.append_message(conv.id, conv.to_dict())
+            except Exception as e:
+                yield "data:" + json.dumps({"retcode": 500, "retmsg": str(e),
+                                            "data": {"answer": "**ERROR**: "+str(e), "reference": []}},
+                                           ensure_ascii=False) + "\n\n"
+            yield "data:"+json.dumps({"retcode": 0, "retmsg": "", "data": True}, ensure_ascii=False) + "\n\n"
+
+        if req.get("stream", True):
+            resp = Response(stream(), mimetype="text/event-stream")
+            resp.headers.add_header("Cache-control", "no-cache")
+            resp.headers.add_header("Connection", "keep-alive")
+            resp.headers.add_header("X-Accel-Buffering", "no")
+            resp.headers.add_header("Content-Type", "text/event-stream; charset=utf-8")
+            return resp
+        else:
+            answer = None
+            for ans in chat(dia, msg, **req):
+                answer = ans
+                fillin_conv(ans)
+                API4ConversationService.append_message(conv.id, conv.to_dict())
+                break
+            return get_json_result(data=answer)
+
     except Exception as e:
         return server_error_response(e)
 
@@ -233,6 +276,13 @@ def upload():
     if file.filename == '':
         return get_json_result(
             data=False, retmsg='No file selected!', retcode=RetCode.ARGUMENT_ERROR)
+
+    root_folder = FileService.get_root_folder(tenant_id)
+    pf_id = root_folder["id"]
+    FileService.init_knowledgebase_docs(pf_id, tenant_id)
+    kb_root_folder = FileService.get_kb_folder(tenant_id)
+    kb_folder = FileService.new_a_file_from_kb(kb.tenant_id, kb.name, kb_root_folder["id"])
+
     try:
         if DocumentService.get_doc_count(kb.tenant_id) >= int(os.environ.get('MAX_FILE_NUM_PER_USER', 8192)):
             return get_data_error_result(
@@ -264,11 +314,82 @@ def upload():
             "size": len(blob),
             "thumbnail": thumbnail(filename, blob)
         }
+
+        form_data=request.form
+        if "parser_id" in form_data.keys():
+            if request.form.get("parser_id").strip() in list(vars(ParserType).values())[1:-3]:
+                doc["parser_id"] = request.form.get("parser_id").strip()
         if doc["type"] == FileType.VISUAL:
             doc["parser_id"] = ParserType.PICTURE.value
         if re.search(r"\.(ppt|pptx|pages)$", filename):
             doc["parser_id"] = ParserType.PRESENTATION.value
-        doc = DocumentService.insert(doc)
-        return get_json_result(data=doc.to_json())
+
+        doc_result = DocumentService.insert(doc)
+        FileService.add_file_from_kb(doc, kb_folder["id"], kb.tenant_id)
     except Exception as e:
         return server_error_response(e)
+
+    if "run" in form_data.keys():
+        if request.form.get("run").strip() == "1":
+            try:
+                info = {"run": 1, "progress": 0}
+                info["progress_msg"] = ""
+                info["chunk_num"] = 0
+                info["token_num"] = 0
+                DocumentService.update_by_id(doc["id"], info)
+                # if str(req["run"]) == TaskStatus.CANCEL.value:
+                tenant_id = DocumentService.get_tenant_id(doc["id"])
+                if not tenant_id:
+                    return get_data_error_result(retmsg="Tenant not found!")
+
+                #e, doc = DocumentService.get_by_id(doc["id"])
+                TaskService.filter_delete([Task.doc_id == doc["id"]])
+                e, doc = DocumentService.get_by_id(doc["id"])
+                doc = doc.to_dict()
+                doc["tenant_id"] = tenant_id
+                bucket, name = File2DocumentService.get_minio_address(doc_id=doc["id"])
+                queue_tasks(doc, bucket, name)
+            except Exception as e:
+                 return server_error_response(e)
+
+    return get_json_result(data=doc_result.to_json())
+
+
+@manager.route('/list_chunks', methods=['POST'])
+# @login_required
+def list_chunks():
+    token = request.headers.get('Authorization').split()[1]
+    objs = APIToken.query(token=token)
+    if not objs:
+        return get_json_result(
+            data=False, retmsg='Token is not valid!"', retcode=RetCode.AUTHENTICATION_ERROR)
+
+    form_data = request.form
+
+    try:
+        if "doc_name" in form_data.keys():
+            tenant_id = DocumentService.get_tenant_id_by_name(form_data['doc_name'])
+            q = Q("match", docnm_kwd=form_data['doc_name'])
+
+        elif "doc_id" in form_data.keys():
+            tenant_id = DocumentService.get_tenant_id(form_data['doc_id'])
+            q = Q("match", doc_id=form_data['doc_id'])
+        else:
+            return get_json_result(
+                data=False,retmsg="Can't find doc_name or doc_id"
+            )
+
+        res_es_search = ELASTICSEARCH.search(q,idxnm=search.index_name(tenant_id),timeout="600s")
+
+        res = [{} for _ in range(len(res_es_search['hits']['hits']))]
+
+        for index , chunk in enumerate(res_es_search['hits']['hits']):
+            res[index]['doc_name'] = chunk['_source']['docnm_kwd']
+            res[index]['content'] = chunk['_source']['content_with_weight']
+            if 'img_id' in chunk['_source'].keys():
+                res[index]['img_id'] = chunk['_source']['img_id']
+
+    except Exception as e:
+        return server_error_response(e)
+
+    return get_json_result(data=res)

api/apps/chunk_app.py

@@ -38,7 +38,7 @@ import re
 @manager.route('/list', methods=['POST'])
 @login_required
 @validate_request("doc_id")
-def list():
+def list_chunk():
     req = request.json
     doc_id = req["doc_id"]
     page = int(req.get("page", 1))

api/apps/conversation_app.py

@@ -13,12 +13,13 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 #
-from flask import request
+from flask import request, Response, jsonify
 from flask_login import login_required
 from api.db.services.dialog_service import DialogService, ConversationService, chat
 from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
 from api.utils import get_uuid
 from api.utils.api_utils import get_json_result
+import json
 
 
 @manager.route('/set', methods=['POST'])
@@ -103,9 +104,12 @@ def list_convsersation():
 
 @manager.route('/completion', methods=['POST'])
 @login_required
-@validate_request("conversation_id", "messages")
+#@validate_request("conversation_id", "messages")
 def completion():
     req = request.json
+    #req = {"conversation_id": "9aaaca4c11d311efa461fa163e197198", "messages": [
+    #    {"role": "user", "content": "上海有吗？"}
+    #]}
     msg = []
     for m in req["messages"]:
         if m["role"] == "system":
@@ -123,13 +127,48 @@ def completion():
             return get_data_error_result(retmsg="Dialog not found!")
         del req["conversation_id"]
         del req["messages"]
-        ans = chat(dia, msg, **req)
+
         if not conv.reference:
             conv.reference = []
-        conv.reference.append(ans["reference"])
-        conv.message.append({"role": "assistant", "content": ans["answer"]})
-        ConversationService.update_by_id(conv.id, conv.to_dict())
-        return get_json_result(data=ans)
+        conv.message.append({"role": "assistant", "content": ""})
+        conv.reference.append({"chunks": [], "doc_aggs": []})
+
+        def fillin_conv(ans):
+            nonlocal conv
+            if not conv.reference:
+                conv.reference.append(ans["reference"])
+            else: conv.reference[-1] = ans["reference"]
+            conv.message[-1] = {"role": "assistant", "content": ans["answer"]}
+
+        def stream():
+            nonlocal dia, msg, req, conv
+            try:
+                for ans in chat(dia, msg, True, **req):
+                    fillin_conv(ans)
+                    yield "data:"+json.dumps({"retcode": 0, "retmsg": "", "data": ans}, ensure_ascii=False) + "\n\n"
+                ConversationService.update_by_id(conv.id, conv.to_dict())
+            except Exception as e:
+                yield "data:" + json.dumps({"retcode": 500, "retmsg": str(e),
+                                            "data": {"answer": "**ERROR**: "+str(e), "reference": []}},
+                                           ensure_ascii=False) + "\n\n"
+            yield "data:"+json.dumps({"retcode": 0, "retmsg": "", "data": True}, ensure_ascii=False) + "\n\n"
+
+        if req.get("stream", True):
+            resp = Response(stream(), mimetype="text/event-stream")
+            resp.headers.add_header("Cache-control", "no-cache")
+            resp.headers.add_header("Connection", "keep-alive")
+            resp.headers.add_header("X-Accel-Buffering", "no")
+            resp.headers.add_header("Content-Type", "text/event-stream; charset=utf-8")
+            return resp
+
+        else:
+            answer = None
+            for ans in chat(dia, msg, **req):
+                answer = ans
+                fillin_conv(ans)
+                ConversationService.update_by_id(conv.id, conv.to_dict())
+                break
+            return get_json_result(data=answer)
     except Exception as e:
         return server_error_response(e)
 

api/apps/dialog_app.py

@@ -136,7 +136,7 @@ def get_kb_names(kb_ids):
 
 @manager.route('/list', methods=['GET'])
 @login_required
-def list():
+def list_dialogs():
     try:
         diags = DialogService.query(
             tenant_id=current_user.id,

api/apps/document_app.py

@@ -23,7 +23,7 @@ from elasticsearch_dsl import Q
 from flask import request
 from flask_login import login_required, current_user
 
-from api.db.db_models import Task
+from api.db.db_models import Task, File
 from api.db.services.file2document_service import File2DocumentService
 from api.db.services.file_service import FileService
 from api.db.services.task_service import TaskService, queue_tasks
@@ -33,7 +33,7 @@ from api.db.services import duplicate_name
 from api.db.services.knowledgebase_service import KnowledgebaseService
 from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
 from api.utils import get_uuid
-from api.db import FileType, TaskStatus, ParserType
+from api.db import FileType, TaskStatus, ParserType, FileSource
 from api.db.services.document_service import DocumentService
 from api.settings import RetCode
 from api.utils.api_utils import get_json_result
@@ -59,12 +59,19 @@ def upload():
             return get_json_result(
                 data=False, retmsg='No file selected!', retcode=RetCode.ARGUMENT_ERROR)
 
+    e, kb = KnowledgebaseService.get_by_id(kb_id)
+    if not e:
+        raise LookupError("Can't find this knowledgebase!")
+
+    root_folder = FileService.get_root_folder(current_user.id)
+    pf_id = root_folder["id"]
+    FileService.init_knowledgebase_docs(pf_id, current_user.id)
+    kb_root_folder = FileService.get_kb_folder(current_user.id)
+    kb_folder = FileService.new_a_file_from_kb(kb.tenant_id, kb.name, kb_root_folder["id"])
+
     err = []
     for file in file_objs:
         try:
-            e, kb = KnowledgebaseService.get_by_id(kb_id)
-            if not e:
-                raise LookupError("Can't find this knowledgebase!")
             MAX_FILE_NUM_PER_USER = int(os.environ.get('MAX_FILE_NUM_PER_USER', 0))
             if MAX_FILE_NUM_PER_USER > 0 and DocumentService.get_doc_count(kb.tenant_id) >= MAX_FILE_NUM_PER_USER:
                 raise RuntimeError("Exceed the maximum file number of a free user!")
@@ -99,6 +106,8 @@ def upload():
             if re.search(r"\.(ppt|pptx|pages)$", filename):
                 doc["parser_id"] = ParserType.PRESENTATION.value
             DocumentService.insert(doc)
+
+            FileService.add_file_from_kb(doc, kb_folder["id"], kb.tenant_id)
         except Exception as e:
             err.append(file.filename + ": " + str(e))
     if err:
@@ -145,7 +154,7 @@ def create():
 
 @manager.route('/list', methods=['GET'])
 @login_required
-def list():
+def list_docs():
     kb_id = request.args.get("kb_id")
     if not kb_id:
         return get_json_result(
@@ -228,34 +237,36 @@ def rm():
     req = request.json
     doc_ids = req["doc_id"]
     if isinstance(doc_ids, str): doc_ids = [doc_ids]
+    root_folder = FileService.get_root_folder(current_user.id)
+    pf_id = root_folder["id"]
+    FileService.init_knowledgebase_docs(pf_id, current_user.id)
     errors = ""
     for doc_id in doc_ids:
         try:
             e, doc = DocumentService.get_by_id(doc_id)
-
             if not e:
                 return get_data_error_result(retmsg="Document not found!")
             tenant_id = DocumentService.get_tenant_id(doc_id)
             if not tenant_id:
                 return get_data_error_result(retmsg="Tenant not found!")
 
-            ELASTICSEARCH.deleteByQuery(
-                Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
-            DocumentService.increment_chunk_num(
-                doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-            if not DocumentService.delete(doc):
+            b, n = File2DocumentService.get_minio_address(doc_id=doc_id)
+
+            if not DocumentService.remove_document(doc, tenant_id):
                 return get_data_error_result(
                     retmsg="Database error (Document removal)!")
 
-            informs = File2DocumentService.get_by_document_id(doc_id)
-            if not informs:
-                MINIO.rm(doc.kb_id, doc.location)
-            else:
-                File2DocumentService.delete_by_document_id(doc_id)
+            f2d = File2DocumentService.get_by_document_id(doc_id)
+            FileService.filter_delete([File.source_type == FileSource.KNOWLEDGEBASE, File.id == f2d[0].file_id])
+            File2DocumentService.delete_by_document_id(doc_id)
+
+            MINIO.rm(b, n)
         except Exception as e:
             errors += str(e)
 
-    if errors: return server_error_response(e)
+    if errors:
+        return get_json_result(data=False, retmsg=errors, retcode=RetCode.SERVER_ERROR)
+
     return get_json_result(data=True)
 
 
@@ -307,9 +318,10 @@ def rename():
                 data=False,
                 retmsg="The extension of file can't be changed",
                 retcode=RetCode.ARGUMENT_ERROR)
-        if DocumentService.query(name=req["name"], kb_id=doc.kb_id):
-            return get_data_error_result(
-                retmsg="Duplicated document name in the same knowledgebase.")
+        for d in DocumentService.query(name=req["name"], kb_id=doc.kb_id):
+            if d.name == req["name"]:
+                return get_data_error_result(
+                    retmsg="Duplicated document name in the same knowledgebase.")
 
         if not DocumentService.update_by_id(
                 req["doc_id"], {"name": req["name"]}):
@@ -334,12 +346,8 @@ def get(doc_id):
         if not e:
             return get_data_error_result(retmsg="Document not found!")
 
-        informs = File2DocumentService.get_by_document_id(doc_id)
-        if not informs:
-            response = flask.make_response(MINIO.get(doc.kb_id, doc.location))
-        else:
-            e, file = FileService.get_by_id(informs[0].file_id)
-            response = flask.make_response(MINIO.get(file.parent_id, doc.location))
+        b,n = File2DocumentService.get_minio_address(doc_id=doc_id)
+        response = flask.make_response(MINIO.get(b, n))
 
         ext = re.search(r"\.([^.]+)$", doc.name)
         if ext:

api/apps/file2document_app.py

@@ -58,11 +58,7 @@ def convert():
                     tenant_id = DocumentService.get_tenant_id(doc_id)
                     if not tenant_id:
                         return get_data_error_result(retmsg="Tenant not found!")
-                    ELASTICSEARCH.deleteByQuery(
-                        Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
-                    DocumentService.increment_chunk_num(
-                        doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-                    if not DocumentService.delete(doc):
+                    if not DocumentService.remove_document(doc, tenant_id):
                         return get_data_error_result(
                             retmsg="Database error (Document removal)!")
                 File2DocumentService.delete_by_file_id(id)
@@ -125,11 +121,7 @@ def rm():
                 tenant_id = DocumentService.get_tenant_id(doc_id)
                 if not tenant_id:
                     return get_data_error_result(retmsg="Tenant not found!")
-                ELASTICSEARCH.deleteByQuery(
-                    Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
-                DocumentService.increment_chunk_num(
-                    doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-                if not DocumentService.delete(doc):
+                if not DocumentService.remove_document(doc, tenant_id):
                     return get_data_error_result(
                         retmsg="Database error (Document removal)!")
         return get_json_result(data=True)

api/apps/file_app.py

@@ -26,7 +26,7 @@ from api.db.services.document_service import DocumentService
 from api.db.services.file2document_service import File2DocumentService
 from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
 from api.utils import get_uuid
-from api.db import FileType
+from api.db import FileType, FileSource
 from api.db.services import duplicate_name
 from api.db.services.file_service import FileService
 from api.settings import RetCode
@@ -45,7 +45,7 @@ def upload():
 
     if not pf_id:
         root_folder = FileService.get_root_folder(current_user.id)
-        pf_id = root_folder.id
+        pf_id = root_folder["id"]
 
     if 'file' not in request.files:
         return get_json_result(
@@ -132,7 +132,7 @@ def create():
     input_file_type = request.json.get("type")
     if not pf_id:
         root_folder = FileService.get_root_folder(current_user.id)
-        pf_id = root_folder.id
+        pf_id = root_folder["id"]
 
     try:
         if not FileService.is_parent_folder_exist(pf_id):
@@ -165,7 +165,7 @@ def create():
 
 @manager.route('/list', methods=['GET'])
 @login_required
-def list():
+def list_files():
     pf_id = request.args.get("parent_id")
 
     keywords = request.args.get("keywords", "")
@@ -176,7 +176,8 @@ def list():
     desc = request.args.get("desc", True)
     if not pf_id:
         root_folder = FileService.get_root_folder(current_user.id)
-        pf_id = root_folder.id
+        pf_id = root_folder["id"]
+        FileService.init_knowledgebase_docs(pf_id, current_user.id)
     try:
         e, file = FileService.get_by_id(pf_id)
         if not e:
@@ -199,7 +200,7 @@ def list():
 def get_root_folder():
     try:
         root_folder = FileService.get_root_folder(current_user.id)
-        return get_json_result(data={"root_folder": root_folder.to_json()})
+        return get_json_result(data={"root_folder": root_folder})
     except Exception as e:
         return server_error_response(e)
 
@@ -250,6 +251,8 @@ def rm():
                 return get_data_error_result(retmsg="File or Folder not found!")
             if not file.tenant_id:
                 return get_data_error_result(retmsg="Tenant not found!")
+            if file.source_type == FileSource.KNOWLEDGEBASE:
+                continue
 
             if file.type == FileType.FOLDER.value:
                 file_id_list = FileService.get_all_innermost_file_ids(file_id, [])
@@ -274,11 +277,7 @@ def rm():
                 tenant_id = DocumentService.get_tenant_id(doc_id)
                 if not tenant_id:
                     return get_data_error_result(retmsg="Tenant not found!")
-                ELASTICSEARCH.deleteByQuery(
-                    Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
-                DocumentService.increment_chunk_num(
-                    doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-                if not DocumentService.delete(doc):
+                if not DocumentService.remove_document(doc, tenant_id):
                     return get_data_error_result(
                         retmsg="Database error (Document removal)!")
             File2DocumentService.delete_by_file_id(file_id)
@@ -303,9 +302,10 @@ def rename():
                 data=False,
                 retmsg="The extension of file can't be changed",
                 retcode=RetCode.ARGUMENT_ERROR)
-        if FileService.query(name=req["name"], pf_id=file.parent_id):
-            return get_data_error_result(
-                retmsg="Duplicated file name in the same folder.")
+        for file in FileService.query(name=req["name"], pf_id=file.parent_id):
+            if file.name == req["name"]:
+                return get_data_error_result(
+                    retmsg="Duplicated file name in the same folder.")
 
         if not FileService.update_by_id(
                 req["file_id"], {"name": req["name"]}):

api/apps/kb_app.py

@@ -19,12 +19,14 @@ from flask_login import login_required, current_user
 
 from api.db.services import duplicate_name
 from api.db.services.document_service import DocumentService
+from api.db.services.file2document_service import File2DocumentService
+from api.db.services.file_service import FileService
 from api.db.services.user_service import TenantService, UserTenantService
 from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
 from api.utils import get_uuid, get_format_time
-from api.db import StatusEnum, UserTenantRole
+from api.db import StatusEnum, UserTenantRole, FileSource
 from api.db.services.knowledgebase_service import KnowledgebaseService
-from api.db.db_models import Knowledgebase
+from api.db.db_models import Knowledgebase, File
 from api.settings import stat_logger, RetCode
 from api.utils.api_utils import get_json_result
 from rag.nlp import search
@@ -109,7 +111,7 @@ def detail():
 
 @manager.route('/list', methods=['GET'])
 @login_required
-def list():
+def list_kbs():
     page_number = request.args.get("page", 1)
     items_per_page = request.args.get("page_size", 150)
     orderby = request.args.get("orderby", "create_time")
@@ -136,17 +138,14 @@ def rm():
                 data=False, retmsg=f'Only owner of knowledgebase authorized for this operation.', retcode=RetCode.OPERATING_ERROR)
 
         for doc in DocumentService.query(kb_id=req["kb_id"]):
-            ELASTICSEARCH.deleteByQuery(
-                Q("match", doc_id=doc.id), idxnm=search.index_name(kbs[0].tenant_id))
-
-            DocumentService.increment_chunk_num(
-                doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-            if not DocumentService.delete(doc):
+            if not DocumentService.remove_document(doc, kbs[0].tenant_id):
                 return get_data_error_result(
                     retmsg="Database error (Document removal)!")
+            f2d = File2DocumentService.get_by_document_id(doc.id)
+            FileService.filter_delete([File.source_type == FileSource.KNOWLEDGEBASE, File.id == f2d[0].file_id])
+            File2DocumentService.delete_by_document_id(doc.id)
 
