chore: Update version to v1.1.3

.github/release-drafter.yml

@@ -0,0 +1,22 @@
+name-template: 'v$RESOLVED_VERSION'
+tag-template: 'v$RESOLVED_VERSION'
+categories:
+  - title: '🚀 Features'
+    labels:
+      - 'feat'
+  - title: '🐛 Bug Fixes'
+    labels:
+      - 'fix'
+  - title: '📦 Code Refactoring'
+    labels:
+      - 'refactor'
+  - title: '📄 Documentation'
+    labels:
+      - 'docs'
+  - title: '🧰 Maintenance'
+    label: 'chore'
+change-template: '- $TITLE @$AUTHOR (#$NUMBER)'
+template: |
+  ## Changes
+
+  $CHANGES

.github/workflows/docker.yml

@@ -32,7 +32,7 @@ jobs:
             VERSION=${GITHUB_REF#refs/tags/}
             echo "DOCKER_TAGS=ghcr.io/${REPO_LC}:${VERSION},ghcr.io/${REPO_LC}:latest" >> $GITHUB_ENV
           else
-            echo "DOCKER_TAGS=ghcr.io/${REPO_LC}:latest" >> $GITHUB_ENV
+            echo "DOCKER_TAGS=ghcr.io/${REPO_LC}:dev" >> $GITHUB_ENV
           fi
 
       - name: Log in to GitHub Container Registry

.github/workflows/release-drafter.yml

@@ -0,0 +1,17 @@
+name: Release Drafter
+
+on:
+  workflow_dispatch:
+
+jobs:
+  update_release_draft:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      - uses: release-drafter/release-drafter@v5
+        with:
+          config-name: release-drafter.yml
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

CONTRIBUTING.md

@@ -82,7 +82,24 @@
     pip install -r requirements.txt
     ```
 
-4. **启动开发服务器**
+4. **初始化环境**
+
+    在启动服务前，请进行以下准备：
+
+    - **创建数据目录**:
+      在项目根目录执行 `mkdir -p data/db`。这将创建用于存放数据库等文件的目录。
+      > [!IMPORTANT]
+      > 请确保应用拥有对 `data/db` 目录的读写权限。
+
+    - **创建 `.env` 配置文件**:
+      在项目根目录创建名为 `.env` 的文件，并填入以下内容。这些密钥用于保障应用安全，您可以按需修改。
+
+      ```dotenv
+      SECRET_KEY=EnsRhL9NFPxgFVc+7t96/y70DIOR+9SpntcIqQa90TU=
+      TEMP_LINK_SECRET_KEY=EnsRhL9NFPxgFVc+7t96/y70DIOR+9SpntcIqQa90TU=
+      ```
+
+5. **启动开发服务器**
 
     ```bash
     uvicorn main:app --reload --host 0.0.0.0 --port 8000

README.md

@@ -1,8 +1,12 @@
+<div align="right">
+  <b>English</b> | <a href="./README_zh.md">简体中文</a>
+</div>
+
 <div align="center">
 
 # Foxel
 
-**一个面向个人和团队的、高度可扩展的私有云盘解决方案，支持 AI 语义搜索。**
+**A highly extensible private cloud storage solution for individuals and teams, featuring AI-powered semantic search.**
 
 ![Python Version](https://img.shields.io/badge/Python-3.13+-blue.svg)
 ![React](https://img.shields.io/badge/React-19.0-blue.svg)
@@ -11,32 +15,31 @@
 
 ---
   <blockquote>
-    <em><strong>数据之洋浩瀚无涯，当以洞察之目引航，然其脉络深隐，非表象所能尽窥。</strong></em><br>
     <em><strong>The ocean of data is boundless, let the eye of insight guide the voyage, yet its intricate connections lie deep, not fully discernible from the surface.</strong></em>
   </blockquote>
 </div>
 
-## 👀 在线体验
+## 👀 Online Demo
 
 > [https://demo.foxel.cc](https://demo.foxel.cc)
 >
-> 账号/密码：`admin` / `admin`
+> Account/Password: `admin` / `admin`
 
-## ✨ 核心功能
+## ✨ Core Features
 
-- **统一文件管理**：集中管理分布于不同存储后端的文件。
-- **插件化存储后端**：采用可扩展的适配器模式，方便集成多种存储类型。
-- **语义搜索**：支持自然语言描述搜索图片、文档等非结构化数据内容。
-- **内置文件预览**：可直接预览图片、视频、PDF、Office 文档及文本、代码文件，无需下载。
-- **权限与分享**：支持公开或私密分享链接，便于文件共享。
-- **任务处理中心**：支持异步任务处理，如文件索引和数据备份，不影响主应用运行。
+- **Unified File Management**: Centralize management of files distributed across different storage backends.
+- **Pluggable Storage Backends**: Utilizes an extensible adapter pattern to easily integrate various storage types.
+- **Semantic Search**: Supports natural language search for content within unstructured data like images and documents.
+- **Built-in File Preview**: Preview images, videos, PDFs, Office documents, text, and code files directly without downloading.
+- **Permissions and Sharing**: Supports public or private sharing links for easy file distribution.
+- **Task Processing Center**: Supports asynchronous task processing, such as file indexing and data backups, without impacting the main application.
 
-## 🚀 快速开始
+## 🚀 Quick Start
 
-使用 Docker Compose 是启动 Foxel 最推荐的方式。
+Using Docker Compose is the most recommended way to start Foxel.
 
-1. **创建数据目录**:
-新建 `data` 文件夹用于持久化数据：
+1. **Create Data Directories**:
+Create a `data` folder for persistent data:
 
 ```bash
 mkdir -p data/db
@@ -44,40 +47,40 @@ mkdir -p data/mount
 chmod 777 data/db data/mount
 ```
 
-2. **下载 Docker Compose 文件**：
+2. **Download Docker Compose File**:
 
   ```bash
   curl -L -O https://github.com/DrizzleTime/Foxel/raw/main/compose.yaml
   ```
 
-  下载完成后，**强烈建议**修改 `compose.yaml` 文件中的环境变量以确保安全：
+  After downloading, it is **strongly recommended** to modify the environment variables in the `compose.yaml` file to ensure security:
 
-- 修改 `SECRET_KEY` 和 `TEMP_LINK_SECRET_KEY`：将默认的密钥替换为随机生成的强密钥
+- Modify `SECRET_KEY` and `TEMP_LINK_SECRET_KEY`: Replace the default keys with randomly generated strong keys.
 
-3. **启动服务**:
+3. **Start the Services**:
 
   ```bash
   docker-compose up -d
   ```
 
-4. **访问应用**:
+4. **Access the Application**:
 
-  服务启动后，在浏览器中打开页面。
+  Once the services are running, open the page in your browser.
 
-  > 首次启动，请根据引导页面完成管理员账号的初始化设置。
+  > On the first launch, please follow the setup guide to initialize the administrator account.
 
-## 🤝 如何贡献
+## 🤝 How to Contribute
 
-我们非常欢迎来自社区的贡献！无论是提交 Bug、建议新功能还是直接贡献代码。
+We welcome contributions from the community! Whether it's submitting bugs, suggesting new features, or contributing code directly.
 
-在开始之前，请先阅读我们的 [`CONTRIBUTING.md`](CONTRIBUTING.md) 文件，它会指导你如何设置开发环境以及提交流程。
+Before you start, please read our [`CONTRIBUTING.md`](CONTRIBUTING.md) file, which will guide you on how to set up your development environment and the submission process.
 
-## 🌐 社区
+## 🌐 Community
 
-加入我们的交流社区：[Telegram 群组](https://t.me/+thDsBfyqJxZkNTU1)，与开发者和用户一起讨论！
+Join our community on [Telegram](https://t.me/+thDsBfyqJxZkNTU1) to discuss with developers and other users!
 
-你也可以加入我们的微信群，获取更多实时交流与支持。请扫描下方二维码加入：
+You can also join our WeChat group for more real-time communication and support. Please scan the QR code below to join:
 
-<img src="https://foxel.cc/image/wechat.png" alt="微信群二维码" width="180">
+<img src="https://foxel.cc/image/wechat.png" alt="WeChat Group QR Code" width="180">
 
-> 如果二维码失效，请添加微信号 **drizzle2001**，我们会邀请你加入群聊。
+> If the QR code is invalid, please add WeChat ID **drizzle2001**, and we will invite you to the group.

README_zh.md

@@ -0,0 +1,87 @@
+<div align="right">
+  <a href="./README.md">English</a> | <b>简体中文</b>
+</div>
+
+<div align="center">
+
+# Foxel
+
+**一个面向个人和团队的、高度可扩展的私有云盘解决方案，支持 AI 语义搜索。**
+
+![Python Version](https://img.shields.io/badge/Python-3.13+-blue.svg)
+![React](https://img.shields.io/badge/React-19.0-blue.svg)
+![License](https://img.shields.io/badge/license-MIT-green.svg)
+![GitHub stars](https://img.shields.io/github/stars/DrizzleTime/foxel?style=social)
+
+---
+  <blockquote>
+    <em><strong>数据之洋浩瀚无涯，当以洞察之目引航，然其脉络深隐，非表象所能尽窥。</strong></em><br>
+    <em><strong>The ocean of data is boundless, let the eye of insight guide the voyage, yet its intricate connections lie deep, not fully discernible from the surface.</strong></em>
+  </blockquote>
+</div>
+
+## 👀 在线体验
+
+> [https://demo.foxel.cc](https://demo.foxel.cc)
+>
+> 账号/密码：`admin` / `admin`
+
+## ✨ 核心功能
+
+- **统一文件管理**：集中管理分布于不同存储后端的文件。
+- **插件化存储后端**：采用可扩展的适配器模式，方便集成多种存储类型。
+- **语义搜索**：支持自然语言描述搜索图片、文档等非结构化数据内容。
+- **内置文件预览**：可直接预览图片、视频、PDF、Office 文档及文本、代码文件，无需下载。
+- **权限与分享**：支持公开或私密分享链接，便于文件共享。
+- **任务处理中心**：支持异步任务处理，如文件索引和数据备份，不影响主应用运行。
+
+## 🚀 快速开始
+
+使用 Docker Compose 是启动 Foxel 最推荐的方式。
+
+1. **创建数据目录**:
+新建 `data` 文件夹用于持久化数据：
+
+```bash
+mkdir -p data/db
+mkdir -p data/mount
+chmod 777 data/db data/mount
+```
+
+2. **下载 Docker Compose 文件**：
+
+  ```bash
+  curl -L -O https://github.com/DrizzleTime/Foxel/raw/main/compose.yaml
+  ```
+
+  下载完成后，**强烈建议**修改 `compose.yaml` 文件中的环境变量以确保安全：
+
+- 修改 `SECRET_KEY` 和 `TEMP_LINK_SECRET_KEY`：将默认的密钥替换为随机生成的强密钥
+
+3. **启动服务**:
+
+  ```bash
+  docker-compose up -d
+  ```
+
+4. **访问应用**:
+
+  服务启动后，在浏览器中打开页面。
+
+  > 首次启动，请根据引导页面完成管理员账号的初始化设置。
+
+## 🤝 如何贡献
+
+我们非常欢迎来自社区的贡献！无论是提交 Bug、建议新功能还是直接贡献代码。
+
+在开始之前，请先阅读我们的 [`CONTRIBUTING.md`](CONTRIBUTING.md) 文件，它会指导你如何设置开发环境以及提交流程。
+
+## 🌐 社区
+
+加入我们的交流社区：[Telegram 群组](https://t.me/+thDsBfyqJxZkNTU1)，与开发者和用户一起讨论！
+
+你也可以加入我们的微信群，获取更多实时交流与支持。请扫描下方二维码加入：
+
+<img src="https://foxel.cc/image/wechat.png" alt="微信群二维码" width="180">
+
+> 如果二维码失效，请添加微信号 **drizzle2001**，我们会邀请你加入群聊。

api/routes/adapters.py

@@ -39,7 +39,7 @@ async def create_adapter(
     data: AdapterCreate,
     current_user: Annotated[User, Depends(get_current_active_user)]
 ):
-    norm_path = AdapterCreate.normalize_mount_path(data.mount_path)
+    norm_path = AdapterCreate.normalize_mount_path(data.path)
     exists = await StorageAdapter.get_or_none(path=norm_path)
     if exists:
         raise HTTPException(400, detail="Mount path already exists")
@@ -108,7 +108,7 @@ async def update_adapter(
     if not rec:
         raise HTTPException(404, detail="Not found")
 
-    norm_path = AdapterCreate.normalize_mount_path(data.mount_path)
+    norm_path = AdapterCreate.normalize_mount_path(data.path)
     existing = await StorageAdapter.get_or_none(path=norm_path)
     if existing and existing.id != adapter_id:
         raise HTTPException(400, detail="Mount path already exists")

api/routes/config.py

@@ -41,7 +41,9 @@ async def get_system_status():
         "version": VERSION,
         "title": await ConfigCenter.get("APP_NAME", "Foxel"),
         "logo": await ConfigCenter.get("APP_LOGO", "/logo.svg"),
-        "is_initialized": await has_users()
+        "is_initialized": await has_users(),
+        "app_domain": await ConfigCenter.get("APP_DOMAIN"),
+        "file_domain": await ConfigCenter.get("FILE_DOMAIN"),
     }
     return success(system_info)
 

api/routes/processors.py

@@ -1,7 +1,7 @@
 from fastapi import APIRouter, Depends, Body
 from typing import Annotated
 from services.processors.registry import get_config_schemas
-from services.virtual_fs import process_file
+from services.task_queue import task_queue_service
 from services.auth import get_current_active_user, User
 from api.response import success
 from pydantic import BaseModel
@@ -21,7 +21,7 @@ async def list_processors(
             "name": meta["name"],
             "supported_exts": meta.get("supported_exts", []),
             "config_schema": meta["config_schema"],
-            "produces_file": meta.get("produces_file", False), 
+            "produces_file": meta.get("produces_file", False),
         })
     return success(out)
 
@@ -40,5 +40,13 @@ async def process_file_with_processor(
     req: ProcessRequest = Body(...)
 ):
     save_to = req.path if req.overwrite else req.save_to
-    result = await process_file(req.path, req.processor_type, req.config, save_to)
-    return success(result)
+    task = await task_queue_service.add_task(
+        "process_file",
+        {
+            "path": req.path,
+            "processor_type": req.processor_type,
+            "config": req.config,
+            "save_to": save_to,
+        },
+    )
+    return success({"task_id": task.id})

api/routes/tasks.py

@@ -6,6 +6,7 @@ from schemas.tasks import AutomationTaskCreate, AutomationTaskUpdate
 from api.response import success
 from services.auth import get_current_active_user, User
 from services.logging import LogService
+from services.task_queue import task_queue_service
 
 router = APIRouter(
     prefix="/api/tasks",
@@ -15,6 +16,25 @@ router = APIRouter(
 )
 
