feat: 添加便捷小工具页面 - AI检索图片/数据清洗功能

- 新增便捷工具页面 (/tools)，提供AI检索图片/数据清洗功能
- 支持JSONL格式数据的URI匹配去重处理
- 实现实时进度显示和批量数据处理
- 添加完整的错误处理和用户反馈机制
- 遵循Tauri开发规范和UI/UX设计标准
- 包含测试数据和功能文档

技术实现：
- 后端: Rust异步处理，流式文件读取，进度事件发送
- 前端: React + TypeScript，文件选择对话框，进度条显示
- 导航: 新增便捷工具菜单项，集成到主导航栏

apps/desktop/TOOLS_FEATURE.md

@@ -0,0 +1,100 @@
+# 便捷小工具功能 - v0.1.33
+
+## 功能概述
+
+新增便捷小工具页面，提供AI检索图片/数据清洗功能，支持JSONL格式数据的去重处理。
+
+## 功能特性
+
+### AI检索图片/数据清洗工具
+
+- **文件格式支持**: JSONL (JSON Lines) 格式
+- **数据去重**: 基于URI字段进行精确匹配去重
+- **进度显示**: 实时显示处理进度和状态
+- **批量处理**: 支持大数据量文件处理
+- **结果统计**: 显示原始数据量、去除数量、最终结果数量
+
+## 使用方法
+
+1. **选择全部数据文件**: 包含所有数据的JSONL文件
+2. **选择要去除的数据文件**: 包含需要去除数据的JSONL文件
+3. **选择输出文件**: 指定处理结果的保存位置
+4. **开始处理**: 点击"开始处理"按钮执行数据清洗
+
+## 数据格式说明
+
+### 全部数据文件格式
+```json
+{
+  "kind": "storage#object",
+  "id": "...",
+  "uri": "gs://bucket/path/to/file.jpg",
+  // 其他字段...
+}
+```
+
+### 要去除的数据文件格式
+```json
+{
+  "id": "uuid",
+  "schema_id": "default_schema",
+  "json_data": "{\"uri\":\"gs://bucket/path/to/file.jpg\", ...}"
+}
+```
+
+## 处理逻辑
+
+1. **解析要去除的数据**: 从`json_data`字段中提取`uri`值
+2. **构建URI集合**: 将所有要去除的URI存储在HashSet中
+3. **过滤全部数据**: 遍历全部数据文件，过滤掉URI匹配的记录
+4. **保存结果**: 将过滤后的数据保存到输出文件
+
+## 技术实现
+
+### 后端 (Rust)
+- **命令**: `clean_jsonl_data`
+- **文件**: `src-tauri/src/presentation/commands/tools_commands.rs`
+- **特性**: 
+  - 异步处理
+  - 进度事件发送
+  - 内存优化的流式处理
+  - 错误处理和恢复
+
+### 前端 (React + TypeScript)
+- **页面**: `src/pages/Tools.tsx`
+- **路由**: `/tools`
+- **特性**:
+  - 文件选择对话框
+  - 实时进度显示
+  - 结果统计展示
+  - 错误处理和用户反馈
+
+## 导航集成
+
+在主导航栏中新增"便捷工具"菜单项，使用扳手图标，提供快速访问入口。
+
+## 测试数据
+
+项目包含测试数据文件：
+- `test_data/sample_all_data.jsonl`: 示例全部数据文件
+- `test_data/sample_remove_data.jsonl`: 示例要去除的数据文件
+
+## 性能优化
+
+- **流式处理**: 逐行读取文件，避免内存溢出
+- **进度更新**: 每处理100行更新一次进度，平衡性能和用户体验
+- **错误容错**: 解析失败的行会被跳过并记录警告
+
+## 开发规范遵循
+
+- 遵循Tauri开发规范的四层架构设计
+- 使用TypeScript确保类型安全
+- 遵循UI/UX设计标准，提供优雅的用户界面
+- 实现完整的错误处理和用户反馈机制
+
+## 未来扩展
+
+- 支持更多数据格式 (CSV, JSON等)
+- 添加更多数据清洗规则
+- 支持自定义匹配字段
+- 添加数据预览功能