-        if not KnowledgebaseService.update_by_id(
-                req["kb_id"], {"status": StatusEnum.INVALID.value}):
+        if not KnowledgebaseService.delete_by_id(req["kb_id"]):
             return get_data_error_result(
                 retmsg="Database error (Knowledgebase removal)!")
         return get_json_result(data=True)

api/apps/llm_app.py

@@ -142,6 +142,16 @@ def add_llm():
     return get_json_result(data=True)
 
 
+@manager.route('/delete_llm', methods=['POST'])
+@login_required
+@validate_request("llm_factory", "llm_name")
+def delete_llm():
+    req = request.json
+    TenantLLMService.filter_delete(
+            [TenantLLM.tenant_id == current_user.id, TenantLLM.llm_factory == req["llm_factory"], TenantLLM.llm_name == req["llm_name"]])
+    return get_json_result(data=True)
+
+
 @manager.route('/my_llms', methods=['GET'])
 @login_required
 def my_llms():
@@ -165,7 +175,7 @@ def my_llms():
 
 @manager.route('/list', methods=['GET'])
 @login_required
-def list():
+def list_app():
     model_type = request.args.get("model_type")
     try:
         objs = TenantLLMService.query(tenant_id=current_user.id)
@@ -184,7 +194,7 @@ def list():
 
         res = {}
         for m in llms:
-            if model_type and m["model_type"] != model_type:
+            if model_type and m["model_type"].find(model_type)<0:
                 continue
             if m["fid"] not in res:
                 res[m["fid"]] = []

api/apps/system_app.py

@@ -0,0 +1,67 @@
+#
+#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License
+#
+from flask_login import login_required
+
+from api.db.services.knowledgebase_service import KnowledgebaseService
+from api.utils.api_utils import get_json_result
+from api.versions import get_rag_version
+from rag.settings import SVR_QUEUE_NAME
+from rag.utils.es_conn import ELASTICSEARCH
+from rag.utils.minio_conn import MINIO
+from timeit import default_timer as timer
+
+from rag.utils.redis_conn import REDIS_CONN
+
+
+@manager.route('/version', methods=['GET'])
+@login_required
+def version():
+    return get_json_result(data=get_rag_version())
+
+
+@manager.route('/status', methods=['GET'])
+@login_required
+def status():
+    res = {}
+    st = timer()
+    try:
+        res["es"] = ELASTICSEARCH.health()
+        res["es"]["elapsed"] = "{:.1f}".format((timer() - st)*1000.)
+    except Exception as e:
+        res["es"] = {"status": "red", "elapsed": "{:.1f}".format((timer() - st)*1000.), "error": str(e)}
+
+    st = timer()
+    try:
+        MINIO.health()
+        res["minio"] = {"status": "green", "elapsed": "{:.1f}".format((timer() - st)*1000.)}
+    except Exception as e:
+        res["minio"] = {"status": "red", "elapsed": "{:.1f}".format((timer() - st)*1000.), "error": str(e)}
+
+    st = timer()
+    try:
+        KnowledgebaseService.get_by_id("x")
+        res["mysql"] = {"status": "green", "elapsed": "{:.1f}".format((timer() - st)*1000.)}
+    except Exception as e:
+        res["mysql"] = {"status": "red", "elapsed": "{:.1f}".format((timer() - st)*1000.), "error": str(e)}
+
+    st = timer()
+    try:
+        qinfo = REDIS_CONN.health(SVR_QUEUE_NAME)
+        res["redis"] = {"status": "green", "elapsed": "{:.1f}".format((timer() - st)*1000.), "pending": qinfo["pending"]}
+    except Exception as e:
+        res["redis"] = {"status": "red", "elapsed": "{:.1f}".format((timer() - st)*1000.), "error": str(e)}
+
+    return get_json_result(data=res)

api/apps/user_app.py

@@ -122,6 +122,79 @@ def github_callback():
     return redirect("/?auth=%s" % user.get_id())
 
 
+@manager.route('/feishu_callback', methods=['GET'])
+def feishu_callback():
+    import requests
+    app_access_token_res = requests.post(FEISHU_OAUTH.get("app_access_token_url"), data=json.dumps({
+        "app_id": FEISHU_OAUTH.get("app_id"),
+        "app_secret": FEISHU_OAUTH.get("app_secret")
+    }), headers={"Content-Type": "application/json; charset=utf-8"})
+    app_access_token_res = app_access_token_res.json()
+    if app_access_token_res['code'] != 0:
+        return redirect("/?error=%s" % app_access_token_res)
+
+    res = requests.post(FEISHU_OAUTH.get("user_access_token_url"), data=json.dumps({
+        "grant_type": FEISHU_OAUTH.get("grant_type"),
+        "code": request.args.get('code')
+    }), headers={"Content-Type": "application/json; charset=utf-8",
+                 'Authorization': f"Bearer {app_access_token_res['app_access_token']}"})
+    res = res.json()
+    if res['code'] != 0:
+        return redirect("/?error=%s" % res["message"])
+
+    if "contact:user.email:readonly" not in res["data"]["scope"].split(" "):
+        return redirect("/?error=contact:user.email:readonly not in scope")
+    session["access_token"] = res["data"]["access_token"]
+    session["access_token_from"] = "feishu"
+    userinfo = user_info_from_feishu(session["access_token"])
+    users = UserService.query(email=userinfo["email"])
+    user_id = get_uuid()
+    if not users:
+        try:
+            try:
+                avatar = download_img(userinfo["avatar_url"])
+            except Exception as e:
+                stat_logger.exception(e)
+                avatar = ""
+            users = user_register(user_id, {
+                "access_token": session["access_token"],
+                "email": userinfo["email"],
+                "avatar": avatar,
+                "nickname": userinfo["en_name"],
+                "login_channel": "feishu",
+                "last_login_time": get_format_time(),
+                "is_superuser": False,
+            })
+            if not users:
+                raise Exception('Register user failure.')
+            if len(users) > 1:
+                raise Exception('Same E-mail exist!')
+            user = users[0]
+            login_user(user)
+            return redirect("/?auth=%s" % user.get_id())
+        except Exception as e:
+            rollback_user_registration(user_id)
+            stat_logger.exception(e)
+            return redirect("/?error=%s" % str(e))
+    user = users[0]
+    user.access_token = get_uuid()
+    login_user(user)
+    user.save()
+    return redirect("/?auth=%s" % user.get_id())
+
+
+def user_info_from_feishu(access_token):
+    import requests
+    headers = {"Content-Type": "application/json; charset=utf-8",
+               'Authorization': f"Bearer {access_token}"}
+    res = requests.get(
+        f"https://open.feishu.cn/open-apis/authen/v1/user_info",
+        headers=headers)
+    user_info = res.json()["data"]
+    user_info["email"] = None if user_info.get("email") == "" else user_info["email"]
+    return user_info
+
+
 def user_info_from_github(access_token):
     import requests
     headers = {"Accept": "application/json",
@@ -200,7 +273,7 @@ def rollback_user_registration(user_id):
     except Exception as e:
         pass
     try:
-        TenantLLM.delete().where(TenantLLM.tenant_id == user_id).excute()
+        TenantLLM.delete().where(TenantLLM.tenant_id == user_id).execute()
     except Exception as e:
         pass
 

api/db/__init__.py

@@ -83,3 +83,11 @@ class ParserType(StrEnum):
     NAIVE = "naive"
     PICTURE = "picture"
     ONE = "one"
+
+
+class FileSource(StrEnum):
+    LOCAL = ""
+    KNOWLEDGEBASE = "knowledgebase"
+    S3 = "s3"
+
+KNOWLEDGEBASE_FOLDER_NAME=".knowledgebase"

api/db/db_models.py

@@ -21,14 +21,13 @@ import operator
 from functools import wraps
 from itsdangerous.url_safe import URLSafeTimedSerializer as Serializer
 from flask_login import UserMixin
-
+from playhouse.migrate import MySQLMigrator, migrate
 from peewee import (
-    BigAutoField, BigIntegerField, BooleanField, CharField,
-    CompositeKey, Insert, IntegerField, TextField, FloatField, DateTimeField,
+    BigIntegerField, BooleanField, CharField,
+    CompositeKey, IntegerField, TextField, FloatField, DateTimeField,
     Field, Model, Metadata
 )
 from playhouse.pool import PooledMySQLDatabase
-
 from api.db import SerializedType, ParserType
 from api.settings import DATABASE, stat_logger, SECRET_KEY
 from api.utils.log_utils import getLogger
@@ -344,7 +343,7 @@ class DataBaseModel(BaseModel):
 
 
 @DB.connection_context()
-def init_database_tables():
+def init_database_tables(alter_fields=[]):
     members = inspect.getmembers(sys.modules[__name__], inspect.isclass)
     table_objs = []
     create_failed_list = []
@@ -361,6 +360,7 @@ def init_database_tables():
     if create_failed_list:
         LOGGER.info(f"create tables failed: {create_failed_list}")
         raise Exception(f"create tables failed: {create_failed_list}")
+    migrate_db()
 
 
 def fill_db_model_object(model_object, human_model_dict):
@@ -386,7 +386,7 @@ class User(DataBaseModel, UserMixin):
         max_length=32,
         null=True,
         help_text="English|Chinese",
-        default="English")
+        default="Chinese" if "zh_CN" in os.getenv("LANG", "") else "English")     
     color_schema = CharField(
         max_length=32,
         null=True,
@@ -578,7 +578,7 @@ class Knowledgebase(DataBaseModel):
     language = CharField(
         max_length=32,
         null=True,
-        default="English",
+        default="Chinese" if "zh_CN" in os.getenv("LANG", "") else "English",
         help_text="English|Chinese")
     description = TextField(null=True, help_text="KB description")
     embd_id = CharField(
@@ -699,6 +699,11 @@ class File(DataBaseModel):
         help_text="where dose it store")
     size = IntegerField(default=0)
     type = CharField(max_length=32, null=False, help_text="file extension")
+    source_type = CharField(
+        max_length=128,
+        null=False,
+        default="",
+        help_text="where dose this document come from")
 
     class Meta:
         db_table = "file"
@@ -750,11 +755,11 @@ class Dialog(DataBaseModel):
     language = CharField(
         max_length=32,
         null=True,
-        default="Chinese",
+        default="Chinese" if "zh_CN" in os.getenv("LANG", "") else "English",
         help_text="English|Chinese")
     llm_id = CharField(max_length=128, null=False, help_text="default llm ID")
     llm_setting = JSONField(null=False, default={"temperature": 0.1, "top_p": 0.3, "frequency_penalty": 0.7,
-                                                 "presence_penalty": 0.4, "max_tokens": 215})
+                                                 "presence_penalty": 0.4, "max_tokens": 512})
     prompt_type = CharField(
         max_length=16,
         null=False,
@@ -817,3 +822,14 @@ class API4Conversation(DataBaseModel):
 
     class Meta:
         db_table = "api_4_conversation"
+
+
+def migrate_db():
+    try:
+        with DB.transaction():
+            migrator = MySQLMigrator(DB)
+            migrate(
+                migrator.add_column('file', 'source_type', CharField(max_length=128, null=False, default="", help_text="where dose this document come from"))
+            )
+    except Exception as e:
+        pass

api/db/init_data.py

@@ -16,10 +16,13 @@
 import os
 import time
 import uuid
+from copy import deepcopy
 
 from api.db import LLMType, UserTenantRole
 from api.db.db_models import init_database_tables as init_web_db, LLMFactories, LLM, TenantLLM
 from api.db.services import UserService
+from api.db.services.document_service import DocumentService
+from api.db.services.knowledgebase_service import KnowledgebaseService
 from api.db.services.llm_service import LLMFactoriesService, LLMService, TenantLLMService, LLMBundle
 from api.db.services.user_service import TenantService, UserTenantService
 from api.settings import CHAT_MDL, EMBEDDING_MDL, ASR_MDL, IMAGE2TEXT_MDL, PARSERS, LLM_FACTORY, API_KEY, LLM_BASE_URL
@@ -143,6 +146,12 @@ def init_llm_factory():
     llm_infos = [
         # ---------------------- OpenAI ------------------------
         {
+            "fid": factory_infos[0]["name"],
+            "llm_name": "gpt-4o",
+            "tags": "LLM,CHAT,128K",
+            "max_tokens": 128000,
+            "model_type": LLMType.CHAT.value + "," + LLMType.IMAGE2TEXT.value
+        }, {
             "fid": factory_infos[0]["name"],
             "llm_name": "gpt-3.5-turbo",
             "tags": "LLM,CHAT,4K",
@@ -160,6 +169,18 @@ def init_llm_factory():
             "tags": "TEXT EMBEDDING,8K",
             "max_tokens": 8191,
             "model_type": LLMType.EMBEDDING.value
+        }, {
+            "fid": factory_infos[0]["name"],
+            "llm_name": "text-embedding-3-small",
+            "tags": "TEXT EMBEDDING,8K",
+            "max_tokens": 8191,
+            "model_type": LLMType.EMBEDDING.value
+        }, {
+            "fid": factory_infos[0]["name"],
+            "llm_name": "text-embedding-3-large",
+            "tags": "TEXT EMBEDDING,8K",
+            "max_tokens": 8191,
+            "model_type": LLMType.EMBEDDING.value
         }, {
             "fid": factory_infos[0]["name"],
             "llm_name": "whisper-1",
@@ -370,6 +391,25 @@ def init_llm_factory():
     LLMFactoriesService.filter_delete([LLMFactoriesService.model.name == "QAnything"])
     LLMService.filter_delete([LLMService.model.fid == "QAnything"])
     TenantLLMService.filter_update([TenantLLMService.model.llm_factory == "QAnything"], {"llm_factory": "Youdao"})
+    ## insert openai two embedding models to the current openai user.
+    print("Start to insert 2 OpenAI embedding models...")
+    tenant_ids = set([row["tenant_id"] for row in TenantLLMService.get_openai_models()])
+    for tid in tenant_ids:
+        for row in TenantLLMService.query(llm_factory="OpenAI", tenant_id=tid):
+            row = row.to_dict()
+            row["model_type"] = LLMType.EMBEDDING.value
+            row["llm_name"] = "text-embedding-3-small"
+            row["used_tokens"] = 0
+            try:
+                TenantLLMService.save(**row)
+                row = deepcopy(row)
+                row["llm_name"] = "text-embedding-3-large"
+                TenantLLMService.save(**row)
+            except Exception as e:
+                pass
+            break
+    for kb_id in KnowledgebaseService.get_all_ids():
+        KnowledgebaseService.update_by_id(kb_id, {"doc_num": DocumentService.get_kb_doc_count(kb_id)})
     """
     drop table llm;
     drop table llm_factories;

api/db/services/dialog_service.py

@@ -14,6 +14,7 @@
 #  limitations under the License.
 #
 import re
+from copy import deepcopy
 
 from api.db import LLMType
 from api.db.db_models import Dialog, Conversation
@@ -71,7 +72,7 @@ def message_fit_in(msg, max_length=4000):
     return max_length, msg
 
 
-def chat(dialog, messages, **kwargs):
+def chat(dialog, messages, stream=True, **kwargs):
     assert messages[-1]["role"] == "user", "The last content of this conversation is not from user."
     llm = LLMService.query(llm_name=dialog.llm_id)
     if not llm:
@@ -82,7 +83,9 @@ def chat(dialog, messages, **kwargs):
     else: max_tokens = llm[0].max_tokens
     kbs = KnowledgebaseService.get_by_ids(dialog.kb_ids)
     embd_nms = list(set([kb.embd_id for kb in kbs]))
-    assert len(embd_nms) == 1, "Knowledge bases use different embedding models."
+    if len(embd_nms) != 1:
+        yield {"answer": "**ERROR**: Knowledge bases use different embedding models.", "reference": []}
+        return {"answer": "**ERROR**: Knowledge bases use different embedding models.", "reference": []}
 
     questions = [m["content"] for m in messages if m["role"] == "user"]
     embd_mdl = LLMBundle(dialog.tenant_id, LLMType.EMBEDDING, embd_nms[0])
@@ -94,7 +97,9 @@ def chat(dialog, messages, **kwargs):
     if field_map:
         chat_logger.info("Use SQL to retrieval:{}".format(questions[-1]))
         ans = use_sql(questions[-1], field_map, dialog.tenant_id, chat_mdl, prompt_config.get("quote", True))
-        if ans: return ans
+        if ans:
+            yield ans
+            return
 
     for p in prompt_config["parameters"]:
         if p["key"] == "knowledge":
@@ -112,14 +117,16 @@ def chat(dialog, messages, **kwargs):
     else:
         kbinfos = retrievaler.retrieval(" ".join(questions), embd_mdl, dialog.tenant_id, dialog.kb_ids, 1, dialog.top_n,
                                         dialog.similarity_threshold,
-                                        dialog.vector_similarity_weight, top=1024, aggs=False)
+                                        dialog.vector_similarity_weight,
+                                        doc_ids=kwargs["doc_ids"].split(",") if "doc_ids" in kwargs else None,
+                                        top=1024, aggs=False)
     knowledges = [ck["content_with_weight"] for ck in kbinfos["chunks"]]
     chat_logger.info(
         "{}->{}".format(" ".join(questions), "\n->".join(knowledges)))
 
     if not knowledges and prompt_config.get("empty_response"):
-        return {
-            "answer": prompt_config["empty_response"], "reference": kbinfos}
+        yield {"answer": prompt_config["empty_response"], "reference": kbinfos}
+        return {"answer": prompt_config["empty_response"], "reference": kbinfos}
 
     kwargs["knowledge"] = "\n".join(knowledges)
     gen_conf = dialog.llm_setting
@@ -130,33 +137,45 @@ def chat(dialog, messages, **kwargs):
         gen_conf["max_tokens"] = min(
             gen_conf["max_tokens"],
             max_tokens - used_token_count)
-    answer = chat_mdl.chat(
-        prompt_config["system"].format(
-            **kwargs), msg, gen_conf)
-    chat_logger.info("User: {}|Assistant: {}".format(
-        msg[-1]["content"], answer))
 
-    if knowledges and (prompt_config.get("quote", True) and kwargs.get("quote", True)):
-        answer, idx = retrievaler.insert_citations(answer,
-                                                   [ck["content_ltks"]
-                                                       for ck in kbinfos["chunks"]],
-                                                   [ck["vector"]
-                                                       for ck in kbinfos["chunks"]],
-                                                   embd_mdl,
-                                                   tkweight=1 - dialog.vector_similarity_weight,
-                                                   vtweight=dialog.vector_similarity_weight)
-        idx = set([kbinfos["chunks"][int(i)]["doc_id"] for i in idx])
-        recall_docs = [
-            d for d in kbinfos["doc_aggs"] if d["doc_id"] in idx]
-        if not recall_docs: recall_docs = kbinfos["doc_aggs"]
-        kbinfos["doc_aggs"] = recall_docs
+    def decorate_answer(answer):
+        nonlocal prompt_config, knowledges, kwargs, kbinfos
+        if knowledges and (prompt_config.get("quote", True) and kwargs.get("quote", True)):
+            answer, idx = retrievaler.insert_citations(answer,
+                                                       [ck["content_ltks"]
+                                                           for ck in kbinfos["chunks"]],
+                                                       [ck["vector"]
+                                                           for ck in kbinfos["chunks"]],
+                                                       embd_mdl,
+                                                       tkweight=1 - dialog.vector_similarity_weight,
+                                                       vtweight=dialog.vector_similarity_weight)
+            idx = set([kbinfos["chunks"][int(i)]["doc_id"] for i in idx])
+            recall_docs = [
+                d for d in kbinfos["doc_aggs"] if d["doc_id"] in idx]
+            if not recall_docs: recall_docs = kbinfos["doc_aggs"]
+            kbinfos["doc_aggs"] = recall_docs
 
-    for c in kbinfos["chunks"]:
-        if c.get("vector"):
-            del c["vector"]
-    if answer.lower().find("invalid key") >= 0 or answer.lower().find("invalid api")>=0:
-        answer += " Please set LLM API-Key in 'User Setting -> Model Providers -> API-Key'"
-    return {"answer": answer, "reference": kbinfos}
+        refs = deepcopy(kbinfos)
+        for c in refs["chunks"]:
+            if c.get("vector"):
+                del c["vector"]
+        if answer.lower().find("invalid key") >= 0 or answer.lower().find("invalid api")>=0:
+            answer += " Please set LLM API-Key in 'User Setting -> Model Providers -> API-Key'"
+        return {"answer": answer, "reference": refs}
+
+    if stream:
+        answer = ""
+        for ans in chat_mdl.chat_streamly(prompt_config["system"].format(**kwargs), msg, gen_conf):
+            answer = ans
+            yield {"answer": answer, "reference": {}}
+        yield decorate_answer(answer)
+    else:
+        answer = chat_mdl.chat(
+            prompt_config["system"].format(
+                **kwargs), msg, gen_conf)
+        chat_logger.info("User: {}|Assistant: {}".format(
+            msg[-1]["content"], answer))
+        yield decorate_answer(answer)
 
 
 def use_sql(question, field_map, tenant_id, chat_mdl, quota=True):

api/db/services/document_service.py

@@ -16,6 +16,7 @@
 import random
 from datetime import datetime
 from elasticsearch_dsl import Q
+from peewee import fn
 
 from api.settings import stat_logger
 from api.utils import current_timestamp, get_format_time
@@ -40,8 +41,9 @@ class DocumentService(CommonService):
                      orderby, desc, keywords):
         if keywords:
             docs = cls.model.select().where(
-                cls.model.kb_id == kb_id,
-                cls.model.name.like(f"%%{keywords}%%"))
+                (cls.model.kb_id == kb_id),
+                (fn.LOWER(cls.model.name).contains(keywords.lower()))
+            )
         else:
             docs = cls.model.select().where(cls.model.kb_id == kb_id)
         count = docs.count()
@@ -68,27 +70,12 @@ class DocumentService(CommonService):
             raise RuntimeError("Database error (Knowledgebase)!")
         return doc
 