 
+@router.get("/queue")
+async def get_task_queue_status(
+    current_user: Annotated[User, Depends(get_current_active_user)],
+):
+    tasks = task_queue_service.get_all_tasks()
+    return success([task.dict() for task in tasks])
+
+
+@router.get("/queue/{task_id}")
+async def get_task_status(
+    task_id: str,
+    current_user: Annotated[User, Depends(get_current_active_user)],
+):
+    task = task_queue_service.get_task(task_id)
+    if not task:
+        raise HTTPException(status_code=404, detail="Task not found")
+    return success(task.dict())
+
+
 @router.post("/")
 async def create_task(
     task_in: AutomationTaskCreate,

api/routes/virtual_fs.py

@@ -19,6 +19,7 @@ from services.virtual_fs import (
 from services.thumbnail import is_image_filename, get_or_create_thumb, is_raw_filename
 from schemas import MkdirRequest, MoveRequest
 from api.response import success
+from services.config import ConfigCenter
 
 router = APIRouter(prefix='/api/fs', tags=["virtual-fs"])
 
@@ -151,7 +152,13 @@ async def get_temp_link(
     """获取文件的临时公开访问令牌"""
     full_path = '/' + full_path if not full_path.startswith('/') else full_path
     token = await generate_temp_link_token(full_path, expires_in=expires_in)
-    return success({"token": token, "path": full_path})
+    file_domain = await ConfigCenter.get("FILE_DOMAIN")
+    if file_domain:
+        file_domain = file_domain.rstrip('/')
+        url = f"{file_domain}/api/fs/public/{token}"
+    else:
+        url = f"/api/fs/public/{token}"
+    return success({"token": token, "path": full_path, "url": url})
 
 
 @router.get("/public/{token}")

main.py

@@ -8,6 +8,7 @@ from fastapi import FastAPI
 from services.middleware.logging_middleware import LoggingMiddleware
 from services.middleware.exception_handler import global_exception_handler
 from dotenv import load_dotenv
+from services.task_queue import task_queue_service
 
 load_dotenv()
 
@@ -17,9 +18,11 @@ async def lifespan(app: FastAPI):
     await init_db()
     await runtime_registry.refresh()
     await ConfigCenter.set("APP_VERSION", VERSION)
+    await task_queue_service.start_worker()
     try:
         yield
     finally:
+        await task_queue_service.stop_worker()
         await close_db()
 
 

requirements.txt

@@ -1,14 +1,27 @@
+aioboto3==15.1.0
+aiobotocore==2.24.0
+aiofiles==24.1.0
+aiohappyeyeballs==2.6.1
+aiohttp==3.12.15
+aioitertools==0.12.0
+aiosignal==1.4.0
 aiosqlite==0.21.0
 annotated-types==0.7.0
 anyio==4.10.0
+asyncclick==8.2.2.2
+attrs==25.3.0
 bcrypt==4.3.0
+boto3==1.39.11
+botocore==1.39.11
 certifi==2025.8.3
 click==8.2.1
+dictdiffer==0.9.0
 dnspython==2.7.0
 email_validator==2.2.0
 fastapi==0.116.1
 fastapi-cli==0.0.8
 fastapi-cloud-cli==0.1.5
+frozenlist==1.7.0
 grpcio==1.74.0
 h11==0.16.0
 httpcore==1.0.9
@@ -18,14 +31,17 @@ idna==3.10
 imageio==2.37.0
 iso8601==2.1.0
 Jinja2==3.1.6
+jmespath==1.0.1
 markdown-it-py==4.0.0
 MarkupSafe==3.0.2
 mdurl==0.1.2
 milvus-lite==2.5.1
+multidict==6.6.4
 numpy==2.3.2
 pandas==2.3.1
 passlib==1.7.4
 pillow==11.3.0
+propcache==0.3.2
 protobuf==6.32.0
 pyaes==1.6.1
 pyasn1==0.6.1
@@ -46,6 +62,7 @@ rich==14.1.0
 rich-toolkit==0.15.0
 rignore==0.6.4
 rsa==4.9.1
+s3transfer==0.13.1
 sentry-sdk==2.35.0
 setuptools==80.9.0
 shellingham==1.5.4
@@ -65,3 +82,5 @@ uvicorn==0.35.0
 uvloop==0.21.0
 watchfiles==1.1.0
 websockets==15.0.1
+wrapt==1.17.3
+yarl==1.20.1

schemas/adapters.py

@@ -1,5 +1,5 @@
 from typing import Dict, Optional
-from pydantic import BaseModel, Field, validator
+from pydantic import BaseModel, Field, field_validator
 
 
 class AdapterBase(BaseModel):
@@ -7,12 +7,11 @@ class AdapterBase(BaseModel):
     type: str = Field(pattern=r"^[a-zA-Z0-9_]+$")
     config: Dict = Field(default_factory=dict)
     enabled: bool = True
+    path: str = None
     sub_path: Optional[str] = None
 
 
 class AdapterCreate(AdapterBase):
-    mount_path: str
-
     @staticmethod
     def normalize_mount_path(p: str) -> str:
         p = p.strip()
@@ -21,7 +20,7 @@ class AdapterCreate(AdapterBase):
         p = p.rstrip('/')
         return p or '/'
 
-    @validator("mount_path")
+    @field_validator("path")
     def _v_mount(cls, v: str):
         if not v:
             raise ValueError("mount_path required")
@@ -30,7 +29,8 @@ class AdapterCreate(AdapterBase):
 
 class AdapterOut(AdapterBase):
     id: int
-    mount_path: str = Field(alias='path')
+    path: str = None
+    sub_path: Optional[str] = None
 
     class Config:
         from_attributes = True

services/adapters/onedrive.py

@@ -0,0 +1,431 @@
+from __future__ import annotations
+from datetime import datetime, timezone, timedelta
+from typing import List, Dict, Tuple, AsyncIterator
+import httpx
+from fastapi.responses import StreamingResponse
+from fastapi import HTTPException
+from models import StorageAdapter
+
+MS_GRAPH_URL = "https://graph.microsoft.com/v1.0"
+MS_OAUTH_URL = "https://login.microsoftonline.com/common/oauth2/v2.0/token"
+
+
+class OneDriveAdapter:
+    """OneDrive 存储适配器"""
+
+    def __init__(self, record: StorageAdapter):
+        self.record = record
+        cfg = record.config
+        self.client_id = cfg.get("client_id")
+        self.client_secret = cfg.get("client_secret")
+        self.refresh_token = cfg.get("refresh_token")
+        self.root = cfg.get("root", "/").strip("/")
+
+        if not all([self.client_id, self.client_secret, self.refresh_token]):
+            raise ValueError(
+                "OneDrive 适配器需要 client_id, client_secret, 和 refresh_token")
+
+        self._access_token: str | None = None
+        self._token_expiry: datetime | None = None
+
+    def get_effective_root(self, sub_path: str | None) -> str:
+        """
+        获取有效根路径。
+        :param sub_path: 子路径。
+        :return: 完整的有效路径。
+        """
+        if sub_path:
+            return f"/{self.root.strip('/')}/{sub_path.strip('/')}".strip()
+        return f"/{self.root.strip('/')}".strip()
+
+    def _get_api_path(self, rel_path: str) -> str:
+        """
+        将用户可见的相对路径转换为 Graph API 路径段。
+        :param rel_path: 相对路径。
+        :return: Graph API 路径段。
+        """
+        full_path = self.get_effective_root(rel_path).strip('/')
+        if not full_path:
+            return ""
+        return f":/{full_path}"
+
+    async def _get_access_token(self) -> str:
+        """
+        获取或刷新 access token。
+        :return: access token。
+        """
+        if self._access_token and self._token_expiry and datetime.now(timezone.utc) < self._token_expiry:
+            return self._access_token
+
+        data = {
+            "client_id": self.client_id,
+            "client_secret": self.client_secret,
+            "refresh_token": self.refresh_token,
+            "grant_type": "refresh_token",
+        }
+        async with httpx.AsyncClient() as client:
+            resp = await client.post(MS_OAUTH_URL, data=data)
+            resp.raise_for_status()
+            token_data = resp.json()
+            self._access_token = token_data["access_token"]
+            self._token_expiry = datetime.now(
+                timezone.utc) + timedelta(seconds=token_data["expires_in"] - 300)
+            return self._access_token
+
+    async def _request(self, method: str, api_path_segment: str | None = None, *, full_url: str | None = None, **kwargs):
+        """
+        向 Microsoft Graph API 发送请求。
+        :param method: HTTP 方法。
+        :param api_path_segment: API 路径段 (与 full_url 互斥)。
+        :param full_url: 完整的请求 URL (与 api_path_segment 互斥)。
+        :param kwargs: 其他请求参数。
+        :return: 响应对象。
+        """
+        if not ((api_path_segment is not None) ^ (full_url is not None)):
+            raise ValueError("必须提供 api_path_segment 或 full_url 中的一个，且仅一个")
+
+        token = await self._get_access_token()
+        headers = {"Authorization": f"Bearer {token}"}
+        if "headers" in kwargs:
+            headers.update(kwargs.pop("headers"))
+
+        url = full_url if full_url else f"{MS_GRAPH_URL}/me/drive/root{api_path_segment}"
+        async with httpx.AsyncClient() as client:
+            resp = await client.request(method, url, headers=headers, **kwargs)
+            # 如果 token 过期 (401)，刷新并重试一次
+            if resp.status_code == 401:
+                self._access_token = None  # 强制刷新
+                token = await self._get_access_token()
+                headers["Authorization"] = f"Bearer {token}"
+                resp = await client.request(method, url, headers=headers, **kwargs)
+            return resp
+
+    def _format_item(self, item: Dict) -> Dict:
+        """
+        将 Graph API 返回的 item 格式化为统一的格式。
+        :param item: Graph API 返回的 item 字典。
+        :return: 格式化后的字典。
+        """
+        is_dir = "folder" in item
+        return {
+            "name": item["name"],
+            "is_dir": is_dir,
+            "size": 0 if is_dir else item.get("size", 0),
+            "mtime": int(datetime.fromisoformat(item["lastModifiedDateTime"].replace("Z", "+00:00")).timestamp()),
+            "type": "dir" if is_dir else "file",
+        }
+
+    async def list_dir(self, root: str, rel: str, page_num: int = 1, page_size: int = 50) -> Tuple[List[Dict], int]:
+        """
+        列出目录内容。
+        :param root: 根路径 (在此适配器中未使用，通过配置的 root 确定)。
+        :param rel: 相对路径。
+        :param page_num: 页码。
+        :param page_size: 每页大小。
+        :return: 文件/目录列表和总数。
+        """
+        api_path = self._get_api_path(rel)
+        children_path = f"{api_path}:/children" if api_path else "/children"
+
+        # Graph API 的分页是基于 @odata.nextLink token 的。
+        # 为了支持自定义排序（文件夹在前），我们必须获取所有项目，
+        # 然后在内存中进行排序和分页。此版本通过处理分页链接来稳健地获取所有项目。
+        all_items = []
+
+        # 初始请求
+        resp = await self._request("GET", api_path_segment=children_path, params={"$top": 200})
+
+        while True:
+            if resp.status_code == 404 and not all_items:
+                return [], 0
+            resp.raise_for_status()
+
+            try:
+                data = resp.json()
+            except Exception as e:
+                raise IOError(f"解析 Graph API 响应失败: {e}") from e
+
+            all_items.extend(data.get("value", []))
+            next_link = data.get("@odata.nextLink")
+
+            if not next_link:
+                break
+
+            # 后续分页请求
+            resp = await self._request("GET", full_url=next_link)
+
+        formatted_items = [self._format_item(item) for item in all_items]
+        # 排序：文件夹在前，然后按名称排序
+        formatted_items.sort(key=lambda x: (
+            not x["is_dir"], x["name"].lower()))
+
+        total_count = len(formatted_items)
+        start_idx = (page_num - 1) * page_size
+        end_idx = start_idx + page_size
+
+        return formatted_items[start_idx:end_idx], total_count
+
+    async def read_file(self, root: str, rel: str) -> bytes:
+        """
+        读取文件内容。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :return: 文件内容的字节流。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            raise IsADirectoryError("不能将根目录作为文件读取")
+
+        resp = await self._request("GET", api_path_segment=f"{api_path}:/content")
+        if resp.status_code == 404:
+            raise FileNotFoundError(rel)
+        resp.raise_for_status()
+        return resp.content
+
+    async def write_file(self, root: str, rel: str, data: bytes):
+        """
+        写入文件。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :param data: 文件内容的字节流。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            raise ValueError("不能直接写入根路径")
+        resp = await self._request("PUT", api_path_segment=f"{api_path}:/content", content=data)
+        resp.raise_for_status()
+
+    async def write_file_stream(self, root: str, rel: str, data_iter: AsyncIterator[bytes]):
+        """
+        以流式方式写入文件。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :param data_iter: 文件内容的异步迭代器。
+        :return: 文件大小。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            raise ValueError("不能直接写入根路径")
+
+        resp = await self._request("PUT", api_path_segment=f"{api_path}:/content", content=data_iter)
+        resp.raise_for_status()
+        return resp.json().get("size", 0)
+
+    async def mkdir(self, root: str, rel: str):
+        """
+        创建目录。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        """
+        parent_path_str, new_dir_name = rel.rstrip(
+            '/').rsplit('/', 1) if '/' in rel.rstrip('/') else ('', rel)
+        parent_api_path = self._get_api_path(parent_path_str)
+
+        children_path = f"{parent_api_path}:/children" if parent_api_path else "/children"
+
+        payload = {
+            "name": new_dir_name,
+            "folder": {},
+            "@microsoft.graph.conflictBehavior": "fail"  # 如果已存在则失败
+        }
+        resp = await self._request("POST", api_path_segment=children_path, json=payload)
+        resp.raise_for_status()
+
+    async def delete(self, root: str, rel: str):
+        """
+        删除文件或目录。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            raise ValueError("不能删除根目录")
+
+        resp = await self._request("DELETE", api_path_segment=api_path)
+        if resp.status_code not in (204, 404):
+            resp.raise_for_status()
+
+    async def move(self, root: str, src_rel: str, dst_rel: str):
+        """
+        移动或重命名文件/目录。
+        :param root: 根路径。
+        :param src_rel: 源相对路径。
+        :param dst_rel: 目标相对路径。
+        """
+        src_api_path = self._get_api_path(src_rel)
+        if not src_api_path:
+            raise ValueError("不能移动根目录")
+
+        dst_parent_rel, dst_name = dst_rel.rstrip(
+            '/').rsplit('/', 1) if '/' in dst_rel.rstrip('/') else ('', dst_rel)
+        dst_parent_api_path = self._get_api_path(dst_parent_rel)
+
+        # 获取父项目的 ID
+        parent_resp = await self._request("GET", api_path_segment=dst_parent_api_path)
+        parent_resp.raise_for_status()
+        parent_id = parent_resp.json()["id"]
+
+        payload = {
+            "parentReference": {"id": parent_id},
+            "name": dst_name
+        }
+        resp = await self._request("PATCH", api_path_segment=src_api_path, json=payload)
+        resp.raise_for_status()
+
+    async def rename(self, root: str, src_rel: str, dst_rel: str):
+        """
+        重命名文件或目录。
+        在 Graph API 中，移动和重命名是同一个 PATCH 操作。
+        """
+        await self.move(root, src_rel, dst_rel)
+
+    async def copy(self, root: str, src_rel: str, dst_rel: str, overwrite: bool = False):
+        """
+        复制文件或目录。
+        :param root: 根路径。
+        :param src_rel: 源相对路径。
+        :param dst_rel: 目标相对路径。
+        :param overwrite: 是否覆盖 (在此 API 中未直接使用)。
+        """
+        src_api_path = self._get_api_path(src_rel)
+        if not src_api_path:
+            raise ValueError("不能复制根目录")
+
+        dst_parent_rel, dst_name = dst_rel.rstrip(
+            '/').rsplit('/', 1) if '/' in dst_rel.rstrip('/') else ('', dst_rel)
+        dst_parent_api_path = self._get_api_path(dst_parent_rel)
+
+        parent_resp = await self._request("GET", api_path_segment=dst_parent_api_path)
+        parent_resp.raise_for_status()
+        parent_id = parent_resp.json()["id"]
+
+        payload = {"parentReference": {"id": parent_id}, "name": dst_name}
+        copy_path = f"{src_api_path}:/copy"
+        resp = await self._request("POST", api_path_segment=copy_path, json=payload)
+        resp.raise_for_status()
+
+    async def stream_file(self, root: str, rel: str, range_header: str | None):
+        """
+        流式传输文件（支持范围请求）。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :param range_header: HTTP Range 头。
+        :return: FastAPI StreamingResponse 对象。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            raise IsADirectoryError("不能对目录进行流式传输")
+
+        resp = await self._request("GET", api_path_segment=api_path)
+        if resp.status_code == 404:
+            raise FileNotFoundError(rel)
+        resp.raise_for_status()
+        item_data = resp.json()
+
+        download_url = item_data.get("@microsoft.graph.downloadUrl")
+        if not download_url:
+            raise Exception("无法获取下载 URL")
+
+        file_size = item_data.get("size", 0)
+        content_type = item_data.get("file", {}).get(
+            "mimeType", "application/octet-stream")
+
+        start = 0
+        end = file_size - 1
+        status = 200
+        headers = {
+            "Accept-Ranges": "bytes",
+            "Content-Type": content_type,
+            "Content-Disposition": f"inline; filename=\"{item_data.get('name')}\""
+        }
+
+        if range_header and range_header.startswith("bytes="):
+            try:
+                part = range_header.removeprefix("bytes=")
+                s, e = part.split("-", 1)
+                if s.strip():
+                    start = int(s)
+                if e.strip():
+                    end = int(e)
+                if start >= file_size:
+                    raise HTTPException(416, "Requested Range Not Satisfiable")
+                if end >= file_size:
+                    end = file_size - 1
+                status = 206
+            except ValueError:
+                raise HTTPException(400, "Invalid Range header")
+
+            headers["Content-Range"] = f"bytes {start}-{end}/{file_size}"
+            headers["Content-Length"] = str(end - start + 1)
+        else:
+            headers["Content-Length"] = str(file_size)
+
+        async def file_iterator():
+            nonlocal start, end
+            async with httpx.AsyncClient() as client:
+                req_headers = {'Range': f'bytes={start}-{end}'}
+                async with client.stream("GET", download_url, headers=req_headers) as stream_resp:
+                    stream_resp.raise_for_status()
+                    async for chunk in stream_resp.aiter_bytes():
+                        yield chunk
+
+        return StreamingResponse(file_iterator(), status_code=status, headers=headers, media_type=content_type)
+
+    async def get_thumbnail(self, root: str, rel: str, size: str = "medium"):
+        """
+        获取文件的缩略图。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :param size: 缩略图大小 (large, medium, small)。
+        :return: 缩略图内容的字节流，或在不支持时返回 None。
+        """
+        api_path = self._get_api_path(rel)
+        if not api_path:
+            return None
+
+        thumb_path = f"{api_path}:/thumbnails/0/{size}"
+
+        try:
+            resp = await self._request("GET", api_path_segment=thumb_path)
+            if resp.status_code == 200:
+                thumb_data = resp.json()
+                async with httpx.AsyncClient() as client:
+                    thumb_resp = await client.get(thumb_data['url'])
+                    thumb_resp.raise_for_status()
+                    return thumb_resp.content
+            elif resp.status_code == 404:
+                return None
+            else:
+                resp.raise_for_status()
+        except Exception:
+            return None
+
+    async def stat_file(self, root: str, rel: str):
+        """
+        获取文件或目录的元数据。
+        :param root: 根路径。
+        :param rel: 相对路径。
+        :return: 格式化后的文件/目录信息。
+        """
+        api_path = self._get_api_path(rel)
+        resp = await self._request("GET", api_path_segment=api_path)
+        if resp.status_code == 404:
+            raise FileNotFoundError(rel)
+        resp.raise_for_status()
+        return self._format_item(resp.json())
+
+
+ADAPTER_TYPE = "OneDrive"
+
+CONFIG_SCHEMA = [
+    {"key": "client_id", "label": "Client ID", "type": "string", "required": True},
+    {"key": "client_secret", "label": "Client Secret",
+        "type": "password", "required": True},
+    {"key": "refresh_token", "label": "Refresh Token", "type": "password",
+        "required": True, "help_text": "可以通过运行 'python -m services.adapters.onedrive' 获取"},
+    {"key": "root", "label": "根目录 (Root Path)", "type": "string",
+     "required": False, "placeholder": "默认为根目录 /"},
+]
+
+
+def ADAPTER_FACTORY(rec): return OneDriveAdapter(rec)