apps/desktop/src-tauri/src/lib.rs

@@ -253,7 +253,9 @@ pub fn run() {
             commands::test_commands::test_logging,
             // 调试命令
             commands::debug_commands::test_parse_draft_file,
-            commands::debug_commands::validate_template_structure
+            commands::debug_commands::validate_template_structure,
+            // 便捷工具命令
+            commands::tools_commands::clean_jsonl_data
         ])
         .setup(|app| {
             // 初始化日志系统

apps/desktop/src-tauri/src/presentation/commands/mod.rs

@@ -16,3 +16,4 @@ pub mod material_matching_commands;
 pub mod template_matching_result_commands;
 pub mod export_record_commands;
 pub mod video_generation_commands;
+pub mod tools_commands;

apps/desktop/src-tauri/src/presentation/commands/tools_commands.rs

@@ -0,0 +1,222 @@
+use tauri::{command, Emitter};
+use serde::{Deserialize, Serialize};
+use std::collections::HashSet;
+use std::fs::File;
+use std::io::{BufRead, BufReader, Write};
+use std::path::Path;
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct DataCleaningProgress {
+    pub current: usize,
+    pub total: usize,
+    pub percentage: f64,
+    pub status: String,
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct DataCleaningResult {
+    pub success: bool,
+    pub message: String,
+    pub original_count: usize,
+    pub removed_count: usize,
+    pub final_count: usize,
+    pub output_file: String,
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+pub struct StorageObject {
+    pub uri: String,
+    // 其他字段可以忽略，我们只关心 uri
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+pub struct ProcessedData {
+    pub json_data: String,
+}
+
+/// AI检索图片/数据清洗 - JSONL格式处理
+#[command]
+pub async fn clean_jsonl_data(
+    all_data_file: String,
+    remove_data_file: String,
+    output_file: String,
+    window: tauri::Window,
+) -> Result<DataCleaningResult, String> {
+    // 验证文件存在
+    if !Path::new(&all_data_file).exists() {
+        return Err(format!("全部数据文件不存在: {}", all_data_file));
+    }
+    
+    if !Path::new(&remove_data_file).exists() {
+        return Err(format!("要去除的数据文件不存在: {}", remove_data_file));
+    }
+
+    // 发送开始处理的进度
+    let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+        current: 0,
+        total: 0,
+        percentage: 0.0,
+        status: "开始处理...".to_string(),
+    });
+
+    // 第一步：读取要去除的数据文件，提取所有URI
+    let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+        current: 0,
+        total: 0,
+        percentage: 10.0,
+        status: "读取要去除的数据文件...".to_string(),
+    });
+
+    let remove_uris = match extract_remove_uris(&remove_data_file).await {
+        Ok(uris) => uris,
+        Err(e) => return Err(format!("读取要去除的数据文件失败: {}", e)),
+    };
+
+    let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+        current: 0,
+        total: 0,
+        percentage: 30.0,
+        status: format!("找到 {} 个要去除的URI", remove_uris.len()),
+    });
+
+    // 第二步：处理全部数据文件
+    let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+        current: 0,
+        total: 0,
+        percentage: 40.0,
+        status: "开始处理全部数据文件...".to_string(),
+    });
+
+    let result = match process_all_data_file(&all_data_file, &remove_uris, &output_file, &window).await {
+        Ok(result) => result,
+        Err(e) => return Err(format!("处理全部数据文件失败: {}", e)),
+    };
+
+    let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+        current: result.final_count,
+        total: result.original_count,
+        percentage: 100.0,
+        status: "处理完成".to_string(),
+    });
+
+    Ok(result)
+}
+
+/// 从要去除的数据文件中提取所有URI
+async fn extract_remove_uris(file_path: &str) -> Result<HashSet<String>, String> {
+    let file = File::open(file_path)
+        .map_err(|e| format!("无法打开文件: {}", e))?;
+    
+    let reader = BufReader::new(file);
+    let mut uris = HashSet::new();
+
+    for line in reader.lines() {
+        let line = line.map_err(|e| format!("读取行失败: {}", e))?;
+        
+        if line.trim().is_empty() {
+            continue;
+        }
+
+        // 解析JSON