-    @classmethod
-    @DB.connection_context()
-    def delete(cls, doc):
-        e, kb = KnowledgebaseService.get_by_id(doc.kb_id)
-        if not KnowledgebaseService.update_by_id(
-                kb.id, {"doc_num": kb.doc_num - 1}):
-            raise RuntimeError("Database error (Knowledgebase)!")
-        return cls.delete_by_id(doc.id)
-
     @classmethod
     @DB.connection_context()
     def remove_document(cls, doc, tenant_id):
         ELASTICSEARCH.deleteByQuery(
-            Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
-
-        cls.increment_chunk_num(
-            doc.id, doc.kb_id, doc.token_num * -1, doc.chunk_num * -1, 0)
-        if not cls.delete(doc):
-            raise RuntimeError("Database error (Document removal)!")
-
-        MINIO.rm(doc.kb_id, doc.location)
+                Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
+        cls.clear_chunk_num(doc.id)
         return cls.delete_by_id(doc.id)
 
     @classmethod
@@ -150,6 +137,22 @@ class DocumentService(CommonService):
             Knowledgebase.id == kb_id).execute()
         return num
 
+    @classmethod
+    @DB.connection_context()
+    def clear_chunk_num(cls, doc_id):
+        doc = cls.model.get_by_id(doc_id)
+        assert doc, "Can't fine document in database."
+
+        num = Knowledgebase.update(
+            token_num=Knowledgebase.token_num -
+            doc.token_num,
+            chunk_num=Knowledgebase.chunk_num -
+            doc.chunk_num,
+            doc_num=Knowledgebase.doc_num-1
+        ).where(
+            Knowledgebase.id == doc.kb_id).execute()
+        return num
+
     @classmethod
     @DB.connection_context()
     def get_tenant_id(cls, doc_id):
@@ -163,6 +166,19 @@ class DocumentService(CommonService):
             return
         return docs[0]["tenant_id"]
 
+    @classmethod
+    @DB.connection_context()
+    def get_tenant_id_by_name(cls, name):
+        docs = cls.model.select(
+            Knowledgebase.tenant_id).join(
+            Knowledgebase, on=(
+                    Knowledgebase.id == cls.model.kb_id)).where(
+            cls.model.name == name, Knowledgebase.status == StatusEnum.VALID.value)
+        docs = docs.dicts()
+        if not docs:
+            return
+        return docs[0]["tenant_id"]
+
     @classmethod
     @DB.connection_context()
     def get_thumbnails(cls, docids):
@@ -249,3 +265,9 @@ class DocumentService(CommonService):
             except Exception as e:
                 stat_logger.error("fetch task exception:" + str(e))
 
+    @classmethod
+    @DB.connection_context()
+    def get_kb_doc_count(cls, kb_id):
+        return len(cls.model.select(cls.model.id).where(
+            cls.model.kb_id == kb_id).dicts())
+

api/db/services/file2document_service.py

@@ -15,12 +15,12 @@
 #
 from datetime import datetime
 
+from api.db import FileSource
 from api.db.db_models import DB
-from api.db.db_models import File, Document, File2Document
+from api.db.db_models import File, File2Document
 from api.db.services.common_service import CommonService
 from api.db.services.document_service import DocumentService
-from api.db.services.file_service import FileService
-from api.utils import current_timestamp, datetime_format
+from api.utils import current_timestamp, datetime_format, get_uuid
 
 
 class File2DocumentService(CommonService):
@@ -71,13 +71,15 @@ class File2DocumentService(CommonService):
     @DB.connection_context()
     def get_minio_address(cls, doc_id=None, file_id=None):
         if doc_id:
-            ids = File2DocumentService.get_by_document_id(doc_id)
+            f2d = cls.get_by_document_id(doc_id)
         else:
-            ids = File2DocumentService.get_by_file_id(file_id)
-        if ids:
-            e, file = FileService.get_by_id(ids[0].file_id)
-            return file.parent_id, file.location
-        else:
-            assert doc_id, "please specify doc_id"
-            e, doc = DocumentService.get_by_id(doc_id)
-            return doc.kb_id, doc.location
+            f2d = cls.get_by_file_id(file_id)
+        if f2d:
+            file = File.get_by_id(f2d[0].file_id)
+            if file.source_type == FileSource.LOCAL:
+                return file.parent_id, file.location
+            doc_id = f2d[0].document_id
+
+        assert doc_id, "please specify doc_id"
+        e, doc = DocumentService.get_by_id(doc_id)
+        return doc.kb_id, doc.location

api/db/services/file_service.py

@@ -16,10 +16,12 @@
 from flask_login import current_user
 from peewee import fn
 
-from api.db import FileType
+from api.db import FileType, KNOWLEDGEBASE_FOLDER_NAME, FileSource
 from api.db.db_models import DB, File2Document, Knowledgebase
 from api.db.db_models import File, Document
 from api.db.services.common_service import CommonService
+from api.db.services.document_service import DocumentService
+from api.db.services.file2document_service import File2DocumentService
 from api.utils import get_uuid
 
 
@@ -32,11 +34,16 @@ class FileService(CommonService):
                      orderby, desc, keywords):
         if keywords:
             files = cls.model.select().where(
-                (cls.model.tenant_id == tenant_id)
-                & (cls.model.parent_id == pf_id), (fn.LOWER(cls.model.name).like(f"%%{keywords.lower()}%%")))
+                (cls.model.tenant_id == tenant_id),
+                (cls.model.parent_id == pf_id),
+                (fn.LOWER(cls.model.name).contains(keywords.lower())),
+                ~(cls.model.id == pf_id)
+            )
         else:
-            files = cls.model.select().where((cls.model.tenant_id == tenant_id)
-                                             & (cls.model.parent_id == pf_id))
+            files = cls.model.select().where((cls.model.tenant_id == tenant_id),
+                                             (cls.model.parent_id == pf_id),
+                                             ~(cls.model.id == pf_id)
+                                             )
         count = files.count()
         if desc:
             files = files.order_by(cls.model.getter_by(orderby).desc())
@@ -135,29 +142,69 @@ class FileService(CommonService):
     @classmethod
     @DB.connection_context()
     def get_root_folder(cls, tenant_id):
-        file = cls.model.select().where(cls.model.tenant_id == tenant_id and
-                                        cls.model.parent_id == cls.model.id)
-        if not file:
-            file_id = get_uuid()
-            file = {
-                "id": file_id,
-                "parent_id": file_id,
-                "tenant_id": tenant_id,
-                "created_by": tenant_id,
-                "name": "/",
-                "type": FileType.FOLDER.value,
-                "size": 0,
-                "location": "",
-            }
-            cls.save(**file)
-        else:
-            file_id = file[0].id
+        for file in cls.model.select().where((cls.model.tenant_id == tenant_id),
+                                        (cls.model.parent_id == cls.model.id)
+                                        ):
+            return file.to_dict()
 
-        e, file = cls.get_by_id(file_id)
-        if not e:
-            raise RuntimeError("Database error (File retrieval)!")
+        file_id = get_uuid()
+        file = {
+            "id": file_id,
+            "parent_id": file_id,
+            "tenant_id": tenant_id,
+            "created_by": tenant_id,
+            "name": "/",
+            "type": FileType.FOLDER.value,
+            "size": 0,
+            "location": "",
+        }
+        cls.save(**file)
         return file
 
+    @classmethod
+    @DB.connection_context()
+    def get_kb_folder(cls, tenant_id):
+        for root in cls.model.select().where(cls.model.tenant_id == tenant_id and
+                                             cls.model.parent_id == cls.model.id):
+            for folder in cls.model.select().where(cls.model.tenant_id == tenant_id and
+                                     cls.model.parent_id == root.id and
+                                     cls.model.name == KNOWLEDGEBASE_FOLDER_NAME
+                                     ):
+                return folder.to_dict()
+        assert False, "Can't find the KB folder. Database init error."
+
+    @classmethod
+    @DB.connection_context()
+    def new_a_file_from_kb(cls, tenant_id, name, parent_id, ty=FileType.FOLDER.value, size=0, location=""):
+        for file in cls.query(tenant_id=tenant_id, parent_id=parent_id, name=name):
+            return file.to_dict()
+        file = {
+            "id": get_uuid(),
+            "parent_id": parent_id,
+            "tenant_id": tenant_id,
+            "created_by": tenant_id,
+            "name": name,
+            "type": ty,
+            "size": size,
+            "location": location,
+            "source_type": FileSource.KNOWLEDGEBASE
+        }
+        cls.save(**file)
+        return file
+
+    @classmethod
+    @DB.connection_context()
+    def init_knowledgebase_docs(cls, root_id, tenant_id):
+        for _ in cls.model.select().where((cls.model.name == KNOWLEDGEBASE_FOLDER_NAME)\
+                                          & (cls.model.parent_id == root_id)):
+            return
+        folder = cls.new_a_file_from_kb(tenant_id, KNOWLEDGEBASE_FOLDER_NAME, root_id)
+
+        for kb in Knowledgebase.select(*[Knowledgebase.id, Knowledgebase.name]).where(Knowledgebase.tenant_id==tenant_id):
+            kb_folder = cls.new_a_file_from_kb(tenant_id, kb.name, folder["id"])
+            for doc in DocumentService.query(kb_id=kb.id):
+                FileService.add_file_from_kb(doc.to_dict(), kb_folder["id"], tenant_id)
+
     @classmethod
     @DB.connection_context()
     def get_parent_folder(cls, file_id):
@@ -241,3 +288,20 @@ class FileService(CommonService):
         dfs(folder_id)
         return size
 
+    @classmethod
+    @DB.connection_context()
+    def add_file_from_kb(cls, doc, kb_folder_id, tenant_id):
+        for _ in File2DocumentService.get_by_document_id(doc["id"]): return
+        file = {
+            "id": get_uuid(),
+            "parent_id": kb_folder_id,
+            "tenant_id": tenant_id,
+            "created_by": tenant_id,
+            "name": doc["name"],
+            "type": doc["type"],
+            "size": doc["size"],
+            "location": doc["location"],
+            "source_type": FileSource.KNOWLEDGEBASE
+        }
+        cls.save(**file)
+        File2DocumentService.save(**{"id": get_uuid(), "file_id": file["id"], "document_id": doc["id"]})

api/db/services/kb_service.py

@@ -1,67 +0,0 @@
-#
-#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
-#
-#  Licensed under the Apache License, Version 2.0 (the "License");
-#  you may not use this file except in compliance with the License.
-#  You may obtain a copy of the License at
-#
-#      http://www.apache.org/licenses/LICENSE-2.0
-#
-#  Unless required by applicable law or agreed to in writing, software
-#  distributed under the License is distributed on an "AS IS" BASIS,
-#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#  See the License for the specific language governing permissions and
-#  limitations under the License.
-#
-
-from api.db import TenantPermission
-from api.db.db_models import DB, Tenant
-from api.db.db_models import Knowledgebase
-from api.db.services.common_service import CommonService
-from api.db import StatusEnum
-
-
-class KnowledgebaseService(CommonService):
-    model = Knowledgebase
-
-    @classmethod
-    @DB.connection_context()
-    def get_by_tenant_ids(cls, joined_tenant_ids, user_id,
-                          page_number, items_per_page, orderby, desc):
-        kbs = cls.model.select().where(
-            ((cls.model.tenant_id.in_(joined_tenant_ids) & (cls.model.permission ==
-             TenantPermission.TEAM.value)) | (cls.model.tenant_id == user_id))
-            & (cls.model.status == StatusEnum.VALID.value)
-        )
-        if desc:
-            kbs = kbs.order_by(cls.model.getter_by(orderby).desc())
-        else:
-            kbs = kbs.order_by(cls.model.getter_by(orderby).asc())
-
-        kbs = kbs.paginate(page_number, items_per_page)
-
-        return list(kbs.dicts())
-
-    @classmethod
-    @DB.connection_context()
-    def get_detail(cls, kb_id):
-        fields = [
-            cls.model.id,
-            Tenant.embd_id,
-            cls.model.avatar,
-            cls.model.name,
-            cls.model.description,
-            cls.model.permission,
-            cls.model.doc_num,
-            cls.model.token_num,
-            cls.model.chunk_num,
-            cls.model.parser_id]
-        kbs = cls.model.select(*fields).join(Tenant, on=((Tenant.id == cls.model.tenant_id)&(Tenant.status== StatusEnum.VALID.value))).where(
-            (cls.model.id == kb_id),
-            (cls.model.status == StatusEnum.VALID.value)
-        )
-        if not kbs:
-            return
-        d = kbs[0].to_dict()
-        d["embd_id"] = kbs[0].tenant.embd_id
-        return d

api/db/services/knowledgebase_service.py

@@ -112,3 +112,8 @@ class KnowledgebaseService(CommonService):
         if kb:
             return True, kb[0]
         return False, None
+
+    @classmethod
+    @DB.connection_context()
+    def get_all_ids(cls):
+        return [m["id"] for m in cls.model.select(cls.model.id).dicts()]

api/db/services/llm_service.py

@@ -81,7 +81,7 @@ class TenantLLMService(CommonService):
         if not model_config:
             if llm_type == LLMType.EMBEDDING.value:
                 llm = LLMService.query(llm_name=llm_name)
-                if llm and llm[0].fid in ["Youdao", "FastEmbed"]:
+                if llm and llm[0].fid in ["Youdao", "FastEmbed", "DeepSeek"]:
                     model_config = {"llm_factory": llm[0].fid, "api_key":"", "llm_name": llm_name, "api_base": ""}
             if not model_config:
                 if llm_name == "flag-embedding":
@@ -135,6 +135,16 @@ class TenantLLMService(CommonService):
                 .execute()
         return num
 
+    @classmethod
+    @DB.connection_context()
+    def get_openai_models(cls):
+        objs = cls.model.select().where(
+            (cls.model.llm_factory == "OpenAI"),
+            ~(cls.model.llm_name == "text-embedding-3-small"),
+            ~(cls.model.llm_name == "text-embedding-3-large")
+        ).dicts()
+        return list(objs)
+
 
 class LLMBundle(object):
     def __init__(self, tenant_id, llm_type, llm_name=None, lang="Chinese"):
@@ -172,8 +182,18 @@ class LLMBundle(object):
 
     def chat(self, system, history, gen_conf):
         txt, used_tokens = self.mdl.chat(system, history, gen_conf)
-        if TenantLLMService.increase_usage(
+        if not TenantLLMService.increase_usage(
                 self.tenant_id, self.llm_type, used_tokens, self.llm_name):
             database_logger.error(
                 "Can't update token usage for {}/CHAT".format(self.tenant_id))
         return txt
+
+    def chat_streamly(self, system, history, gen_conf):
+        for txt in self.mdl.chat_streamly(system, history, gen_conf):
+            if isinstance(txt, int):
+                if not TenantLLMService.increase_usage(
+                        self.tenant_id, self.llm_type, txt, self.llm_name):
+                    database_logger.error(
+                        "Can't update token usage for {}/CHAT".format(self.tenant_id))
+                return
+            yield txt

api/db/services/task_service.py

@@ -96,7 +96,7 @@ class TaskService(CommonService):
             return doc.run == TaskStatus.CANCEL.value or doc.progress < 0
         except Exception as e:
             pass
-        return True
+        return False
 
     @classmethod
     @DB.connection_context()
@@ -159,4 +159,4 @@ def queue_tasks(doc, bucket, name):
     DocumentService.begin2parse(doc["id"])
 
     for t in tsks:
-        REDIS_CONN.queue_product(SVR_QUEUE_NAME, message=t)
+        assert REDIS_CONN.queue_product(SVR_QUEUE_NAME, message=t), "Can't access Redis. Please check the Redis' status."

api/settings.py

@@ -86,6 +86,12 @@ default_llm = {
         "embedding_model": "",
         "image2text_model": "",
         "asr_model": "",
+    },
+    "DeepSeek": {
+        "chat_model": "deepseek-chat",
+        "embedding_model": "BAAI/bge-large-zh-v1.5",
+        "image2text_model": "",
+        "asr_model": "",
     }
 }
 LLM = get_base_config("user_default_llm", {})
@@ -152,6 +158,7 @@ CLIENT_AUTHENTICATION = AUTHENTICATION_CONF.get(
         "switch", False)
 HTTP_APP_KEY = AUTHENTICATION_CONF.get("client", {}).get("http_app_key")
 GITHUB_OAUTH = get_base_config("oauth", {}).get("github")
+FEISHU_OAUTH = get_base_config("oauth", {}).get("feishu")
 WECHAT_OAUTH = get_base_config("oauth", {}).get("wechat")
 
 # site

api/utils/api_utils.py

@@ -25,7 +25,6 @@ from flask import (
 from werkzeug.http import HTTP_STATUS_CODES
 
 from api.utils import json_dumps
-from api.versions import get_rag_version
 from api.settings import RetCode
 from api.settings import (
     REQUEST_MAX_WAIT_SEC, REQUEST_WAIT_SEC,
@@ -84,9 +83,6 @@ def request(**kwargs):
     return sess.send(prepped, stream=stream, timeout=timeout)
 
 
-rag_version = get_rag_version() or ''
-
-
 def get_exponential_backoff_interval(retries, full_jitter=False):
     """Calculate the exponential backoff wait time."""
     # Will be zero if factor equals 0

api/utils/file_utils.py

@@ -156,7 +156,7 @@ def filename_type(filename):
         return FileType.PDF.value
 
     if re.match(
-            r".*\.(doc|docx|ppt|pptx|yml|xml|htm|json|csv|txt|ini|xls|xlsx|wps|rtf|hlp|pages|numbers|key|md)$", filename):
+            r".*\.(doc|docx|ppt|pptx|yml|xml|htm|json|csv|txt|ini|xls|xlsx|wps|rtf|hlp|pages|numbers|key|md|py|js|java|c|cpp|h|php|go|ts|sh|cs|kt)$", filename):
         return FileType.DOC.value
 
     if re.match(
@@ -174,7 +174,7 @@ def thumbnail(filename, blob):
     if re.match(r".*\.pdf$", filename):
         pdf = pdfplumber.open(BytesIO(blob))
         buffered = BytesIO()
-        pdf.pages[0].to_image().annotated.save(buffered, format="png")
+        pdf.pages[0].to_image(resolution=32).annotated.save(buffered, format="png")
         return "data:image/png;base64," + \
             base64.b64encode(buffered.getvalue()).decode("utf-8")
 

api/versions.py

@@ -14,17 +14,15 @@
 #  limitations under the License.
 #
 import os
-
 import dotenv
 import typing
-
 from api.utils.file_utils import get_project_base_directory
 
 
 def get_versions() -> typing.Mapping[str, typing.Any]:
-    return dotenv.dotenv_values(
-        dotenv_path=os.path.join(get_project_base_directory(), "rag.env")
-    )
+    dotenv.load_dotenv(dotenv.find_dotenv())
+    return dotenv.dotenv_values()
+
 
 def get_rag_version() -> typing.Optional[str]:
-    return get_versions().get("RAG")
+    return get_versions().get("RAGFLOW_VERSION", "dev")

conf/service_conf.yaml

@@ -28,6 +28,12 @@ oauth:
     client_id: xxxxxxxxxxxxxxxxxxxxxxxxx
     secret_key: xxxxxxxxxxxxxxxxxxxxxxxxxxxx
     url: https://github.com/login/oauth/access_token
+  feishu:
+    app_id: cli_xxxxxxxxxxxxxxxxxxx
+    app_secret: xxxxxxxxxxxxxxxxxxxxxxxxxxxx
+    app_access_token_url: https://open.feishu.cn/open-apis/auth/v3/app_access_token/internal
+    user_access_token_url: https://open.feishu.cn/open-apis/authen/v1/oidc/access_token
+    grant_type: 'authorization_code'
 authentication:
   client:
     switch: false
@@ -38,4 +44,4 @@ authentication:
 permission:
   switch: false
   component: false
-  dataset: false
+  dataset: false

deepdoc/parser/excel_parser.py

@@ -7,30 +7,39 @@ from rag.nlp import find_codec
 
 
 class RAGFlowExcelParser:
-    def html(self, fnm):
+    def html(self, fnm, chunk_rows=256):
         if isinstance(fnm, str):
             wb = load_workbook(fnm)
         else:
             wb = load_workbook(BytesIO(fnm))
-        tb = ""
+
+        tb_chunks = []
         for sheetname in wb.sheetnames:
             ws = wb[sheetname]
             rows = list(ws.rows)
-            if not rows:continue
-            tb += f"<table><caption>{sheetname}</caption><tr>"
+            if not rows: continue
+
+            tb_rows_0 = "<tr>"
             for t in list(rows[0]):
-                tb += f"<th>{t.value}</th>"
-            tb += "</tr>"
-            for r in list(rows[1:]):
-                tb += "<tr>"
-                for i, c in enumerate(r):
-                    if c.value is None:
-                        tb += "<td></td>"
-                    else:
-                        tb += f"<td>{c.value}</td>"
-                tb += "</tr>"
-            tb += "</table>\n"
-        return tb
+                tb_rows_0 += f"<th>{t.value}</th>"
+            tb_rows_0 += "</tr>"
+
+            for chunk_i in range((len(rows) - 1) // chunk_rows + 1):
+                tb = ""
+                tb += f"<table><caption>{sheetname}</caption>"
+                tb += tb_rows_0
+                for r in list(rows[1 + chunk_i * chunk_rows:1 + (chunk_i + 1) * chunk_rows]):
+                    tb += "<tr>"
+                    for i, c in enumerate(r):
+                        if c.value is None:
+                            tb += "<td></td>"
+                        else:
+                            tb += f"<td>{c.value}</td>"
+                    tb += "</tr>"
+                tb += "</table>\n"
+                tb_chunks.append(tb)
+
+        return tb_chunks
 
     def __call__(self, fnm):
         if isinstance(fnm, str):

deepdoc/parser/pdf_parser.py

@@ -749,6 +749,7 @@ class RAGFlowPdfParser:
                             "layoutno", "")))
 
                 left, top, right, bott = b["x0"], b["top"], b["x1"], b["bottom"]
+                if right < left: right = left + 1
                 poss.append((pn + self.page_from, left, right, top, bott))
                 return self.page_images[pn] \
                     .crop((left * ZM, top * ZM,

docker/.env

@@ -29,7 +29,7 @@ REDIS_PASSWORD=infini_rag_flow
 
 SVR_HTTP_PORT=9380
 
-RAGFLOW_VERSION=dev
+RAGFLOW_VERSION=0.6.0
 
 TIMEZONE='Asia/Shanghai'
 

docker/entrypoint.sh

@@ -4,18 +4,20 @@
 
 export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu/
 
-PY=/root/miniconda3/envs/py11/bin/python
+PY=python3
+if [[ -z "$WS" || $WS -lt 1 ]]; then
+  WS=1
+fi
 
 function task_exe(){
     while [ 1 -eq 1 ];do
-      $PY rag/svr/task_executor.py $1 $2;
+      $PY rag/svr/task_executor.py ;
     done
 }
 