services/adapters/s3.py

@@ -0,0 +1,366 @@
+from __future__ import annotations
+import asyncio
+import mimetypes
+from datetime import datetime
+from typing import List, Dict, Tuple, AsyncIterator
+from urllib.parse import quote
+
+import aioboto3
+from botocore.exceptions import ClientError
+from fastapi import HTTPException
+from fastapi.responses import StreamingResponse
+from models import StorageAdapter
+from services.logging import LogService
+
+
+class S3Adapter:
+    """S3 兼容对象存储适配器"""
+
+    def __init__(self, record: StorageAdapter):
+        self.record = record
+        cfg = record.config
+        self.bucket_name = cfg.get("bucket_name")
+        self.aws_access_key_id = cfg.get("access_key_id")
+        self.aws_secret_access_key = cfg.get("secret_access_key")
+        self.region_name = cfg.get("region_name")
+        self.endpoint_url = cfg.get("endpoint_url")
+        self.root = cfg.get("root", "").strip("/")
+
+        if not all([self.bucket_name, self.aws_access_key_id, self.aws_secret_access_key]):
+            raise ValueError(
+                "S3 适配器需要 bucket_name, access_key_id, 和 secret_access_key")
+
+        self.session = aioboto3.Session(
+            aws_access_key_id=self.aws_access_key_id,
+            aws_secret_access_key=self.aws_secret_access_key,
+            region_name=self.region_name,
+        )
+
+    def get_effective_root(self, sub_path: str | None) -> str:
+        """获取 S3 中的有效根路径 (key prefix)"""
+        if sub_path:
+            return f"{self.root}/{sub_path.strip('/')}".strip("/")
+        return self.root
+
+    def _get_s3_key(self, rel_path: str) -> str:
+        """将相对路径转换为 S3 key"""
+        rel_path = rel_path.strip("/")
+        if self.root:
+            return f"{self.root}/{rel_path}"
+        return rel_path
+
+    def _get_client(self):
+        return self.session.client("s3", endpoint_url=self.endpoint_url)
+
+    async def list_dir(self, root: str, rel: str, page_num: int = 1, page_size: int = 50) -> Tuple[List[Dict], int]:
+        prefix = self._get_s3_key(rel)
+        if prefix and not prefix.endswith("/"):
+            prefix += "/"
+
+        all_items = []
+
+        async with self._get_client() as s3:
+            paginator = s3.get_paginator("list_objects_v2")
+            async for result in paginator.paginate(Bucket=self.bucket_name, Prefix=prefix, Delimiter="/"):
+                # 添加子目录
+                for common_prefix in result.get("CommonPrefixes", []):
+                    dir_name = common_prefix.get(
+                        "Prefix").removeprefix(prefix).strip("/")
+                    if dir_name:
+                        all_items.append({
+                            "name": dir_name,
+                            "is_dir": True,
+                            "size": 0,
+                            "mtime": 0,
+                            "type": "dir",
+                        })
+
+                # 添加文件
+                for content in result.get("Contents", []):
+                    file_key = content.get("Key")
+                    if file_key == prefix:  # 忽略目录本身
+                        continue
+                    file_name = file_key.removeprefix(prefix)
+                    if file_name:
+                        all_items.append({
+                            "name": file_name,
+                            "is_dir": False,
+                            "size": content.get("Size", 0),
+                            "mtime": int(content.get("LastModified", datetime.now()).timestamp()),
+                            "type": "file",
+                        })
+
+        # 在内存中排序和分页
+        all_items.sort(key=lambda x: (not x["is_dir"], x["name"].lower()))
+        total_count = len(all_items)
+        start_idx = (page_num - 1) * page_size
+        end_idx = start_idx + page_size
+
+        return all_items[start_idx:end_idx], total_count
+
+    async def read_file(self, root: str, rel: str) -> bytes:
+        key = self._get_s3_key(rel)
+        async with self._get_client() as s3:
+            try:
+                resp = await s3.get_object(Bucket=self.bucket_name, Key=key)
+                return await resp["Body"].read()
+            except ClientError as e:
+                if e.response["Error"]["Code"] == "NoSuchKey":
+                    raise FileNotFoundError(rel)
+                raise
+
+    async def write_file(self, root: str, rel: str, data: bytes):
+        key = self._get_s3_key(rel)
+        async with self._get_client() as s3:
+            await s3.put_object(Bucket=self.bucket_name, Key=key, Body=data)
+            await LogService.info(
+                "adapter:s3", f"Wrote file to {rel}",
+                details={"adapter_id": self.record.id,
+                         "bucket": self.bucket_name, "key": key, "size": len(data)}
+            )
+
+    async def write_file_stream(self, root: str, rel: str, data_iter: AsyncIterator[bytes]):
+        key = self._get_s3_key(rel)
+        MIN_PART_SIZE = 5 * 1024 * 1024
+        
+        async with self._get_client() as s3:
+            mpu = await s3.create_multipart_upload(Bucket=self.bucket_name, Key=key)
+            upload_id = mpu['UploadId']
+            
+            parts = []
+            part_number = 1
+            total_size = 0
+            buffer = bytearray()
+            
+            try:
+                async for chunk in data_iter:
+                    if not chunk:
+                        continue
+                    buffer.extend(chunk)
+                    
+                    while len(buffer) >= MIN_PART_SIZE:
+                        part_data = buffer[:MIN_PART_SIZE]
+                        del buffer[:MIN_PART_SIZE]
+                        
+                        part = await s3.upload_part(
+                            Bucket=self.bucket_name,
+                            Key=key,
+                            PartNumber=part_number,
+                            UploadId=upload_id,
+                            Body=part_data
+                        )
+                        
+                        parts.append({'PartNumber': part_number, 'ETag': part['ETag']})
+                        total_size += len(part_data)
+                        part_number += 1
+
+                if buffer:
+                    part = await s3.upload_part(
+                        Bucket=self.bucket_name,
+                        Key=key,
+                        PartNumber=part_number,
+                        UploadId=upload_id,
+                        Body=bytes(buffer)
+                    )
+                    parts.append({'PartNumber': part_number, 'ETag': part['ETag']})
+                    total_size += len(buffer)
+                
+                await s3.complete_multipart_upload(
+                    Bucket=self.bucket_name,
+                    Key=key,
+                    UploadId=upload_id,
+                    MultipartUpload={'Parts': parts}
+                )
+            except Exception as e:
+                await s3.abort_multipart_upload(
+                    Bucket=self.bucket_name,
+                    Key=key,
+                    UploadId=upload_id
+                )
+                raise IOError(f"S3 stream upload failed: {e}") from e
+
+        await LogService.info(
+            "adapter:s3", f"Wrote file stream to {rel}",
+            details={"adapter_id": self.record.id, "bucket": self.bucket_name, "key": key, "size": total_size}
+        )
+        return total_size
+
+    async def mkdir(self, root: str, rel: str):
+        key = self._get_s3_key(rel)
+        if not key.endswith("/"):
+            key += "/"
+        async with self._get_client() as s3:
+            await s3.put_object(Bucket=self.bucket_name, Key=key, Body=b"")
+            await LogService.info(
+                "adapter:s3", f"Created directory {rel}",
+                details={"adapter_id": self.record.id,
+                         "bucket": self.bucket_name, "key": key}
+            )
+
+    async def delete(self, root: str, rel: str):
+        key = self._get_s3_key(rel)
+        async with self._get_client() as s3:
+            is_dir_like = False
+            try:
+                head = await s3.head_object(Bucket=self.bucket_name, Key=key)
+                if head['ContentLength'] == 0 and key.endswith('/'):
+                    is_dir_like = True
+            except ClientError as e:
+                if e.response['Error']['Code'] != '404': 
+                    raise
+
+            # 如果是目录，删除目录下的所有对象
+            if is_dir_like or not await self.stat_file(root, rel):  
+                dir_key = key if key.endswith('/') else key + '/'
+                paginator = s3.get_paginator("list_objects_v2")
+                objects_to_delete = []
+                async for result in paginator.paginate(Bucket=self.bucket_name, Prefix=dir_key):
+                    for content in result.get("Contents", []):
+                        objects_to_delete.append({"Key": content["Key"]})
+                if objects_to_delete:
+                    await s3.delete_objects(Bucket=self.bucket_name, Delete={"Objects": objects_to_delete})
+            # 如果是文件，直接删除
+            else:
+                await s3.delete_object(Bucket=self.bucket_name, Key=key)
+
+            await LogService.info(
+                "adapter:s3", f"Deleted {rel}",
+                details={"adapter_id": self.record.id,
+                         "bucket": self.bucket_name, "key": key}
+            )
+
+    async def move(self, root: str, src_rel: str, dst_rel: str):
+        await self.copy(root, src_rel, dst_rel, overwrite=True)
+        await self.delete(root, src_rel)
+        await LogService.info(
+            "adapter:s3", f"Moved {src_rel} to {dst_rel}",
+            details={"adapter_id": self.record.id, "bucket": self.bucket_name,
+                     "src_key": self._get_s3_key(src_rel), "dst_key": self._get_s3_key(dst_rel)}
+        )
+
+    async def rename(self, root: str, src_rel: str, dst_rel: str):
+        await self.move(root, src_rel, dst_rel)
+
+    async def copy(self, root: str, src_rel: str, dst_rel: str, overwrite: bool = False):
+        src_key = self._get_s3_key(src_rel)
+        dst_key = self._get_s3_key(dst_rel)
+
+        async with self._get_client() as s3:
+            if not overwrite:
+                try:
+                    await s3.head_object(Bucket=self.bucket_name, Key=dst_key)
+                    raise FileExistsError(dst_rel)
+                except ClientError as e:
+                    if e.response["Error"]["Code"] != "404":
+                        raise
+
+            copy_source = {"Bucket": self.bucket_name, "Key": src_key}
+            await s3.copy_object(CopySource=copy_source, Bucket=self.bucket_name, Key=dst_key)
+            await LogService.info(
+                "adapter:s3", f"Copied {src_rel} to {dst_rel}",
+                details={"adapter_id": self.record.id, "bucket": self.bucket_name,
+                         "src_key": src_key, "dst_key": dst_key}
+            )
+
+    async def stat_file(self, root: str, rel: str):
+        key = self._get_s3_key(rel)
+        async with self._get_client() as s3:
+            try:
+                head = await s3.head_object(Bucket=self.bucket_name, Key=key)
+                return {
+                    "name": rel.split("/")[-1],
+                    "is_dir": False,
+                    "size": head["ContentLength"],
+                    "mtime": int(head["LastModified"].timestamp()),
+                    "type": "file",
+                }
+            except ClientError as e:
+                if e.response["Error"]["Code"] == "404":
+                    # 检查是否为一个 "目录"
+                    dir_key = key if key.endswith('/') else key + '/'
+                    resp = await s3.list_objects_v2(Bucket=self.bucket_name, Prefix=dir_key, MaxKeys=1)
+                    if resp.get('KeyCount', 0) > 0:
+                        return {
+                            "name": rel.split("/")[-1],
+                            "is_dir": True,
+                            "size": 0,
+                            "mtime": 0, 
+                            "type": "dir",
+                        }
+                    raise FileNotFoundError(rel)
+                raise
+
+    async def stream_file(self, root: str, rel: str, range_header: str | None):
+        key = self._get_s3_key(rel)
+        async with self._get_client() as s3:
+            try:
+                head = await s3.head_object(Bucket=self.bucket_name, Key=key)
+                file_size = head["ContentLength"]
+                content_type = head.get("ContentType", mimetypes.guess_type(key)[
+                                        0] or "application/octet-stream")
+            except ClientError as e:
+                if e.response["Error"]["Code"] == "404":
+                    raise HTTPException(
+                        status_code=404, detail="File not found")
+                raise
+
+            start = 0
+            end = file_size - 1
+            status = 200
+            headers = {
+                "Accept-Ranges": "bytes",
+                "Content-Type": content_type,
+                "Content-Length": str(file_size),
+                "Content-Disposition": f"inline; filename=\"{quote(rel.split('/')[-1])}\""
+            }
+
+            if range_header:
+                range_val = range_header.strip().partition("=")[2]
+                s, _, e = range_val.partition("-")
+                try:
+                    start = int(s) if s else 0
+                    end = int(e) if e else file_size - 1
+                    if start >= file_size or end >= file_size or start > end:
+                        raise HTTPException(
+                            status_code=416, detail="Requested Range Not Satisfiable")
+                    status = 206
+                    headers["Content-Length"] = str(end - start + 1)
+                    headers["Content-Range"] = f"bytes {start}-{end}/{file_size}"
+                except ValueError:
+                    raise HTTPException(
+                        status_code=400, detail="Invalid Range header")
+
+            range_arg = f"bytes={start}-{end}"
+
+            async def iterator():
+                try:
+                    resp = await s3.get_object(Bucket=self.bucket_name, Key=key, Range=range_arg)
+                    body = resp["Body"]
+                    while chunk := await body.read(65536):
+                        yield chunk
+                except Exception as e:
+                    LogService.error(
+                        "adapter:s3", f"Error streaming file {key}: {e}")
+
+        return StreamingResponse(iterator(), status_code=status, headers=headers, media_type=content_type)
+
+
+ADAPTER_TYPE = "S3"
+
+CONFIG_SCHEMA = [
+    {"key": "bucket_name", "label": "Bucket 名称",
+        "type": "string", "required": True},
+    {"key": "access_key_id", "label": "Access Key ID",
+        "type": "string", "required": True},
+    {"key": "secret_access_key", "label": "Secret Access Key",
+        "type": "password", "required": True},
+    {"key": "region_name", "label": "区域 (Region)", "type": "string",
+     "required": False, "placeholder": "例如 us-east-1"},
+    {"key": "endpoint_url", "label": "Endpoint URL", "type": "string",
+        "required": False, "placeholder": "对于 S3 兼容存储, 例如 https://minio.example.com"},
+    {"key": "root", "label": "根路径 (Root Path)", "type": "string",
+     "required": False, "placeholder": "在 bucket 内的路径前缀"},
+]
+
+
+def ADAPTER_FACTORY(rec): return S3Adapter(rec)