+        let processed_data: ProcessedData = serde_json::from_str(&line)
+            .map_err(|e| format!("解析JSON失败: {} - 行内容: {}", e, line))?;
+
+        // 从json_data字段中解析内部JSON
+        let inner_json: serde_json::Value = serde_json::from_str(&processed_data.json_data)
+            .map_err(|e| format!("解析内部JSON失败: {}", e))?;
+
+        // 提取uri字段
+        if let Some(uri) = inner_json.get("uri").and_then(|v| v.as_str()) {
+            uris.insert(uri.to_string());
+        }
+    }
+
+    Ok(uris)
+}
+
+/// 处理全部数据文件，过滤掉要去除的URI
+async fn process_all_data_file(
+    input_file: &str,
+    remove_uris: &HashSet<String>,
+    output_file: &str,
+    window: &tauri::Window,
+) -> Result<DataCleaningResult, String> {
+    let input_file_handle = File::open(input_file)
+        .map_err(|e| format!("无法打开输入文件: {}", e))?;
+    
+    let reader = BufReader::new(input_file_handle);
+    
+    let mut output_file_handle = File::create(output_file)
+        .map_err(|e| format!("无法创建输出文件: {}", e))?;
+
+    let mut original_count = 0;
+    let mut removed_count = 0;
+    let mut final_count = 0;
+
+    // 首先计算总行数用于进度显示
+    let total_lines = count_lines(input_file)?;
+
+    for (line_number, line) in reader.lines().enumerate() {
+        let line = line.map_err(|e| format!("读取行失败: {}", e))?;
+        
+        if line.trim().is_empty() {
+            continue;
+        }
+
+        original_count += 1;
+
+        // 解析JSON
+        let storage_object: StorageObject = match serde_json::from_str(&line) {
+            Ok(obj) => obj,
+            Err(e) => {
+                // 如果解析失败，跳过这一行但记录警告
+                eprintln!("警告: 解析JSON失败，跳过行 {}: {} - 内容: {}", line_number + 1, e, line);
+                continue;
+            }
+        };
+
+        // 检查URI是否在要去除的列表中
+        if remove_uris.contains(&storage_object.uri) {
+            removed_count += 1;
+        } else {
+            // 写入输出文件
+            writeln!(output_file_handle, "{}", line)
+                .map_err(|e| format!("写入输出文件失败: {}", e))?;
+            final_count += 1;
+        }
+
+        // 每处理100行发送一次进度更新
+        if line_number % 100 == 0 {
+            let percentage = 40.0 + (line_number as f64 / total_lines as f64) * 50.0;
+            let _ = window.emit("data-cleaning-progress", DataCleaningProgress {
+                current: line_number,
+                total: total_lines,
+                percentage,
+                status: format!("已处理 {} 行，已去除 {} 个重复项", line_number, removed_count),
+            });
+        }
+    }
+
+    // 确保文件被正确写入
+    output_file_handle.flush()
+        .map_err(|e| format!("刷新输出文件失败: {}", e))?;
+
+    Ok(DataCleaningResult {
+        success: true,
+        message: "数据清洗完成".to_string(),
+        original_count,
+        removed_count,
+        final_count,
+        output_file: output_file.to_string(),
+    })
+}
+
+/// 计算文件行数
+fn count_lines(file_path: &str) -> Result<usize, String> {
+    let file = File::open(file_path)
+        .map_err(|e| format!("无法打开文件计算行数: {}", e))?;
+    
+    let reader = BufReader::new(file);
+    let count = reader.lines().count();
+    Ok(count)
+}

apps/desktop/src/App.tsx

@@ -8,6 +8,7 @@ import ModelDetail from './pages/ModelDetail';
 import AiClassificationSettings from './pages/AiClassificationSettings';
 import TemplateManagement from './pages/TemplateManagement';
 import { MaterialModelBinding } from './pages/MaterialModelBinding';
+import Tools from './pages/Tools';
 import Navigation from './components/Navigation';
 import { NotificationSystem, useNotifications } from './components/NotificationSystem';
 import { useProjectStore } from './store/projectStore';
@@ -78,6 +79,7 @@ function App() {
               <Route path="/ai-classification-settings" element={<AiClassificationSettings />} />
               <Route path="/templates" element={<TemplateManagement />} />
               <Route path="/material-model-binding" element={<MaterialModelBinding />} />
+              <Route path="/tools" element={<Tools />} />
             </Routes>
           </div>
         </main>