-WS=1
 for ((i=0;i<WS;i++))
 do
-  task_exe $i $WS &
+  task_exe  &
 done
 
 while [ 1 -eq 1 ];do

docs/configure_knowledge_base.md

@@ -0,0 +1,132 @@
+# Configure a knowledge base
+
+Knowledge base, hallucination-free chat, and file management are three pillars of RAGFlow. RAGFlow's AI chats are based on knowledge bases. Each of RAGFlow's knowledge bases serves as a knowledge source, *parsing* files uploaded from your local machine and file references generated in **File Management** into the real 'knowledge' for future AI chats. This guide demonstrates some basic usages of the knowledge base feature, covering the following topics:
+
+- Create a knowledge base
+- Configure a knowledge base
+- Search for a knowledge base
+- Delete a knowledge base
+
+## Create knowledge base
+
+With multiple knowledge bases, you can build more flexible, diversified question answering. To create your first knowledge base:
+
+![create knowledge base](https://github.com/infiniflow/ragflow/assets/93570324/110541ed-6cea-4a03-a11c-414a0948ba80)
+
+_Each time a knowledge base is created, a folder with the same name is generated in the **root/.knowledgebase** directory._
+
+## Configure knowledge base
+
+The following screen shot shows the configuration page of a knowledge base. A proper configuration of your knowledge base is crucial for future AI chats. For example, choosing the wrong embedding model or chunk method would cause unexpected semantic loss or mismatched answers in chats. 
+
+![knowledge base configuration](https://github.com/infiniflow/ragflow/assets/93570324/384c671a-8b9c-468c-b1c9-1401128a9b65)
+
+This section covers the following topics: 
+
+- Select chunk method
+- Select embedding model
+- Upload file
+- Parse file
+- Intervene with file parsing results
+- Run retrieval testing
+
+### Select chunk method
+
+RAGFlow offers multiple chunking template to facilitate chunking files of different layouts and ensure semantic integrity. In **Chunk method**, you can choose the default template that suits the layouts and formats of your files. The following table shows the descriptions and the compatible file formats of each supported chunk template:
+
+| **Template** | Description                                                  | File format                                          |
+| ------------ | ------------------------------------------------------------ | ---------------------------------------------------- |
+| General      | Files are consecutively chunked based on a preset chunk token number. | DOCX, EXCEL, PPT, PDF, TXT, JPEG, JPG, PNG, TIF, GIF |
+| Q&A          |                                                              | EXCEL, CSV/TXT                                       |
+| Manual       |                                                              | PDF                                                  |
+| Table        |                                                              | EXCEL, CSV/TXT                                       |
+| Paper        |                                                              | PDF                                                  |
+| Book         |                                                              | DOCX, PDF, TXT                                       |
+| Laws         |                                                              | DOCX, PDF, TXT                                       |
+| Presentation |                                                              | PDF, PPTX                                            |
+| Picture      |                                                              | JPEG, JPG, PNG, TIF, GIF                             |
+| One          | The entire document is chunked as one.                       | DOCX, EXCEL, PDF, TXT                                |
+
+You can also change the chunk template for a particular file on the **Datasets** page. 
+
+![change chunk method](https://github.com/infiniflow/ragflow/assets/93570324/ac116353-2793-42b2-b181-65e7082bed42)
+
+### Select embedding model
+
+An embedding model builds vector index on file chunks. Once you have chosen an embedding model and used it to parse a file, you are no longer allowed to change it. To switch to a different embedding model, you *must* deletes all completed file chunks in the knowledge base. The obvious reason is that we must *ensure* that all files in a specific knowledge base are parsed using the *same* embedding model (ensure that they are compared in the same embedding space). 
+
+The following embedding models can be deployed locally:
+
+- BAAI/bge-base-en-v1.5
+- BAAI/bge-large-en-v1.5
+- BAAI/bge-small-en-v1.5
+- BAAI/bge-small-zh-v1.5
+- jinaai/jina-embeddings-v2-base-en
+- jinaai/jina-embeddings-v2-small-en
+- nomic-ai/nomic-embed-text-v1.5
+- sentence-transformers/all-MiniLM-L6-v2
+- maidalun1020/bce-embedding-base_v1
+
+### Upload file
+
+- RAGFlow's **File Management** allows you to link a file to multiple knowledge bases, in which case each target knowledge base holds a reference to the file.
+- In **Knowledge Base**, you are also given the option of uploading a single file or a folder of files (bulk upload) from your local machine to a knowledge base, in which case the knowledge base holds file copies. 
+
+While uploading files directly to a knowledge base seems more convenient, we *highly* recommend uploading files to **File Management** and then linking them to the target knowledge bases. This way, you can avoid permanently deleting files uploaded to the knowledge base. 
+
+### Parse file
+
+File parsing is a crucial topic in knowledge base configuration. The meaning of file parsing in RAGFlow is twofold: chunking files based on file layout and building embedding and full-text (keyword) indexes on these chunks. After having selected the chunk method and embedding model, you can start parsing an file:
+
+![parse file](https://github.com/infiniflow/ragflow/assets/93570324/5311f166-6426-447f-aa1f-bd488f1cfc7b)
+
+- Click the play button next to **UNSTART** to start file parsing.
+- Click the red-cross icon and then refresh, if your file parsing stalls for a long time. 
+- As shown above, RAGFlow allows you to use a different chunk method for a particular file, offering flexibility beyond the default method. 
+- As shown above, RAGFlow allows you to enable or disable individual files, offering finer control over knowledge base-based AI chats. 
+
+### Intervene with file parsing results
+
+RAGFlow features visibility and explainability, allowing you to view the chunking results and intervene where necessary. To do so: 
+
+1. Click on the file that completes file parsing to view the chunking results: 
+
+   _You are taken to the **Chunk** page:_
+
+   ![chunks](https://github.com/infiniflow/ragflow/assets/93570324/0547fd0e-e71b-41f8-8e0e-31649c85fd3d)
+
+2. Hover over each snapshot for a quick view of each chunk.
+
+3. Double click the chunked texts to add keywords or make *manual* changes where necessary:
+
+   ![update chunk](https://github.com/infiniflow/ragflow/assets/93570324/1d84b408-4e9f-46fd-9413-8c1059bf9c76)
+
+4. In Retrieval testing, ask a quick question in **Test text** to double check if your configurations work:
+
+   _As you can tell from the following, RAGFlow responds with truthful citations._
+
+   ![retrieval test](https://github.com/infiniflow/ragflow/assets/93570324/c03f06f6-f41f-4b20-a97e-ae405d3a950c)
+
+### Run retrieval testing
+
+RAGFlow uses multiple recall of both full-text search and vector search in its chats. Prior to setting up an AI chat, consider adjusting the following parameters to ensure that the intended information always turns up in answers:
+
+- Similarity threshold: Chunks with similarities below the threshold will be filtered. Defaultly set to 0.2.
+- Vector similarity weight: The percentage by which vector similarity contributes to the overall score. Defaultly set to 0.3.
+
+![retrieval test](https://github.com/infiniflow/ragflow/assets/93570324/c03f06f6-f41f-4b20-a97e-ae405d3a950c)
+
+## Search for knowledge base
+
+As of RAGFlow v0.5.0, the search feature is still in a rudimentary form, supporting only knowledge base search by name.
+
+![search knowledge base](https://github.com/infiniflow/ragflow/assets/93570324/836ae94c-2438-42be-879e-c7ad2a59693e)
+
+## Delete knowledge base
+
+You are allowed to delete a knowledge base. Hover your mouse over the three dot of the intended knowledge base card and the **Delete** option appears. Once you delete a knowledge base, the associated folder under **root/.knowledge** directory is AUTOMATICALLY REMOVED. The consequence is: 
+
+- The files uploaded directly to the knowledge base are gone;  
+- The file references, which you created from within **File Management**, are gone, but the associated files still exist in **File Management**. 
+
+![delete knowledge base](https://github.com/infiniflow/ragflow/assets/93570324/fec7a508-6cfe-4bca-af90-81d3fdb94098)

docs/conversation_api.md

@@ -220,8 +220,10 @@ This will be called to get the answer to users' questions.
 | name | type | optional | description|
 |------|-------|----|----|
 | conversation_id| string | No | This is from calling /new_conversation.|
-| messages| json | No | All the conversation history stored here including the latest user's question.|
+| messages| json | No | The latest question, such as `[{"role": "user", "content": "How are you doing!"}]`|
 | quote | bool | Yes | Default: true |
+| stream | bool | Yes | Default: true |
+| doc_ids | string | Yes | Document IDs which is delimited by comma, like `c790da40ea8911ee928e0242ac180005,c790da40ea8911ee928e0242ac180005`. The retrieved content is limited in these documents. |
 