services/adapters/telegram.py

@@ -0,0 +1,319 @@
+from __future__ import annotations
+from typing import List, Dict, Tuple, AsyncIterator
+import io
+import os
+from models import StorageAdapter
+from telethon import TelegramClient
+from telethon.sessions import StringSession
+import socks
+
+# 适配器类型标识
+ADAPTER_TYPE = "Telegram"
+
+# 适配器配置项定义
+CONFIG_SCHEMA = [
+    {"key": "api_id", "label": "API ID", "type": "string", "required": True, "help_text": "从 my.telegram.org 获取"},
+    {"key": "api_hash", "label": "API Hash", "type": "password", "required": True, "help_text": "从 my.telegram.org 获取"},
+    {"key": "session_string", "label": "Session String", "type": "password", "required": True, "help_text": "通过 generate_session.py 生成"},
+    {"key": "chat_id", "label": "Chat ID", "type": "string", "required": True, "placeholder": "频道/群组的ID或用户名, 例如: -100123456789 或 'channel_username'"},
+    {"key": "proxy_protocol", "label": "代理协议", "type": "string", "required": False, "placeholder": "例如: socks5, http"},
+    {"key": "proxy_host", "label": "代理主机", "type": "string", "required": False, "placeholder": "例如: 127.0.0.1"},
+    {"key": "proxy_port", "label": "代理端口", "type": "number", "required": False, "placeholder": "例如: 1080"},
+]
+
+class TelegramAdapter:
+    """Telegram 存储适配器 (使用用户 Session)"""
+
+    def __init__(self, record: StorageAdapter):
+        self.record = record
+        cfg = record.config
+        self.api_id = int(cfg.get("api_id"))
+        self.api_hash = cfg.get("api_hash")
+        self.session_string = cfg.get("session_string")
+        self.chat_id_str = cfg.get("chat_id")
+        
+        # 代理设置
+        self.proxy_protocol = cfg.get("proxy_protocol")
+        self.proxy_host = cfg.get("proxy_host")
+        self.proxy_port = cfg.get("proxy_port")
+        
+        self.proxy = None
+        if self.proxy_protocol and self.proxy_host and self.proxy_port:
+            proto_map = {
+                "socks5": socks.SOCKS5,
+                "http": socks.HTTP,
+            }
+            proxy_type = proto_map.get(self.proxy_protocol.lower())
+            if proxy_type:
+                self.proxy = (proxy_type, self.proxy_host, int(self.proxy_port))
+
+        try:
+            self.chat_id = int(self.chat_id_str)
+        except (ValueError, TypeError):
+            self.chat_id = self.chat_id_str
+
+        if not all([self.api_id, self.api_hash, self.session_string, self.chat_id]):
+            raise ValueError("Telegram 适配器需要 api_id, api_hash, session_string 和 chat_id")
+
+    def _get_client(self) -> TelegramClient:
+        """创建一个新的 TelegramClient 实例"""
+        return TelegramClient(StringSession(self.session_string), self.api_id, self.api_hash, proxy=self.proxy)
+
+    def get_effective_root(self, sub_path: str | None) -> str:
+        return ""
+
+    async def list_dir(self, root: str, rel: str, page_num: int = 1, page_size: int = 50) -> Tuple[List[Dict], int]:
+        if rel:
+            return [], 0
+
+        client = self._get_client()
+        entries = []
+        try:
+            await client.connect()
+            messages = await client.get_messages(self.chat_id, limit=50)
+            for message in messages:
+                if not message:
+                    continue
+                
+                media = message.document or message.video or message.photo
+                if not media:
+                    continue
+
+                filename = None
+                size = 0
+                
+                if message.photo:
+                    photo_size = message.photo.sizes[-1]
+                    size = photo_size.size if hasattr(photo_size, 'size') else 0
+                    filename = f"photo_{message.id}.jpg"
+
+                elif message.document or message.video:
+                    size = media.size
+                    if hasattr(media, 'attributes'):
+                        for attr in media.attributes:
+                            if hasattr(attr, 'file_name') and attr.file_name:
+                                filename = attr.file_name
+                                break
+                
+                if not filename:
+                    if message.text and '.' in message.text and len(message.text) < 256 and '\n' not in message.text:
+                        filename = message.text
+                
+                if not filename:
+                    filename = f"unknown_{message.id}"
+
+                entries.append({
+                    "name": f"{message.id}_{filename}",
+                    "is_dir": False,
+                    "size": size,
+                    "mtime": int(message.date.timestamp()),
+                    "type": "file",
+                })
+        finally:
+            if client.is_connected():
+                await client.disconnect()
+
+        return entries, len(entries)
+
+    async def read_file(self, root: str, rel: str) -> bytes:
+        try:
+            message_id_str, _ = rel.split('_', 1)
+            message_id = int(message_id_str)
+        except (ValueError, IndexError):
+            raise FileNotFoundError(f"无效的文件路径格式: {rel}")
+
+        client = self._get_client()
+        try:
+            await client.connect()
+            message = await client.get_messages(self.chat_id, ids=message_id)
+            if not message or not (message.document or message.video or message.photo):
+                raise FileNotFoundError(f"在频道 {self.chat_id} 中未找到消息ID为 {message_id} 的文件")
+            
+            file_bytes = await client.download_media(message, file=bytes)
+            return file_bytes
+        finally:
+            if client.is_connected():
+                await client.disconnect()
+
+    async def write_file(self, root: str, rel: str, data: bytes):
+        """将字节数据作为文件上传"""
+        client = self._get_client()
+        file_like = io.BytesIO(data)
+        file_like.name = os.path.basename(rel) or "file"
+
+        try:
+            await client.connect()
+            await client.send_file(self.chat_id, file_like, caption=file_like.name)
+        finally:
+            if client.is_connected():
+                await client.disconnect()
+
+    async def write_file_stream(self, root: str, rel: str, data_iter: AsyncIterator[bytes]):
+        """以流式方式上传文件"""
+        client = self._get_client()
+        filename = os.path.basename(rel) or "file"
+        import tempfile
+        temp_dir = tempfile.gettempdir()
+        temp_path = os.path.join(temp_dir, filename)
+        
+        total_size = 0
+        try:
+            with open(temp_path, "wb") as f:
+                async for chunk in data_iter:
+                    if chunk:
+                        f.write(chunk)
+                        total_size += len(chunk)
+            
+            await client.connect()
+            await client.send_file(self.chat_id, temp_path, caption=filename)
+
+        finally:
+            if os.path.exists(temp_path):
+                os.remove(temp_path)
+            if client.is_connected():
+                await client.disconnect()
+        return total_size
+
+    async def mkdir(self, root: str, rel: str):
+        raise NotImplementedError("Telegram 适配器不支持创建目录。")
+
+    async def delete(self, root: str, rel: str):
+        """删除一个文件 (即一条消息)"""
+        try:
+            message_id_str, _ = rel.split('_', 1)
+            message_id = int(message_id_str)
+        except (ValueError, IndexError):
+            raise FileNotFoundError(f"无效的文件路径格式，无法解析消息ID: {rel}")
+
+        client = self._get_client()
+        try:
+            await client.connect()
+            result = await client.delete_messages(self.chat_id, [message_id])
+            if not result or not result[0].pts:
+                 raise FileNotFoundError(f"在 {self.chat_id} 中删除消息 {message_id} 失败，可能消息不存在或无权限")
+        finally:
+            if client.is_connected():
+                await client.disconnect()
+
+    async def move(self, root: str, src_rel: str, dst_rel: str):
+        raise NotImplementedError("Telegram 适配器不支持移动。")
+
+    async def rename(self, root: str, src_rel: str, dst_rel: str):
+        raise NotImplementedError("Telegram 适配器不支持重命名。")
+
+    async def copy(self, root: str, src_rel: str, dst_rel: str, overwrite: bool = False):
+        raise NotImplementedError("Telegram 适配器不支持复制。")
+
+    async def stream_file(self, root: str, rel: str, range_header: str | None):
+        from fastapi.responses import StreamingResponse
+        from fastapi import HTTPException
+
+        try:
+            message_id_str, _ = rel.split('_', 1)
+            message_id = int(message_id_str)
+        except (ValueError, IndexError):
+            raise HTTPException(status_code=400, detail=f"无效的文件路径格式: {rel}")
+
+        client = self._get_client()
+        
+        try:
+            await client.connect()
+            message = await client.get_messages(self.chat_id, ids=message_id)
+            media = message.document or message.video or message.photo
+            if not message or not media:
+                raise FileNotFoundError(f"在频道 {self.chat_id} 中未找到消息ID为 {message_id} 的文件")
+
+            if message.photo:
+                photo_size = media.sizes[-1]
+                file_size = photo_size.size if hasattr(photo_size, 'size') else 0
+                mime_type = "image/jpeg"
+            else:
+                file_size = media.size
+                mime_type = media.mime_type or "application/octet-stream"
+
+            start = 0
+            end = file_size - 1
+            status = 200
+            
+            headers = {
+                "Accept-Ranges": "bytes",
+                "Content-Type": mime_type,
+                "Content-Length": str(file_size),
+            }
+
+            if range_header:
+                try:
+                    range_val = range_header.strip().partition("=")[2]
+                    s, _, e = range_val.partition("-")
+                    start = int(s) if s else 0
+                    end = int(e) if e else file_size - 1
+                    if start >= file_size or end >= file_size or start > end:
+                        raise HTTPException(status_code=416, detail="Requested Range Not Satisfiable")
+                    status = 206
+                    headers["Content-Length"] = str(end - start + 1)
+                    headers["Content-Range"] = f"bytes {start}-{end}/{file_size}"
+                except ValueError:
+                    raise HTTPException(status_code=400, detail="Invalid Range header")
+
+            async def iterator():
+                try:
+                    limit = end - start + 1
+                    downloaded = 0
+                    
+                    async for chunk in client.iter_download(media, offset=start):
+                        if downloaded + len(chunk) > limit:
+                            yield chunk[:limit - downloaded]
+                            break
+                        yield chunk
+                        downloaded += len(chunk)
+                        if downloaded >= limit:
+                            break
+                finally:
+                    if client.is_connected():
+                        await client.disconnect()
+
+            return StreamingResponse(iterator(), status_code=status, headers=headers)
+
+        except FileNotFoundError as e:
+            if client.is_connected():
+                await client.disconnect()
+            raise HTTPException(status_code=404, detail=str(e))
+        except Exception as e:
+            if client.is_connected():
+                await client.disconnect()
+            raise HTTPException(status_code=500, detail=f"Streaming failed: {str(e)}")
+
+    async def stat_file(self, root: str, rel: str):
+        try:
+            message_id_str, filename = rel.split('_', 1)
+            message_id = int(message_id_str)
+        except (ValueError, IndexError):
+            raise FileNotFoundError(f"无效的文件路径格式: {rel}")
+
+        client = self._get_client()
+        try:
+            await client.connect()
+            message = await client.get_messages(self.chat_id, ids=message_id)
+            media = message.document or message.video or message.photo
+            if not message or not media:
+                raise FileNotFoundError(f"在频道 {self.chat_id} 中未找到消息ID为 {message_id} 的文件")
+
+            if message.photo:
+                photo_size = media.sizes[-1]
+                size = photo_size.size if hasattr(photo_size, 'size') else 0
+            else:
+                size = media.size
+
+            return {
+                "name": rel,
+                "is_dir": False,
+                "size": size,
+                "mtime": int(message.date.timestamp()),
+                "type": "file",
+            }
+        finally:
+            if client.is_connected():
+                await client.disconnect()
+
+def ADAPTER_FACTORY(rec: StorageAdapter) -> TelegramAdapter:
+    return TelegramAdapter(rec)

services/ai.py

@@ -2,13 +2,14 @@ import httpx
 from typing import List
 from services.config import ConfigCenter
 
