zhang_crawl_bio/tests/mytest/docling_t.py

import os
from docling.document_converter import DocumentConverter
from docx import Document
from bs4 import BeautifulSoup

def html_to_docx(html_content, output_docx_path):
    # 创建一个新的 DOCX 文档
    doc = Document()
    
    # 使用 BeautifulSoup 解析 HTML 内容
    soup = BeautifulSoup(html_content, 'html.parser')
    
    # 将 HTML 内容添加到 DOCX 文档中
    for element in soup.find_all(['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'ul', 'ol', 'li']):
        if element.name == 'p':
            doc.add_paragraph(element.get_text())
        elif element.name.startswith('h'):
            level = int(element.name[1])
            doc.add_heading(element.get_text(), level=level)
        elif element.name in ['ul', 'ol']:
            for li in element.find_all('li'):
                doc.add_paragraph(li.get_text(), style='ListBullet' if element.name == 'ul' else 'ListNumber')
    
    # 保存 DOCX 文档
    doc.save(output_docx_path)

def main():
    # 本地 HTML 文件路径
    file_path = r"K:\code\upwork\zhang_crawl_bio\output\Acalypha malabarica essential oil\all_paper\0.html"
    # source = "https://arxiv.org/pdf/2408.09869"  # document per local path or URL
    source = file_path
    converter = DocumentConverter()
    result = converter.convert(source)
    
    # 导出为 Markdown
    markdown_content = result.document.export_to_markdown()
    print(markdown_content)
    

if __name__ == "__main__":
    main()