 ### Response 
 ```json
@@ -315,10 +317,12 @@ This is usually used when upload a file to.
 
 ### Parameter:
 
-| name    | type   | optional | description                            |
-|---------|--------|----------|----------------------------------------|
-| file    | file   | No       | Upload file.                           |
-| kb_name | string | No       | Choose the upload knowledge base name. |
+| name      | type   | optional | description                                             |
+|-----------|--------|----------|---------------------------------------------------------|
+| file      | file   | No       | Upload file.                                            |
+| kb_name   | string | No       | Choose the upload knowledge base name.                  |
+| parser_id | string | Yes      | Choose the parsing method.                              |
+| run       | string | Yes      | Parsing will start automatically when the value is "1". |
 
 ### Response 
 ```json
@@ -362,3 +366,38 @@ This is usually used when upload a file to.
 }
 
 ```
+
+## Get document chunks
+
+Get the chunks of the document based on doc_name or doc_id.
+### Path: /api/list_chunks/
+### Method: POST
+
+### Parameter:
+
+| Name     | Type   | Optional | Description                     |
+|----------|--------|----------|---------------------------------|
+| `doc_name` | string | Yes      | The name of the document in the knowledge base. It must not be empty if `doc_id` is not set.|
+| `doc_id`   | string | Yes      | The ID of the document in the knowledge base. It must not be empty if `doc_name` is not set.|
+
+
+### Response 
+```json
+{
+    "data": [
+        {
+            "content": "Figure 14: Per-request neural-net processingof RL-Cache.\n103\n(sn)\nCPU\n 102\nGPU\n8101\n100\n8\n16 64 256 1K\n4K",
+            "doc_name": "RL-Cache.pdf",
+            "img_id": "0335167613f011ef91240242ac120006-b46c3524952f82dbe061ce9b123f2211"
+        },
+        {
+            "content": "4.3 ProcessingOverheadof RL-CacheACKNOWLEDGMENTSThis section evaluates how e￿ectively our RL-Cache implemen-tation leverages modern multi-core CPUs and GPUs to keep the per-request neural-net processing overhead low. Figure 14 depictsThis researchwas supported inpart by the Regional Government of Madrid (grant P2018/TCS-4499, EdgeData-CM)andU.S. National Science Foundation (grants CNS-1763617 andCNS-1717179).REFERENCES",
+            "doc_name": "RL-Cache.pdf",
+            "img_id": "0335167613f011ef91240242ac120006-d4c12c43938eb55d2d8278eea0d7e6d7"
+        }
+    ],
+    "retcode": 0,
+    "retmsg": "success"
+}
+
+```

docs/faq.md

@@ -186,12 +186,14 @@ Parsing requests have to wait in queue due to limited server resources. We are c
 
 If your RAGFlow is deployed *locally*, try the following: 
 
-1. Check the log of your RAGFlow server to see if it is running properly:
-```bash
-docker logs -f ragflow-server
-```
-2. Check if the **task_executor.py** process exists.
-3. Check if your RAGFlow server can access hf-mirror.com or huggingface.com.
+1. Click the red cross icon next to **Parsing Status** and refresh the file parsing process. 
+2. If the issue still persists, try the following: 
+   - check the log of your RAGFlow server to see if it is running properly:
+   ```bash
+   docker logs -f ragflow-server
+   ```
+   - Check if the **task_executor.py** process exists.
+   - Check if your RAGFlow server can access hf-mirror.com or huggingface.com.
 
 #### 4.5 Why does my pdf parsing stall near completion, while the log does not show any error?
 
@@ -264,7 +266,7 @@ This is because you forgot to update the `vm.max_map_count` value in **/etc/sysc
 
 #### 4.11 `{"data":null,"retcode":100,"retmsg":"<NotFound '404: Not Found'>"}`
 
-Your IP address or port number may be incorrect. If you are using the default configurations, enter http://<IP_OF_YOUR_MACHINE> (**NOT 9380, AND NO PORT NUMBER REQUIRED!**) in your browser. This should work.
+Your IP address or port number may be incorrect. If you are using the default configurations, enter `http://<IP_OF_YOUR_MACHINE>` (**NOT 9380, AND NO PORT NUMBER REQUIRED!**) in your browser. This should work.
 
 #### 4.12 `Ollama - Mistral instance running at 127.0.0.1:11434 but cannot add Ollama as model in RagFlow`
 
@@ -367,11 +369,11 @@ You can use Ollama to deploy local LLM. See [here](https://github.com/infiniflow
 2. Right click the desired knowledge base to display the **Configuration** dialogue. 
 3. Choose **Q&A** as the chunk method and click **Save** to confirm your change. 
 
-### 7 Do I need to connect to Redis?
+### 7. Do I need to connect to Redis?
 
 No, connecting to Redis is not required. 
 
-### 8 `Error: Range of input length should be [1, 30000]`
+### 8. `Error: Range of input length should be [1, 30000]`
 
 This error occurs because there are too many chunks matching your search criteria. Try reducing the **TopN** and increasing **Similarity threshold** to fix this issue: 
 
@@ -382,7 +384,15 @@ This error occurs because there are too many chunks matching your search criteri
 
 ![topn](https://github.com/infiniflow/ragflow/assets/93570324/7ec72ab3-0dd2-4cff-af44-e2663b67b2fc)
 
-### 9 How to update RAGFlow to the latest version?
+### 9. How to upgrade RAGFlow?
+   
+You can upgrade RAGFlow to either the dev version or the latest version:
+
+- Dev versions are for developers and contributors. They are published on a nightly basis and may crash because they are not fully tested. We cannot guarantee their validity and you are at your own risk trying out latest, untested features.
+- The latest version refers to the most recent, officially published release. It is stable and works best with regular users.
+
+
+To upgrade RAGFlow to the dev version:
 
 1. Pull the latest source code
    ```bash
@@ -403,3 +413,30 @@ This error occurs because there are too many chunks matching your search criteri
    ```bash
    docker compose -f docker-compose-CN.yml up -d
    ```
+   
+To upgrade RAGFlow to the latest version:
+
+1. Update **ragflow/docker/.env** as follows:
+   ```bash
+   RAGFLOW_VERSION=latest
+   ```
+2. Pull the latest source code:
+   ```bash
+   cd ragflow
+   git pull
+   ```   
+
+3. If you used `docker compose up -d` to start up RAGFlow server:
+   ```bash
+   docker pull infiniflow/ragflow:latest
+   ```
+   ```bash
+   docker compose up ragflow -d
+   ```
+4. If you used `docker compose -f docker-compose-CN.yml up -d` to start up RAGFlow server:
+   ```bash
+   docker pull swr.cn-north-4.myhuaweicloud.com/infiniflow/ragflow:latest
+   ```
+   ```bash
+   docker compose -f docker-compose-CN.yml up -d
+   ```

docs/manage_files.md

@@ -0,0 +1,79 @@
+#  Manage files
+
+Knowledge base, hallucination-free chat, and file management are three pillars of RAGFlow. RAGFlow's file management allows you to upload files individually or in bulk. You can then link an uploaded file to multiple target knowledge bases. This guide showcases some basic usages of the file management feature. 
+
+## Create folder
+
+RAGFlow's file management allows you to establish your file system with nested folder structures. To create a folder in the root directory of RAGFlow: 
+
+![create new folder](https://github.com/infiniflow/ragflow/assets/93570324/3a37a5f4-43a6-426d-a62a-e5cd2ff7a533)
+
+> Each knowledge base in RAGFlow has a corresponding folder under the **root/.knowledgebase** directory. You are not allowed to create a subfolder within it.
+
+## Upload file
+
+RAGFlow's file management supports file uploads from your local machine, allowing both individual and bulk uploads: 
+
+![upload file](https://github.com/infiniflow/ragflow/assets/93570324/5d7ded14-ce2b-4703-8567-9356a978f45c)
+
+![bulk upload](https://github.com/infiniflow/ragflow/assets/93570324/def0db55-824c-4236-b809-a98d8c8674e3)
+
+## Preview file
+
+RAGFlow's file management supports previewing files in the following formats:
+
+- Documents (PDF, DOCS)
+- Tables (XLSX)
+- Pictures (JPEG, JPG, PNG, TIF, GIF)
+
+![preview](https://github.com/infiniflow/ragflow/assets/93570324/2e931362-8bbf-482c-ac86-b68b09d331bc)
+
+## Link file to knowledge bases
+
+RAGFlow's file management allows you to *link* an uploaded file to multiple knowledge bases, creating a file reference in each target knowledge base. Therefore, deleting a file in your file management will AUTOMATICALLY REMOVE all related file references across the knowledge bases. 
+
+![link knowledgebase](https://github.com/infiniflow/ragflow/assets/93570324/6c6b8db4-3269-4e35-9434-6089887e3e3f)
+
+You can link your file to one knowledge base or multiple knowledge bases at one time: 
+
+![link multiple kb](https://github.com/infiniflow/ragflow/assets/93570324/6c508803-fb1f-435d-b688-683066fd7fff)
+
+## Move file to specified folder
+
+As of RAGFlow v0.5.0, this feature is *not* available.
+
+## Search files or folders
+
+As of RAGFlow v0.5.0, the search feature is still in a rudimentary form, supporting only file and folder search in the current directory by name (files or folders in the child directory will not be retrieved).
+
+![search file](https://github.com/infiniflow/ragflow/assets/93570324/77ffc2e5-bd80-4ed1-841f-068e664efffe)
+
+## Rename file or folder
+
+RAGFlow's file management allows you to rename a file or folder:
+
+![rename_file](https://github.com/infiniflow/ragflow/assets/93570324/5abb0704-d9e9-4b43-9ed4-5750ccee011f)
+
+
+## Delete files or folders
+
+RAGFlow's file management allows you to delete files or folders individually or in bulk. 
+
+To delete a file or folder: 
+
+![delete file](https://github.com/infiniflow/ragflow/assets/93570324/85872728-125d-45e9-a0ee-21e9d4cedb8b)
+
+To bulk delete files or folders:
+
+![bulk delete](https://github.com/infiniflow/ragflow/assets/93570324/519b99ab-ec7f-4c8a-8cea-e0b6dcb3cb46)
+
+> - You are not allowed to delete the **root/.knowledgebase** folder. 
+> - Deleting files that have been linked to knowledge bases will AUTOMATICALLY REMOVE all associated file references across the knowledge bases. 
+
+## Download uploaded file
+
+RAGFlow's file management allows you to download an uploaded file:
+
+![download_file](https://github.com/infiniflow/ragflow/assets/93570324/cf3b297f-7d9b-4522-bf5f-4f45743e4ed5)
+
+> As of RAGFlow v0.5.0, bulk download is not supported, nor can you download an entire folder. 

docs/quickstart.md

@@ -0,0 +1,203 @@
+# Quickstart
+
+RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. When integrated with LLMs, it is capable of providing truthful question-answering capabilities, backed by well-founded citations from various complex formatted data.
+
+This quick start guide describes a general process from:
+
+- Starting up a local RAGFlow server, 
+- Creating a knowledge base, 
+- Intervening with file parsing, to 
+- Establishing an AI chat based on your datasets. 
+
+## Prerequisites
+
+- CPU >= 4 cores
+- RAM >= 16 GB
+- Disk >= 50 GB
+- Docker >= 24.0.0 & Docker Compose >= v2.26.1
+
+  > If you have not installed Docker on your local machine (Windows, Mac, or Linux), see [Install Docker Engine](https://docs.docker.com/engine/install/).
+
+## Start up the server
+
+1. Ensure `vm.max_map_count` >= 262144 ([more](./docs/max_map_count.md)):
+
+   > To check the value of `vm.max_map_count`:
+   >
+   > ```bash
+   > $ sysctl vm.max_map_count
+   > ```
+   >
+   > Reset `vm.max_map_count` to a value at least 262144 if it is not.
+   >
+   > ```bash
+   > # In this case, we set it to 262144:
+   > $ sudo sysctl -w vm.max_map_count=262144
+   > ```
+   >
+   > This change will be reset after a system reboot. To ensure your change remains permanent, add or update the `vm.max_map_count` value in **/etc/sysctl.conf** accordingly:
+   >
+   > ```bash
+   > vm.max_map_count=262144
+   > ```
+
+2. Clone the repo:
+
+   ```bash
+   $ git clone https://github.com/infiniflow/ragflow.git
+   ```
+
+3. Build the pre-built Docker images and start up the server:
+
+   > Running the following commands automatically downloads the *dev* version RAGFlow Docker image. To download and run a specified Docker version, update `RAGFLOW_VERSION` in **docker/.env** to the intended version, for example `RAGFLOW_VERSION=v0.6.0`, before running the following commands.
+
+   ```bash
+   $ cd ragflow/docker
+   $ chmod +x ./entrypoint.sh
+   $ docker compose up -d
+   ```
+
+> The core image is about 9 GB in size and may take a while to load.
+
+4. Check the server status after having the server up and running:
+
+   ```bash
+   $ docker logs -f ragflow-server
+   ```
+
+   _The following output confirms a successful launch of the system:_
+
+   ```bash
+       ____                 ______ __
+      / __ \ ____ _ ____ _ / ____// /____  _      __
+     / /_/ // __ `// __ `// /_   / // __ \| | /| / /
+    / _, _// /_/ // /_/ // __/  / // /_/ /| |/ |/ /
+   /_/ |_| \__,_/ \__, //_/    /_/ \____/ |__/|__/
+                 /____/
+   
+    * Running on all addresses (0.0.0.0)
+    * Running on http://127.0.0.1:9380
+    * Running on http://x.x.x.x:9380
+    INFO:werkzeug:Press CTRL+C to quit
+   ```
+
+   > If you skip this confirmation step and directly log in to RAGFlow, your browser may prompt a `network anomaly` error because, at that moment, your RAGFlow may not be fully initialized.  
+
+5. In your web browser, enter the IP address of your server and log in to RAGFlow.
+
+   > - With default settings, you only need to enter `http://IP_OF_YOUR_MACHINE` (**sans** port number) as the default HTTP serving port `80` can be omitted when using the default configurations.
+
+## Configure LLMs
+
+RAGFlow is a RAG engine, and it needs to work with an LLM to offer grounded, hallucination-free question-answering capabilities. For now, RAGFlow supports the following LLMs, and the list is expanding:
+
+- OpenAI
+- Tongyi-Qianwen
+- Moonshot
+- DeepSeek-V2
+
+>  RAGFlow also supports deploying LLMs locally using Ollama or Xinference, but this part is not covered in this quick start guide. 
+
+To add and configure an LLM: 
+
+1. Click on your logo on the top right of the page **>** **Model Providers**:
+
+   ![2 add llm](https://github.com/infiniflow/ragflow/assets/93570324/10635088-028b-4b3d-add9-5c5a6e626814)
+
+   > Each RAGFlow account is able to use **text-embedding-v2** for free, a embedding model of Tongyi-Qianwen. This is why you can see Tongyi-Qianwen in the **Added models** list. And you may need to update your Tongyi-Qianwen API key at a later point.
+
+2. Click on the desired LLM and update the API key accordingly (DeepSeek-V2 in this case):
+
+   ![update api key](https://github.com/infiniflow/ragflow/assets/93570324/4e5e13ef-a98d-42e6-bcb1-0c6045fc1666)
+
+   *Your added models appear as follows:* 
+
+   ![added available models](https://github.com/infiniflow/ragflow/assets/93570324/d08b80e4-f921-480a-b41d-11832489c916)
+
+3. Click **System Model Settings** to select the default models: 
+
+   - Chat model, 
+   - Embedding model, 
+   - Image-to-text model. 
+
+   ![system model settings](https://github.com/infiniflow/ragflow/assets/93570324/cdcc1da5-4494-44cd-ad5b-1222ed6acc3f)
+
+> Some of the models, such as the image-to-text model **qwen-vl-max**, are subsidiary to a particular LLM. And you may need to update your API key accordingly to use these models. 
+
+## Create your first knowledge base
+
+You are allowed to upload files to a knowledge base in RAGFlow and parse them into datasets. A knowledge base is virtually a collection of datasets. Question answering in RAGFlow can be based on a particular knowledge base or multiple knowledge bases. File formats that RAGFlow supports include documents (PDF, DOC, DOCX, TXT, MD), tables (CSV, XLSX, XLS), pictures (JPEG, JPG, PNG, TIF, GIF), and slides (PPT, PPTX).
+
+To create your first knowledge base:
+
+1. Click the **Knowledge Base** tab in the top middle of the page **>** **Create knowledge base**.
+
+2. Input the name of your knowledge base and click **OK** to confirm your changes.
+
+   _You are taken to the **Configuration** page of your knowledge base._
+
+   ![knowledge base configuration](https://github.com/infiniflow/ragflow/assets/93570324/384c671a-8b9c-468c-b1c9-1401128a9b65)
+
+3. RAGFlow offers multiple chunk templates that cater to different document layouts and file formats. Select the embedding model and chunk method (template) for your knowledge base. 
+
+   > IMPORTANT: Once you have selected an embedding model and used it to parse a file, you are no longer allowed to change it. The obvious reason is that we must ensure that all files in a specific knowledge base are parsed using the *same* embedding model (ensure that they are being compared in the same embedding space). 
+
+   _You are taken to the **Dataset** page of your knowledge base._
+
+4. Click **+ Add file** **>** **Local files** to start uploading a particular file to the knowledge base. 
+
+5. In the uploaded file entry, click the play button to start file parsing:
+
+   ![file parsing](https://github.com/infiniflow/ragflow/assets/93570324/19f273fa-0ab0-435e-bdf4-a47fb080a078)
+
+   _When the file parsing completes, its parsing status changes to **SUCCESS**._
+
+## Intervene with file parsing 
+
+RAGFlow features visibility and explainability, allowing you to view the chunking results and intervene where necessary. To do so: 
+
+1. Click on the file that completes file parsing to view the chunking results: 
+
+   _You are taken to the **Chunk** page:_
+
+   ![chunks](https://github.com/infiniflow/ragflow/assets/93570324/0547fd0e-e71b-41f8-8e0e-31649c85fd3d)
+
+2. Hover over each snapshot for a quick view of each chunk.
+
+3. Double click the chunked texts to add keywords or make *manual* changes where necessary:
+
+   ![update chunk](https://github.com/infiniflow/ragflow/assets/93570324/1d84b408-4e9f-46fd-9413-8c1059bf9c76)
+
+4. In Retrieval testing, ask a quick question in **Test text** to double check if your configurations work:
+
+   _As you can tell from the following, RAGFlow responds with truthful citations._
+
+   ![retrieval test](https://github.com/infiniflow/ragflow/assets/93570324/c03f06f6-f41f-4b20-a97e-ae405d3a950c)
+
+## Set up an AI chat
+
+Conversations in RAGFlow are based on a particular knowledge base or multiple knowledge bases. Once you have created your knowledge base and finished file parsing, you can go ahead and start an AI conversation. 
+
+1. Click the **Chat** tab in the middle top of the mage **>** **Create an assistant** to show the **Chat Configuration** dialogue *of your next dialogue*.
+   > RAGFlow offer the flexibility of choosing a different chat model for each dialogue, while allowing you to set the default models in **System Model Settings**.
+
+2. Update **Assistant Setting**: 
+
+   - Name your assistant and specify your knowledge bases.
+   - **Empty response**:
+     - If you wish to *confine* RAGFlow's answers to your knowledge bases, leave a response here. Then when it doesn't retrieve an answer, it *uniformly* responds with what you set here. 
+     - If you wish RAGFlow to *improvise* when it doesn't retrieve an answer from your knowledge bases, leave it blank, which may give rise to hallucinations. 
+
+3. Update **Prompt Engine** or leave it as is for the beginning.
+
+4. Update **Model Setting**.
+
+5. RAGFlow also offers conversation APIs. Hover over your dialogue **>** **Chat Bot API** to integrate RAGFlow's chat capabilities into your applications:
+
+   ![chatbot api](https://github.com/infiniflow/ragflow/assets/93570324/fec23715-f9af-4ac2-81e5-942c5035c5e6)
+
+6. Now, let's start the show:
+
+   ![question1](https://github.com/infiniflow/ragflow/assets/93570324/bb72dd67-b35e-4b2a-87e9-4e4edbd6e677)
+
+   ![question2](https://github.com/infiniflow/ragflow/assets/93570324/7cc585ae-88d0-4aa2-817d-0370b2ad7230)

docs/start_chat.md

@@ -0,0 +1,54 @@
+# Start an AI chat
+
+Knowledge base, hallucination-free chat, and file management are three pillars of RAGFlow. Chats in RAGFlow are based on a particular knowledge base or multiple knowledge bases. Once you have created your knowledge base and finished file parsing, you can go ahead and start an AI conversation. 
+
+## Start an AI chat
+
+You start an AI conversation by creating an assistant. 
+
+1. Click the **Chat** tab in the middle top of the page **>** **Create an assistant** to show the **Chat Configuration** dialogue *of your next dialogue*.
+
+   > RAGFlow offers you the flexibility of choosing a different chat model for each dialogue, while allowing you to set the default models in **System Model Settings**.
+
+2. Update **Assistant Setting**: 
+
+   - **Assistant name** is the name of your chat assistant. Each assistant corresponds to a dialogue with a unique combination of knowledge bases, prompts, hybrid search configurations, and large model settings. 
+   - **Empty response**:
+     - If you wish to *confine* RAGFlow's answers to your knowledge bases, leave a response here. Then when it doesn't retrieve an answer, it *uniformly* responds with what you set here. 
+     - If you wish RAGFlow to *improvise* when it doesn't retrieve an answer from your knowledge bases, leave it blank, which may give rise to hallucinations. 
+   - **Show Quote**: This is a key feature of RAGFlow and enabled by default. RAGFlow does not work like a black box. instead, it clearly shows the sources of information that its responses are based on. 
+   -  Select the corresponding knowledge bases. You can select one or multiple knowledge bases, but ensure that they use the same embedding model, otherwise an error would occur. 
+
+3. Update **Prompt Engine**: 
+
+   - In **System**, you fill in the prompts for your LLM, you can also leave the default prompt as-is for the beginning. 
+   - **Similarity threshold** sets the similarity "bar" for each chunk of text. The default is 0.2. Text chunks with lower similarity scores are filtered out of the final response. 
+   - **Vector similarity weight** is set to 0.3 by default. RAGFlow uses a hybrid score system, combining keyword similarity and vector similarity, for evaluating the relevance of different text chunks. This value sets the weight assigned to the vector similarity component in the hybrid score. 
+   - **Top N** determines the *maximum* number of chunks to feed to the LLM. In other words, even if more chunks are retrieved, only the top N chunks are provided as input. 
+   - **Variable**: 
+
+4. Update **Model Setting**:
+
+   - In **Model**: you select the chat model. Though you have selected the default chat model in **System Model Settings**, RAGFlow allows you to choose an alternative chat model for your dialogue. 
+   - **Freedom** refers to the level that the LLM improvises. From **Improvise**, **Precise**, to **Balance**, each freedom level corresponds to a unique combination of **Temperature**, **Top P**, **Presence Penalty**, and **Frequency Penalty**.