+
 async def describe_image_base64(base64_image: str, detail: str = "high") -> str:
     """
     传入base64图片和文本提示，返回图片描述文本。
     """
-    OAI_API_URL = await ConfigCenter.get("AI_API_URL", "https://api.siliconflow.cn/v1/chat/completions")
-    VISION_MODEL = await ConfigCenter.get("AI_VISION_MODEL", "Qwen/Qwen2.5-VL-32B-Instruct")
-    API_KEY = await ConfigCenter.get("AI_API_KEY", "")
+    OAI_API_URL = await ConfigCenter.get("AI_VISION_API_URL")
+    VISION_MODEL = await ConfigCenter.get("AI_VISION_MODEL")
+    API_KEY = await ConfigCenter.get("AI_VISION_API_KEY")
     payload = {
         "model": VISION_MODEL,
         "messages": [
@@ -42,13 +43,14 @@ async def describe_image_base64(base64_image: str, detail: str = "high") -> str:
     except Exception as e:
         return f"请求失败: {str(e)}"
 
+
 async def get_text_embedding(text: str) -> List[float]:
     """
     传入文本，返回嵌入向量。
     """
-    OAI_API_URL = await ConfigCenter.get("AI_API_URL", "https://api.siliconflow.cn/v1/chat/completions")
-    EMBED_MODEL = await ConfigCenter.get("AI_EMBED_MODEL", "Qwen/Qwen3-Embedding-8B")
-    API_KEY = await ConfigCenter.get("AI_API_KEY", "")
+    OAI_API_URL = await ConfigCenter.get("AI_EMBED_API_URL")
+    EMBED_MODEL = await ConfigCenter.get("AI_EMBED_MODEL")
+    API_KEY = await ConfigCenter.get("AI_EMBED_API_KEY")
     payload = {
         "model": EMBED_MODEL,
         "input": text
@@ -58,7 +60,11 @@ async def get_text_embedding(text: str) -> List[float]:
         "Content-Type": "application/json"
     }
     async with httpx.AsyncClient() as client:
-        resp = await client.post(OAI_API_URL.replace("chat/completions", "embeddings"), headers=headers, json=payload)
+        if OAI_API_URL.endswith("chat/completions"):
+            url = OAI_API_URL.replace("chat/completions", "embeddings")
+        else:
+            url = OAI_API_URL
+        resp = await client.post(url, headers=headers, json=payload)
         resp.raise_for_status()
         result = resp.json()
         return result["data"][0]["embedding"]

services/config.py

@@ -4,7 +4,7 @@ from typing import Any, Optional, Dict
 from dotenv import load_dotenv
 from models.database import Configuration
 load_dotenv(dotenv_path=".env")
-VERSION = "v1.1.0"
+VERSION = "v1.1.3"
 
 class ConfigCenter:
     _cache: Dict[str, Any] = {}

services/task_queue.py

@@ -0,0 +1,122 @@
+import asyncio
+from typing import Dict, Any
+from pydantic import BaseModel, Field
+import uuid
+from services.logging import LogService
+from enum import Enum
+
+
+class TaskStatus(str, Enum):
+    PENDING = "pending"
+    RUNNING = "running"
+    SUCCESS = "success"
+    FAILED = "failed"
+
+
+class Task(BaseModel):
+    id: str = Field(default_factory=lambda: uuid.uuid4().hex)
+    name: str
+    status: TaskStatus = TaskStatus.PENDING
+    result: Any = None
+    error: str | None = None
+    task_info: Dict[str, Any] = {}
+
+
+class TaskQueueService:
+    def __init__(self):
+        self._queue = asyncio.Queue()
+        self._tasks: Dict[str, Task] = {}
+        self._worker_task: asyncio.Task | None = None
+
+    async def add_task(self, name: str, task_info: Dict[str, Any]) -> Task:
+        task = Task(name=name, task_info=task_info)
+        self._tasks[task.id] = task
+        await self._queue.put(task)
+        await LogService.info("task_queue", f"Task {name} ({task.id}) enqueued", {"task_id": task.id, "name": name})
+        return task
+
+    def get_task(self, task_id: str) -> Task | None:
+        return self._tasks.get(task_id)
+
+    def get_all_tasks(self) -> list[Task]:
+        return list(self._tasks.values())
+
+    async def _execute_task(self, task: Task):
+        from services.virtual_fs import process_file
+
+        task.status = TaskStatus.RUNNING
+        await LogService.info("task_queue", f"Task {task.name} ({task.id}) started", {"task_id": task.id, "name": task.name})
+
+        try:
+            if task.name == "process_file":
+                params = task.task_info
+                result = await process_file(
+                    path=params["path"],
+                    processor_type=params["processor_type"],
+                    config=params["config"],
+                    save_to=params["save_to"]
+                )
+                task.result = result
+            elif task.name == "automation_task":
+                from models.database import AutomationTask
+                from services.processors.registry import get as get_processor
+                from services.virtual_fs import read_file, write_file
+
+                params = task.task_info
+                auto_task = await AutomationTask.get(id=params["task_id"])
+                path = params["path"]
+
+                processor = get_processor(auto_task.processor_type)
+                if not processor:
+                    raise ValueError(f"Processor {auto_task.processor_type} not found for task {auto_task.id}")
+
+                file_content = await read_file(path)
+                result = await processor.process(file_content, path, auto_task.processor_config)
+                
+                save_to = auto_task.processor_config.get("save_to")
+                if save_to and getattr(processor, "produces_file", False):
+                    await write_file(save_to, result)
+                task.result = "Automation task completed"
+            else:
+                raise ValueError(f"Unknown task name: {task.name}")
+            
+            task.status = TaskStatus.SUCCESS
+            await LogService.info("task_queue", f"Task {task.name} ({task.id}) succeeded", {"task_id": task.id, "name": task.name})
+
+        except Exception as e:
+            task.status = TaskStatus.FAILED
+            task.error = str(e)
+            await LogService.error("task_queue", f"Task {task.name} ({task.id}) failed: {e}", {"task_id": task.id, "name": task.name})
+
+    async def worker(self):
+        await LogService.info("task_queue", "Task worker started")
+        while True:
+            try:
+                task = await self._queue.get()
+                await self._execute_task(task)
+            except asyncio.CancelledError:
+                await LogService.info("task_queue", "Task worker stopped")
+                break
+            except Exception as e:
+                await LogService.error("task_queue", f"Error in task worker: {e}", exc_info=True)
+            finally:
+                self._queue.task_done()
+
+    async def start_worker(self):
+        if self._worker_task is None or self._worker_task.done():
+            self._worker_task = asyncio.create_task(self.worker())
+            await LogService.info("task_queue", "Task worker created.")
+
+    async def stop_worker(self):
+        if self._worker_task and not self._worker_task.done():
+            self._worker_task.cancel()
+            try:
+                await self._worker_task
+            except asyncio.CancelledError:
+                pass
+            finally:
+                self._worker_task = None
+                await LogService.info("task_queue", "Task worker has been stopped.")
+
+
+task_queue_service = TaskQueueService()

services/tasks.py

@@ -4,6 +4,9 @@ from models.database import AutomationTask
 from services.processors.registry import get as get_processor
 from services.logging import LogService
 
+from services.task_queue import task_queue_service
+
+
 class TaskService:
     async def trigger_tasks(self, event: str, path: str):
         tasks = await AutomationTask.filter(event=event, enabled=True)
@@ -21,28 +24,12 @@ class TaskService:
         return True
 
     async def execute(self, task: AutomationTask, path: str):
-        from services.virtual_fs import read_file, write_file
-
-        processor = get_processor(task.processor_type)
-        if not processor:
-            print(f"Processor {task.processor_type} not found for task {task.id}")
-            return
-
-        try:
-            file_content = await read_file(path)
-            result = await processor.process(file_content, path, task.processor_config)
-            
-            save_to = task.processor_config.get("save_to")
-            if save_to and getattr(processor, "produces_file", False):
-                await write_file(save_to, result)
-
-        except Exception as e:
-            error_message = f"Error executing task {task.id} for path {path}: {e}"
-            print(error_message)
-            await LogService.error(
-                source=f"task_executor:{task.id}",
-                message=error_message,
-                details={"task_name": task.name, "event": task.event, "path": path, "processor": task.processor_type}
-            )
+        await task_queue_service.add_task(
+            "automation_task",
+            {
+                "task_id": task.id,
+                "path": path,
+            },
+        )
 
 task_service = TaskService()

services/thumbnail.py

@@ -5,7 +5,8 @@ from pathlib import Path
 from typing import Tuple
 from fastapi import HTTPException
 
-ALLOWED_EXT = {"jpg", "jpeg", "png", "webp", "gif", "bmp", "tiff", "arw", "cr2", "cr3", "nef", "rw2", "orf", "pef", "dng"}
+ALLOWED_EXT = {"jpg", "jpeg", "png", "webp", "gif", "bmp",
+               "tiff", "arw", "cr2", "cr3", "nef", "rw2", "orf", "pef", "dng"}
 RAW_EXT = {"arw", "cr2", "cr3", "nef", "rw2", "orf", "pef", "dng"}
 MAX_SOURCE_SIZE = 200 * 1024 * 1024
 CACHE_ROOT = Path('data/.thumb_cache')
@@ -49,11 +50,12 @@ def generate_thumb(data: bytes, w: int, h: int, fit: str, is_raw: bool = False)
                     thumb = raw.extract_thumb()
                 except rawpy.LibRawNoThumbnailError:
                     thumb = None
-                
+
                 if thumb is not None and thumb.format in [rawpy.ThumbFormat.JPEG, rawpy.ThumbFormat.BITMAP]:
                     im = Image.open(io.BytesIO(thumb.data))
                 else:
-                    rgb = raw.postprocess(use_camera_wb=False, use_auto_wb=True, output_bps=8)
+                    rgb = raw.postprocess(
+                        use_camera_wb=False, use_auto_wb=True, output_bps=8)
                     im = Image.fromarray(rgb)
         except Exception as e:
             print(f"rawpy processing failed: {e}")
@@ -87,18 +89,48 @@ def generate_thumb(data: bytes, w: int, h: int, fit: str, is_raw: bool = False)
 async def get_or_create_thumb(adapter, adapter_id: int, root: str, rel: str, w: int, h: int, fit: str = 'cover'):
     stat = await adapter.stat_file(root, rel)
     if stat['size'] > MAX_SOURCE_SIZE:
-         raise HTTPException(400, detail="Image too large for thumbnail")
-    
-    key = _cache_key(adapter_id, rel, stat['size'], int(stat['mtime']), w, h, fit)
+        raise HTTPException(400, detail="Image too large for thumbnail")
+
+    key = _cache_key(adapter_id, rel, stat['size'], int(
+        stat['mtime']), w, h, fit)
     path = _cache_path(key)
     if path.exists():
         return path.read_bytes(), 'image/webp', key
+
     _ensure_cache_dir(path)
-    read_data = await adapter.read_file(root, rel)
-    try:
-        thumb_bytes, mime = generate_thumb(read_data, w, h, fit, is_raw=is_raw_filename(rel))
-    except Exception as e:
-        print(e)
-        raise HTTPException(500, detail=f"Thumbnail generation failed: {e}")
-    path.write_bytes(thumb_bytes)
-    return thumb_bytes, mime, key
+    thumb_bytes, mime = None, None
+
+    get_thumb_impl = getattr(adapter, "get_thumbnail", None)
+    if callable(get_thumb_impl):
+        size_str = "large" if w > 400 else "medium" if w > 100 else "small"
+        native_thumb_bytes = await get_thumb_impl(root, rel, size_str)
+
+        if native_thumb_bytes:
+            try:
+                from PIL import Image
+                im = Image.open(io.BytesIO(native_thumb_bytes))
+                buf = io.BytesIO()
+                im.save(buf, 'WEBP', quality=85)
+                thumb_bytes = buf.getvalue()
+                mime = 'image/webp'
+            except Exception as e:
+                print(
+                    f"Failed to convert native thumbnail to WebP: {e}, falling back.")
+                thumb_bytes, mime = None, None
+
+    if not thumb_bytes:
+        read_data = await adapter.read_file(root, rel)
+        try:
+            thumb_bytes, mime = generate_thumb(
+                read_data, w, h, fit, is_raw=is_raw_filename(rel))
+        except Exception as e:
+            print(e)
+            raise HTTPException(
+                500, detail=f"Thumbnail generation failed: {e}")
+
+    if thumb_bytes:
+        path.write_bytes(thumb_bytes)
+        return thumb_bytes, mime, key
+
+    raise HTTPException(
+        500, detail="Failed to generate thumbnail by any means")

services/virtual_fs.py

@@ -42,7 +42,12 @@ async def resolve_adapter_and_rel(path: str):
         raise e
     adapter_instance = runtime_registry.get(adapter_model.id)
     if not adapter_instance:
-        raise HTTPException(404, detail=f"Adapter instance not found for ID {adapter_model.id}")
+        await runtime_registry.refresh()
+        adapter_instance = runtime_registry.get(adapter_model.id)
+        if not adapter_instance:
+            raise HTTPException(
+                404, detail=f"Adapter instance for ID {adapter_model.id} not found or failed to load."
+            )
     effective_root = adapter_instance.get_effective_root(adapter_model.sub_path)
     return adapter_instance, adapter_model, effective_root, rel
 
@@ -72,7 +77,16 @@ async def list_virtual_dir(path: str, page_num: int = 1, page_size: int = 50) ->
     try:
         adapter_model, rel = await resolve_adapter_by_path(norm)
         adapter_instance = runtime_registry.get(adapter_model.id)
-        effective_root = adapter_instance.get_effective_root(adapter_model.sub_path)
+        if not adapter_instance:
+            await runtime_registry.refresh()
+            adapter_instance = runtime_registry.get(adapter_model.id)
+        
+        if adapter_instance:
+            effective_root = adapter_instance.get_effective_root(adapter_model.sub_path)
+        else:
+            adapter_model = None
+            effective_root = ""
+            rel = ""
     except HTTPException:
         adapter_model = None
         adapter_instance = None

web/bun.lock

@@ -8,6 +8,7 @@
         "@ant-design/v5-patch-for-react-19": "^1.0.3",
         "@uiw/react-md-editor": "^4.0.8",
         "antd": "^5.27.0",
+        "artplayer": "^5.2.5",
         "date-fns": "^4.1.0",
         "react": "^19.1.1",
         "react-dom": "^19.1.1",
@@ -316,6 +317,8 @@
 
     "argparse": ["argparse@2.0.1", "", {}, "sha512-8+9WqebbFzpX9OR+Wa6O29asIogeRMzcGtAINdpMHHyAg10f05aSFVBbcEqGf/PXw1EjAZ+q2/bEBg3DvurK3Q=="],
 
+    "artplayer": ["artplayer@5.2.5", "", { "dependencies": { "option-validator": "^2.0.6" } }, "sha512-Ogym5rvkAJ4VLncM4Apl3TJ/a/ozM3csvY4IKuuMR++hUmEZgj/HaGsNonwx8r56nsqiZYE7O4vS1HFZl+NBSg=="],
+
     "bail": ["bail@2.0.2", "", {}, "sha512-0xO6mYd7JB2YesxDKplafRpsiOzPt9V02ddPCLbY1xYGPOX24NTyN50qnUxgCPcSoYMhKpAuBTjQoRZCAkUDRw=="],
 
