zhangyice
/
ax-frontend-app


			
				
					
						
						
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335
							<!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>文档重复检查工具</title>
    <style>
        body {
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
            max-width: 1200px;
            margin: 0 auto;
            padding: 20px;
            background: #f5f5f5;
        }
        .container {
            background: white;
            padding: 30px;
            border-radius: 8px;
            box-shadow: 0 2px 8px rgba(0,0,0,0.1);
        }
        h1 {
            color: #333;
            border-bottom: 3px solid #1890ff;
            padding-bottom: 10px;
        }
        .section {
            margin: 20px 0;
            padding: 15px;
            background: #fafafa;
            border-left: 4px solid #1890ff;
            border-radius: 4px;
        }
        .input-group {
            margin: 15px 0;
        }
        label {
            display: block;
            margin-bottom: 5px;
            font-weight: 600;
            color: #555;
        }
        input, select {
            width: 100%;
            padding: 10px;
            border: 1px solid #d9d9d9;
            border-radius: 4px;
            font-size: 14px;
            box-sizing: border-box;
        }
        button {
            background: #1890ff;
            color: white;
            border: none;
            padding: 12px 24px;
            border-radius: 4px;
            cursor: pointer;
            font-size: 14px;
            font-weight: 600;
            margin-right: 10px;
            margin-top: 10px;
        }
        button:hover {
            background: #40a9ff;
        }
        button:disabled {
            background: #d9d9d9;
            cursor: not-allowed;
        }
        .results {
            margin-top: 20px;
        }
        .result-box {
            background: white;
            border: 1px solid #e8e8e8;
            padding: 15px;
            margin: 10px 0;
            border-radius: 4px;
            white-space: pre-wrap;
            font-family: 'Courier New', monospace;
            font-size: 12px;
            max-height: 400px;
            overflow-y: auto;
        }
        .success {
            color: #52c41a;
            font-weight: 600;
        }
        .error {
            color: #ff4d4f;
            font-weight: 600;
        }
        .warning {
            color: #faad14;
            font-weight: 600;
        }
        .info {
            color: #1890ff;
            font-weight: 600;
        }
        .stat {
            display: inline-block;
            background: #e6f7ff;
            padding: 5px 15px;
            margin: 5px 5px 5px 0;
            border-radius: 20px;
            font-size: 13px;
            font-weight: 600;
            color: #0050b3;
        }
    </style>
</head>
<body>
    <div class="container">
        <h1>🔍 文档重复检查工具</h1>
        
        <div class="section">
            <h3>步骤 1: 从数据库获取文档</h3>
            <div class="input-group">
                <label>后端 API 地址:</label>
                <input type="text" id="apiBase" value="http://192.168.0.195:8000" />
            </div>
            <div class="input-group">
                <label>文档 ID:</label>
                <input type="text" id="documentId" placeholder="例如: doc-7791bc371ad7" />
            </div>
            <button onclick="fetchDocument()">获取文档</button>
            <button onclick="fetchLatestDocument()">获取最新文档</button>
        </div>

        <div class="section">
            <h3>步骤 2: 检查导出的Word文件</h3>
            <div class="input-group">
                <label>导出记录 ID:</label>
                <input type="text" id="recordId" placeholder="例如: rec-xxx" />
            </div>
            <div class="input-group">
                <label>用户 ID:</label>
                <input type="text" id="userId" value="default-user" />
            </div>
            <button onclick="checkExportedWord()">下载并检查Word文件</button>
        </div>

        <div id="results" class="results"></div>
    </div>

    <script>
        const API_BASE = () => document.getElementById('apiBase').value;

        function showResult(message, type = 'info') {
            const results = document.getElementById('results');
            const box = document.createElement('div');
            box.className = 'result-box';
            box.innerHTML = `<span class="${type}">[${type.toUpperCase()}]</span> ${message}`;
            results.appendChild(box);
            results.scrollTop = results.scrollHeight;
        }

        async function fetchLatestDocument() {
            showResult('正在获取最新文档...', 'info');
            
            try {
                const response = await fetch(`${API_BASE()}/api/v1/documents?userId=default-user&page=1&pageSize=1&sortBy=createdAt&sortOrder=desc`);
                
                if (!response.ok) {
                    throw new Error(`HTTP ${response.status}: ${response.statusText}`);
                }

                const data = await response.json();
                
                if (data.code === 200 && data.data.documents.length > 0) {
                    const doc = data.data.documents[0];
                    document.getElementById('documentId').value = doc.id;
                    showResult(`最新文档 ID: ${doc.id}`, 'success');
                    