+   - **Temperature**: Level of the prediction randomness of the LLM. The higher the value, the more creative the LLM is. 
+   - **Top P** is also known as "nucleus sampling". See [here](https://en.wikipedia.org/wiki/Top-p_sampling) for more information. 
+   - **Max Tokens**: The maximum length of the LLM's responses. Note that the responses may be curtailed if this value is set too low. 
+
+5. Now, let's start the show:
+
+   ![question1](https://github.com/infiniflow/ragflow/assets/93570324/bb72dd67-b35e-4b2a-87e9-4e4edbd6e677)
+
+   ![question2](https://github.com/infiniflow/ragflow/assets/93570324/7cc585ae-88d0-4aa2-817d-0370b2ad7230)
+
+## Update settings of an existing dialogue
+
+Hover over an intended dialogue **>** **Edit** to show the chat configuration dialogue:
+
+![update chat configuration](https://github.com/infiniflow/ragflow/assets/93570324/e08397c7-2a4c-44e1-9032-13d30e99d741)
+
+## Integrate chat capabilities into your application
+
+RAGFlow also offers conversation APIs. Hover over your dialogue **>** **Chat Bot API** to integrate RAGFlow's chat capabilities into your application:
+
+![chatbot api](https://github.com/infiniflow/ragflow/assets/93570324/fec23715-f9af-4ac2-81e5-942c5035c5e6)

rag/app/naive.py

@@ -134,9 +134,9 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
     elif re.search(r"\.xlsx?$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")
         excel_parser = ExcelParser()
-        sections = [(excel_parser.html(binary), "")]
+        sections = [(l, "") for l in excel_parser.html(binary) if l]
 
-    elif re.search(r"\.(txt|md)$", filename, re.IGNORECASE):
+    elif re.search(r"\.(txt|md|py|js|java|c|cpp|h|php|go|ts|sh|cs|kt)$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")
         txt = ""
         if binary:

rag/app/one.py

@@ -78,7 +78,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
     elif re.search(r"\.xlsx?$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")
         excel_parser = ExcelParser()
-        sections = [excel_parser.html(binary)]
+        sections = excel_parser.html(binary, 1000000000)
 
     elif re.search(r"\.txt$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")

rag/llm/__init__.py

@@ -25,7 +25,8 @@ EmbeddingModel = {
     "Tongyi-Qianwen": DefaultEmbedding, #QWenEmbed,
     "ZHIPU-AI": ZhipuEmbed,
     "FastEmbed": FastEmbed,
-    "Youdao": YoudaoEmbed
+    "Youdao": YoudaoEmbed,
+    "DeepSeek": DefaultEmbedding
 }
 
 

rag/llm/chat_model.py

@@ -44,6 +44,31 @@ class Base(ABC):
         except openai.APIError as e:
             return "**ERROR**: " + str(e), 0
 
+    def chat_streamly(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        ans = ""
+        total_tokens = 0
+        try:
+            response = self.client.chat.completions.create(
+                model=self.model_name,
+                messages=history,
+                stream=True,
+                **gen_conf)
+            for resp in response:
+                if not resp.choices[0].delta.content:continue
+                ans += resp.choices[0].delta.content
+                total_tokens += 1
+                if resp.choices[0].finish_reason == "length":
+                    ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
+                        [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                yield ans
+
+        except openai.APIError as e:
+            yield ans + "\n**ERROR**: " + str(e)
+
+        yield total_tokens
+
 
 class GptTurbo(Base):
     def __init__(self, key, model_name="gpt-3.5-turbo", base_url="https://api.openai.com/v1"):
@@ -97,6 +122,35 @@ class QWenChat(Base):
 
         return "**ERROR**: " + response.message, tk_count
 
+    def chat_streamly(self, system, history, gen_conf):
+        from http import HTTPStatus
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        ans = ""
+        try:
+            response = Generation.call(
+                self.model_name,
+                messages=history,
+                result_format='message',
+                stream=True,
+                **gen_conf
+            )
+            tk_count = 0
+            for resp in response:
+                if resp.status_code == HTTPStatus.OK:
+                    ans = resp.output.choices[0]['message']['content']
+                    tk_count = resp.usage.total_tokens
+                    if resp.output.choices[0].get("finish_reason", "") == "length":
+                        ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
+                            [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                    yield ans
+                else:
+                    yield ans + "\n**ERROR**: " + resp.message if str(resp.message).find("Access")<0 else "Out of credit. Please set the API key in **settings > Model providers.**"
+        except Exception as e:
+            yield ans + "\n**ERROR**: " + str(e)
+
+        yield tk_count
+
 
 class ZhipuChat(Base):
     def __init__(self, key, model_name="glm-3-turbo", **kwargs):
@@ -122,6 +176,35 @@ class ZhipuChat(Base):
         except Exception as e:
             return "**ERROR**: " + str(e), 0
 
+    def chat_streamly(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        if "presence_penalty" in gen_conf: del gen_conf["presence_penalty"]
+        if "frequency_penalty" in gen_conf: del gen_conf["frequency_penalty"]
+        ans = ""
+        try:
+            response = self.client.chat.completions.create(
+                model=self.model_name,
+                messages=history,
+                stream=True,
+                **gen_conf
+            )
+            tk_count = 0
+            for resp in response:
+                if not resp.choices[0].delta.content:continue
+                delta = resp.choices[0].delta.content
+                ans += delta
+                if resp.choices[0].finish_reason == "length":
+                    ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
+                        [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                    tk_count = resp.usage.total_tokens
+                if resp.choices[0].finish_reason == "stop": tk_count = resp.usage.total_tokens
+                yield ans
+        except Exception as e:
+            yield ans + "\n**ERROR**: " + str(e)
+
+        yield tk_count
+
 
 class OllamaChat(Base):
     def __init__(self, key, model_name, **kwargs):
@@ -148,3 +231,86 @@ class OllamaChat(Base):
         except Exception as e:
             return "**ERROR**: " + str(e), 0
 
+    def chat_streamly(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        options = {}
+        if "temperature" in gen_conf: options["temperature"] = gen_conf["temperature"]
+        if "max_tokens" in gen_conf: options["num_predict"] = gen_conf["max_tokens"]
+        if "top_p" in gen_conf: options["top_k"] = gen_conf["top_p"]
+        if "presence_penalty" in gen_conf: options["presence_penalty"] = gen_conf["presence_penalty"]
+        if "frequency_penalty" in gen_conf: options["frequency_penalty"] = gen_conf["frequency_penalty"]
+        ans = ""
+        try:
+            response = self.client.chat(
+                model=self.model_name,
+                messages=history,
+                stream=True,
+                options=options
+            )
+            for resp in response:
+                if resp["done"]:
+                    yield resp.get("prompt_eval_count", 0) + resp.get("eval_count", 0)
+                ans += resp["message"]["content"]
+                yield ans
+        except Exception as e:
+            yield ans + "\n**ERROR**: " + str(e)
+        yield 0
+
+
+class LocalLLM(Base):
+    class RPCProxy:
+        def __init__(self, host, port):
+            self.host = host
+            self.port = int(port)
+            self.__conn()
+
+        def __conn(self):
+            from multiprocessing.connection import Client
+            self._connection = Client(
+                (self.host, self.port), authkey=b'infiniflow-token4kevinhu')
+
+        def __getattr__(self, name):
+            import pickle
+
+            def do_rpc(*args, **kwargs):
+                for _ in range(3):
+                    try:
+                        self._connection.send(
+                            pickle.dumps((name, args, kwargs)))
+                        return pickle.loads(self._connection.recv())
+                    except Exception as e:
+                        self.__conn()
+                raise Exception("RPC connection lost!")
+
+            return do_rpc
+
+    def __init__(self, key, model_name="glm-3-turbo"):
+        self.client = LocalLLM.RPCProxy("127.0.0.1", 7860)
+
+    def chat(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        try:
+            ans = self.client.chat(
+                history,
+                gen_conf
+            )
+            return ans, num_tokens_from_string(ans)
+        except Exception as e:
+            return "**ERROR**: " + str(e), 0
+
+    def chat_streamly(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        token_count = 0
+        answer = ""
+        try:
+            for ans in self.client.chat_streamly(history, gen_conf):
+                answer += ans
+                token_count += 1
+                yield answer
+        except Exception as e:
+            yield answer + "\n**ERROR**: " + str(e)
+
+        yield token_count

rag/llm/embedding_model.py

@@ -27,8 +27,7 @@ import torch
 import numpy as np
 
 from api.utils.file_utils import get_project_base_directory, get_home_cache_dir
-from rag.utils import num_tokens_from_string
-
+from rag.utils import num_tokens_from_string, truncate
 
 try:
     flag_model = FlagModel(os.path.join(get_home_cache_dir(), "bge-large-zh-v1.5"),
@@ -70,7 +69,7 @@ class DefaultEmbedding(Base):
         self.model = flag_model
 
     def encode(self, texts: list, batch_size=32):
-        texts = [t[:2000] for t in texts]
+        texts = [truncate(t, 2048) for t in texts]
         token_count = 0
         for t in texts:
             token_count += num_tokens_from_string(t)
@@ -93,12 +92,14 @@ class OpenAIEmbed(Base):
         self.model_name = model_name
 
     def encode(self, texts: list, batch_size=32):
+        texts = [truncate(t, 8196) for t in texts]
         res = self.client.embeddings.create(input=texts,
                                             model=self.model_name)
-        return np.array([d.embedding for d in res.data]), res.usage.total_tokens
+        return np.array([d.embedding for d in res.data]
+                        ), res.usage.total_tokens
 
     def encode_queries(self, text):
-        res = self.client.embeddings.create(input=[text],
+        res = self.client.embeddings.create(input=[truncate(text, 8196)],
                                             model=self.model_name)
         return np.array(res.data[0].embedding), res.usage.total_tokens
 
@@ -112,7 +113,7 @@ class QWenEmbed(Base):
         import dashscope
         res = []
         token_count = 0
-        texts = [txt[:2048] for txt in texts]
+        texts = [truncate(t, 2048) for t in texts]
         for i in range(0, len(texts), batch_size):
             resp = dashscope.TextEmbedding.call(
                 model=self.model_name,

rag/llm/rpc_server.py

@@ -2,9 +2,10 @@ import argparse
 import pickle
 import random
 import time
+from copy import deepcopy
 from multiprocessing.connection import Listener
 from threading import Thread
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
 
 
 def torch_gc():
@@ -95,6 +96,32 @@ def chat(messages, gen_conf):
         return str(e)
 
 
+def chat_streamly(messages, gen_conf):
+    global tokenizer
+    model = Model()
+    try:
+        torch_gc()
+        conf = deepcopy(gen_conf)
+        print(messages, conf)
+        text = tokenizer.apply_chat_template(
+            messages,
+            tokenize=False,
+            add_generation_prompt=True
+        )
+        model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+        streamer = TextStreamer(tokenizer)
+        conf["inputs"] = model_inputs.input_ids
+        conf["streamer"] = streamer
+        conf["max_new_tokens"] = conf["max_tokens"]
+        del conf["max_tokens"]
+        thread = Thread(target=model.generate, kwargs=conf)
+        thread.start()
+        for _, new_text in enumerate(streamer):
+            yield new_text
+    except Exception as e:
+        yield "**ERROR**: " + str(e)
+
+
 def Model():
     global models
     random.seed(time.time())
@@ -113,6 +140,7 @@ if __name__ == "__main__":
 
     handler = RPCHandler()
     handler.register_function(chat)
+    handler.register_function(chat_streamly)
 
     models = []
     for _ in range(1):

rag/nlp/query.py

@@ -36,7 +36,7 @@ class EsQueryer:
         patts = [
             (r"是*(什么样的|哪家|一下|那家|啥样|咋样了|什么时候|何时|何地|何人|是否|是不是|多少|哪里|怎么|哪儿|怎么样|如何|哪些|是啥|啥是|啊|吗|呢|吧|咋|什么|有没有|呀)是*", ""),
             (r"(^| )(what|who|how|which|where|why)('re|'s)? ", " "),
-            (r"(^| )('s|'re|is|are|were|was|do|does|did|don't|doesn't|didn't|has|have|be|there|you|me|your|my|mine|just|please|may|i|should|would|wouldn't|will|won't|done|go|for|with|so|the|a|an|by|i'm|it's|he's|she's|they|they're|you're|as|by|on|in|at|up|out|down)", " ")
+            (r"(^| )('s|'re|is|are|were|was|do|does|did|don't|doesn't|didn't|has|have|be|there|you|me|your|my|mine|just|please|may|i|should|would|wouldn't|will|won't|done|go|for|with|so|the|a|an|by|i'm|it's|he's|she's|they|they're|you're|as|by|on|in|at|up|out|down) ", " ")
         ]
         for r, p in patts:
             txt = re.sub(r, p, txt, flags=re.IGNORECASE)
@@ -44,7 +44,7 @@ class EsQueryer:
 
     def question(self, txt, tbl="qa", min_match="60%"):
         txt = re.sub(
-            r"[ \r\n\t,，。？?/`!！&]+",
+            r"[ \r\n\t,，。？?/`!！&\^%%]+",
             " ",
             rag_tokenizer.tradi2simp(
                 rag_tokenizer.strQ2B(
@@ -53,9 +53,10 @@ class EsQueryer:
 
         if not self.isChinese(txt):
             tks = rag_tokenizer.tokenize(txt).split(" ")
-            q = copy.deepcopy(tks)
-            for i in range(1, len(tks)):
-                q.append("\"%s %s\"^2" % (tks[i - 1], tks[i]))
+            tks_w = self.tw.weights(tks)
+            q = [re.sub(r"[ \\\"']+", "", tk)+"^{:.4f}".format(w) for tk, w in tks_w]
+            for i in range(1, len(tks_w)):
+                q.append("\"%s %s\"^%.4f" % (tks_w[i - 1][0], tks_w[i][0], max(tks_w[i - 1][1], tks_w[i][1])*2))
             if not q:
                 q.append(txt)
             return Q("bool",

rag/nlp/search.py

@@ -52,16 +52,21 @@ class Dealer:
     def search(self, req, idxnm, emb_mdl=None):
         qst = req.get("question", "")
         bqry, keywords = self.qryr.question(qst)
-        if req.get("kb_ids"):
-            bqry.filter.append(Q("terms", kb_id=req["kb_ids"]))
-        if req.get("doc_ids"):
-            bqry.filter.append(Q("terms", doc_id=req["doc_ids"]))
-        if "available_int" in req:
-            if req["available_int"] == 0:
-                bqry.filter.append(Q("range", available_int={"lt": 1}))
-            else:
-                bqry.filter.append(
-                    Q("bool", must_not=Q("range", available_int={"lt": 1})))
+        def add_filters(bqry):
+            nonlocal req
+            if req.get("kb_ids"):
+                bqry.filter.append(Q("terms", kb_id=req["kb_ids"]))
+            if req.get("doc_ids"):
+                bqry.filter.append(Q("terms", doc_id=req["doc_ids"]))
+            if "available_int" in req:
+                if req["available_int"] == 0:
+                    bqry.filter.append(Q("range", available_int={"lt": 1}))
+                else:
+                    bqry.filter.append(
+                        Q("bool", must_not=Q("range", available_int={"lt": 1})))
+            return bqry
+
+        bqry = add_filters(bqry)
         bqry.boost = 0.05
 
         s = Search()
@@ -117,8 +122,7 @@ class Dealer:
         es_logger.info("TOTAL: {}".format(self.es.getTotal(res)))
         if self.es.getTotal(res) == 0 and "knn" in s:
             bqry, _ = self.qryr.question(qst, min_match="10%")
-            if req.get("kb_ids"):
-                bqry.filter.append(Q("terms", kb_id=req["kb_ids"]))
+            bqry = add_filters(bqry)
             s["query"] = bqry.to_dict()
             s["knn"]["filter"] = bqry.to_dict()
             s["knn"]["similarity"] = 0.17

rag/svr/task_executor.py

@@ -80,7 +80,7 @@ def set_progress(task_id, from_page=0, to_page=-1,
 
     if to_page > 0:
         if msg:
-            msg = f"Page({from_page+1}~{to_page+1}): " + msg
+            msg = f"Page({from_page + 1}~{to_page + 1}): " + msg
     d = {"progress_msg": msg}
     if prog is not None:
         d["progress"] = prog
@@ -109,6 +109,7 @@ def collect():
     if not msg: return pd.DataFrame()
 
     if TaskService.do_cancel(msg["id"]):
+        cron_logger.info("Task {} has been canceled.".format(msg["id"]))
         return pd.DataFrame()
     tasks = TaskService.get_tasks(msg["id"])
     assert tasks, "{} empty task!".format(msg["id"])
@@ -123,7 +124,7 @@ def get_minio_binary(bucket, name):
 def build(row):
     if row["size"] > DOC_MAXIMUM_SIZE:
         set_progress(row["id"], prog=-1, msg="File size exceeds( <= %dMb )" %
-                     (int(DOC_MAXIMUM_SIZE / 1024 / 1024)))
+                                             (int(DOC_MAXIMUM_SIZE / 1024 / 1024)))
         return []
 
     callback = partial(
@@ -137,12 +138,12 @@ def build(row):
         bucket, name = File2DocumentService.get_minio_address(doc_id=row["doc_id"])
         binary = get_minio_binary(bucket, name)
         cron_logger.info(
-            "From minio({}) {}/{}".format(timer()-st, row["location"], row["name"]))
+            "From minio({}) {}/{}".format(timer() - st, row["location"], row["name"]))
         cks = chunker.chunk(row["name"], binary=binary, from_page=row["from_page"],
                             to_page=row["to_page"], lang=row["language"], callback=callback,
                             kb_id=row["kb_id"], parser_config=row["parser_config"], tenant_id=row["tenant_id"])
         cron_logger.info(
-            "Chunkking({}) {}/{}".format(timer()-st, row["location"], row["name"]))
+            "Chunkking({}) {}/{}".format(timer() - st, row["location"], row["name"]))
     except TimeoutError as e:
         callback(-1, f"Internal server error: Fetch file timeout. Could you try it again.")
         cron_logger.error(
@@ -172,7 +173,7 @@ def build(row):
         d.update(ck)
         md5 = hashlib.md5()
         md5.update((ck["content_with_weight"] +
-                   str(d["doc_id"])).encode("utf-8"))
+                    str(d["doc_id"])).encode("utf-8"))
         d["_id"] = md5.hexdigest()
         d["create_time"] = str(datetime.datetime.now()).replace("T", " ")[:19]
         d["create_timestamp_flt"] = datetime.datetime.now().timestamp()
@@ -254,13 +255,13 @@ def main():
         try:
             embd_mdl = LLMBundle(r["tenant_id"], LLMType.EMBEDDING, llm_name=r["embd_id"], lang=r["language"])
         except Exception as e:
-            traceback.print_stack(e)
-            callback(prog=-1, msg=str(e))
+            callback(-1, msg=str(e))
+            cron_logger.error(str(e))
             continue
 
         st = timer()
         cks = build(r)
-        cron_logger.info("Build chunks({}): {}".format(r["name"], timer()-st))
+        cron_logger.info("Build chunks({}): {}".format(r["name"], timer() - st))
         if cks is None:
             continue
         if not cks:
@@ -270,7 +271,7 @@ def main():
         ## set_progress(r["did"], -1, "ERROR: ")
         callback(
             msg="Finished slicing files(%d). Start to embedding the content." %
-            len(cks))
+                len(cks))
         st = timer()
         try:
             tk_count = embedding(cks, embd_mdl, r["parser_config"], callback)
@@ -278,14 +279,19 @@ def main():
             callback(-1, "Embedding error:{}".format(str(e)))
             cron_logger.error(str(e))
             tk_count = 0
-        cron_logger.info("Embedding elapsed({}): {}".format(r["name"], timer()-st))
+        cron_logger.info("Embedding elapsed({}): {:.2f}".format(r["name"], timer() - st))
 
-        callback(msg="Finished embedding({})! Start to build index!".format(timer()-st))
+        callback(msg="Finished embedding({:.2f})! Start to build index!".format(timer() - st))
         init_kb(r)
         chunk_count = len(set([c["_id"] for c in cks]))
         st = timer()
-        es_r = ELASTICSEARCH.bulk(cks, search.index_name(r["tenant_id"]))
-        cron_logger.info("Indexing elapsed({}): {}".format(r["name"], timer()-st))
+        es_r = ""
+        for b in range(0, len(cks), 32):
+            es_r = ELASTICSEARCH.bulk(cks[b:b + 32], search.index_name(r["tenant_id"]))
+            if b % 128 == 0:
+                callback(prog=0.8 + 0.1 * (b + 1) / len(cks), msg="")
+
+        cron_logger.info("Indexing elapsed({}): {:.2f}".format(r["name"], timer() - st))
         if es_r:
             callback(-1, "Index failure!")
             ELASTICSEARCH.deleteByQuery(
@@ -300,9 +306,8 @@ def main():
             DocumentService.increment_chunk_num(
                 r["doc_id"], r["kb_id"], tk_count, chunk_count, 0)
             cron_logger.info(
-                "Chunk doc({}), token({}), chunks({}), elapsed:{}".format(
-                    r["id"], tk_count, len(cks), timer()-st))
-
+                "Chunk doc({}), token({}), chunks({}), elapsed:{:.2f}".format(
+                    r["id"], tk_count, len(cks), timer() - st))
 
 
 if __name__ == "__main__":

rag/utils/__init__.py

@@ -63,3 +63,7 @@ def num_tokens_from_string(string: str) -> int:
     num_tokens = len(encoder.encode(string))
     return num_tokens
 
+
+def truncate(string: str, max_len: int) -> int:
+    """Returns truncated text if the length of text exceed max_len."""
+    return encoder.decode(encoder.encode(string)[:max_len])

rag/utils/es_conn.py

@@ -43,6 +43,9 @@ class ESConnection:
         v = v["number"].split(".")[0]
         return int(v) >= 7
 
+    def health(self):
+        return dict(self.es.cluster.health())
+
     def upsert(self, df, idxnm=""):
         res = []
         for d in df:

rag/utils/minio_conn.py

@@ -34,6 +34,16 @@ class RAGFlowMinio(object):
         del self.conn
         self.conn = None
 
+    def health(self):
+        bucket, fnm, binary = "txtxtxtxt1", "txtxtxtxt1", b"_t@@@1"
+        if not self.conn.bucket_exists(bucket):
+            self.conn.make_bucket(bucket)
+        r = self.conn.put_object(bucket, fnm,
+                                 BytesIO(binary),
+                                 len(binary)
+                                 )
+        return r
+
     def put(self, bucket, fnm, binary):
         for _ in range(3):
             try:

rag/utils/redis_conn.py

@@ -44,6 +44,10 @@ class RedisDB:
             logging.warning("Redis can't be connected.")