apps/desktop/src/components/Navigation.tsx

@@ -5,7 +5,8 @@ import {
   UserGroupIcon,
   CpuChipIcon,
   DocumentDuplicateIcon,
-  LinkIcon
+  LinkIcon,
+  WrenchScrewdriverIcon
 } from '@heroicons/react/24/outline';
 
 const Navigation: React.FC = () => {
@@ -41,6 +42,12 @@ const Navigation: React.FC = () => {
       href: '/ai-classification-settings',
       icon: CpuChipIcon,
       description: '管理AI视频分类规则'
+    },
+    {
+      name: '便捷工具',
+      href: '/tools',
+      icon: WrenchScrewdriverIcon,
+      description: 'AI检索图片/数据清洗工具'
     }
   ];
 

apps/desktop/src/pages/Tools.tsx

@@ -0,0 +1,354 @@
+import React, { useState } from 'react';
+import { 
+  Wrench, 
+  FileText, 
+  Upload, 
+  Download, 
+  Loader2, 
+  CheckCircle, 
+  AlertCircle,
+  Info,
+  Trash2
+} from 'lucide-react';
+import { invoke } from '@tauri-apps/api/core';
+import { open, save } from '@tauri-apps/plugin-dialog';
+import { listen } from '@tauri-apps/api/event';
+import { useNotifications } from '../components/NotificationSystem';
+
+interface DataCleaningProgress {
+  current: number;
+  total: number;
+  percentage: number;
+  status: string;
+}
+
+interface DataCleaningResult {
+  success: boolean;
+  message: string;
+  original_count: number;
+  removed_count: number;
+  final_count: number;
+  output_file: string;
+}
+
+/**
+ * 便捷小工具页面
+ * 遵循 Tauri 开发规范和 UI/UX 设计标准
+ */
+const Tools: React.FC = () => {
+  const [allDataFile, setAllDataFile] = useState<string>('');
+  const [removeDataFile, setRemoveDataFile] = useState<string>('');
+  const [outputFile, setOutputFile] = useState<string>('');
+  const [isProcessing, setIsProcessing] = useState(false);
+  const [progress, setProgress] = useState<DataCleaningProgress | null>(null);
+  const [result, setResult] = useState<DataCleaningResult | null>(null);
+  
+  const { success, error } = useNotifications();
+
+  // 选择全部数据文件
+  const selectAllDataFile = async () => {
+    try {
+      const selected = await open({
+        multiple: false,
+        filters: [{
+          name: 'JSONL Files',
+          extensions: ['jsonl']
+        }]
+      });
+
+      if (selected && typeof selected === 'string') {
+        setAllDataFile(selected);
+      }
+    } catch (err) {
+      error('文件选择失败', '无法选择全部数据文件');
+    }
+  };
+
+  // 选择要去除的数据文件
+  const selectRemoveDataFile = async () => {
+    try {
+      const selected = await open({
+        multiple: false,
+        filters: [{
+          name: 'JSONL Files',
+          extensions: ['jsonl']
+        }]
+      });
+
+      if (selected && typeof selected === 'string') {
+        setRemoveDataFile(selected);
+      }
+    } catch (err) {
+      error('文件选择失败', '无法选择要去除的数据文件');
+    }
+  };
+
+  // 选择输出文件
+  const selectOutputFile = async () => {
+    try {
+      const selected = await save({
+        filters: [{
+          name: 'JSONL Files',
+          extensions: ['jsonl']
+        }],
+        defaultPath: 'cleaned_data.jsonl'
+      });
+
+      if (selected) {
+        setOutputFile(selected);
+      }
+    } catch (err) {
+      error('文件选择失败', '无法选择输出文件');
+    }
+  };
+
+  // 开始数据清洗
+  const startDataCleaning = async () => {
+    if (!allDataFile || !removeDataFile || !outputFile) {
+      error('参数错误', '请选择所有必需的文件');
+      return;
+    }
+
+    setIsProcessing(true);
+    setProgress(null);
+    setResult(null);
+
+    try {
+      // 监听进度事件
+      const unlisten = await listen<DataCleaningProgress>('data-cleaning-progress', (event) => {
+        setProgress(event.payload);
+      });
+
+      // 调用后端命令
+      const cleaningResult = await invoke<DataCleaningResult>('clean_jsonl_data', {
+        allDataFile,
+        removeDataFile,
+        outputFile
+      });
+
+      setResult(cleaningResult);
+      
+      if (cleaningResult.success) {
+        success('数据清洗完成', 
+          `原始数据: ${cleaningResult.original_count} 条，` +
+          `去除重复: ${cleaningResult.removed_count} 条，` +
+          `最终结果: ${cleaningResult.final_count} 条`
+        );
+      } else {
+        error('数据清洗失败', cleaningResult.message);
+      }
+
+      // 清理事件监听器
+      unlisten();
+    } catch (err) {
+      const errorMessage = err instanceof Error ? err.message : '未知错误';
+      error('数据清洗失败', errorMessage);
+    } finally {
+      setIsProcessing(false);