                    // 自动获取文档详情
                    await fetchDocument();
                } else {
                    showResult('未找到文档', 'warning');
                }
            } catch (error) {
                showResult(`错误: ${error.message}`, 'error');
            }
        }

        async function fetchDocument() {
            const docId = document.getElementById('documentId').value;
            
            if (!docId) {
                showResult('请输入文档ID', 'warning');
                return;
            }

            showResult(`正在获取文档 ${docId}...`, 'info');
            
            try {
                const response = await fetch(`${API_BASE()}/api/v1/documents/${docId}`);
                
                if (!response.ok) {
                    throw new Error(`HTTP ${response.status}: ${response.statusText}`);
                }

                const data = await response.json();
                
                if (data.code === 200) {
                    const doc = data.data;
                    const content = doc.content;
                    
                    showResult(`文档ID: ${doc.id}`, 'info');
                    showResult(`格式: ${doc.format}`, 'info');
                    showResult(`创建时间: ${new Date(doc.createdAt).toLocaleString()}`, 'info');
                    
                    // 统计信息
                    const totalLength = content.length;
                    const lines = content.split('\n');
                    const totalLines = lines.length;
                    
                    showResult(`
<span class="stat">总长度: ${totalLength} 字符</span>
<span class="stat">总行数: ${totalLines} 行</span>
                    `, 'info');
                    
                    // 检查是否有重复
                    const halfLength = Math.floor(totalLength / 2);
                    const firstHalf = content.substring(0, halfLength);
                    const secondHalf = content.substring(halfLength);
                    
                    // 检查前500字符是否在后半部分重复出现
                    const sample = content.substring(0, Math.min(500, halfLength));
                    const isDuplicated = secondHalf.includes(sample);
                    
                    if (isDuplicated) {
                        showResult('⚠️ 警告: 检测到内容可能重复！前500字符在后半部分也出现了', 'warning');
                    } else {
                        showResult('✓ 数据库内容没有重复', 'success');
                    }
                    
                    // 显示前1000字符
                    showResult('数据库中的前1000字符:\n' + content.substring(0, 1000), 'info');
                    
                    // 显示后500字符
                    showResult('数据库中的后500字符:\n' + content.substring(totalLength - 500), 'info');
                    
                } else {
                    showResult(`API错误: ${data.message}`, 'error');
                }
            } catch (error) {
                showResult(`错误: ${error.message}`, 'error');
            }
        }

        async function checkExportedWord() {
            const recordId = document.getElementById('recordId').value;
            const userId = document.getElementById('userId').value;
            
            if (!recordId || !userId) {
                showResult('请输入导出记录ID和用户ID', 'warning');
                return;
            }

            showResult(`正在下载并检查Word文件...`, 'info');
            
            try {
                const downloadUrl = `${API_BASE()}/api/v1/export/records/${recordId}/download?userId=${encodeURIComponent(userId)}`;
                showResult(`下载URL: ${downloadUrl}`, 'info');
                
                const response = await fetch(downloadUrl);
                
                if (!response.ok) {
                    throw new Error(`HTTP ${response.status}: ${response.statusText}`);
                }

                const blob = await response.blob();
                const arrayBuffer = await blob.arrayBuffer();
                
                showResult(`Word文件大小: ${arrayBuffer.byteLength} bytes`, 'info');
                
                // 使用mammoth解析Word文件
                showResult('正在解析Word文档...', 'info');
                
                // 动态加载mammoth库
                if (typeof mammoth === 'undefined') {
                    const script = document.createElement('script');
                    script.src = 'https://cdn.jsdelivr.net/npm/mammoth@1.6.0/mammoth.browser.min.js';
                    document.head.appendChild(script);
                    
                    await new Promise((resolve, reject) => {
                        script.onload = resolve;
                        script.onerror = reject;
                    });
                }
                
                const result = await mammoth.convertToHtml({ arrayBuffer });
                const html = result.value;
                
                showResult(`HTML长度: ${html.length} 字符`, 'info');
                
                // 提取纯文本
                const parser = new DOMParser();
                const doc = parser.parseFromString(html, 'text/html');
                const text = doc.body.textContent || '';
                
                showResult(`纯文本长度: ${text.length} 字符`, 'info');
                
                // 检查是否有重复
                const lines = text.split('\n').filter(line => line.trim());
                const halfPoint = Math.floor(lines.length / 2);
                
                let matchCount = 0;
                for (let i = 0; i < Math.min(10, halfPoint); i++) {
                    if (lines[i] === lines[halfPoint + i]) {
                        matchCount++;
                    }
                }
                
                if (matchCount >= 8) {
                    showResult(`⚠️ 警告: Word文件中检测到重复内容！前10行有${matchCount}行与后半部分匹配`, 'warning');
                } else {
                    showResult('✓ Word文件内容没有明显重复', 'success');
                }
                
                // 显示前1000字符
                showResult('Word文件的前1000字符:\n' + text.substring(0, 1000), 'info');
                
            } catch (error) {
                showResult(`错误: ${error.message}`, 'error');
                console.error(error);
            }
        }

        // 初始化
        showResult('工具已就绪。请使用上面的按钮检查文档。', 'success');
    </script>
</body>
</html>