         return self.REDIS
 
+    def health(self, queue_name):
+        self.REDIS.ping()
+        return self.REDIS.xinfo_groups(queue_name)[0]
+
     def is_alive(self):
         return self.REDIS is not None
 

requirements_dev.txt

@@ -78,8 +78,6 @@ pycryptodomex==3.20.0
 pydantic==2.6.2
 pydantic_core==2.16.3
 PyJWT==2.8.0
-PyMuPDF==1.23.25
-PyMuPDFb==1.23.22
 PyMySQL==1.1.0
 PyPDF2==3.0.1
 pypdfium2==4.27.0

web/.env

@@ -0,0 +1 @@
+PORT=9222

web/.umirc.ts

@@ -1,7 +1,9 @@
 import { defineConfig } from 'umi';
+import { appName } from './src/conf.json';
 import routes from './src/routes';
 
 export default defineConfig({
+  title: appName,
   outputPath: 'dist',
   // alias: { '@': './src' },
   npmClient: 'npm',
@@ -25,10 +27,13 @@ export default defineConfig({
     },
   },
   devtool: 'source-map',
+  copy: ['src/conf.json'],
   proxy: {
     '/v1': {
-      target: 'http://123.60.95.134:9380/',
+      target: '',
       changeOrigin: true,
+      ws: true,
+      logger: console,
       // pathRewrite: { '^/v1': '/v1' },
     },
   },

web/package.json

@@ -3,7 +3,7 @@
   "author": "zhaofengchao <13723060510@163.com>",
   "scripts": {
     "build": "umi build",
-    "dev": "cross-env PORT=9200 umi dev",
+    "dev": "cross-env UMI_DEV_SERVER_COMPRESS=none umi dev",
     "postinstall": "umi setup",
     "lint": "umi lint --eslint-only",
     "setup": "umi setup",
@@ -19,14 +19,16 @@
     "axios": "^1.6.3",
     "classnames": "^2.5.1",
     "dayjs": "^1.11.10",
+    "eventsource-parser": "^1.1.2",
     "i18next": "^23.7.16",
+    "i18next-browser-languagedetector": "^8.0.0",
     "js-base64": "^3.7.5",
     "jsencrypt": "^3.3.2",
     "lodash": "^4.17.21",
+    "mammoth": "^1.7.2",
     "rc-tween-one": "^3.0.6",
     "react-chat-elements": "^12.0.13",
     "react-copy-to-clipboard": "^5.1.0",
-    "react-file-viewer": "^1.2.1",
     "react-i18next": "^14.0.0",
     "react-infinite-scroll-component": "^6.1.0",
     "react-markdown": "^9.0.1",

web/src/assets/svg/es.svg

@@ -0,0 +1,24 @@
+<svg t="1716195941333" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="7780"
+    width="200" height="200">
+    <path
+        d="M1024 534.4c0-85.76-53.12-160.96-133.44-190.08 3.52-17.92 5.44-36.16 5.44-55.04C896 129.92 766.4 0 606.72 0c-93.12 0-179.84 44.8-234.24 120A153.632 153.632 0 0 0 278.4 87.68a153.632 153.632 0 0 0-144 207.04c-79.68 28.8-134.4 105.6-134.4 190.72 0 86.4 53.44 161.6 133.76 190.72-3.52 17.92-5.12 36.48-5.12 55.04 0 159.04 129.6 288.64 288.64 288.64 93.44 0 180.16-44.8 234.24-120.64a152 152 0 0 0 94.08 32.64 153.632 153.632 0 0 0 144-207.04c79.68-28.48 134.4-105.28 134.4-190.4"
+        fill="#FFFFFF" p-id="7781"></path>
+    <path
+        d="M402.56 439.36l224 102.08 225.92-198.08c3.2-16.32 4.8-32.64 4.8-49.6 0-139.52-113.28-252.8-252.8-252.8-83.52 0-161.28 40.96-208.32 109.76l-37.76 195.2 44.16 93.44z"
+        fill="#FFD00A" p-id="7782"></path>
+    <path
+        d="M170.56 676.48c-3.2 16.32-4.8 33.28-4.8 50.56 0 139.84 113.6 253.44 253.44 253.44 84.16 0 162.24-41.28 209.28-111.04l37.44-194.56-49.92-95.04-224.96-102.4-220.48 199.04z"
+        fill="#20B9AF" p-id="7783"></path>
+    <path
+        d="M169.28 288.96l153.6 36.16 33.6-174.72c-21.12-16-47.04-24.96-73.6-24.96-66.88 0-120.96 54.4-120.96 120.96 0 15.04 2.56 29.12 7.36 42.56"
+        fill="#EE5096" p-id="7784"></path>
+    <path
+        d="M155.84 325.44c-68.48 22.72-116.16 88.64-116.16 160.96 0 70.4 43.52 133.44 108.8 158.08l215.36-194.88-39.68-84.48-168.32-39.68z"
+        fill="#12A5DF" p-id="7785"></path>
+    <path
+        d="M667.84 869.44c21.12 16.32 46.72 24.96 73.28 24.96 66.88 0 120.96-54.4 120.96-120.96 0-14.72-2.56-28.8-7.36-42.24l-153.28-35.84-33.6 174.08z"
+        fill="#90C640" p-id="7786"></path>
+    <path
+        d="M699.2 655.36l168.96 39.36c68.48-22.72 116.48-88.32 116.48-160.96 0-70.4-43.52-133.12-109.12-158.08l-220.8 193.6 44.48 86.08z"
+        fill="#05799F" p-id="7787"></path>
+</svg>

web/src/assets/svg/logo.svg

@@ -1,29 +0,0 @@
-<svg width="32" height="34" viewBox="0 0 32 34" fill="none" xmlns="http://www.w3.org/2000/svg">
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M3.43265 20.7677C4.15835 21.5062 4.15834 22.7035 3.43262 23.4419L3.39546 23.4797C2.66974 24.2182 1.49312 24.2182 0.767417 23.4797C0.0417107 22.7412 0.0417219 21.544 0.767442 20.8055L0.804608 20.7677C1.53033 20.0292 2.70694 20.0293 3.43265 20.7677Z"
-        fill="#B2DDFF" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M12.1689 21.3375C12.8933 22.0773 12.8912 23.2746 12.1641 24.0117L7.01662 29.2307C6.2896 29.9678 5.11299 29.9657 4.38859 29.2259C3.66419 28.4861 3.66632 27.2888 4.39334 26.5517L9.54085 21.3327C10.2679 20.5956 11.4445 20.5977 12.1689 21.3375Z"
-        fill="#53B1FD" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M19.1551 30.3217C19.7244 29.4528 20.8781 29.218 21.7321 29.7973L21.8436 29.8729C22.6975 30.4522 22.9283 31.6262 22.359 32.4952C21.7897 33.3641 20.6359 33.5989 19.782 33.0196L19.6705 32.944C18.8165 32.3647 18.5858 31.1907 19.1551 30.3217Z"
-        fill="#B2DDFF" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M31.4184 20.6544C32.1441 21.3929 32.1441 22.5902 31.4184 23.3286L28.8911 25.9003C28.1654 26.6388 26.9887 26.6388 26.263 25.9003C25.5373 25.1619 25.5373 23.9646 26.263 23.2261L28.7903 20.6544C29.516 19.916 30.6927 19.916 31.4184 20.6544Z"
-        fill="#53B1FD" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M31.4557 11.1427C32.1814 11.8812 32.1814 13.0785 31.4557 13.8169L12.7797 32.8209C12.054 33.5594 10.8774 33.5594 10.1517 32.8209C9.42599 32.0825 9.42599 30.8852 10.1517 30.1467L28.8277 11.1427C29.5534 10.4043 30.73 10.4043 31.4557 11.1427Z"
-        fill="#1570EF" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M27.925 5.29994C28.6508 6.0384 28.6508 7.23568 27.925 7.97414L17.184 18.9038C16.4583 19.6423 15.2817 19.6423 14.556 18.9038C13.8303 18.1653 13.8303 16.9681 14.556 16.2296L25.297 5.29994C26.0227 4.56148 27.1993 4.56148 27.925 5.29994Z"
-        fill="#1570EF" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M22.256 1.59299C22.9822 2.33095 22.983 3.52823 22.2578 4.26718L8.45055 18.3358C7.72533 19.0748 6.54871 19.0756 5.82251 18.3376C5.09631 17.5996 5.09552 16.4024 5.82075 15.6634L19.6279 1.59478C20.3532 0.855827 21.5298 0.855022 22.256 1.59299Z"
-        fill="#1570EF" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M8.58225 6.09619C9.30671 6.83592 9.30469 8.0332 8.57772 8.77038L3.17006 14.2541C2.4431 14.9913 1.26649 14.9893 0.542025 14.2495C-0.182438 13.5098 -0.180413 12.3125 0.546548 11.5753L5.95421 6.09159C6.68117 5.3544 7.85778 5.35646 8.58225 6.09619Z"
-        fill="#53B1FD" />
-    <path fill-rule="evenodd" clip-rule="evenodd"
-        d="M11.893 0.624023C12.9193 0.624023 13.7513 1.47063 13.7513 2.51497V2.70406C13.7513 3.7484 12.9193 4.59501 11.893 4.59501C10.8667 4.59501 10.0347 3.7484 10.0347 2.70406V2.51497C10.0347 1.47063 10.8667 0.624023 11.893 0.624023Z"
-        fill="#B2DDFF" />
-</svg>

web/src/assets/svg/minio.svg

@@ -0,0 +1,10 @@
+<svg t="1716195854453" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="5857"
+    width="200" height="200">
+    <path
+        d="M638.855218 56.525807s99.268136 159.838523 132.357514 216.623262a2.523766 2.523766 0 0 1 0 2.944394 2.383557 2.383557 0 0 1-3.50523 0L596.231612 97.326693l42.623606-40.800886z"
+        fill="#dd113c" p-id="5858"></path>
+    <path
+        d="M346.518971 639.655999a588.878771 588.878771 0 0 1 116.654081-165.446893 597.291325 597.291325 0 0 1 58.32704-51.176369v126.188308L346.518971 639.655999zM245.568325 756.590498l275.931767-140.209231v321.079139l62.112689 80.760517v-434.648616l37.716283-19.489084a187.179324 187.179324 0 0 0 51.456788-296.121896L530.753901 119.479752a31.547077 31.547077 0 0 1 1.542302-44.446327 31.687286 31.687286 0 0 1 44.586535 1.542302l19.909711 20.750966 42.062769-40.941095c-50.335114-65.337502-112.167385-57.065157-147.64032-24.396407a90.575163 90.575163 0 0 0-3.925859 127.870819l143.574253 149.60325a128.151237 128.151237 0 0 1-28.041846 197.414597l-19.489083 10.095065V314.090164A649.589368 649.589368 0 0 0 245.568325 755.889452v0.701046z"
+        fill="#dd113c" p-id="5859"></path>
+    <path d="M583.612781 583.432097v65.617921l-62.112689 31.547077v-65.197293z" fill="#dd113c" p-id="5860"></path>
+</svg>

web/src/assets/svg/mysql.svg

@@ -0,0 +1,9 @@
+<svg t="1716195691568" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="4834"
+    width="200" height="200">
+    <path
+        d="M1001.632 793.792c-7.84-13.856-26.016-37.536-93.12-83.2a1096.224 1096.224 0 0 0-125.152-74.144c-30.592-82.784-89.824-190.112-176.256-319.36-93.056-139.168-201.12-197.792-321.888-174.56a756.608 756.608 0 0 0-40.928-37.696C213.824 78.688 139.2 56.48 96.32 60.736c-19.424 1.952-34.016 9.056-43.36 21.088-21.664 27.904-14.432 68.064 85.504 198.912 19.008 55.616 23.072 84.672 23.072 99.296 0 30.912 15.968 66.368 49.984 110.752l-32 109.504c-28.544 97.792 23.328 224.288 71.616 268.384 25.76 23.552 47.456 20.032 58.176 15.84 21.504-8.448 38.848-29.472 50.048-89.504 5.728 14.112 11.808 29.312 18.208 45.6 34.56 87.744 68.352 136.288 106.336 152.736a32.032 32.032 0 0 0 25.44-58.688c-9.408-4.096-35.328-23.712-72.288-117.504-31.168-79.136-53.856-132.064-69.376-161.856a32.224 32.224 0 0 0-35.328-16.48 32.032 32.032 0 0 0-25.024 29.92c-3.872 91.04-13.056 130.4-19.2 147.008-26.496-30.464-68.128-125.984-47.232-197.536 20.768-71.232 32.992-112.928 36.64-125.248a31.936 31.936 0 0 0-5.888-29.28c-41.664-51.168-46.176-75.584-46.176-83.712 0-29.472-9.248-70.4-28.288-125.152a31.104 31.104 0 0 0-4.768-8.896c-53.824-70.112-73.6-105.216-80.832-121.888 25.632 1.216 74.336 15.04 91.008 29.376a660.8 660.8 0 0 1 49.024 46.304c8 8.448 19.968 11.872 31.232 8.928 100.192-25.92 188.928 21.152 271.072 144 87.808 131.328 146.144 238.048 173.408 317.216a32 32 0 0 0 16.384 18.432 1004.544 1004.544 0 0 1 128.8 75.264c7.392 5.024 14.048 9.696 20.064 14.016h-98.848a32.032 32.032 0 0 0-24.352 52.736 3098.752 3098.752 0 0 0 97.856 110.464 32 32 0 1 0 46.56-43.872 2237.6 2237.6 0 0 1-50.08-55.328h110.08a32.032 32.032 0 0 0 27.84-47.776z"
+        p-id="4835"></path>
+    <path
+        d="M320 289.472c12.672 21.76 22.464 37.344 29.344 46.784 8.288 16.256 21.184 29.248 29.44 45.536l2.016-1.984c14.528-9.952 25.92-49.504 2.752-75.488-12.032-18.176-51.04-17.664-63.552-14.848z"
+        p-id="4836"></path>
+</svg>

web/src/assets/svg/redis.svg

@@ -0,0 +1,6 @@
+<svg t="1716195575286" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="3818"
+    width="200" height="200">
+    <path
+        d="M959.744 602.16l0.256 0.064v101.952c0 10.24-10.752 21.44-35.072 35.84-22.976 13.696-91.968 47.616-163.328 82.624l-35.712 17.536c-65.088 32-126.016 62.208-149.184 76.032-52.8 31.36-82.048 31.104-123.712 8.32-41.6-22.72-305.28-144.256-352.704-170.176-23.744-12.992-36.224-23.936-36.224-34.24v-103.424c0.384 10.368 12.48 21.248 36.224 34.24C147.776 676.8 411.328 798.4 452.992 821.12c41.664 22.784 70.912 23.04 123.712-8.32 52.672-31.36 300.416-147.712 348.224-176.128 23.232-13.824 34.56-24.768 34.88-34.56l-0.064 0.064z m0-168.576h0.192v101.952c0 10.24-10.752 21.44-35.072 35.968-47.808 28.416-295.552 144.768-348.224 176.128-52.8 31.36-82.048 31.04-123.712 8.32-41.6-22.72-305.28-144.32-352.704-170.24C76.48 572.8 64 561.92 64 551.536v-103.424c0.384 10.24 12.48 21.248 36.224 34.176 47.488 25.92 311.04 147.52 352.704 170.24 41.664 22.72 70.912 23.04 123.712-8.32 52.672-31.36 300.416-147.712 348.224-176.192 23.168-13.824 34.56-24.704 34.88-34.432zM462.656 81.84c55.36-22.72 74.56-23.488 121.664-3.776 47.168 19.776 293.376 131.648 339.968 151.104 24 10.048 35.84 19.2 35.456 29.632H960v101.952c0 10.176-10.816 21.44-35.072 35.904C877.056 425.072 629.376 541.44 576.64 572.8c-52.736 31.36-81.984 31.104-123.648 8.32-41.664-22.656-305.28-144.32-352.768-170.24C76.544 397.936 64 387.056 64 376.688V273.28c-0.32-10.304 11.072-19.968 34.368-30.464 46.656-20.8 308.8-138.24 364.288-160.896v-0.064z m129.792 238.4l-207.552 36.352 144.832 68.608 62.72-104.96z m128.704-113.6l-135.936 61.44 122.688 55.36 13.376-5.952 122.752-55.424-122.88-55.424z m-392.32 13.44c-61.248 0-110.912 22.016-110.912 49.152 0 27.072 49.664 49.088 110.976 49.088s110.912-21.952 110.912-49.088-49.6-49.088-110.912-49.088l-0.064-0.064z m134.656-101.888l20.096 42.304-66.88 27.52 89.6 9.216 28.032 53.248 17.408-47.744 77.632-9.216-60.16-25.728 16-43.712-59.136 22.08-62.592-27.968z"
+        fill="#D82A1F" p-id="3819"></path>
+</svg>

web/src/components/chunk-method-modal/index.tsx

@@ -15,7 +15,6 @@ import {
   Modal,
   Select,
   Space,
-  Switch,
   Tooltip,
 } from 'antd';
 import omit from 'lodash/omit';
@@ -23,6 +22,7 @@ import React, { useEffect, useMemo } from 'react';
 import { useFetchParserListOnMount } from './hooks';
 
 import { useTranslate } from '@/hooks/commonHooks';
+import LayoutRecognize from '../layout-recognize';
 import styles from './index.less';
 
 interface IProps extends Omit<IModalManagerChildrenProps, 'showModal'> {
@@ -228,17 +228,7 @@ const ChunkMethodModal: React.FC<IProps> = ({
             </Form.List>
           </>
         )}
-        {showOne && (
-          <Form.Item
-            name={['parser_config', 'layout_recognize']}
-            label={t('layoutRecognize')}
-            initialValue={true}
-            valuePropName="checked"
-            tooltip={t('layoutRecognizeTip')}
-          >
-            <Switch />
-          </Form.Item>
-        )}
+        {showOne && <LayoutRecognize></LayoutRecognize>}
         {showPages && (
           <Form.Item
             noStyle

web/src/components/layout-recognize.tsx

@@ -0,0 +1,19 @@
+import { useTranslate } from '@/hooks/commonHooks';
+import { Form, Switch } from 'antd';
+
+const LayoutRecognize = () => {
+  const { t } = useTranslate('knowledgeDetails');
+  return (
+    <Form.Item
+      name={['parser_config', 'layout_recognize']}
+      label={t('layoutRecognize')}
+      initialValue={true}
+      valuePropName="checked"
+      tooltip={t('layoutRecognizeTip')}
+    >
+      <Switch />
+    </Form.Item>
+  );
+};
+
+export default LayoutRecognize;

web/src/components/new-document-link.tsx

@@ -1,22 +1,24 @@
-import { api_host } from '@/utils/api';
 import React from 'react';
 
 interface IProps extends React.PropsWithChildren {
-  documentId: string;
+  link: string;
   preventDefault?: boolean;
+  color?: string;
 }
 
 const NewDocumentLink = ({
   children,
-  documentId,
+  link,
   preventDefault = false,
+  color = 'rgb(15, 79, 170)',
 }: IProps) => {
   return (
     <a
       target="_blank"
       onClick={!preventDefault ? undefined : (e) => e.preventDefault()}
-      href={`${api_host}/document/get/${documentId}`}
+      href={link}
       rel="noreferrer"
+      style={{ color, wordBreak: 'break-all' }}
     >
       {children}
     </a>

web/src/components/pdf-previewer/hooks.ts

@@ -0,0 +1,18 @@
+import axios from 'axios';
+import { useCallback, useEffect, useState } from 'react';
+
+export const useCatchDocumentError = (url: string) => {
+  const [error, setError] = useState<string>('');
+
+  const fetchDocument = useCallback(async () => {
+    const { data } = await axios.get(url);
+    if (data.retcode !== 0) {
+      setError(data?.retmsg);
+    }
+  }, [url]);
+  useEffect(() => {
+    fetchDocument();
+  }, [fetchDocument]);
+
+  return error;
+};

web/src/components/pdf-previewer/index.tsx

@@ -14,6 +14,8 @@ import {
   Popup,
 } from 'react-pdf-highlighter';
 
+import FileError from '@/pages/document-viewer/file-error';
+import { useCatchDocumentError } from './hooks';
 import styles from './index.less';
 
 interface IProps {
@@ -34,10 +36,12 @@ const HighlightPopup = ({
   ) : null;
 
 const DocumentPreviewer = ({ chunk, documentId, visible }: IProps) => {
-  const url = useGetDocumentUrl(documentId);
+  const getDocumentUrl = useGetDocumentUrl(documentId);
   const { highlights: state, setWidthAndHeight } = useGetChunkHighlights(chunk);
   const ref = useRef<(highlight: IHighlight) => void>(() => {});
   const [loaded, setLoaded] = useState(false);
+  const url = getDocumentUrl();
+  const error = useCatchDocumentError(url);
 
   const resetHash = () => {};
 
@@ -58,6 +62,7 @@ const DocumentPreviewer = ({ chunk, documentId, visible }: IProps) => {
         url={url}
         beforeLoad={<Skeleton active />}
         workerSrc="/pdfjs-dist/pdf.worker.min.js"
+        errorMessage={<FileError>{error}</FileError>}
       >
         {(pdfDocument) => {
           pdfDocument.getPage(1).then((page) => {

web/src/conf.json

@@ -0,0 +1,3 @@
+{
+  "appName": "RAGFlow"
+}

web/src/constants/common.ts

@@ -68,3 +68,25 @@ export const FileMimeTypeMap = {
   xlsx: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
   mp4: 'video/mp4',
 };
+
+export const Domain = 'demo.ragflow.io';
+
+//#region file preview
+export const Images = [
+  'jpg',
+  'jpeg',
+  'png',
+  'gif',
+  'bmp',
+  'tif',
+  'tiff',
+  'webp',
+  // 'svg',
+  'ico',
+];
+
+// Without FileViewer
+export const ExceptiveType = ['xlsx', 'xls', 'pdf', 'docx', ...Images];
+
+export const SupportedPreviewDocumentTypes = [...ExceptiveType];
+//#endregion

web/src/constants/knowledge.ts

@@ -31,14 +31,14 @@ export const settledModelVariableMap = {
     top_p: 0.3,
     frequency_penalty: 0.7,
     presence_penalty: 0.4,
-    max_tokens: 215,
+    max_tokens: 512,
   },
   [ModelVariableType.Balance]: {
     temperature: 0.5,
     top_p: 0.5,
     frequency_penalty: 0.7,
     presence_penalty: 0.4,
-    max_tokens: 215,
+    max_tokens: 512,
   },
 };
 

web/src/constants/setting.ts

@@ -4,6 +4,7 @@ export enum UserSettingRouteKey {
   Profile = 'profile',
   Password = 'password',
   Model = 'model',
+  System = 'system',
   Team = 'team',
   Logout = 'logout',
 }
@@ -12,6 +13,7 @@ export const UserSettingRouteMap = {
   [UserSettingRouteKey.Profile]: 'Profile',
   [UserSettingRouteKey.Password]: 'Password',
   [UserSettingRouteKey.Model]: 'Model Providers',
+  [UserSettingRouteKey.System]: 'System Version',
   [UserSettingRouteKey.Team]: 'Team',
   [UserSettingRouteKey.Logout]: 'Log out',
 };

web/src/hooks/chatHooks.ts

@@ -154,6 +154,9 @@ export const useRemoveConversation = () => {
   return removeConversation;
 };
 
+/*
+@deprecated
+ */
 export const useCompleteConversation = () => {
   const dispatch = useDispatch();
 
@@ -283,20 +286,4 @@ export const useFetchSharedConversation = () => {
   return fetchSharedConversation;
 };
 
-export const useCompleteSharedConversation = () => {
-  const dispatch = useDispatch();
-
-  const completeSharedConversation = useCallback(
-    (payload: any) => {
-      return dispatch<any>({
-        type: 'chatModel/completeExternalConversation',
-        payload: payload,
-      });
-    },
-    [dispatch],
-  );
-
-  return completeSharedConversation;
-};
-
 //#endregion

web/src/hooks/documentHooks.ts

@@ -9,12 +9,15 @@ import { useDispatch, useSelector } from 'umi';
 import { useGetKnowledgeSearchParams } from './routeHook';
 import { useOneNamespaceEffectsLoading } from './storeHooks';
 
-export const useGetDocumentUrl = (documentId: string) => {
-  const url = useMemo(() => {
-    return `${api_host}/document/get/${documentId}`;
-  }, [documentId]);
+export const useGetDocumentUrl = (documentId?: string) => {
+  const getDocumentUrl = useCallback(
+    (id?: string) => {
+      return `${api_host}/document/get/${documentId || id}`;
+    },
+    [documentId],
+  );
 
-  return url;
+  return getDocumentUrl;
 };
 
 export const useGetChunkHighlights = (selectedChunk: IChunk) => {

web/src/hooks/llmHooks.ts

@@ -4,7 +4,10 @@ import {
   IMyLlmValue,
   IThirdOAIModelCollection,
 } from '@/interfaces/database/llm';
-import { IAddLlmRequestBody } from '@/interfaces/request/llm';
+import {
+  IAddLlmRequestBody,
+  IDeleteLlmRequestBody,
+} from '@/interfaces/request/llm';
 import { sortLLmFactoryListBySpecifiedOrder } from '@/utils/commonUtil';
 import { useCallback, useEffect, useMemo } from 'react';
 import { useDispatch, useSelector } from 'umi';
@@ -64,13 +67,15 @@ export const useSelectLlmOptionsByModelType = () => {
   const groupOptionsByModelType = (modelType: LlmModelType) => {
     return Object.entries(llmInfo)
       .filter(([, value]) =>
-        modelType ? value.some((x) => x.model_type === modelType) : true,
+        modelType ? value.some((x) => x.model_type.includes(modelType)) : true,
       )
       .map(([key, value]) => {
         return {
           label: key,
           options: value
-            .filter((x) => (modelType ? x.model_type === modelType : true))
+            .filter((x) =>
+              modelType ? x.model_type.includes(modelType) : true,
+            )
             .map((x) => ({
               label: x.llm_name,
               value: x.llm_name,
@@ -211,7 +216,7 @@ export const useSaveTenantInfo = () => {
 export const useAddLlm = () => {
   const dispatch = useDispatch();
 
-  const saveTenantInfo = useCallback(
+  const addLlm = useCallback(
     (requestBody: IAddLlmRequestBody) => {
       return dispatch<any>({
         type: 'settingModel/add_llm',
@@ -221,5 +226,21 @@ export const useAddLlm = () => {
     [dispatch],
   );
 
-  return saveTenantInfo;
+  return addLlm;
+};
+
+export const useDeleteLlm = () => {
+  const dispatch = useDispatch();
+
+  const deleteLlm = useCallback(
+    (requestBody: IDeleteLlmRequestBody) => {
+      return dispatch<any>({
+        type: 'settingModel/delete_llm',
+        payload: requestBody,
+      });
+    },
+    [dispatch],
+  );
+
+  return deleteLlm;
 };

web/src/hooks/logicHooks.ts

@@ -1,9 +1,15 @@
+import { Authorization } from '@/constants/authorization';
 import { LanguageTranslationMap } from '@/constants/common';
 import { Pagination } from '@/interfaces/common';
+import { IAnswer } from '@/interfaces/database/chat';
 import { IKnowledgeFile } from '@/interfaces/database/knowledge';
 import { IChangeParserConfigRequestBody } from '@/interfaces/request/document';
+import api from '@/utils/api';
+import { getAuthorization } from '@/utils/authorizationUtil';
 import { PaginationProps } from 'antd';
-import { useCallback, useMemo, useState } from 'react';
+import axios from 'axios';
+import { EventSourceParserStream } from 'eventsource-parser/stream';
+import { useCallback, useEffect, useMemo, useState } from 'react';
 import { useTranslation } from 'react-i18next';
 import { useDispatch } from 'umi';
 import { useSetModalState, useTranslate } from './commonHooks';
@@ -113,3 +119,80 @@ export const useSetPagination = (namespace: string) => {
 
   return setPagination;
 };
+
+export interface AppConf {
+  appName: string;
+}
+
+export const useFetchAppConf = () => {
+  const [appConf, setAppConf] = useState<AppConf>({} as AppConf);
+  const fetchAppConf = useCallback(async () => {
+    const ret = await axios.get('/conf.json');
+
+    setAppConf(ret.data);
+  }, []);
+
+  useEffect(() => {