+    }
+  };
+
+  // 重置表单
+  const resetForm = () => {
+    setAllDataFile('');
+    setRemoveDataFile('');
+    setOutputFile('');
+    setProgress(null);
+    setResult(null);
+  };
+
+  return (
+    <div className="space-y-6">
+      {/* 页面标题 */}
+      <div className="flex items-center gap-3">
+        <div className="w-10 h-10 bg-gradient-to-br from-purple-500 to-purple-600 rounded-lg flex items-center justify-center shadow-sm">
+          <Wrench className="w-5 h-5 text-white" />
+        </div>
+        <div>
+          <h1 className="text-2xl font-bold text-gray-900">便捷小工具</h1>
+          <p className="text-gray-600">AI检索图片/数据清洗工具集</p>
+        </div>
+      </div>
+
+      {/* AI检索图片/数据清洗工具 */}
+      <div className="bg-white rounded-xl shadow-sm border border-gray-200 overflow-hidden">
+        <div className="p-6 border-b border-gray-200">
+          <div className="flex items-center gap-3">
+            <FileText className="w-6 h-6 text-purple-600" />
+            <div>
+              <h2 className="text-lg font-semibold text-gray-900">AI检索图片/数据清洗</h2>
+              <p className="text-sm text-gray-600">JSONL格式数据去重处理</p>
+            </div>
+          </div>
+        </div>
+
+        <div className="p-6 space-y-6">
+          {/* 使用说明 */}
+          <div className="bg-blue-50 border border-blue-200 rounded-lg p-4">
+            <div className="flex items-start gap-3">
+              <Info className="w-5 h-5 text-blue-600 mt-0.5 flex-shrink-0" />
+              <div className="text-sm text-blue-800">
+                <p className="font-medium mb-2">使用说明：</p>
+                <ul className="space-y-1 list-disc list-inside">
+                  <li>选择包含所有数据的JSONL文件（全部数据）</li>
+                  <li>选择包含要去除数据的JSONL文件（要去除的数据）</li>
+                  <li>系统将根据URI字段进行匹配，去除重复项</li>
+                  <li>处理结果将保存到指定的输出文件</li>
+                </ul>
+              </div>
+            </div>
+          </div>
+
+          {/* 文件选择区域 */}
+          <div className="grid grid-cols-1 md:grid-cols-2 gap-6">
+            {/* 全部数据文件 */}
+            <div className="space-y-2">
+              <label className="block text-sm font-medium text-gray-700">
+                全部数据文件 <span className="text-red-500">*</span>
+              </label>
+              <div className="flex gap-2">
+                <button
+                  onClick={selectAllDataFile}
+                  disabled={isProcessing}
+                  className="flex items-center gap-2 px-4 py-2 bg-gray-100 hover:bg-gray-200 disabled:bg-gray-50 disabled:text-gray-400 text-gray-700 rounded-lg border border-gray-300 transition-colors"
+                >
+                  <Upload className="w-4 h-4" />
+                  选择文件
+                </button>
+              </div>
+              {allDataFile && (
+                <p className="text-xs text-gray-600 break-all bg-gray-50 p-2 rounded">
+                  {allDataFile}
+                </p>
+              )}
+            </div>
+
+            {/* 要去除的数据文件 */}
+            <div className="space-y-2">
+              <label className="block text-sm font-medium text-gray-700">
+                要去除的数据文件 <span className="text-red-500">*</span>
+              </label>
+              <div className="flex gap-2">
+                <button
+                  onClick={selectRemoveDataFile}
+                  disabled={isProcessing}
+                  className="flex items-center gap-2 px-4 py-2 bg-gray-100 hover:bg-gray-200 disabled:bg-gray-50 disabled:text-gray-400 text-gray-700 rounded-lg border border-gray-300 transition-colors"
+                >
+                  <Upload className="w-4 h-4" />
+                  选择文件
+                </button>
+              </div>
+              {removeDataFile && (
+                <p className="text-xs text-gray-600 break-all bg-gray-50 p-2 rounded">
+                  {removeDataFile}
+                </p>
+              )}
+            </div>
+          </div>
+
+          {/* 输出文件选择 */}
+          <div className="space-y-2">
+            <label className="block text-sm font-medium text-gray-700">
+              输出文件 <span className="text-red-500">*</span>
+            </label>
+            <div className="flex gap-2">
+              <button