     "balanced-match": ["balanced-match@1.0.2", "", {}, "sha512-3oSeUO0TMV67hN1AmbXsK4yaqU7tjiHlbxRDZOpH0KW9+CeX4bRAaX0Anxt0tx2MrpRpWwQaPwIlISEJhYU5Pw=="],
@@ -532,6 +535,8 @@
 
     "keyv": ["keyv@4.5.4", "", { "dependencies": { "json-buffer": "3.0.1" } }, "sha512-oxVHkHR/EJf2CNXnWxRLW6mg7JyCCUcG0DtEGmL2ctUo1PNTin1PUil+r/+4r5MpVgC/fn1kjsx7mjSujKqIpw=="],
 
+    "kind-of": ["kind-of@6.0.3", "", {}, "sha512-dcS1ul+9tmeD95T+x28/ehLgd9mENa3LsvDTtzm3vyBEO7RPptvAD+t44WVXaUjTBRcrpFeFlC8WCruUR456hw=="],
+
     "levn": ["levn@0.4.1", "", { "dependencies": { "prelude-ls": "^1.2.1", "type-check": "~0.4.0" } }, "sha512-+bT2uH4E5LGE7h/n3evcS/sQlJXCpIp6ym8OWJ5eV6+67Dsql/LaaT7qJBAt2rzfoa/5QBGBhxDix1dMt2kQKQ=="],
 
     "locate-path": ["locate-path@6.0.0", "", { "dependencies": { "p-locate": "^5.0.0" } }, "sha512-iPZK6eYjbxRu3uB4/WZ3EsEIMJFMqAoopl3R+zuq0UjcAm/MO6KCweDgPfP3elTztoKP3KtnVHxTn2NHBSDVUw=="],
@@ -646,6 +651,8 @@
 
     "nth-check": ["nth-check@2.1.1", "", { "dependencies": { "boolbase": "^1.0.0" } }, "sha512-lqjrjmaOoAnWfMmBPL+XNnynZh2+swxiX3WUE0s4yEHI6m+AwrK2UZOimIRl3X/4QctVqS8AiZjFqyOGrMXb/w=="],
 
+    "option-validator": ["option-validator@2.0.6", "", { "dependencies": { "kind-of": "^6.0.3" } }, "sha512-tmZDan2LRIRQyhUGvkff68/O0R8UmF+Btmiiz0SmSw2ng3CfPZB9wJlIjHpe/MKUZqyIZkVIXCrwr1tIN+0Dzg=="],
+
     "optionator": ["optionator@0.9.4", "", { "dependencies": { "deep-is": "^0.1.3", "fast-levenshtein": "^2.0.6", "levn": "^0.4.1", "prelude-ls": "^1.2.1", "type-check": "^0.4.0", "word-wrap": "^1.2.5" } }, "sha512-6IpQ7mKUxRcZNLIObR0hz7lxsapSSIYNZJwXPGeF0mTVqGKFIXj1DQcMoT22S3ROcLyY/rz0PWaWZ9ayWmad9g=="],
 
     "p-limit": ["p-limit@3.1.0", "", { "dependencies": { "yocto-queue": "^0.1.0" } }, "sha512-TYOanM3wGwNGsZN2cVTYPArw454xnXj5qmWF1bEoAc4+cU/ol7GVh7odevjp1FNHduHc3KZMcFduxU5Xc6uJRQ=="],

web/package.json

@@ -14,6 +14,7 @@
     "@ant-design/v5-patch-for-react-19": "^1.0.3",
     "@uiw/react-md-editor": "^4.0.8",
     "antd": "^5.27.0",
+    "artplayer": "^5.2.5",
     "date-fns": "^4.1.0",
     "react": "^19.1.1",
     "react-dom": "^19.1.1",

web/src/api/adapters.ts

@@ -6,7 +6,7 @@ export interface AdapterItem {
   type: string;
   config: any;
   enabled: boolean;
-  mount_path?: string | null;
+  path?: string | null;
   sub_path?: string | null;
 }
 

web/src/api/config.ts

@@ -20,6 +20,8 @@ export interface SystemStatus {
   title: string;
   logo: string;
   is_initialized: boolean;
+  app_domain?: string;
+  file_domain?: string;
 }
 
 export async function status() {

web/src/api/vfs.ts

@@ -51,7 +51,7 @@ export const vfsApi = {
   streamUrl: (path: string) => `${API_BASE_URL}/fs/stream/${encodeURI(path.replace(/^\/+/, ''))}`,
   stat: (path: string) => request(`/fs/stat/${encodeURI(path.replace(/^\/+/, ''))}`),
   getTempLinkToken: (path: string, expiresIn: number = 3600) =>
-    request<{token: string}>(`/fs/temp-link/${encodeURI(path.replace(/^\/+/, ''))}?expires_in=${expiresIn}`),
+    request<{token: string, path: string, url: string}>(`/fs/temp-link/${encodeURI(path.replace(/^\/+/, ''))}?expires_in=${expiresIn}`),
   getTempPublicUrl: (token: string) => `${API_BASE_URL}/fs/public/${token}`,
   uploadStream: (fullPath: string, file: File, overwrite: boolean = true, onProgress?: (loaded: number, total: number) => void) => {
     const enc = encodeURI(fullPath.replace(/^\/+/, ''));

web/src/apps/OfficeViewer/OfficeViewer.tsx

@@ -2,8 +2,10 @@ import React, { useEffect, useState } from 'react';
 import { vfsApi } from '../../api/client';
 import type { AppComponentProps } from '../types';
 import { Spin, Result, Button } from 'antd';
+import { useSystemStatus } from '../../contexts/SystemContext';
 
 export const OfficeViewerApp: React.FC<AppComponentProps> = ({ filePath, onRequestClose }) => {
+  const systemStatus = useSystemStatus();
   const [url, setUrl] = useState<string>();
   const [loading, setLoading] = useState(true);
   const [err, setErr] = useState<string>();
@@ -17,8 +19,8 @@ export const OfficeViewerApp: React.FC<AppComponentProps> = ({ filePath, onReque
     vfsApi.getTempLinkToken(filePath.replace(/^\/+/, ''))
       .then(res => {
         if (cancelled) return;
-        // 注意：vfsApi.getTempPublicUrl 返回的是相对路径，我们需要构建完整的 URL
-        const fullUrl = new URL(vfsApi.getTempPublicUrl(res.token), window.location.origin).href;
+        const baseUrl = systemStatus?.file_domain || window.location.origin;
+        const fullUrl = new URL(res.url, baseUrl).href;
         const officeUrl = `https://view.officeapps.live.com/op/embed.aspx?src=${encodeURIComponent(fullUrl)}`;
         setUrl(officeUrl);
       })

web/src/apps/VideoPlayer/VideoPlayer.tsx

@@ -1,408 +1,46 @@
-import React, { useEffect, useRef, useState } from 'react';
+import React, { useEffect, useRef } from 'react';
+import Artplayer from 'artplayer';
 import { vfsApi } from '../../api/client';
 import type { AppComponentProps } from '../types';
-import { Spin, Button } from 'antd';
-import {
-  PauseOutlined,
-  CaretRightOutlined,
-  SoundOutlined,
-  FullscreenOutlined,
-  ReloadOutlined
-} from '@ant-design/icons';
+
 
 export const VideoPlayerApp: React.FC<AppComponentProps> = ({ filePath }) => {
-  const containerRef = useRef<HTMLDivElement | null>(null);
-  const videoRef = useRef<HTMLVideoElement | null>(null);
-  const progressBarRef = useRef<HTMLDivElement | null>(null);
-  const progressRef = useRef<HTMLDivElement | null>(null);
-  const isMountedRef = useRef(true);
-  const [isPlaying, setIsPlaying] = useState(false);
-  const [volume, setVolume] = useState(0.7);
-  const [isMuted, setIsMuted] = useState(false);
-  const [currentTime, setCurrentTime] = useState(0);
-  const [duration, setDuration] = useState(0);
-  const [loading, setLoading] = useState(true);
-  const [err, setErr] = useState<string>();
-  const [url, setUrl] = useState<string>();
-  const [showControls, setShowControls] = useState(true);
-  const [retryKey, setRetryKey] = useState(0);
-  const controlsTimerRef = useRef<number | undefined>(undefined);
-
-  useEffect(() => {
-    isMountedRef.current = true;
-
-    return () => {
-      isMountedRef.current = false;
-      if (controlsTimerRef.current) {
-        window.clearTimeout(controlsTimerRef.current);
-      }
-    };
-  }, []);
+  const artRef = useRef<HTMLDivElement | null>(null);
+  const artInstance = useRef<Artplayer | null>(null);
 
   useEffect(() => {
+    //
     const safePath = filePath.replace(/^\/+/, '').split('#').map((seg, idx) => idx === 0 ? seg : encodeURIComponent('#') + seg).join('');
-    const u = vfsApi.streamUrl(safePath);
-    setUrl(u);
-    setErr(undefined);
-    setLoading(true);
-  }, [filePath, retryKey]);
+    const videoUrl = vfsApi.streamUrl(safePath);
 
-  // 处理视频事件
-  useEffect(() => {
-    const video = videoRef.current;
-    if (!video || !url) return;
-
-    const onLoadedMetadata = () => {
-      if (isMountedRef.current) {
-        setDuration(video.duration);
-      }
-    };
-
-    const onTimeUpdate = () => {
-      if (isMountedRef.current) {
-        setCurrentTime(video.currentTime);
-        updateProgressBar();
-      }
-    };
-
-    const onCanPlay = () => {
-      if (isMountedRef.current) {
-        setLoading(false);
-      }
-    };
-
-    const onEnded = () => {
-      if (isMountedRef.current) {
-        setIsPlaying(false);
-      }
-    };
-
-    const onError = () => {
-      if (isMountedRef.current) {
-        setLoading(false);
-        setErr('视频加载失败');
-      }
-    };
-
-    const onPlay = () => {
-      if (isMountedRef.current) {
-        setIsPlaying(true);
-      }
-    };
-
-    const onPause = () => {
-      if (isMountedRef.current) {
-        setIsPlaying(false);
-      }
-    };
-
-    const onProgress = () => {
-      // 监听缓冲进度
-      if (video.buffered.length > 0) {
-        const bufferedEnd = video.buffered.end(video.buffered.length - 1);
-        if (progressBarRef.current) {
-          const bufferProgress = bufferedEnd / video.duration * 100;
-          progressBarRef.current.style.setProperty('--buffer-width', `${bufferProgress}%`);
-        }
-      }
-    };
-
-    video.addEventListener('loadedmetadata', onLoadedMetadata);
-    video.addEventListener('timeupdate', onTimeUpdate);
-    video.addEventListener('canplay', onCanPlay);
-    video.addEventListener('ended', onEnded);
-    video.addEventListener('error', onError);
-    video.addEventListener('play', onPlay);
-    video.addEventListener('pause', onPause);
-    video.addEventListener('progress', onProgress);
+    if (artRef.current) {
+      artInstance.current = new Artplayer({
+        container: artRef.current,
+        url: videoUrl,
+        autoplay: true,
+        fullscreen: true,
+        fullscreenWeb: true,
+        pip: true,
+        setting: true,
+        playbackRate: true,
+      });
+    }
 
     return () => {
-      video.removeEventListener('loadedmetadata', onLoadedMetadata);
-      video.removeEventListener('timeupdate', onTimeUpdate);
-      video.removeEventListener('canplay', onCanPlay);
-      video.removeEventListener('ended', onEnded);
-      video.removeEventListener('error', onError);
-      video.removeEventListener('play', onPlay);
-      video.removeEventListener('pause', onPause);
-      video.removeEventListener('progress', onProgress);
-    };
-  }, [url]);
-
-  // 处理进度条更新
-  const updateProgressBar = () => {
-    const video = videoRef.current;
-    const progress = progressRef.current;
-
-    if (video && progress && duration > 0) {
-      const percentage = (video.currentTime / duration) * 100;
-      progress.style.width = `${percentage}%`;
-    }
-  };
-
-  // 处理进度条点击
-  const handleProgressBarClick = (e: React.MouseEvent<HTMLDivElement>) => {
-    const progressBar = progressBarRef.current;
-    const video = videoRef.current;
-
-    if (progressBar && video) {
-      const rect = progressBar.getBoundingClientRect();
-      const clickPosition = e.clientX - rect.left;
-      const percentage = clickPosition / rect.width;
-      const newTime = percentage * duration;
-
-      video.currentTime = newTime;
-      setCurrentTime(newTime);
-    }
-  };
-
-  // 播放/暂停
-  const togglePlay = () => {
-    const video = videoRef.current;
-    if (!video) return;
-
-    if (isPlaying) {
-      video.pause();
-    } else {
-      video.play().catch(error => {
-        console.error('播放失败:', error);
-        setErr('播放失败');
-      });
-    }
-  };
-
-  // 全屏
-  const toggleFullscreen = () => {
-    const container = containerRef.current;
-    if (!container) return;
-
-    if (!document.fullscreenElement) {
-      container.requestFullscreen().catch(err => {
-        console.error('全屏失败:', err);
-      });
-    } else {
-      document.exitFullscreen();
-    }
-  };
-
-  // 音量控制
-  const handleVolumeChange = (e: React.ChangeEvent<HTMLInputElement>) => {
-    const newVolume = parseFloat(e.target.value);
-    setVolume(newVolume);
-
-    const video = videoRef.current;
-    if (video) {
-      video.volume = newVolume;
-      setIsMuted(newVolume === 0);
-    }
-  };
-
-  // 静音切换
-  const toggleMute = () => {
-    const video = videoRef.current;
-    if (!video) return;
-
-    const newMuted = !isMuted;
-    setIsMuted(newMuted);
-    video.muted = newMuted;
-  };
-
-  // 格式化时间显示
-  const formatTime = (seconds: number): string => {
-    if (isNaN(seconds)) return '00:00';
-
-    const mins = Math.floor(seconds / 60);
-    const secs = Math.floor(seconds % 60);
-    return `${mins.toString().padStart(2, '0')}:${secs.toString().padStart(2, '0')}`;
-  };
-
-  // 控制栏自动隐藏
-  const resetControlsTimer = () => {
-    if (controlsTimerRef.current) {
-      window.clearTimeout(controlsTimerRef.current);
-    }
-
-    setShowControls(true);
-
-    controlsTimerRef.current = window.setTimeout(() => {
-      if (isPlaying && isMountedRef.current) {
-        setShowControls(false);
+      if (artInstance.current) {
+        artInstance.current.destroy();
       }
-    }, 3000);
-  };
-
-  const handleMouseMove = () => {
-    resetControlsTimer();
-  };
-
-  const retry = () => setRetryKey(k => k + 1);
+    };
+  }, [filePath]);
 
