zhengchun
/
MinerU


			
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187
							"""
HTML 生成器模块

提供 HTML 输出功能：
- 表格 HTML 生成（带样式）
- 单元格坐标展示
"""
import json
from pathlib import Path
from typing import Dict, Any, List
from loguru import logger


class HTMLGenerator:
    """HTML 生成器类"""
    
    @staticmethod
    def save_table_htmls(
        results: Dict[str, Any],
        output_dir: Path,
        doc_name: str
    ) -> Path:
        """
        保存表格 HTML 文件
        
        Args:
            results: 处理结果
            output_dir: 输出目录
            doc_name: 文档名称
            
        Returns:
            表格目录路径
        """
        tables_dir = output_dir / 'tables'
        tables_dir.mkdir(exist_ok=True)
        
        table_count = 0
        
        for page in results.get('pages', []):
            page_idx = page.get('page_idx', 0)
            
            for element in page.get('elements', []):
                if element.get('type') in ['table', 'table_body']:
                    table_count += 1
                    content = element.get('content', {})
                    html = content.get('html', '')
                    cells = content.get('cells', [])
                    
                    if html:
                        full_html = HTMLGenerator._generate_table_html_with_styles(
                            html, cells, doc_name, page_idx, table_count
                        )
                        
                        html_path = tables_dir / f"{doc_name}_table_{table_count}_page_{page_idx + 1}.html"
                        with open(html_path, 'w', encoding='utf-8') as f:
                            f.write(full_html)
        
        if table_count > 0:
            logger.info(f"📊 {table_count} tables saved to: {tables_dir}")
        
        return tables_dir
    
    @staticmethod
    def _generate_table_html_with_styles(
        table_html: str,
        cells: List[Dict],
        doc_name: str,
        page_idx: int,
        table_idx: int
    ) -> str:
        """
        生成带样式的完整 HTML
        
        Args:
            table_html: 表格 HTML 内容
            cells: 单元格列表
            doc_name: 文档名称
            page_idx: 页码
            table_idx: 表格序号
            
        Returns:
            完整的 HTML 字符串
        """
        cells_json = json.dumps(cells, ensure_ascii=False, indent=2) if cells else "[]"
        
        return f"""<!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>{doc_name} - Table {table_idx}</title>
    <style>
        body {{
            font-family: Arial, "Microsoft YaHei", sans-serif;
            margin: 20px;
            background-color: #f5f5f5;
        }}
        .container {{
            max-width: 1400px;
            margin: 0 auto;
            background-color: white;
            padding: 20px;
            box-shadow: 0 0 10px rgba(0,0,0,0.1);
            border-radius: 8px;
        }}
        .meta {{
            color: #666;
            font-size: 0.9em;
            margin-bottom: 20px;
            padding-bottom: 10px;
            border-bottom: 1px solid #ddd;
        }}
        table {{
            border-collapse: collapse;
            width: 100%;
            margin: 20px 0;
        }}
        th, td {{
            border: 1px solid #ddd;
            padding: 8px 12px;
            text-align: left;
        }}
        th {{
            background-color: #f2f2f2;
            font-weight: bold;
        }}
        tr:hover {{
            background-color: #f9f9f9;
        }}
        td[data-bbox], th[data-bbox] {{
            position: relative;
        }}
        td[data-bbox]:hover::after, th[data-bbox]:hover::after {{
            content: attr(data-bbox);
            position: absolute;
            bottom: 100%;
            left: 0;
            background: #333;
            color: white;
            padding: 2px 6px;
            font-size: 10px;
            border-radius: 3px;
            white-space: nowrap;
            z-index: 100;
        }}
        .cells-info {{
            margin-top: 30px;
            padding: 15px;
            background-color: #f8f9fa;
            border-radius: 5px;
        }}
        .cells-info summary {{
            cursor: pointer;
            font-weight: bold;
            color: #333;
        }}
        .cells-info pre {{
            background-color: #2d2d2d;
            color: #f8f8f2;
            padding: 15px;
            border-radius: 5px;
            overflow-x: auto;
            font-size: 12px;
        }}
    </style>
</head>
<body>
    <div class="container">
        <div class="meta">
            <p><strong>Document:</strong> {doc_name}</p>
            <p><strong>Page:</strong> {page_idx + 1}</p>
            <p><strong>Table:</strong> {table_idx}</p>
            <p><strong>Cells with coordinates:</strong> {len(cells)}</p>
        </div>
        
        {table_html}
        
        <div class="cells-info">
            <details>
                <summary>📍 单元格坐标数据 (JSON)</summary>
                <pre>{cells_json}</pre>
            </details>
        </div>
    </div>
</body>
</html>"""