+                onClick={selectOutputFile}
+                disabled={isProcessing}
+                className="flex items-center gap-2 px-4 py-2 bg-gray-100 hover:bg-gray-200 disabled:bg-gray-50 disabled:text-gray-400 text-gray-700 rounded-lg border border-gray-300 transition-colors"
+              >
+                <Download className="w-4 h-4" />
+                选择保存位置
+              </button>
+            </div>
+            {outputFile && (
+              <p className="text-xs text-gray-600 break-all bg-gray-50 p-2 rounded">
+                {outputFile}
+              </p>
+            )}
+          </div>
+
+          {/* 进度显示 */}
+          {progress && (
+            <div className="bg-gray-50 rounded-lg p-4">
+              <div className="flex items-center gap-3 mb-3">
+                <Loader2 className="w-5 h-5 text-purple-600 animate-spin" />
+                <span className="text-sm font-medium text-gray-900">处理中...</span>
+              </div>
+              <div className="space-y-2">
+                <div className="flex justify-between text-sm text-gray-600">
+                  <span>{progress.status}</span>
+                  <span>{progress.percentage.toFixed(1)}%</span>
+                </div>
+                <div className="w-full bg-gray-200 rounded-full h-2">
+                  <div 
+                    className="bg-purple-600 h-2 rounded-full transition-all duration-300"
+                    style={{ width: `${progress.percentage}%` }}
+                  />
+                </div>
+                {progress.total > 0 && (
+                  <div className="text-xs text-gray-500">
+                    {progress.current.toLocaleString()} / {progress.total.toLocaleString()}
+                  </div>
+                )}
+              </div>
+            </div>
+          )}
+
+          {/* 结果显示 */}
+          {result && (
+            <div className={`rounded-lg p-4 ${result.success ? 'bg-green-50 border border-green-200' : 'bg-red-50 border border-red-200'}`}>
+              <div className="flex items-start gap-3">
+                {result.success ? (
+                  <CheckCircle className="w-5 h-5 text-green-600 mt-0.5 flex-shrink-0" />
+                ) : (
+                  <AlertCircle className="w-5 h-5 text-red-600 mt-0.5 flex-shrink-0" />
+                )}
+                <div className="flex-1">
+                  <p className={`font-medium ${result.success ? 'text-green-800' : 'text-red-800'}`}>
+                    {result.message}
+                  </p>
+                  {result.success && (
+                    <div className="mt-2 text-sm text-green-700 space-y-1">
+                      <p>原始数据条数: {result.original_count.toLocaleString()}</p>
+                      <p>去除重复条数: {result.removed_count.toLocaleString()}</p>
+                      <p>最终结果条数: {result.final_count.toLocaleString()}</p>
+                      <p className="break-all">输出文件: {result.output_file}</p>
+                    </div>
+                  )}
+                </div>
+              </div>
+            </div>
+          )}
+
+          {/* 操作按钮 */}
+          <div className="flex gap-3 pt-4 border-t border-gray-200">
+            <button
+              onClick={startDataCleaning}
+              disabled={!allDataFile || !removeDataFile || !outputFile || isProcessing}
+              className="flex items-center gap-2 px-6 py-2 bg-purple-600 hover:bg-purple-700 disabled:bg-gray-300 disabled:cursor-not-allowed text-white rounded-lg font-medium transition-colors"
+            >
+              {isProcessing ? (
+                <Loader2 className="w-4 h-4 animate-spin" />
+              ) : (
+                <FileText className="w-4 h-4" />
+              )}
+              {isProcessing ? '处理中...' : '开始处理'}
+            </button>
+
+            <button
+              onClick={resetForm}
+              disabled={isProcessing}
+              className="flex items-center gap-2 px-6 py-2 bg-gray-100 hover:bg-gray-200 disabled:bg-gray-50 disabled:text-gray-400 text-gray-700 rounded-lg font-medium transition-colors"
+            >
+              <Trash2 className="w-4 h-4" />
+              重置
+            </button>
+          </div>
+        </div>
+      </div>
+    </div>
+  );
+};
+
+export default Tools;