   return (
     <div
-      style={{ width: '100%', height: '100%', display: 'flex', flexDirection: 'column', background: '#000' }}
-      ref={containerRef}
-      onMouseMove={handleMouseMove}
-    >
-      <div style={{ flex: 1, position: 'relative', backgroundColor: '#000', overflow: 'hidden' }}>
-        {/* 视频元素 */}
-        <video
-          ref={videoRef}
-          style={{
-            width: '100%',
-            height: '100%',
-            objectFit: 'contain'
-          }}
-          src={url}
-          controlsList="nodownload"
-          crossOrigin="anonymous"
-          preload="metadata"
-          onClick={togglePlay}
-        />
-
-        {/* 加载指示器 */}
-        {loading && !err && (
-          <div style={{ position: 'absolute', inset: 0, display: 'flex', flexDirection: 'column', alignItems: 'center', justifyContent: 'center', background: 'rgba(0,0,0,0.35)', gap: 12 }}>
-            <Spin />
-            <span style={{ fontSize: 12, color: '#aaa' }}>正在缓冲...</span>
-          </div>
-        )}
-
-        {/* 错误显示 */}
-        {err && (
-          <div style={{ position: 'absolute', inset: 0, display: 'flex', flexDirection: 'column', alignItems: 'center', justifyContent: 'center', background: 'rgba(0,0,0,0.5)', gap: 12 }}>
-            <span style={{ color: '#ff4d4f', fontSize: 13 }}>{err}</span>
-            <Button icon={<ReloadOutlined />} size="small" onClick={retry}>重试</Button>
-          </div>
-        )}
-
-        {/* 控制栏 */}
-        {showControls && (
-          <div
-            style={{
-              position: 'absolute',
-              bottom: 0,
-              left: 0,
-              right: 0,
-              background: 'linear-gradient(transparent, rgba(0,0,0,0.7))',
-              padding: '30px 15px 10px',
-              transition: 'opacity 0.3s',
-              opacity: showControls ? 1 : 0,
-              display: 'flex',
-              flexDirection: 'column',
-              gap: '8px'
-            }}
-          >
-            {/* 进度条 */}
-            <div
-              ref={progressBarRef}
-              onClick={handleProgressBarClick}
-              style={{
-                height: '4px',
-                backgroundColor: 'rgba(255,255,255,0.2)',
-                cursor: 'pointer',
-                position: 'relative',
-                borderRadius: '2px',
-                '--buffer-width': '0%'
-              } as React.CSSProperties}
-            >
-              <div
-                style={{
-                  position: 'absolute',
-                  height: '100%',
-                  width: 'var(--buffer-width)',
-                  backgroundColor: 'rgba(255,255,255,0.4)',
-                  borderRadius: '2px'
-                }}
-              />
-              <div
-                ref={progressRef}
-                style={{
-                  height: '100%',
-                  width: '0%',
-                  backgroundColor: '#1890ff',
-                  position: 'relative',
-                  borderRadius: '2px',
-                }}
-              >
-                <div
-                  style={{
-                    position: 'absolute',
-                    right: '-6px',
-                    top: '-4px',
-                    width: '12px',
-                    height: '12px',
-                    borderRadius: '50%',
-                    backgroundColor: '#1890ff',
-                  }}
-                />
-              </div>
-            </div>
-
-            <div style={{ display: 'flex', alignItems: 'center', justifyContent: 'space-between' }}>
-              <div style={{ display: 'flex', alignItems: 'center', gap: '16px' }}>
-                <Button
-                  type="text"
-                  icon={isPlaying ? <PauseOutlined /> : <CaretRightOutlined />}
-                  onClick={togglePlay}
-                  style={{ color: '#fff' }}
-                />
-
-                <div style={{ display: 'flex', alignItems: 'center', gap: '8px', width: '100px' }}>
-                  <Button
-                    type="text"
-                    icon={<SoundOutlined />}
-                    onClick={toggleMute}
-                    style={{ color: isMuted ? '#888' : '#fff' }}
-                  />
-                  <input
-                    type="range"
-                    min="0"
-                    max="1"
-                    step="0.01"
-                    value={isMuted ? 0 : volume}
-                    onChange={handleVolumeChange}
-                    style={{ width: '60px' }}
-                  />
-                </div>
-
-                <div style={{ color: '#fff', fontSize: '12px' }}>
-                  {formatTime(currentTime)} / {formatTime(duration)}
-                </div>
-              </div>
-
-              <div>
-                <Button
-                  type="text"
-                  icon={<FullscreenOutlined />}
-                  onClick={toggleFullscreen}
-                  style={{ color: '#fff' }}
-                />
-              </div>
-            </div>
-          </div>
-        )}
-
-        {!isPlaying && !loading && !err && (
-          <div
-            onClick={togglePlay}
-            style={{
-              position: 'absolute',
-              top: '50%',
-              left: '50%',
-              transform: 'translate(-50%, -50%)',
-              width: '60px',
-              height: '60px',
-              borderRadius: '50%',
-              background: 'rgba(0,0,0,0.5)',
-              display: 'flex',
-              alignItems: 'center',
-              justifyContent: 'center',
-              cursor: 'pointer'
-            }}
-          >
-            <CaretRightOutlined style={{ fontSize: '24px', color: '#fff' }} />
-          </div>
-        )}
-      </div>
-    </div>
+      ref={artRef}
+      style={{
+        width: '100%',
+        height: '100%',
+        backgroundColor: '#000'
+      }}
+    />
   );
 };

web/src/layout/SideNav.tsx

@@ -5,6 +5,7 @@ import { memo, useEffect, useState } from 'react';
 import { useSystemStatus } from '../contexts/SystemContext.tsx';
 import {
   CheckCircleOutlined,
+  FileTextOutlined,
   GithubOutlined,
   MenuFoldOutlined,
   SendOutlined,
@@ -154,8 +155,8 @@ const SideNav = memo(function SideNav({ collapsed, activeKey, onChange, onToggle
           }} onClick={showVersionModal}>
             {hasUpdate ? (
               <Tooltip title={`发现新版本: ${latestVersion?.version}`} placement={collapsed ? 'right' : 'top'}>
-                <a href="https://github.com/DrizzleTime/Foxel/releases" target="_blank" rel="noopener noreferrer"
-                   style={{ textDecoration: 'none' }}>
+                <a rel="noopener noreferrer"
+                  style={{ textDecoration: 'none' }}>
                   {collapsed ? (
                     <Tag icon={<WarningOutlined />} color="warning" style={{ marginInlineEnd: 0 }} />
                   ) : (
@@ -199,8 +200,14 @@ const SideNav = memo(function SideNav({ collapsed, activeKey, onChange, onToggle
               href="https://t.me/+thDsBfyqJxZkNTU1"
               target="_blank"
             />
+            <Button
+              shape="circle"
+              icon={<FileTextOutlined />}
+              href="https://foxel.cc"
+              target="_blank"
+            />
           </div>
-          
+
         </div>
       </Sider>
       <Modal
@@ -241,11 +248,11 @@ const SideNav = memo(function SideNav({ collapsed, activeKey, onChange, onToggle
 
               {hasUpdate && (
                 <Alert
-                  message={`发现新版本: ${latestVersion.version}`}
-                  description="建议尽快更新到最新版本，以获得新功能和安全修复。"
+                  message={<span style={{ color: token.colorText }}>{`发现新版本: ${latestVersion.version}`}</span>}
+                  description={<span style={{ color: token.colorTextSecondary }}>建议尽快更新到最新版本，以获得新功能和安全修复。</span>}
                   type="info"
                   showIcon
-                  style={{ marginTop: 24, marginBottom: 24 }}
+                  style={{ marginTop: 24, marginBottom: 24, background: token.colorInfoBg, borderColor: token.colorInfoBorder }}
                   action={
                     <Button
                       size="small"
@@ -276,7 +283,8 @@ const SideNav = memo(function SideNav({ collapsed, activeKey, onChange, onToggle
                       borderBottom: `1px solid ${token.colorBorderSecondary}`,
                       paddingBottom: 8,
                       marginTop: 24,
-                      marginBottom: 16
+                      marginBottom: 16,
+                      color: token.colorTextHeading
                     }} {...props} />,
                     ul: ({ ...props }) => <ul style={{ paddingLeft: 20 }} {...props} />,
                     li: ({ ...props }) => <li style={{ marginBottom: 8 }} {...props} />,

web/src/pages/AdaptersPage.tsx

@@ -1,17 +1,8 @@
 import { memo, useState, useEffect, useCallback } from 'react';
 import { Table, Button, Space, Drawer, Form, Input, Switch, message, Typography, Popconfirm, Select } from 'antd';
 import PageCard from '../components/PageCard';
-import { adaptersApi } from '../api/client';
+import { adaptersApi, type AdapterItem } from '../api/client';
 
-interface AdapterItem {
-  id: number;
-  name: string;
-  type: string;
-  config: any;
-  enabled: boolean;
-  mount_path?: string | null;
-  sub_path?: string | null;
-}
 
 interface AdapterTypeField {
   key: string;
@@ -65,7 +56,7 @@ const AdaptersPage = memo(function AdaptersPage() {
     form.setFieldsValue({
       name: '',
       type: defaultType,
-      mount_path: '/',
+      path: '/',
       sub_path: '',
       enabled: true,
       config: cfgDefaults
@@ -79,7 +70,7 @@ const AdaptersPage = memo(function AdaptersPage() {
     form.setFieldsValue({
       name: rec.name,
       type: rec.type,
-      mount_path: rec.mount_path || '/',
+      path: rec.path || '/',
       sub_path: rec.sub_path || '',
       enabled: rec.enabled,
       config: rec.config || {}
@@ -105,7 +96,7 @@ const AdaptersPage = memo(function AdaptersPage() {
       const body = {
         name: values.name.trim(),
         type: values.type,
-        mount_path: values.mount_path || '/',
+        path: values.path || '/',
         sub_path: values.sub_path?.trim() || null,
         enabled: values.enabled,
         config: cfg
@@ -155,7 +146,7 @@ const AdaptersPage = memo(function AdaptersPage() {
   const columns = [
     { title: '名称', dataIndex: 'name' },
     { title: '类型', dataIndex: 'type', width: 100 },
-    { title: '挂载路径', dataIndex: 'mount_path', width: 140, render: (v: string) => v || '-' },
+    { title: '挂载路径', dataIndex: 'path', width: 140, render: (v: string) => v || '-' },
     { title: '子路径', dataIndex: 'sub_path', width: 140, render: (v: string) => v || '-' },
     {
       title: '启用',
@@ -251,7 +242,6 @@ const AdaptersPage = memo(function AdaptersPage() {
               placeholder="选择适配器类型"
               options={availableTypes.map(t => ({ value: t.type, label: `${t.name} (${t.type})` }))}
               onChange={() => {
-                // 切换类型时刷新默认 config
                 const t = availableTypes.find(v => v.type === form.getFieldValue('type'));
                 const cfgDefaults: Record<string, any> = {};
                 t?.config_schema.forEach(f => {
@@ -261,7 +251,7 @@ const AdaptersPage = memo(function AdaptersPage() {
               }}
             />
           </Form.Item>
-          <Form.Item name="mount_path" label="挂载路径" rules={[{ required: true, message: '请输入挂载路径' }]}>
+          <Form.Item name="path" label="挂载路径" rules={[{ required: true, message: '请输入挂载路径' }]}>
             <Input placeholder="/或/drive" />
           </Form.Item>
           <Form.Item name="sub_path" label="子路径(可选)">

web/src/pages/FileExplorerPage/FileExplorerPage.tsx

@@ -109,7 +109,7 @@ const FileExplorerPage = memo(function FileExplorerPage() {
 
       <div style={{ flex: 1, overflow: 'auto', paddingBottom: pagination.total > 0 ? '80px' : '0' }} onContextMenu={openBlankContextMenu}>
         {loading && entries.length === 0 ? (
-          <div style={{ textAlign: 'center', padding: 40 }}><EmptyState isRoot={path === '/'} onCreateDir={() => setCreatingDir(true)} onGoUp={goUp} /></div>
+          <div style={{ textAlign: 'center', padding: 40 }}><EmptyState isRoot={path === '/'} /></div>
         ) : viewMode === 'grid' ? (
           <GridView
             entries={entries}
@@ -121,8 +121,6 @@ const FileExplorerPage = memo(function FileExplorerPage() {
             onSelectRange={handleSelectRange}
             onOpen={handleOpenEntry}
             onContextMenu={openContextMenu}
-            onCreateDir={() => setCreatingDir(true)}
-            onGoUp={goUp}
           />
         ) : (
           <FileListView

web/src/pages/FileExplorerPage/components/EmptyState.tsx

@@ -1,14 +1,12 @@
 import React from 'react';
-import { Button, Space, Typography, theme } from 'antd';
-import { PlusOutlined, CloudUploadOutlined, ArrowUpOutlined, FolderOpenOutlined } from '@ant-design/icons';
+import { Typography, theme } from 'antd';
+import { FolderOpenOutlined } from '@ant-design/icons';
 
 interface Props {
   isRoot: boolean;
-  onCreateDir: () => void;
-  onGoUp: () => void;
 }
 
-export const EmptyState: React.FC<Props> = ({ isRoot, onCreateDir, onGoUp }) => {
+export const EmptyState: React.FC<Props> = ({ isRoot }) => {
   const { token } = theme.useToken();
   return (
     <div style={{ display:'flex', flexDirection:'column', alignItems:'center', justifyContent:'center', padding:isRoot? '80px 40px':'60px 40px', minHeight: isRoot? '400px':'300px', color: token.colorTextSecondary }}>
@@ -19,15 +17,6 @@ export const EmptyState: React.FC<Props> = ({ isRoot, onCreateDir, onGoUp }) =>
       <Typography.Text style={{ color: token.colorTextTertiary, marginBottom:24, textAlign:'center', maxWidth:300, lineHeight:1.5 }}>
         {isRoot ? '开始上传文件或创建新目录来组织您的内容' : '您可以在此目录中创建新的文件夹或上传文件'}
       </Typography.Text>
-      <Space size={12}>
-        <Button type="primary" icon={<PlusOutlined />} onClick={onCreateDir}>新建目录</Button>
-        <Button icon={<CloudUploadOutlined />} disabled>上传文件</Button>
-      </Space>
-      {!isRoot && (
-        <div style={{ marginTop:16 }}>
-          <Button type="link" size="small" icon={<ArrowUpOutlined />} onClick={onGoUp} style={{ color: token.colorTextTertiary }}>返回上级目录</Button>
-        </div>
-      )}
     </div>
   );
 };

web/src/pages/FileExplorerPage/components/GridView.tsx

@@ -7,20 +7,14 @@ import { EmptyState } from './EmptyState';
 
 interface Props {
   entries: VfsEntry[];
-  thumbs: Record<string,string>;
-  // ...existing code...
-  // selected was single entry before; now use selectedEntries for multi-select
+  thumbs: Record<string, string>;
   selectedEntries: string[];
   loading: boolean;
   path: string;
-  // onSelect: clicked entry, additive indicates Ctrl/Cmd click to toggle
   onSelect: (e: VfsEntry, additive?: boolean) => void;
-  // onSelectRange: called when marquee/selecting multiple by box
   onSelectRange: (names: string[]) => void;
   onOpen: (e: VfsEntry) => void;
   onContextMenu: (e: React.MouseEvent, entry: VfsEntry) => void;
-  onCreateDir: () => void;
-  onGoUp: () => void;
 }
 
 const formatSize = (size: number) => {
@@ -30,14 +24,12 @@ const formatSize = (size: number) => {
   return (size / 1024 / 1024 / 1024).toFixed(1) + ' GB';
 };
 