+    fetchAppConf();
+  }, [fetchAppConf]);
+
+  return appConf;
+};
+
+export const useSendMessageWithSse = (
+  url: string = api.completeConversation,
+) => {
+  const [answer, setAnswer] = useState<IAnswer>({} as IAnswer);
+  const [done, setDone] = useState(true);
+
+  const send = useCallback(
+    async (body: any) => {
+      try {
+        setDone(false);
+        const response = await fetch(url, {
+          method: 'POST',
+          headers: {
+            [Authorization]: getAuthorization(),
+            'Content-Type': 'application/json',
+          },
+          body: JSON.stringify(body),
+        });
+
+        const reader = response?.body
+          ?.pipeThrough(new TextDecoderStream())
+          .pipeThrough(new EventSourceParserStream())
+          .getReader();
+
+        while (true) {
+          const x = await reader?.read();
+          if (x) {
+            const { done, value } = x;
+            try {
+              const val = JSON.parse(value?.data || '');
+              const d = val?.data;
+              if (typeof d !== 'boolean') {
+                console.info('data:', d);
+                setAnswer(d);
+              }
+            } catch (e) {
+              console.warn(e);
+            }
+            if (done) {
+              console.info('done');
+              break;
+            }
+          }
+        }
+        console.info('done?');
+        setDone(true);
+        return response;
+      } catch (e) {
+        setDone(true);
+        console.warn(e);
+      }
+    },
+    [url],
+  );
+
+  return { send, answer, done };
+};

web/src/hooks/userSettingHook.ts

@@ -1,7 +1,8 @@
 import { ITenantInfo } from '@/interfaces/database/knowledge';
-import { IUserInfo } from '@/interfaces/database/userSetting';
+import { ISystemStatus, IUserInfo } from '@/interfaces/database/userSetting';
+import userService from '@/services/userService';
 import authorizationUtil from '@/utils/authorizationUtil';
-import { useCallback, useEffect, useMemo } from 'react';
+import { useCallback, useEffect, useMemo, useState } from 'react';
 import { history, useDispatch, useSelector } from 'umi';
 
 export const useFetchUserInfo = () => {
@@ -92,3 +93,41 @@ export const useSaveSetting = () => {
 
   return saveSetting;
 };
+
+export const useFetchSystemVersion = () => {
+  const [version, setVersion] = useState('');
+  const [loading, setLoading] = useState(false);
+
+  const fetchSystemVersion = useCallback(async () => {
+    setLoading(true);
+    const { data } = await userService.getSystemVersion();
+    if (data.retcode === 0) {
+      setVersion(data.data);
+      setLoading(false);
+    }
+  }, []);
+
+  return { fetchSystemVersion, version, loading };
+};
+
+export const useFetchSystemStatus = () => {
+  const [systemStatus, setSystemStatus] = useState<ISystemStatus>(
+    {} as ISystemStatus,
+  );
+  const [loading, setLoading] = useState(false);
+
+  const fetchSystemStatus = useCallback(async () => {
+    setLoading(true);
+    const { data } = await userService.getSystemStatus();
+    if (data.retcode === 0) {
+      setSystemStatus(data.data);
+      setLoading(false);
+    }
+  }, []);
+
+  return {
+    systemStatus,
+    fetchSystemStatus,
+    loading,
+  };
+};

web/src/interfaces/database/chat.ts

@@ -72,6 +72,11 @@ export interface IReference {
   total: number;
 }
 
+export interface IAnswer {
+  answer: string;
+  reference: IReference;
+}
+
 export interface Docagg {
   count: number;
   doc_id: string;

web/src/interfaces/database/file-manager.ts

@@ -12,6 +12,7 @@ export interface IFile {
   type: string;
   update_date: string;
   update_time: number;
+  source_type: string;
 }
 
 export interface IFolder {
@@ -27,4 +28,5 @@ export interface IFolder {
   type: string;
   update_date: string;
   update_time: number;
+  source_type: string;
 }

web/src/interfaces/database/userSetting.ts

@@ -19,3 +19,31 @@ export interface IUserInfo {
   update_date: string;
   update_time: number;
 }
+
+export interface ISystemStatus {
+  es: Es;
+  minio: Minio;
+  mysql: Minio;
+  redis: Redis;
+}
+
+interface Redis {
+  status: string;
+  elapsed: number;
+  error: string;
+  pending: number;
+}
+
+export interface Minio {
+  status: string;
+  elapsed: number;
+  error: string;
+}
+
+interface Es {
+  status: string;
+  elapsed: number;
+  error: string;
+  number_of_nodes: number;
+  active_shards: number;
+}

web/src/interfaces/request/llm.ts

@@ -4,3 +4,8 @@ export interface IAddLlmRequestBody {
   model_type: string;
   api_base?: string; // chat|embedding|speech2text|image2text
 }
+
+export interface IDeleteLlmRequestBody {
+  llm_factory: string; // Ollama
+  llm_name: string;
+}

web/src/layouts/components/header/index.less

@@ -18,6 +18,7 @@
 
 .appIcon {
   vertical-align: middle;
+  max-width: 36px;
 }
 
 .appName {

web/src/layouts/components/header/index.tsx

@@ -1,7 +1,6 @@
 import { ReactComponent as StarIon } from '@/assets/svg/chat-star.svg';
 import { ReactComponent as FileIcon } from '@/assets/svg/file-management.svg';
 import { ReactComponent as KnowledgeBaseIcon } from '@/assets/svg/knowledge-base.svg';
-import { ReactComponent as Logo } from '@/assets/svg/logo.svg';
 import { useTranslate } from '@/hooks/commonHooks';
 import { useNavigateWithFromState } from '@/hooks/routeHook';
 import { Layout, Radio, Space, theme } from 'antd';
@@ -9,6 +8,7 @@ import { useCallback, useMemo } from 'react';
 import { useLocation } from 'umi';
 import Toolbar from '../right-toolbar';
 
+import { useFetchAppConf } from '@/hooks/logicHooks';
 import styles from './index.less';
 
 const { Header } = Layout;
@@ -20,6 +20,7 @@ const RagHeader = () => {
   const navigate = useNavigateWithFromState();
   const { pathname } = useLocation();
   const { t } = useTranslate('header');
+  const appConf = useFetchAppConf();
 
   const tagsData = useMemo(
     () => [
@@ -56,8 +57,8 @@ const RagHeader = () => {
       }}
     >
       <Space size={12} onClick={handleLogoClick} className={styles.logoWrapper}>
-        <Logo className={styles.appIcon}></Logo>
-        <span className={styles.appName}>RAGFlow</span>
+        <img src="/logo.svg" alt="" className={styles.appIcon} />
+        <span className={styles.appName}>{appConf.appName}</span>
       </Space>
       <Space size={[0, 8]} wrap>
         <Radio.Group

web/src/locales/config.ts

@@ -1,4 +1,5 @@
 import i18n from 'i18next';
+import LanguageDetector from 'i18next-browser-languagedetector';
 import { initReactI18next } from 'react-i18next';
 
 import translation_en from './en';
@@ -11,13 +12,19 @@ const resources = {
   'zh-TRADITIONAL': translation_zh_traditional,
 };
 
-i18n.use(initReactI18next).init({
-  resources,
-  lng: 'en',
-  fallbackLng: 'en',
-  interpolation: {
-    escapeValue: false,
-  },
-});
+i18n
+  .use(initReactI18next)
+  .use(LanguageDetector)
+  .init({
+    detection: {
+      lookupLocalStorage: 'lng',
+    },
+    supportedLngs: ['en', 'zh', 'zh-TRADITIONAL'],
+    resources,
+    fallbackLng: 'en',
+    interpolation: {
+      escapeValue: false,
+    },
+  });
 
 export default i18n;

web/src/locales/en.ts

@@ -25,6 +25,7 @@ export default {
       comingSoon: 'Coming Soon',
       download: 'Download',
       close: 'Close',
+      preview: 'Preview',
     },
     login: {
       login: 'Sign in',
@@ -381,6 +382,7 @@ export default {
       partialTitle: 'Partial Embed',
       extensionTitle: 'Chrome Extension',
       tokenError: 'Please create API Token first!',
+      searching: 'searching...',
     },
     setting: {
       profile: 'Profile',
@@ -391,6 +393,7 @@ export default {
       model: 'Model Providers',
       modelDescription: 'Set the model parameter and API Key here.',
       team: 'Team',
+      system: 'System',
       logout: 'Log out',
       username: 'Username',
       usernameMessage: 'Please input your username!',
@@ -494,7 +497,7 @@ export default {
       knowledgeBase: 'Knowledge Base',
       size: 'Size',
       action: 'Action',
-      addToKnowledge: 'Add to Knowledge Base',
+      addToKnowledge: 'Link to Knowledge Base',
       pleaseSelect: 'Please select',
       newFolder: 'New Folder',
       file: 'File',
@@ -505,6 +508,8 @@ export default {
         'Support for a single or bulk upload. Strictly prohibited from uploading company data or other banned files.',
       local: 'Local uploads',
       s3: 'S3 uploads',
+      preview: 'Preview',
+      fileError: 'File error',
     },
     footer: {
       profile: 'All rights reserved @ React',

web/src/locales/zh-traditional.ts

@@ -25,6 +25,7 @@ export default {
       comingSoon: '即將推出',
       download: '下載',
       close: '关闭',
+      preview: '預覽',
     },
     login: {
       login: '登入',
@@ -352,6 +353,7 @@ export default {
       partialTitle: '部分嵌入',
       extensionTitle: 'Chrome 插件',
       tokenError: '請先創建 Api Token!',
+      searching: '搜索中',
     },
     setting: {
       profile: '概述',
@@ -362,6 +364,7 @@ export default {
       modelDescription: '在此設置模型參數和 API Key。',
       team: '團隊',
       logout: '登出',
+      system: '系統',
       username: '使用者名稱',
       usernameMessage: '請輸入用戶名',
       photo: '頭像',
@@ -458,7 +461,7 @@ export default {
       knowledgeBase: '知識庫',
       size: '大小',
       action: '操作',
-      addToKnowledge: '添加到知識庫',
+      addToKnowledge: '鏈接知識庫',
       pleaseSelect: '請選擇',
       newFolder: '新建文件夾',
       uploadFile: '上傳文件',
@@ -468,6 +471,8 @@ export default {
       directory: '文件夾',
       local: '本地上傳',
       s3: 'S3 上傳',
+      preview: '預覽',
+      fileError: '文件錯誤',
     },
     footer: {
       profile: '“保留所有權利 @ react”',

web/src/locales/zh.ts

@@ -25,6 +25,7 @@ export default {
       comingSoon: '即将推出',
       download: '下载',
       close: '关闭',
+      preview: '预览',
     },
     login: {
       login: '登录',
@@ -369,6 +370,7 @@ export default {
       partialTitle: '部分嵌入',
       extensionTitle: 'Chrome 插件',
       tokenError: '请先创建 Api Token!',
+      searching: '搜索中',
     },
     setting: {
       profile: '概要',
@@ -378,6 +380,7 @@ export default {
       model: '模型提供商',
       modelDescription: '在此设置模型参数和 API Key。',
       team: '团队',
+      system: '系统',
       logout: '登出',
       username: '用户名',
       usernameMessage: '请输入用户名',
@@ -475,7 +478,7 @@ export default {
       knowledgeBase: '知识库',
       size: '大小',
       action: '操作',
-      addToKnowledge: '添加到知识库',
+      addToKnowledge: '链接知识库',
       pleaseSelect: '请选择',
       newFolder: '新建文件夹',
       uploadFile: '上传文件',
@@ -486,6 +489,8 @@ export default {
       directory: '文件夹',
       local: '本地上传',
       s3: 'S3 上传',
+      preview: '预览',
+      fileError: '文件错误',
     },
     footer: {
       profile: 'All rights reserved @ React',

web/src/pages/add-knowledge/components/knowledge-chunk/components/document-preview/preview.tsx

@@ -11,6 +11,8 @@ import {
 import { useGetChunkHighlights } from '../../hooks';
 import { useGetDocumentUrl } from './hooks';
 
+import { useCatchDocumentError } from '@/components/pdf-previewer/hooks';
+import FileError from '@/pages/document-viewer/file-error';
 import styles from './index.less';
 
 interface IProps {
@@ -30,9 +32,11 @@ const HighlightPopup = ({
 // TODO: merge with DocumentPreviewer
 const Preview = ({ selectedChunkId }: IProps) => {
   const url = useGetDocumentUrl();
+  useCatchDocumentError(url);
   const { highlights: state, setWidthAndHeight } =
     useGetChunkHighlights(selectedChunkId);
   const ref = useRef<(highlight: IHighlight) => void>(() => {});
+  const error = useCatchDocumentError(url);
 
   const resetHash = () => {};
 
@@ -48,6 +52,7 @@ const Preview = ({ selectedChunkId }: IProps) => {
         url={url}
         beforeLoad={<Skeleton active />}
         workerSrc="/pdfjs-dist/pdf.worker.min.js"
+        errorMessage={<FileError>{error}</FileError>}
       >
         {(pdfDocument) => {
           pdfDocument.getPage(1).then((page) => {

web/src/pages/add-knowledge/components/knowledge-file/index.tsx

@@ -106,8 +106,8 @@ const KnowledgeFile = () => {
     },
     {
       title: t('uploadDate'),
-      dataIndex: 'create_date',
-      key: 'create_date',
+      dataIndex: 'create_time',
+      key: 'create_time',
       render(value) {
         return formatDate(value);
       },

web/src/pages/add-knowledge/components/knowledge-file/parsing-status-cell/index.tsx

@@ -55,7 +55,7 @@ const PopoverContent = ({ record }: IProps) => {
     {
       key: 'process_duation',
       label: t('processDuration'),
-      children: record.process_duation,
+      children: `${record.process_duation.toFixed(2)} s`,
     },
     {
       key: 'progress_msg',

web/src/pages/add-knowledge/components/knowledge-setting/configuration.tsx

@@ -6,6 +6,7 @@ import {
   useSubmitKnowledgeConfiguration,
 } from './hooks';
 
+import LayoutRecognize from '@/components/layout-recognize';
 import MaxTokenNumber from '@/components/max-token-number';
 import { useTranslate } from '@/hooks/commonHooks';
 import { FormInstance } from 'antd/lib';
@@ -99,11 +100,17 @@ const ConfigurationForm = ({ form }: { form: FormInstance }) => {
           const parserId = getFieldValue('parser_id');
 
           if (parserId === 'naive') {
-            return <MaxTokenNumber></MaxTokenNumber>;
+            return (
+              <>
+                <MaxTokenNumber></MaxTokenNumber>
+                <LayoutRecognize></LayoutRecognize>
+              </>
+            );
           }
           return null;
         }}
       </Form.Item>
+
       <Form.Item>
         <div className={styles.buttonWrapper}>
           <Space>

web/src/pages/add-knowledge/components/knowledge-testing/testing-control/index.tsx

@@ -1,5 +1,5 @@
 import SimilaritySlider from '@/components/similarity-slider';
-import { Button, Card, Divider, Flex, Form, Input, Slider } from 'antd';
+import { Button, Card, Divider, Flex, Form, Input } from 'antd';
 import { FormInstance } from 'antd/lib';
 
 import { useTranslate } from '@/hooks/commonHooks';
@@ -9,7 +9,6 @@ import styles from './index.less';
 type FieldType = {
   similarity_threshold?: number;
   vector_similarity_weight?: number;
-  top_k?: number;
   question: string;
 };
 
@@ -36,22 +35,8 @@ const TestingControl = ({ form, handleTesting }: IProps) => {
       <p>{t('testingDescription')}</p>
       <Divider></Divider>
       <section>
-        <Form
-          name="testing"
-          layout="vertical"
-          form={form}
-          initialValues={{
-            top_k: 1024,
-          }}
-        >
+        <Form name="testing" layout="vertical" form={form}>
           <SimilaritySlider isTooltipShown></SimilaritySlider>
-          <Form.Item<FieldType>
-            label="Top K"
-            name={'top_k'}
-            tooltip={t('topKTip')}
-          >
-            <Slider marks={{ 0: 0, 2048: 2048 }} max={2048} />
-          </Form.Item>
           <Card size="small" title={t('testText')}>
             <Form.Item<FieldType>
               name={'question'}

web/src/pages/add-knowledge/components/knowledge-testing/testing-result/select-files.tsx

@@ -1,5 +1,6 @@
 import { ReactComponent as NavigationPointerIcon } from '@/assets/svg/navigation-pointer.svg';
 import NewDocumentLink from '@/components/new-document-link';
+import { useGetDocumentUrl } from '@/hooks/documentHooks';
 import { ITestingDocument } from '@/interfaces/database/knowledge';
 import { isPdf } from '@/utils/documentUtils';
 import { Table, TableProps } from 'antd';
@@ -15,6 +16,7 @@ const SelectFiles = ({ handleTesting }: IProps) => {
   );
 
   const dispatch = useDispatch();
+  const getDocumentUrl = useGetDocumentUrl();
 
   const columns: TableProps<ITestingDocument>['columns'] = [
     {
@@ -35,7 +37,10 @@ const SelectFiles = ({ handleTesting }: IProps) => {
       key: 'view',
       width: 50,
       render: (_, { doc_id, doc_name }) => (
-        <NewDocumentLink documentId={doc_id} preventDefault={!isPdf(doc_name)}>
+        <NewDocumentLink
+          link={getDocumentUrl(doc_id)}
+          preventDefault={!isPdf(doc_name)}
+        >
           <NavigationPointerIcon />
         </NewDocumentLink>
       ),

web/src/pages/chat/chat-configuration-modal/assistant-setting.tsx

@@ -22,6 +22,15 @@ const AssistantSetting = ({ show }: ISegmentedContentProps) => {
     return e?.fileList;
   };
 
+  const uploadButtion = (
+    <button style={{ border: 0, background: 'none' }} type="button">
+            <PlusOutlined />
+            <div style={{ marginTop: 8 }}>
+              {t('upload', { keyPrefix: 'common' })}
+            </div>
+          </button>
+  )
+
   return (
     <section
       className={classNames({
@@ -46,12 +55,7 @@ const AssistantSetting = ({ show }: ISegmentedContentProps) => {
           maxCount={1}
           showUploadList={{ showPreviewIcon: false, showRemoveIcon: false }}
         >
-          <button style={{ border: 0, background: 'none' }} type="button">
-            <PlusOutlined />
-            <div style={{ marginTop: 8 }}>
-              {t('upload', { keyPrefix: 'common' })}
-            </div>
-          </button>
+          {show ? uploadButtion : null}
         </Upload>
       </Form.Item>
       <Form.Item

web/src/pages/chat/chat-container/index.less

@@ -36,6 +36,9 @@
   // .referenceIcon {
   //   padding: 0 6px;
   // }
+  .thumbnailImg {
+    max-width: 20px;
+  }
 }
 
 .messageItemLeft {

web/src/pages/chat/chat-container/index.tsx

@@ -6,16 +6,7 @@ import { useSelectFileThumbnails } from '@/hooks/knowledgeHook';
 import { useSelectUserInfo } from '@/hooks/userSettingHook';
 import { IReference, Message } from '@/interfaces/database/chat';
 import { IChunk } from '@/interfaces/database/knowledge';
-import {
-  Avatar,
-  Button,
-  Drawer,
-  Flex,
-  Input,
-  List,
-  Skeleton,
-  Spin,
-} from 'antd';
+import { Avatar, Button, Drawer, Flex, Input, List, Spin } from 'antd';
 import classNames from 'classnames';
 import { useMemo } from 'react';
 import {
@@ -30,20 +21,26 @@ import MarkdownContent from '../markdown-content';
 
 import SvgIcon from '@/components/svg-icon';
 import { useTranslate } from '@/hooks/commonHooks';
+import { useGetDocumentUrl } from '@/hooks/documentHooks';
 import { getExtension, isPdf } from '@/utils/documentUtils';
+import { buildMessageItemReference } from '../utils';
 import styles from './index.less';
 
 const MessageItem = ({
   item,
   reference,
+  loading = false,
   clickDocumentButton,
 }: {
   item: Message;
   reference: IReference;
+  loading?: boolean;
   clickDocumentButton: (documentId: string, chunk: IChunk) => void;
 }) => {
   const userInfo = useSelectUserInfo();
   const fileThumbnails = useSelectFileThumbnails();
+  const getDocumentUrl = useGetDocumentUrl();
+  const { t } = useTranslate('chat');
 
   const isAssistant = item.role === MessageType.Assistant;
 
@@ -51,6 +48,14 @@ const MessageItem = ({
     return reference?.doc_aggs ?? [];
   }, [reference?.doc_aggs]);
 
+  const content = useMemo(() => {
+    let text = item.content;
+    if (text === '') {
+      text = t('searching');
+    }
+    return loading ? text?.concat('~~2$$') : text;
+  }, [item.content, loading, t]);
+
   return (
     <div
       className={classNames(styles.messageItem, {
@@ -83,15 +88,11 @@ const MessageItem = ({
           <Flex vertical gap={8} flex={1}>
             <b>{isAssistant ? '' : userInfo.nickname}</b>
             <div className={styles.messageText}>
-              {item.content !== '' ? (
-                <MarkdownContent
-                  content={item.content}
-                  reference={reference}
-                  clickDocumentButton={clickDocumentButton}
-                ></MarkdownContent>
-              ) : (
-                <Skeleton active className={styles.messageEmpty} />
-              )}
+              <MarkdownContent
+                content={content}
+                reference={reference}
+                clickDocumentButton={clickDocumentButton}
+              ></MarkdownContent>
             </div>
             {isAssistant && referenceDocumentList.length > 0 && (
               <List
@@ -104,7 +105,10 @@ const MessageItem = ({
                     <List.Item>
                       <Flex gap={'small'} align="center">
                         {fileThumbnail ? (
-                          <img src={fileThumbnail}></img>
+                          <img
+                            src={fileThumbnail}
+                            className={styles.thumbnailImg}
+                          ></img>
                         ) : (
                           <SvgIcon
                             name={`file-icon/${fileExtension}`}
@@ -113,7 +117,7 @@ const MessageItem = ({
                         )}
 
                         <NewDocumentLink
-                          documentId={item.doc_id}
+                          link={getDocumentUrl(item.doc_id)}
                           preventDefault={!isPdf(item.doc_name)}
                         >
                           {item.doc_name}
@@ -137,13 +141,19 @@ const ChatContainer = () => {
     currentConversation: conversation,
     addNewestConversation,
     removeLatestMessage,
+    addNewestAnswer,
   } = useFetchConversationOnMount();
   const {
     handleInputChange,
     handlePressEnter,
     value,
     loading: sendLoading,
-  } = useSendMessage(conversation, addNewestConversation, removeLatestMessage);
+  } = useSendMessage(
+    conversation,
+    addNewestConversation,
+    removeLatestMessage,
+    addNewestAnswer,
+  );
   const { visible, hideModal, documentId, selectedChunk, clickDocumentButton } =
     useClickDrawer();
   const disabled = useGetSendButtonDisabled();
@@ -157,19 +167,17 @@ const ChatContainer = () => {
         <Flex flex={1} vertical className={styles.messageContainer}>
           <div>
             <Spin spinning={loading}>
-              {conversation?.message?.map((message) => {
-                const assistantMessages = conversation?.message
-                  ?.filter((x) => x.role === MessageType.Assistant)
-                  .slice(1);
-                const referenceIndex = assistantMessages.findIndex(
-                  (x) => x.id === message.id,
-                );
-                const reference = conversation.reference[referenceIndex];
+              {conversation?.message?.map((message, i) => {
                 return (
                   <MessageItem
+                    loading={
+                      message.role === MessageType.Assistant &&
+                      sendLoading &&
+                      conversation?.message.length - 1 === i
+                    }
                     key={message.id}
                     item={message}
-                    reference={reference}
+                    reference={buildMessageItemReference(conversation, message)}
                     clickDocumentButton={clickDocumentButton}
                   ></MessageItem>
                 );