apps/desktop/test_data/sample_all_data.jsonl

@@ -0,0 +1,5 @@
+{"kind":"storage#object","id":"fashion_image_block/gallery_v2/models/test_image_1.jpg/1752665521922040","selfLink":"https://www.googleapis.com/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_1.jpg","mediaLink":"https://storage.googleapis.com/download/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_1.jpg?generation=1752665521922040&alt=media","name":"gallery_v2/models/test_image_1.jpg","bucket":"fashion_image_block","generation":"1752665521922040","metageneration":"1","contentType":"image/jpeg","storageClass":"STANDARD","size":310505,"md5Hash":"WmzNIJHZxALLVSfn0esl5Q==","crc32c":"6skf3A==","etag":"CPi/4O6jwY4DEAE=","timeCreated":"2025-07-16T11:32:01.981000Z","updated":"2025-07-16T11:32:01.981000Z","uri":"gs://fashion_image_block/gallery_v2/models/test_image_1.jpg"}
+{"kind":"storage#object","id":"fashion_image_block/gallery_v2/models/test_image_2.jpg/1752665524250767","selfLink":"https://www.googleapis.com/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_2.jpg","mediaLink":"https://storage.googleapis.com/download/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_2.jpg?generation=1752665524250767&alt=media","name":"gallery_v2/models/test_image_2.jpg","bucket":"fashion_image_block","generation":"1752665524250767","metageneration":"1","contentType":"image/jpeg","storageClass":"STANDARD","size":488328,"md5Hash":"ksnQY8Wv6g5Q+LkmaxBV+A==","crc32c":"u4zm9A==","etag":"CI/R7u+jwY4DEAE=","timeCreated":"2025-07-16T11:32:04.317000Z","updated":"2025-07-16T11:32:04.317000Z","uri":"gs://fashion_image_block/gallery_v2/models/test_image_2.jpg"}
+{"kind":"storage#object","id":"fashion_image_block/gallery_v2/models/test_image_3.jpg/1752665522362206","selfLink":"https://www.googleapis.com/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_3.jpg","mediaLink":"https://storage.googleapis.com/download/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_3.jpg?generation=1752665522362206&alt=media","name":"gallery_v2/models/test_image_3.jpg","bucket":"fashion_image_block","generation":"1752665522362206","metageneration":"1","contentType":"image/jpeg","storageClass":"STANDARD","size":719213,"md5Hash":"fPmRuc/nFZnN6g5u9pJ99g==","crc32c":"r4eLmg==","etag":"CN6u++6jwY4DEAE=","timeCreated":"2025-07-16T11:32:02.420000Z","updated":"2025-07-16T11:32:02.420000Z","uri":"gs://fashion_image_block/gallery_v2/models/test_image_3.jpg"}
+{"kind":"storage#object","id":"fashion_image_block/gallery_v2/models/test_image_4.jpg/1752665523165046","selfLink":"https://www.googleapis.com/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_4.jpg","mediaLink":"https://storage.googleapis.com/download/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_4.jpg?generation=1752665523165046&alt=media","name":"gallery_v2/models/test_image_4.jpg","bucket":"fashion_image_block","generation":"1752665523165046","metageneration":"1","contentType":"image/jpeg","storageClass":"STANDARD","size":463800,"md5Hash":"G0EjX9ZdkMEXoDR3h+zNJQ==","crc32c":"+wH9Bg==","etag":"CPaurO+jwY4DEAE=","timeCreated":"2025-07-16T11:32:03.230000Z","updated":"2025-07-16T11:32:03.230000Z","uri":"gs://fashion_image_block/gallery_v2/models/test_image_4.jpg"}
+{"kind":"storage#object","id":"fashion_image_block/gallery_v2/models/test_image_5.jpg/1752665521595009","selfLink":"https://www.googleapis.com/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_5.jpg","mediaLink":"https://storage.googleapis.com/download/storage/v1/b/fashion_image_block/o/gallery_v2%2Fmodels%2Ftest_image_5.jpg?generation=1752665521595009&alt=media","name":"gallery_v2/models/test_image_5.jpg","bucket":"fashion_image_block","generation":"1752665521595009","metageneration":"1","contentType":"image/jpeg","storageClass":"STANDARD","size":435049,"md5Hash":"vw3s4DXgN06HPdSnizhPrQ==","crc32c":"kHGLOA==","etag":"CIHFzO6jwY4DEAE=","timeCreated":"2025-07-16T11:32:01.663000Z","updated":"2025-07-16T11:32:01.663000Z","uri":"gs://fashion_image_block/gallery_v2/models/test_image_5.jpg"}