-export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, loading, path, onSelect, onSelectRange, onOpen, onContextMenu, onCreateDir, onGoUp }) => {
+export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, loading, path, onSelect, onSelectRange, onOpen, onContextMenu }) => {
   const { token } = theme.useToken();
-
-  // refs for marquee selection
   const containerRef = useRef<HTMLDivElement | null>(null);
   const itemRefs = useRef<Record<string, HTMLDivElement | null>>({});
-  const startRef = useRef<{x:number,y:number} | null>(null);
-  const [rect, setRect] = useState<{left:number,top:number,width:number,height:number} | null>(null);
+  const startRef = useRef<{ x: number, y: number } | null>(null);
+  const [rect, setRect] = useState<{ left: number, top: number, width: number, height: number } | null>(null);
   const [selecting, setSelecting] = useState(false);
 
   useEffect(() => {
@@ -52,12 +44,11 @@ export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, lo
       const height = Math.abs(cy - s.y);
       setRect({ left, top, width, height });
     };
-    const onUp = () => { // 不需要 MouseEvent 参数，避免未使用警告
+    const onUp = () => {
       if (!startRef.current) return;
       setSelecting(false);
       const r = rect;
       if (r) {
-        // compute intersecting items
         const container = containerRef.current;
         if (container) {
           const sel: string[] = [];
@@ -89,17 +80,14 @@ export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, lo
   }, [selecting, rect, entries, onSelectRange]);
 
   const handleMouseDown = (e: React.MouseEvent) => {
-    // only left button and not on an item actionable element
     if (e.button !== 0) return;
-    // start marquee if click on empty space inside container
     const target = e.target as HTMLElement;
     if (target.closest('.fx-grid-item')) {
-      return; // clicks on item handled separately
+      return;
     }
     startRef.current = { x: e.clientX, y: e.clientY };
     setSelecting(true);
     setRect({ left: e.clientX, top: e.clientY, width: 0, height: 0 });
-    // prevent text selection
     e.preventDefault();
   };
 
@@ -108,29 +96,28 @@ export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, lo
       {entries.map(ent => {
         const isImg = thumbs[ent.name];
         const ext = ent.name.split('.').pop()?.toLowerCase();
-        const isPictureType = ['png','jpg','jpeg','gif','webp','svg'].includes(ext || '');
+        const isPictureType = ['png', 'jpg', 'jpeg', 'gif', 'webp', 'svg'].includes(ext || '');
         const isSelected = selectedEntries.includes(ent.name);
         return (
           <div
             key={ent.name}
-            ref={(el) => { itemRefs.current[ent.name] = el; }} // 确保函数不返回值，匹配 Ref 类型
-            className={['fx-grid-item', isSelected ? 'selected' : '', ent.is_dir? 'dir':'file'].join(' ')}
+            ref={(el) => { itemRefs.current[ent.name] = el; }}
+            className={['fx-grid-item', isSelected ? 'selected' : '', ent.is_dir ? 'dir' : 'file'].join(' ')}
             onClick={(ev) => {
-              // click selection: support ctrl/cmd to toggle
               const additive = ev.ctrlKey || ev.metaKey;
               onSelect(ent, additive);
             }}
             onDoubleClick={() => onOpen(ent)}
-            onContextMenu={(e)=> onContextMenu(e, ent)}
-            style={{ userSelect:'none' }}
+            onContextMenu={(e) => onContextMenu(e, ent)}
+            style={{ userSelect: 'none' }}
           >
             <div className="thumb" style={{ background: ent.is_dir ? 'linear-gradient(#fafafa,#f2f2f2)' : '#fff' }}>
-              {ent.is_dir && <FolderFilled style={{ fontSize:32, color: token.colorPrimary }} />}
-              {!ent.is_dir && (isImg ? <img src={isImg} alt={ent.name} style={{ maxWidth:'100%', maxHeight:'100%'}} /> : isPictureType ? <PictureOutlined style={{ fontSize:32, color:'#8c8c8c' }} /> : getFileIcon(ent.name,32))}
+              {ent.is_dir && <FolderFilled style={{ fontSize: 32, color: token.colorPrimary }} />}
+              {!ent.is_dir && (isImg ? <img src={isImg} alt={ent.name} style={{ maxWidth: '100%', maxHeight: '100%' }} /> : isPictureType ? <PictureOutlined style={{ fontSize: 32, color: '#8c8c8c' }} /> : getFileIcon(ent.name, 32))}
               {ent.type === 'mount' && <span className="badge">M</span>}
             </div>
-            <Tooltip title={ent.name}><div className="name ellipsis" style={{ userSelect:'none' }}>{ent.name}</div></Tooltip>
-            <div className="meta ellipsis" style={{ fontSize:11, color: token.colorTextSecondary, userSelect:'none' }}>{ent.is_dir ? '目录' : formatSize(ent.size)}</div>
+            <Tooltip title={ent.name}><div className="name ellipsis" style={{ userSelect: 'none' }}>{ent.name}</div></Tooltip>
+            <div className="meta ellipsis" style={{ fontSize: 11, color: token.colorTextSecondary, userSelect: 'none' }}>{ent.is_dir ? '目录' : formatSize(ent.size)}</div>
           </div>
         )
       })}
@@ -148,8 +135,8 @@ export const GridView: React.FC<Props> = ({ entries, thumbs, selectedEntries, lo
           }}
         />
       )}
-      {loading && <div style={{ width:'100%', textAlign:'center', padding:40 }}><Spin /></div>}
-      {!loading && entries.length === 0 && <EmptyState isRoot={path==='/' } onCreateDir={onCreateDir} onGoUp={onGoUp} />}
+      {loading && <div style={{ width: '100%', textAlign: 'center', padding: 40 }}><Spin /></div>}
+      {!loading && entries.length === 0 && <EmptyState isRoot={path === '/'} />}
     </div>
   );
 };

web/src/pages/FileExplorerPage/components/Modals/DirectLinkModal.tsx

@@ -29,8 +29,7 @@ export const DirectLinkModal = memo(function DirectLinkModal({ entry, path, open
     try {
       const fullPath = (path === '/' ? '' : path) + '/' + entry.name;
       const res = await vfsApi.getTempLinkToken(fullPath, expiresIn);
-      const tempLink = `${window.location.origin}/api/fs/public/${res.token}`;
-      setLink(tempLink);
+      setLink(res.url);
     } catch (e: any) {
       message.error(e.message || '生成链接失败');
     } finally {

web/src/pages/FileExplorerPage/components/Modals/ShareModal.tsx

@@ -3,6 +3,7 @@ import { Modal, Form, Input, Radio, InputNumber, message, Button, Typography } f
 import { CopyOutlined } from '@ant-design/icons';
 import type { VfsEntry, ShareInfoWithPassword } from '../../../../api/client';
 import { shareApi } from '../../../../api/share';
+import { useSystemStatus } from '../../../../contexts/SystemContext';
 
 interface ShareModalProps {
   entries: VfsEntry[];
@@ -13,6 +14,7 @@ interface ShareModalProps {
 }
 
 export const ShareModal = memo(function ShareModal({ entries, path, open, onOk, onCancel }: ShareModalProps) {
+  const systemStatus = useSystemStatus();
   const [form] = Form.useForm();
   const [loading, setLoading] = useState(false);
   const [accessType, setAccessType] = useState('public');
@@ -66,7 +68,8 @@ export const ShareModal = memo(function ShareModal({ entries, path, open, onOk,
     message.success('已复制到剪贴板');
   };
 
-  const shareUrl = createdShare ? `${window.location.origin}/share/${createdShare.token}` : '';
+  const baseUrl = systemStatus?.app_domain || window.location.origin;
+  const shareUrl = createdShare ? new URL(`/share/${createdShare.token}`, baseUrl).href : '';
 
   const renderForm = () => (
     <Form form={form} layout="vertical" initialValues={{ name: defaultName, accessType: 'public', expiresInDays: 7 }}>

web/src/pages/SetupPage.tsx

@@ -26,7 +26,7 @@ const SetupPage = () => {
             root: values.root_dir
           },
           sub_path: null,
-          mount_path: values.mount_path,
+          path: values.path,
           enabled: true
         });
         window.location.href = '/';
@@ -41,7 +41,7 @@ const SetupPage = () => {
 
   const stepFields = [
     ['db_driver', 'vector_db_driver'],
-    ['adapter_name', 'adapter_type', 'mount_path', 'root_dir'],
+    ['adapter_name', 'adapter_type', 'path', 'root_dir'],
     ['username', 'full_name', 'email', 'password', 'confirm'],
   ]
 
@@ -105,7 +105,7 @@ const SetupPage = () => {
           </Form.Item>
           <Form.Item
             label="挂载路径"
-            name="mount_path"
+            name="path"
             initialValue="/local"
             rules={[{ required: true, message: '请输入挂载路径！' }]}
           >

web/src/pages/SharePage.tsx

@@ -4,8 +4,10 @@ import PageCard from '../components/PageCard';
 import { shareApi, type ShareInfo } from '../api/share';
 import { format, parseISO } from 'date-fns';
 import { LinkOutlined, CopyOutlined, DeleteOutlined } from '@ant-design/icons';
+import { useSystemStatus } from '../contexts/SystemContext';
 
 const SharePage = memo(function SharePage() {
+  const systemStatus = useSystemStatus();
   const [loading, setLoading] = useState(false);
   const [data, setData] = useState<ShareInfo[]>([]);
 
@@ -24,7 +26,8 @@ const SharePage = memo(function SharePage() {
   useEffect(() => { fetchList(); }, [fetchList]);
 
   const doCopy = (rec: ShareInfo) => {
-    const shareUrl = `${window.location.origin}/share/${rec.token}`;
+    const baseUrl = systemStatus?.app_domain || window.location.origin;
+    const shareUrl = new URL(`/share/${rec.token}`, baseUrl).href;
     navigator.clipboard.writeText(shareUrl);
     message.success('链接已复制');
   };

web/src/pages/SystemSettingsPage/SystemSettingsPage.tsx

@@ -1,26 +1,33 @@
-import { Form, Input, Button, message, Tabs, Space } from 'antd';
+import { Form, Input, Button, message, Tabs, Space, Card } from 'antd';
 import { useEffect, useState } from 'react';
 import PageCard from '../../components/PageCard';
 import { getAllConfig, setConfig } from '../../api/config';
-import { API_BASE_URL } from '../../api/client';
 import { AppstoreOutlined, RobotOutlined } from '@ant-design/icons';
 
-const APP_CONFIG_KEYS = [
+const APP_CONFIG_KEYS: {key: string, label: string, default?: string}[] = [
   { key: 'APP_NAME', label: '应用名称' },
   { key: 'APP_LOGO', label: 'LOGO地址' },
-  { key: 'SERVER_URL', label: '服务端URL', default: API_BASE_URL },
+  { key: 'APP_DOMAIN', label: '应用域名' },
+  { key: 'FILE_DOMAIN', label: '文件域名' },
 ];
 
-const AI_CONFIG_KEYS = [
-  { key: 'AI_API_URL', label: 'AI API地址' },
-  { key: 'AI_VISION_MODEL', label: '视觉模型' },
-  { key: 'AI_EMBED_MODEL', label: '嵌入模型' },
-  { key: 'AI_API_KEY', label: 'API Key' },
+const VISION_CONFIG_KEYS = [
+  { key: 'AI_VISION_API_URL', label: '视觉模型 API 地址' },
+  { key: 'AI_VISION_MODEL', label: '视觉模型', default: 'Qwen/Qwen2.5-VL-32B-Instruct' },
+  { key: 'AI_VISION_API_KEY', label: '视觉模型 API Key' },
 ];
 
+const EMBED_CONFIG_KEYS = [
+  { key: 'AI_EMBED_API_URL', label: '嵌入模型 API 地址' },
+  { key: 'AI_EMBED_MODEL', label: '嵌入模型', default: 'Qwen/Qwen3-Embedding-8B' },
+  { key: 'AI_EMBED_API_KEY', label: '嵌入模型 API Key' },
+];
+
+const ALL_AI_KEYS = [...VISION_CONFIG_KEYS, ...EMBED_CONFIG_KEYS];
+
 export default function SystemSettingsPage() {
   const [loading, setLoading] = useState(false);
-  const [config, setConfigState] = useState<Record<string, string> | null>(null); 
+  const [config, setConfigState] = useState<Record<string, string> | null>(null);
   const [activeTab, setActiveTab] = useState('app');
 
   useEffect(() => {
@@ -41,14 +48,13 @@ export default function SystemSettingsPage() {
     setLoading(false);
   };
 
-  // 加载中时不渲染表单
   if (!config) {
     return <PageCard title='系统设置'><div>加载中...</div></PageCard>;
   }
 
   return (
     <PageCard
-    title='系统设置'
+      title='系统设置'
     >
       <Space direction="vertical" style={{ width: '100%' }} size={32}>
         <Tabs
@@ -73,7 +79,7 @@ export default function SystemSettingsPage() {
                   }}
                   onFinish={handleSave}
                   style={{ marginTop: 24 }}
-                  key={JSON.stringify(config)} 
+                  key={JSON.stringify(config)}
                 >
                   {APP_CONFIG_KEYS.map(({ key, label }) => (
                     <Form.Item key={key} name={key} label={label}>
@@ -100,18 +106,27 @@ export default function SystemSettingsPage() {
                 <Form
                   layout="vertical"
                   initialValues={{
-                    ...Object.fromEntries(AI_CONFIG_KEYS.map(({ key }) => [key, config[key] ?? ''])),
+                    ...Object.fromEntries(ALL_AI_KEYS.map(({ key, default: def }) => [key, config[key] ?? def ?? ''])),
                   }}
                   onFinish={handleSave}
                   style={{ marginTop: 24 }}
-                  key={JSON.stringify(config)} // 强制表单重置
+                  key={JSON.stringify(config)}
                 >
-                  {AI_CONFIG_KEYS.map(({ key, label }) => (
-                    <Form.Item key={key} name={key} label={label}>
-                      <Input size="large" />
-                    </Form.Item>
-                  ))}
-                  <Form.Item>
+                  <Card title="视觉模型" style={{ marginBottom: 24 }}>
+                    {VISION_CONFIG_KEYS.map(({ key, label }) => (
+                      <Form.Item key={key} name={key} label={label}>
+                        <Input size="large" />
+                      </Form.Item>
+                    ))}
+                  </Card>
+                  <Card title="嵌入模型">
+                    {EMBED_CONFIG_KEYS.map(({ key, label }) => (
+                      <Form.Item key={key} name={key} label={label}>
+                        <Input size="large" />
+                      </Form.Item>
+                    ))}
+                  </Card>
+                  <Form.Item style={{ marginTop: 24 }}>
                     <Button type="primary" htmlType="submit" loading={loading} block>
                       保存
                     </Button>

web/src/theme.ts

@@ -4,6 +4,7 @@ import type { ThemeConfig } from 'antd/es/config-provider/context';
 export const foxelTheme: ThemeConfig = {
   algorithm: [theme.defaultAlgorithm, theme.compactAlgorithm],
   token: {
+    colorInfoBg: '#efefef',
     colorPrimary: '#111',
     colorInfo: '#111',
     colorText: '#111',
@@ -37,7 +38,7 @@ export const foxelTheme: ThemeConfig = {
     },
     Card: {
       borderRadiusLG: 16,
-     padding: 16
+      padding: 16
     },
     Input: { borderRadius: 8 },
     Dropdown: { controlItemBgHover: '#f2f2f2' },