apps/desktop/test_data/sample_remove_data.jsonl

@@ -0,0 +1,2 @@
+{"id":"0fc38c97-e8a7-4fc2-8228-7598d1b993e9","schema_id":"default_schema","json_data":"{\"dress_color_pattern\":{\"Hue\":0.08,\"Saturation\":0.05,\"Value\":0.95,\"rgb_hex\":\"#f2ebe6\"},\"style_description\":\"测试样例1\",\"environment_tags\":[\"Indoor\",\"Test\"],\"products\":[{\"category\":\"测试\",\"color_pattern\":{\"Hue\":0.08,\"Saturation\":0.05,\"Value\":0.95,\"rgb_hex\":\"#f2ebe6\"},\"design_styles\":[\"Test\"],\"design_styles_embedding\":[]}],\"uri\":\"gs://fashion_image_block/gallery_v2/models/test_image_2.jpg\",\"description\":\"\",\"runtime\":\"\",\"releaseDate\":\"2025-07-16T11:31:48.092583Z\",\"title\":\"model\",\"categories\":[\"测试\"]}"}
+{"id":"e32fcc8d-166c-4da5-991d-422f7c5a47e7","schema_id":"default_schema","json_data":"{\"dress_color_pattern\":{\"Hue\":0.08,\"Saturation\":0.05,\"Value\":0.95,\"rgb_hex\":\"#f2ebe6\"},\"style_description\":\"测试样例2\",\"environment_tags\":[\"Indoor\",\"Test\"],\"products\":[{\"category\":\"测试\",\"color_pattern\":{\"Hue\":0.08,\"Saturation\":0.05,\"Value\":0.95,\"rgb_hex\":\"#f2ebe6\"},\"design_styles\":[\"Test\"],\"design_styles_embedding\":[]}],\"uri\":\"gs://fashion_image_block/gallery_v2/models/test_image_4.jpg\",\"description\":\"\",\"runtime\":\"\",\"releaseDate\":\"2025-07-16T11:31:48.092583Z\",\"title\":\"model\",\"categories\":[\"测试\"]}"}